Merge branch 'drm-next' of git://people.freedesktop.org/~airlied/linux

Pull drm updates from Dave Airlie:
 "This is the main drm merge window pull request, changes all over the
  place, mostly normal levels of churn.

  Highlights:

  Core drm:
     More cleanups, fix race on connector/encoder naming, docs updates,
     object locking rework in prep for atomic modeset

  i915:
     mipi DSI support, valleyview power fixes, cursor size fixes,
     execlist refactoring, vblank improvements, userptr support, OOM
     handling improvements

  radeon:
     GPUVM tuning and large page size support, gart fixes, deep color
     HDMI support, HDMI audio cleanups

  nouveau:
     - displayport rework should fix lots of issues
     - initial gk20a support
     - gk110b support
     - gk208 fixes

  exynos:
     probe order fixes, HDMI changes, IPP consolidation

  msm:
     debugfs updates, misc fixes

  ast:
     ast2400 support, sync with UMS driver

  tegra:
     cleanups, hdmi + hw cursor for Tegra 124.

  panel:
     fixes existing panels add some new ones.

  ipuv3:
     moved from staging to drivers/gpu"

* 'drm-next' of git://people.freedesktop.org/~airlied/linux: (761 commits)
  drm/nouveau/disp/dp: fix tmds passthrough on dp connector
  drm/nouveau/dp: probe dpcd to determine connectedness
  drm/nv50-: trigger update after all connectors disabled
  drm/nv50-: prepare for attaching a SOR to multiple heads
  drm/gf119-/disp: fix debug output on update failure
  drm/nouveau/disp/dp: make use of postcursor when its available
  drm/g94-/disp/dp: take max pullup value across all lanes
  drm/nouveau/bios/dp: parse lane postcursor data
  drm/nouveau/dp: fix support for dpms
  drm/nouveau: register a drm_dp_aux channel for each dp connector
  drm/g94-/disp: add method to power-off dp lanes
  drm/nouveau/disp/dp: maintain link in response to hpd signal
  drm/g94-/disp: bash and wait for something after changing lane power regs
  drm/nouveau/disp/dp: split link config/power into two steps
  drm/nv50/disp: train PIOR-attached DP from second supervisor
  drm/nouveau/disp/dp: make use of existing output data for link training
  drm/gf119/disp: start removing direct vbios parsing from supervisor
  drm/nv50/disp: start removing direct vbios parsing from supervisor
  drm/nouveau/disp/dp: maintain receiver caps in response to hpd signal
  drm/nouveau/disp/dp: create subclass for dp outputs
  ...

447 files changed, 24787 insertions, 9542 deletions

diff --git a/Documentation/DocBook/drm.tmpl b/Documentation/DocBook/drm.tmpl
index ba60d93c1855..7df3134ebc0e 100644
--- a/Documentation/DocBook/drm.tmpl
+++ b/Documentation/DocBook/drm.tmpl
@@ -142,6 +142,12 @@
       to register it with the DRM subsystem.
     </para>
     <para>
+      Newer drivers that no longer require a <structname>drm_bus</structname>
+      structure can alternatively use the low-level device initialization and
+      registration functions such as <function>drm_dev_alloc()</function> and
+      <function>drm_dev_register()</function> directly.
+    </para>
+    <para>
       The <structname>drm_driver</structname> structure contains static
       information that describes the driver and features it supports, and
       pointers to methods that the DRM core will call to implement the DRM API.
@@ -282,6 +288,36 @@ char *date;</synopsis>
       </sect3>
     </sect2>
     <sect2>
+      <title>Device Registration</title>
+      <para>
+        A number of functions are provided to help with device registration.
+        The functions deal with PCI, USB and platform devices, respectively.
+      </para>
+!Edrivers/gpu/drm/drm_pci.c
+!Edrivers/gpu/drm/drm_usb.c
+!Edrivers/gpu/drm/drm_platform.c
+      <para>
+        New drivers that no longer rely on the services provided by the
+        <structname>drm_bus</structname> structure can call the low-level
+        device registration functions directly. The
+        <function>drm_dev_alloc()</function> function can be used to allocate
+        and initialize a new <structname>drm_device</structname> structure.
+        Drivers will typically want to perform some additional setup on this
+        structure, such as allocating driver-specific data and storing a
+        pointer to it in the DRM device's <structfield>dev_private</structfield>
+        field. Drivers should also set the device's unique name using the
+        <function>drm_dev_set_unique()</function> function. After it has been
+        set up a device can be registered with the DRM subsystem by calling
+        <function>drm_dev_register()</function>. This will cause the device to
+        be exposed to userspace and will call the driver's
+        <structfield>.load()</structfield> implementation. When a device is
+        removed, the DRM device can safely be unregistered and freed by calling
+        <function>drm_dev_unregister()</function> followed by a call to
+        <function>drm_dev_unref()</function>.
+      </para>
+!Edrivers/gpu/drm/drm_stub.c
+    </sect2>
+    <sect2>
       <title>Driver Load</title>
       <para>
         The <methodname>load</methodname> method is the driver and device
@@ -342,21 +378,13 @@ char *date;</synopsis>
         <sect4>
           <title>Managed IRQ Registration</title>
           <para>
-            Both the <function>drm_irq_install</function> and
-	    <function>drm_irq_uninstall</function> functions get the device IRQ by
-	    calling <function>drm_dev_to_irq</function>. This inline function will
-	    call a bus-specific operation to retrieve the IRQ number. For platform
-	    devices, <function>platform_get_irq</function>(..., 0) is used to
-	    retrieve the IRQ number.
-          </para>
-          <para>
             <function>drm_irq_install</function> starts by calling the
             <methodname>irq_preinstall</methodname> driver operation. The operation
             is optional and must make sure that the interrupt will not get fired by
             clearing all pending interrupt flags or disabling the interrupt.
           </para>
           <para>
-            The IRQ will then be requested by a call to
+            The passed-in IRQ will then be requested by a call to
             <function>request_irq</function>. If the DRIVER_IRQ_SHARED driver
             feature flag is set, a shared (IRQF_SHARED) IRQ handler will be
             requested.
@@ -1799,6 +1827,12 @@ void intel_crt_init(struct drm_device *dev)
       <title>KMS API Functions</title>
 !Edrivers/gpu/drm/drm_crtc.c
     </sect2>
+    <sect2>
+      <title>KMS Locking</title>
+!Pdrivers/gpu/drm/drm_modeset_lock.c kms locking
+!Iinclude/drm/drm_modeset_lock.h
+!Edrivers/gpu/drm/drm_modeset_lock.c
+    </sect2>
   </sect1>
 
   <!-- Internals: kms helper functions -->
@@ -1903,8 +1937,8 @@ void intel_crt_init(struct drm_device *dev)
           <para>
             The function filters out modes larger than
             <parameter>max_width</parameter> and <parameter>max_height</parameter>
-            if specified. It then calls the connector
-            <methodname>mode_valid</methodname> helper operation for  each mode in
+            if specified. It then calls the optional connector
+            <methodname>mode_valid</methodname> helper operation for each mode in
             the probed list to check whether the mode is valid for the connector.
           </para>
         </listitem>
@@ -2265,7 +2299,7 @@ void intel_crt_init(struct drm_device *dev)
           <para>
             Verify whether a mode is valid for the connector. Return MODE_OK for
             supported modes and one of the enum drm_mode_status values (MODE_*)
-            for unsupported modes. This operation is mandatory.
+            for unsupported modes. This operation is optional.
           </para>
           <para>
             As the mode rejection reason is currently not used beside for
@@ -2450,6 +2484,863 @@ void intel_crt_init(struct drm_device *dev)
       pointer to the target object, a pointer to the previously created property
       and an initial instance value.
     </para>
+    <sect2>
+	<title>Existing KMS Properties</title>
+	<para>
+	The following table gives description of drm properties exposed by various
+	modules/drivers.
+	</para>
+	<table border="1" cellpadding="0" cellspacing="0">
+	<tbody>
+	<tr style="font-weight: bold;">
+	<td valign="top" >Owner Module/Drivers</td>
+	<td valign="top" >Group</td>
+	<td valign="top" >Property Name</td>
+	<td valign="top" >Type</td>
+	<td valign="top" >Property Values</td>
+	<td valign="top" >Object attached</td>
+	<td valign="top" >Description/Restrictions</td>
+	</tr>
+	<tr>
+	<td rowspan="20" valign="top" >DRM</td>
+	<td rowspan="2" valign="top" >Generic</td>
+	<td valign="top" >“EDID”</td>
+	<td valign="top" >BLOB | IMMUTABLE</td>
+	<td valign="top" >0</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >Contains id of edid blob ptr object.</td>
+	</tr>
+	<tr>
+	<td valign="top" >“DPMS”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ “On”, “Standby”, “Suspend”, “Off” }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >Contains DPMS operation mode value.</td>
+	</tr>
+	<tr>
+	<td rowspan="1" valign="top" >Plane</td>
+	<td valign="top" >“type”</td>
+	<td valign="top" >ENUM | IMMUTABLE</td>
+	<td valign="top" >{ "Overlay", "Primary", "Cursor" }</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >Plane type</td>
+	</tr>
+	<tr>
+	<td rowspan="2" valign="top" >DVI-I</td>
+	<td valign="top" >“subconnector”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ “Unknown”, “DVI-D”, “DVI-A” }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“select subconnector”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ “Automatic”, “DVI-D”, “DVI-A” }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="13" valign="top" >TV</td>
+	<td valign="top" >“subconnector”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "Unknown", "Composite", "SVIDEO", "Component", "SCART" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“select subconnector”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "Automatic", "Composite", "SVIDEO", "Component", "SCART" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“mode”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "NTSC_M", "NTSC_J", "NTSC_443", "PAL_B" } etc.</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“left margin”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=100</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“right margin”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=100</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“top margin”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=100</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“bottom margin”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=100</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“brightness”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=100</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“contrast”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=100</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“flicker reduction”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=100</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“overscan”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=100</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“saturation”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=100</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“hue”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=100</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="2" valign="top" >Optional</td>
+	<td valign="top" >“scaling mode”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "None", "Full", "Center", "Full aspect" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“dirty”</td>
+	<td valign="top" >ENUM | IMMUTABLE</td>
+	<td valign="top" >{ "Off", "On", "Annotate" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="21" valign="top" >i915</td>
+	<td rowspan="3" valign="top" >Generic</td>
+	<td valign="top" >"Broadcast RGB"</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "Automatic", "Full", "Limited 16:235" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“audio”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "force-dvi", "off", "auto", "on" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >Standard name as in DRM</td>
+	<td valign="top" >Standard type as in DRM</td>
+	<td valign="top" >Standard value as in DRM</td>
+	<td valign="top" >Standard Object as in DRM</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="17" valign="top" >SDVO-TV</td>
+	<td valign="top" >“mode”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "NTSC_M", "NTSC_J", "NTSC_443", "PAL_B" } etc.</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"left_margin"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"right_margin"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"top_margin"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"bottom_margin"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“hpos”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“vpos”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“contrast”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“saturation”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“hue”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“sharpness”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“flicker_filter”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“flicker_filter_adaptive”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“flicker_filter_2d”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“tv_chroma_filter”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“tv_luma_filter”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“dot_crawl”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=1</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >SDVO-TV/LVDS</td>
+	<td valign="top" >“brightness”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="3" valign="top" >CDV gma-500</td>
+	<td rowspan="3" valign="top" >Generic</td>
+	<td valign="top" >"Broadcast RGB"</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ “Full”, “Limited 16:235” }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"Broadcast RGB"</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ “off”, “auto”, “on” }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >Standard name as in DRM</td>
+	<td valign="top" >Standard type as in DRM</td>
+	<td valign="top" >Standard value as in DRM</td>
+	<td valign="top" >Standard Object as in DRM</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="20" valign="top" >Poulsbo</td>
+	<td rowspan="2" valign="top" >Generic</td>
+	<td valign="top" >“backlight”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=100</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >Standard name as in DRM</td>
+	<td valign="top" >Standard type as in DRM</td>
+	<td valign="top" >Standard value as in DRM</td>
+	<td valign="top" >Standard Object as in DRM</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="17" valign="top" >SDVO-TV</td>
+	<td valign="top" >“mode”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "NTSC_M", "NTSC_J", "NTSC_443", "PAL_B" } etc.</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"left_margin"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"right_margin"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"top_margin"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"bottom_margin"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“hpos”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“vpos”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“contrast”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“saturation”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“hue”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“sharpness”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“flicker_filter”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“flicker_filter_adaptive”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“flicker_filter_2d”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“tv_chroma_filter”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“tv_luma_filter”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“dot_crawl”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=1</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >SDVO-TV/LVDS</td>
+	<td valign="top" >“brightness”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="11" valign="top" >armada</td>
+	<td rowspan="2" valign="top" >CRTC</td>
+	<td valign="top" >"CSC_YUV"</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "Auto" , "CCIR601", "CCIR709" }</td>
+	<td valign="top" >CRTC</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"CSC_RGB"</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "Auto", "Computer system", "Studio" }</td>
+	<td valign="top" >CRTC</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="9" valign="top" >Overlay</td>
+	<td valign="top" >"colorkey"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=0xffffff</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"colorkey_min"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=0xffffff</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"colorkey_max"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=0xffffff</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"colorkey_val"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=0xffffff</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"colorkey_alpha"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=0xffffff</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"colorkey_mode"</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "disabled", "Y component", "U component"
+	, "V component", "RGB", “R component", "G component", "B component" }</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"brightness"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=256 + 255</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"contrast"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=0x7fff</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"saturation"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=0x7fff</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="2" valign="top" >exynos</td>
+	<td valign="top" >CRTC</td>
+	<td valign="top" >“mode”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "normal", "blank" }</td>
+	<td valign="top" >CRTC</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >Overlay</td>
+	<td valign="top" >“zpos”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=MAX_PLANE-1</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="3" valign="top" >i2c/ch7006_drv</td>
+	<td valign="top" >Generic</td>
+	<td valign="top" >“scale”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=2</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="2" valign="top" >TV</td>
+	<td valign="top" >Standard names as in DRM</td>
+	<td valign="top" >Standard types as in DRM</td>
+	<td valign="top" >Standard Values as in DRM</td>
+	<td valign="top" >Standard object as in DRM</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“mode”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "PAL", "PAL-M","PAL-N"}, ”PAL-Nc"
+	, "PAL-60", "NTSC-M", "NTSC-J" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="16" valign="top" >nouveau</td>
+	<td rowspan="6" valign="top" >NV10 Overlay</td>
+	<td valign="top" >"colorkey"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=0x01ffffff</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“contrast”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=8192-1</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“brightness”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=1024</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“hue”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=359</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“saturation”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=8192-1</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“iturbt_709”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=1</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="2" valign="top" >Nv04 Overlay</td>
+	<td valign="top" >“colorkey”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=0x01ffffff</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“brightness”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=1024</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="7" valign="top" >Display</td>
+	<td valign="top" >“dithering mode”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "auto", "off", "on" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“dithering depth”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "auto", "off", "on", "static 2x2", "dynamic 2x2", "temporal" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“underscan”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "auto", "6 bpc", "8 bpc" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“underscan hborder”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=128</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“underscan vborder”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=128</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“vibrant hue”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=180</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“color vibrance”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=200</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >Generic</td>
+	<td valign="top" >Standard name as in DRM</td>
+	<td valign="top" >Standard type as in DRM</td>
+	<td valign="top" >Standard value as in DRM</td>
+	<td valign="top" >Standard Object as in DRM</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="2" valign="top" >omap</td>
+	<td rowspan="2" valign="top" >Generic</td>
+	<td valign="top" >“rotation”</td>
+	<td valign="top" >BITMASK</td>
+	<td valign="top" >{ 0, "rotate-0" },
+	{ 1, "rotate-90" },
+	{ 2, "rotate-180" },
+	{ 3, "rotate-270" },
+	{ 4, "reflect-x" },
+	{ 5, "reflect-y" }</td>
+	<td valign="top" >CRTC, Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“zorder”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=3</td>
+	<td valign="top" >CRTC, Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >qxl</td>
+	<td valign="top" >Generic</td>
+	<td valign="top" >“hotplug_mode_update"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=1</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="10" valign="top" >radeon</td>
+	<td valign="top" >DVI-I</td>
+	<td valign="top" >“coherent”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=1</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >DAC enable load detect</td>
+	<td valign="top" >“load detection”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=1</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >TV Standard</td>
+	<td valign="top" >"tv standard"</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "ntsc", "pal", "pal-m", "pal-60", "ntsc-j"
+	, "scart-pal", "pal-cn", "secam" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >legacy TMDS PLL detect</td>
+	<td valign="top" >"tmds_pll"</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "driver", "bios" }</td>
+	<td valign="top" >-</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="3" valign="top" >Underscan</td>
+	<td valign="top" >"underscan"</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "off", "on", "auto" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"underscan hborder"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=128</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"underscan vborder"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=128</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >Audio</td>
+	<td valign="top" >“audio”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "off", "on", "auto" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >FMT Dithering</td>
+	<td valign="top" >“dither”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "off", "on" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >Generic</td>
+	<td valign="top" >Standard name as in DRM</td>
+	<td valign="top" >Standard type as in DRM</td>
+	<td valign="top" >Standard value as in DRM</td>
+	<td valign="top" >Standard Object as in DRM</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="3" valign="top" >rcar-du</td>
+	<td rowspan="3" valign="top" >Generic</td>
+	<td valign="top" >"alpha"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=255</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"colorkey"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=0x01ffffff</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"zpos"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=1, Max=7</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	</tbody>
+	</table>
+    </sect2>
   </sect1>
 
   <!-- Internals: vertical blanking -->
@@ -2527,6 +3418,10 @@ void (*disable_vblank) (struct drm_device *dev, int crtc);</synopsis>
       with a call to <function>drm_vblank_cleanup</function> in the driver
       <methodname>unload</methodname> operation handler.
     </para>
+    <sect2>
+      <title>Vertical Blanking and Interrupt Handling Functions Reference</title>
+!Edrivers/gpu/drm/drm_irq.c
+    </sect2>
   </sect1>
 
   <!-- Internals: open/close, file operations and ioctls -->
@@ -2869,17 +3764,16 @@ int num_ioctls;</synopsis>
             <term>DRM_IOCTL_MODESET_CTL</term>
             <listitem>
               <para>
-                This should be called by application level drivers before and
-                after mode setting, since on many devices the vertical blank
-                counter is reset at that time.  Internally, the DRM snapshots
-                the last vblank count when the ioctl is called with the
-                _DRM_PRE_MODESET command, so that the counter won't go backwards
-                (which is dealt with when _DRM_POST_MODESET is used).
+		This was only used for user-mode-settind drivers around
+		modesetting changes to allow the kernel to update the vblank
+		interrupt after mode setting, since on many devices the vertical
+		blank counter is reset to 0 at some point during modeset. Modern
+		drivers should not call this any more since with kernel mode
+		setting it is a no-op.
               </para>
             </listitem>
           </varlistentry>
         </variablelist>
-<!--!Edrivers/char/drm/drm_irq.c-->
       </para>
     </sect1>
 
@@ -2942,6 +3836,96 @@ int num_ioctls;</synopsis>
 	  probing, so those sections fully apply.
         </para>
       </sect2>
+      <sect2>
+        <title>DPIO</title>
+!Pdrivers/gpu/drm/i915/i915_reg.h DPIO
+	<table id="dpiox2">
+	  <title>Dual channel PHY (VLV/CHV)</title>
+	  <tgroup cols="8">
+	    <colspec colname="c0" />
+	    <colspec colname="c1" />
+	    <colspec colname="c2" />
+	    <colspec colname="c3" />
+	    <colspec colname="c4" />
+	    <colspec colname="c5" />
+	    <colspec colname="c6" />
+	    <colspec colname="c7" />
+	    <spanspec spanname="ch0" namest="c0" nameend="c3" />
+	    <spanspec spanname="ch1" namest="c4" nameend="c7" />
+	    <spanspec spanname="ch0pcs01" namest="c0" nameend="c1" />
+	    <spanspec spanname="ch0pcs23" namest="c2" nameend="c3" />
+	    <spanspec spanname="ch1pcs01" namest="c4" nameend="c5" />
+	    <spanspec spanname="ch1pcs23" namest="c6" nameend="c7" />
+	    <thead>
+	      <row>
+		<entry spanname="ch0">CH0</entry>
+		<entry spanname="ch1">CH1</entry>
+	      </row>
+	    </thead>
+	    <tbody valign="top" align="center">
+	      <row>
+		<entry spanname="ch0">CMN/PLL/REF</entry>
+		<entry spanname="ch1">CMN/PLL/REF</entry>
+	      </row>
+	      <row>
+		<entry spanname="ch0pcs01">PCS01</entry>
+		<entry spanname="ch0pcs23">PCS23</entry>
+		<entry spanname="ch1pcs01">PCS01</entry>
+		<entry spanname="ch1pcs23">PCS23</entry>
+	      </row>
+	      <row>
+		<entry>TX0</entry>
+		<entry>TX1</entry>
+		<entry>TX2</entry>
+		<entry>TX3</entry>
+		<entry>TX0</entry>
+		<entry>TX1</entry>
+		<entry>TX2</entry>
+		<entry>TX3</entry>
+	      </row>
+	      <row>
+		<entry spanname="ch0">DDI0</entry>
+		<entry spanname="ch1">DDI1</entry>
+	      </row>
+	    </tbody>
+	  </tgroup>
+	</table>
+	<table id="dpiox1">
+	  <title>Single channel PHY (CHV)</title>
+	  <tgroup cols="4">
+	    <colspec colname="c0" />
+	    <colspec colname="c1" />
+	    <colspec colname="c2" />
+	    <colspec colname="c3" />
+	    <spanspec spanname="ch0" namest="c0" nameend="c3" />
+	    <spanspec spanname="ch0pcs01" namest="c0" nameend="c1" />
+	    <spanspec spanname="ch0pcs23" namest="c2" nameend="c3" />
+	    <thead>
+	      <row>
+		<entry spanname="ch0">CH0</entry>
+	      </row>
+	    </thead>
+	    <tbody valign="top" align="center">
+	      <row>
+		<entry spanname="ch0">CMN/PLL/REF</entry>
+	      </row>
+	      <row>
+		<entry spanname="ch0pcs01">PCS01</entry>
+		<entry spanname="ch0pcs23">PCS23</entry>
+	      </row>
+	      <row>
+		<entry>TX0</entry>
+		<entry>TX1</entry>
+		<entry>TX2</entry>
+		<entry>TX3</entry>
+	      </row>
+	      <row>
+		<entry spanname="ch0">DDI2</entry>
+	      </row>
+	    </tbody>
+	  </tgroup>
+	</table>
+      </sect2>
     </sect1>
 
     <sect1>
@@ -2950,6 +3934,11 @@ int num_ioctls;</synopsis>
 	This sections covers all things related to the GEM implementation in the
 	i915 driver.
       </para>
+      <sect2>
+        <title>Batchbuffer Parsing</title>
+!Pdrivers/gpu/drm/i915/i915_cmd_parser.c batch buffer command parser
+!Idrivers/gpu/drm/i915/i915_cmd_parser.c
+      </sect2>
     </sect1>
   </chapter>
 </part>
diff --git a/Documentation/EDID/1024x768.S b/Documentation/EDID/1024x768.S
index 4b486fe31b32..6f3e4b75e49e 100644
--- a/Documentation/EDID/1024x768.S
+++ b/Documentation/EDID/1024x768.S
@@ -36,7 +36,7 @@
 #define DPI 72
 #define VFREQ 60 /* Hz */
 #define TIMING_NAME "Linux XGA"
-#define ESTABLISHED_TIMINGS_BITS 0x08 /* Bit 3 -> 1024x768 @60 Hz */
+#define ESTABLISHED_TIMING2_BITS 0x08 /* Bit 3 -> 1024x768 @60 Hz */
 #define HSYNC_POL 0
 #define VSYNC_POL 0
 #define CRC 0x55
diff --git a/Documentation/EDID/1280x1024.S b/Documentation/EDID/1280x1024.S
index a2799fe33a4d..bd9bef2a65af 100644
--- a/Documentation/EDID/1280x1024.S
+++ b/Documentation/EDID/1280x1024.S
@@ -36,7 +36,7 @@
 #define DPI 72
 #define VFREQ 60 /* Hz */
 #define TIMING_NAME "Linux SXGA"
-#define ESTABLISHED_TIMINGS_BITS 0x00 /* none */
+/* No ESTABLISHED_TIMINGx_BITS */
 #define HSYNC_POL 1
 #define VSYNC_POL 1
 #define CRC 0xa0
diff --git a/Documentation/EDID/1600x1200.S b/Documentation/EDID/1600x1200.S
index 0ded64cfd1f5..a45101c6160c 100644
--- a/Documentation/EDID/1600x1200.S
+++ b/Documentation/EDID/1600x1200.S
@@ -36,7 +36,7 @@
 #define DPI 72
 #define VFREQ 60 /* Hz */
 #define TIMING_NAME "Linux UXGA"
-#define ESTABLISHED_TIMINGS_BITS 0x00 /* none */
+/* No ESTABLISHED_TIMINGx_BITS */
 #define HSYNC_POL 1
 #define VSYNC_POL 1
 #define CRC 0x9d
diff --git a/Documentation/EDID/1680x1050.S b/Documentation/EDID/1680x1050.S
index 96f67cafcf2e..b0d7c69282b4 100644
--- a/Documentation/EDID/1680x1050.S
+++ b/Documentation/EDID/1680x1050.S
@@ -36,7 +36,7 @@
 #define DPI 96
 #define VFREQ 60 /* Hz */
 #define TIMING_NAME "Linux WSXGA"
-#define ESTABLISHED_TIMINGS_BITS 0x00 /* none */
+/* No ESTABLISHED_TIMINGx_BITS */
 #define HSYNC_POL 1
 #define VSYNC_POL 1
 #define CRC 0x26
diff --git a/Documentation/EDID/1920x1080.S b/Documentation/EDID/1920x1080.S
index 36ed5d571d0a..3084355e81e7 100644
--- a/Documentation/EDID/1920x1080.S
+++ b/Documentation/EDID/1920x1080.S
@@ -36,7 +36,7 @@
 #define DPI 96
 #define VFREQ 60 /* Hz */
 #define TIMING_NAME "Linux FHD"
-#define ESTABLISHED_TIMINGS_BITS 0x00 /* none */
+/* No ESTABLISHED_TIMINGx_BITS */
 #define HSYNC_POL 1
 #define VSYNC_POL 1
 #define CRC 0x05
diff --git a/Documentation/EDID/800x600.S b/Documentation/EDID/800x600.S
new file mode 100644
index 000000000000..6644e26d5801
--- /dev/null
+++ b/Documentation/EDID/800x600.S
@@ -0,0 +1,41 @@
+/*
+   800x600.S: EDID data set for standard 800x600 60 Hz monitor
+
+   Copyright (C) 2011 Carsten Emde <C.Emde@osadl.org>
+   Copyright (C) 2014 Linaro Limited
+
+   This program is free software; you can redistribute it and/or
+   modify it under the terms of the GNU General Public License
+   as published by the Free Software Foundation; either version 2
+   of the License, or (at your option) any later version.
+
+   This program is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+   GNU General Public License for more details.
+*/
+
+/* EDID */
+#define VERSION 1
+#define REVISION 3
+
+/* Display */
+#define CLOCK 40000 /* kHz */
+#define XPIX 800
+#define YPIX 600
+#define XY_RATIO XY_RATIO_4_3
+#define XBLANK 256
+#define YBLANK 28
+#define XOFFSET 40
+#define XPULSE 128
+#define YOFFSET (63+1)
+#define YPULSE (63+4)
+#define DPI 72
+#define VFREQ 60 /* Hz */
+#define TIMING_NAME "Linux SVGA"
+#define ESTABLISHED_TIMING1_BITS 0x01 /* Bit 0: 800x600 @ 60Hz */
+#define HSYNC_POL 1
+#define VSYNC_POL 1
+#define CRC 0xc2
+
+#include "edid.S"
diff --git a/Documentation/EDID/HOWTO.txt b/Documentation/EDID/HOWTO.txt
index 7146db1d9e8c..835db332289b 100644
--- a/Documentation/EDID/HOWTO.txt
+++ b/Documentation/EDID/HOWTO.txt
@@ -18,7 +18,7 @@ CONFIG_DRM_LOAD_EDID_FIRMWARE was introduced. It allows to provide an
 individually prepared or corrected EDID data set in the /lib/firmware
 directory from where it is loaded via the firmware interface. The code
 (see drivers/gpu/drm/drm_edid_load.c) contains built-in data sets for
-commonly used screen resolutions (1024x768, 1280x1024, 1600x1200,
+commonly used screen resolutions (800x600, 1024x768, 1280x1024, 1600x1200,
 1680x1050, 1920x1080) as binary blobs, but the kernel source tree does
 not contain code to create these data. In order to elucidate the origin
 of the built-in binary EDID blobs and to facilitate the creation of
diff --git a/Documentation/EDID/edid.S b/Documentation/EDID/edid.S
index ea97ae275fca..7ac03276d7a2 100644
--- a/Documentation/EDID/edid.S
+++ b/Documentation/EDID/edid.S
@@ -33,6 +33,17 @@
 #define XY_RATIO_5_4	0b10
 #define XY_RATIO_16_9	0b11
 
+/* Provide defaults for the timing bits */
+#ifndef ESTABLISHED_TIMING1_BITS
+#define ESTABLISHED_TIMING1_BITS 0x00
+#endif
+#ifndef ESTABLISHED_TIMING2_BITS
+#define ESTABLISHED_TIMING2_BITS 0x00
+#endif
+#ifndef ESTABLISHED_TIMING3_BITS
+#define ESTABLISHED_TIMING3_BITS 0x00
+#endif
+
 #define mfgname2id(v1,v2,v3) \
 	((((v1-'@')&0x1f)<<10)+(((v2-'@')&0x1f)<<5)+((v3-'@')&0x1f))
 #define swap16(v1) ((v1>>8)+((v1&0xff)<<8))
@@ -139,7 +150,7 @@ white_x_y_msb:	.byte	0x50,0x54
    Bit 2	640x480 @ 75 Hz
    Bit 1	800x600 @ 56 Hz
    Bit 0	800x600 @ 60 Hz */
-estbl_timing1:	.byte	0x00
+estbl_timing1:	.byte	ESTABLISHED_TIMING1_BITS
 
 /* Bit 7	800x600 @ 72 Hz
    Bit 6	800x600 @ 75 Hz
@@ -149,11 +160,11 @@ estbl_timing1:	.byte	0x00
    Bit 2	1024x768 @ 72 Hz
    Bit 1	1024x768 @ 75 Hz
    Bit 0	1280x1024 @ 75 Hz */
-estbl_timing2:	.byte	ESTABLISHED_TIMINGS_BITS
+estbl_timing2:	.byte	ESTABLISHED_TIMING2_BITS
 
 /* Bit 7	1152x870 @ 75 Hz (Apple Macintosh II)
    Bits 6-0 	Other manufacturer-specific display mod */
-estbl_timing3:	.byte	0x00
+estbl_timing3:	.byte	ESTABLISHED_TIMING3_BITS
 
 /* Standard timing */
 /* X resolution, less 31, divided by 8 (256-2288 pixels) */
diff --git a/Documentation/devicetree/bindings/gpu/nvidia,tegra20-host1x.txt b/Documentation/devicetree/bindings/gpu/nvidia,tegra20-host1x.txt
index efa8b8451f93..b48f4ef31d93 100644
--- a/Documentation/devicetree/bindings/gpu/nvidia,tegra20-host1x.txt
+++ b/Documentation/devicetree/bindings/gpu/nvidia,tegra20-host1x.txt
@@ -136,6 +136,7 @@ of the following host1x client modules:
   - compatible: "nvidia,tegra<chip>-hdmi"
   - reg: Physical base address and length of the controller's registers.
   - interrupts: The interrupt outputs from the controller.
+  - hdmi-supply: supply for the +5V HDMI connector pin
   - vdd-supply: regulator for supply voltage
   - pll-supply: regulator for PLL
   - clocks: Must contain an entry for each entry in clock-names.
@@ -180,6 +181,7 @@ of the following host1x client modules:
     See ../reset/reset.txt for details.
   - reset-names: Must include the following entries:
     - dsi
+  - avdd-dsi-supply: phandle of a supply that powers the DSI controller
   - nvidia,mipi-calibrate: Should contain a phandle and a specifier specifying
     which pads are used by this DSI output and need to be calibrated. See also
     ../mipi/nvidia,tegra114-mipi.txt.
diff --git a/Documentation/devicetree/bindings/panel/auo,b133xtn01.txt b/Documentation/devicetree/bindings/panel/auo,b133xtn01.txt
new file mode 100644
index 000000000000..7443b7c76769
--- /dev/null
+++ b/Documentation/devicetree/bindings/panel/auo,b133xtn01.txt
@@ -0,0 +1,7 @@
+AU Optronics Corporation 13.3" WXGA (1366x768) TFT LCD panel
+
+Required properties:
+- compatible: should be "auo,b133xtn01"
+
+This binding is compatible with the simple-panel binding, which is specified
+in simple-panel.txt in this directory.
diff --git a/Documentation/devicetree/bindings/panel/edt,et057090dhu.txt b/Documentation/devicetree/bindings/panel/edt,et057090dhu.txt
new file mode 100644
index 000000000000..4903d7b1d947
--- /dev/null
+++ b/Documentation/devicetree/bindings/panel/edt,et057090dhu.txt
@@ -0,0 +1,7 @@
+Emerging Display Technology Corp. 5.7" VGA TFT LCD panel
+
+Required properties:
+- compatible: should be "edt,et057090dhu"
+
+This binding is compatible with the simple-panel binding, which is specified
+in simple-panel.txt in this directory.
diff --git a/Documentation/devicetree/bindings/panel/edt,et070080dh6.txt b/Documentation/devicetree/bindings/panel/edt,et070080dh6.txt
new file mode 100644
index 000000000000..20cb38e836e4
--- /dev/null
+++ b/Documentation/devicetree/bindings/panel/edt,et070080dh6.txt
@@ -0,0 +1,10 @@
+Emerging Display Technology Corp. ET070080DH6 7.0" WVGA TFT LCD panel
+
+Required properties:
+- compatible: should be "edt,et070080dh6"
+
+This panel is the same as ETM0700G0DH6 except for the touchscreen.
+ET070080DH6 is the model with resistive touch.
+
+This binding is compatible with the simple-panel binding, which is specified
+in simple-panel.txt in this directory.
diff --git a/Documentation/devicetree/bindings/panel/edt,etm0700g0dh6.txt b/Documentation/devicetree/bindings/panel/edt,etm0700g0dh6.txt
new file mode 100644
index 000000000000..ee4b18053e40
--- /dev/null
+++ b/Documentation/devicetree/bindings/panel/edt,etm0700g0dh6.txt
@@ -0,0 +1,10 @@
+Emerging Display Technology Corp. ETM0700G0DH6 7.0" WVGA TFT LCD panel
+
+Required properties:
+- compatible: should be "edt,etm0700g0dh6"
+
+This panel is the same as ET070080DH6 except for the touchscreen.
+ETM0700G0DH6 is the model with capacitive multitouch.
+
+This binding is compatible with the simple-panel binding, which is specified
+in simple-panel.txt in this directory.
diff --git a/Documentation/devicetree/bindings/video/exynos_dp.txt b/Documentation/devicetree/bindings/video/exynos_dp.txt
index 57ccdde02c3a..53dbccfa80ca 100644
--- a/Documentation/devicetree/bindings/video/exynos_dp.txt
+++ b/Documentation/devicetree/bindings/video/exynos_dp.txt
@@ -62,6 +62,10 @@ Optional properties for dp-controller:
 	-hsync-active-high:
 		HSYNC polarity configuration.
 			High if defined, Low if not defined
+	-samsung,hpd-gpio:
+		Hotplug detect GPIO.
+			Indicates which GPIO should be used for hotplug
+			detection
 
 Example:
 
diff --git a/Documentation/devicetree/bindings/video/exynos_hdmi.txt b/Documentation/devicetree/bindings/video/exynos_hdmi.txt
index f9187a259259..1fd8cf9cbfac 100644
--- a/Documentation/devicetree/bindings/video/exynos_hdmi.txt
+++ b/Documentation/devicetree/bindings/video/exynos_hdmi.txt
@@ -5,6 +5,7 @@ Required properties:
 	1) "samsung,exynos5-hdmi" <DEPRECATED>
 	2) "samsung,exynos4210-hdmi"
 	3) "samsung,exynos4212-hdmi"
+	4) "samsung,exynos5420-hdmi"
 - reg: physical base address of the hdmi and length of memory mapped
 	region.
 - interrupts: interrupt number to the cpu.
@@ -27,6 +28,7 @@ Required properties:
 	"hdmi", "sclk_hdmi", "sclk_pixel", "sclk_hdmiphy" and "mout_hdmi".
 - ddc: phandle to the hdmi ddc node
 - phy: phandle to the hdmi phy node
+- samsung,syscon-phandle: phandle for system controller node for PMU.
 
 Example:
 
@@ -37,4 +39,5 @@ Example:
 		hpd-gpio = <&gpx3 7 1>;
 		ddc = <&hdmi_ddc_node>;
 		phy = <&hdmi_phy_node>;
+		samsung,syscon-phandle = <&pmu_system_controller>;
 	};
diff --git a/MAINTAINERS b/MAINTAINERS
index b4a66b9d6b4d..c2297fab77c8 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2952,6 +2952,7 @@ L:	dri-devel@lists.freedesktop.org
 T:	git git://people.freedesktop.org/~airlied/linux
 S:	Maintained
 F:	drivers/gpu/drm/
+F:	drivers/gpu/vga/
 F:	include/drm/
 F:	include/uapi/drm/
 
diff --git a/arch/x86/kernel/early-quirks.c b/arch/x86/kernel/early-quirks.c
index 6cda0baeac9d..2e1a6853e00c 100644
--- a/arch/x86/kernel/early-quirks.c
+++ b/arch/x86/kernel/early-quirks.c
@@ -419,7 +419,7 @@ static size_t __init gen6_stolen_size(int num, int slot, int func)
 	return gmch_ctrl << 25; /* 32 MB units */
 }
 
-static size_t gen8_stolen_size(int num, int slot, int func)
+static size_t __init gen8_stolen_size(int num, int slot, int func)
 {
 	u16 gmch_ctrl;
 
@@ -429,48 +429,73 @@ static size_t gen8_stolen_size(int num, int slot, int func)
 	return gmch_ctrl << 25; /* 32 MB units */
 }
 
+static size_t __init chv_stolen_size(int num, int slot, int func)
+{
+	u16 gmch_ctrl;
+
+	gmch_ctrl = read_pci_config_16(num, slot, func, SNB_GMCH_CTRL);
+	gmch_ctrl >>= SNB_GMCH_GMS_SHIFT;
+	gmch_ctrl &= SNB_GMCH_GMS_MASK;
+
+	/*
+	 * 0x0  to 0x10: 32MB increments starting at 0MB
+	 * 0x11 to 0x16: 4MB increments starting at 8MB
+	 * 0x17 to 0x1d: 4MB increments start at 36MB
+	 */
+	if (gmch_ctrl < 0x11)
+		return gmch_ctrl << 25;
+	else if (gmch_ctrl < 0x17)
+		return (gmch_ctrl - 0x11 + 2) << 22;
+	else
+		return (gmch_ctrl - 0x17 + 9) << 22;
+}
 
 struct intel_stolen_funcs {
 	size_t (*size)(int num, int slot, int func);
 	u32 (*base)(int num, int slot, int func, size_t size);
 };
 
-static const struct intel_stolen_funcs i830_stolen_funcs = {
+static const struct intel_stolen_funcs i830_stolen_funcs __initconst = {
 	.base = i830_stolen_base,
 	.size = i830_stolen_size,
 };
 
-static const struct intel_stolen_funcs i845_stolen_funcs = {
+static const struct intel_stolen_funcs i845_stolen_funcs __initconst = {
 	.base = i845_stolen_base,
 	.size = i830_stolen_size,
 };
 
-static const struct intel_stolen_funcs i85x_stolen_funcs = {
+static const struct intel_stolen_funcs i85x_stolen_funcs __initconst = {
 	.base = i85x_stolen_base,
 	.size = gen3_stolen_size,
 };
 
-static const struct intel_stolen_funcs i865_stolen_funcs = {
+static const struct intel_stolen_funcs i865_stolen_funcs __initconst = {
 	.base = i865_stolen_base,
 	.size = gen3_stolen_size,
 };
 
-static const struct intel_stolen_funcs gen3_stolen_funcs = {
+static const struct intel_stolen_funcs gen3_stolen_funcs __initconst = {
 	.base = intel_stolen_base,
 	.size = gen3_stolen_size,
 };
 
-static const struct intel_stolen_funcs gen6_stolen_funcs = {
+static const struct intel_stolen_funcs gen6_stolen_funcs __initconst = {
 	.base = intel_stolen_base,
 	.size = gen6_stolen_size,
 };
 
-static const struct intel_stolen_funcs gen8_stolen_funcs = {
+static const struct intel_stolen_funcs gen8_stolen_funcs __initconst = {
 	.base = intel_stolen_base,
 	.size = gen8_stolen_size,
 };
 
-static struct pci_device_id intel_stolen_ids[] __initdata = {
+static const struct intel_stolen_funcs chv_stolen_funcs __initconst = {
+	.base = intel_stolen_base,
+	.size = chv_stolen_size,
+};
+
+static const struct pci_device_id intel_stolen_ids[] __initconst = {
 	INTEL_I830_IDS(&i830_stolen_funcs),
 	INTEL_I845G_IDS(&i845_stolen_funcs),
 	INTEL_I85X_IDS(&i85x_stolen_funcs),
@@ -496,7 +521,8 @@ static struct pci_device_id intel_stolen_ids[] __initdata = {
 	INTEL_HSW_D_IDS(&gen6_stolen_funcs),
 	INTEL_HSW_M_IDS(&gen6_stolen_funcs),
 	INTEL_BDW_M_IDS(&gen8_stolen_funcs),
-	INTEL_BDW_D_IDS(&gen8_stolen_funcs)
+	INTEL_BDW_D_IDS(&gen8_stolen_funcs),
+	INTEL_CHV_IDS(&chv_stolen_funcs),
 };
 
 static void __init intel_graphics_stolen(int num, int slot, int func)
diff --git a/drivers/gpu/Makefile b/drivers/gpu/Makefile
index d8a22c2a579d..70da9eb52a42 100644
--- a/drivers/gpu/Makefile
+++ b/drivers/gpu/Makefile
@@ -1,2 +1,3 @@
 obj-y			+= drm/ vga/
 obj-$(CONFIG_TEGRA_HOST1X)	+= host1x/
+obj-$(CONFIG_IMX_IPUV3_CORE)	+= ipu-v3/
diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index d1cc2f613a78..f5120046ff80 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -83,6 +83,8 @@ config DRM_KMS_CMA_HELPER
 
 source "drivers/gpu/drm/i2c/Kconfig"
 
+source "drivers/gpu/drm/bridge/Kconfig"
+
 config DRM_TDFX
 	tristate "3dfx Banshee/Voodoo3+"
 	depends on DRM && PCI
@@ -199,5 +201,3 @@ source "drivers/gpu/drm/msm/Kconfig"
 source "drivers/gpu/drm/tegra/Kconfig"
 
 source "drivers/gpu/drm/panel/Kconfig"
-
-source "drivers/gpu/drm/bridge/Kconfig"
diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile
index 48e38ba22783..dd2ba4269740 100644
--- a/drivers/gpu/drm/Makefile
+++ b/drivers/gpu/drm/Makefile
@@ -14,7 +14,7 @@ drm-y       :=	drm_auth.o drm_buffer.o drm_bufs.o drm_cache.o \
 		drm_info.o drm_debugfs.o drm_encoder_slave.o \
 		drm_trace_points.o drm_global.o drm_prime.o \
 		drm_rect.o drm_vma_manager.o drm_flip_work.o \
-		drm_plane_helper.o
+		drm_modeset_lock.o
 
 drm-$(CONFIG_COMPAT) += drm_ioc32.o
 drm-$(CONFIG_DRM_GEM_CMA_HELPER) += drm_gem_cma_helper.o
@@ -23,7 +23,8 @@ drm-$(CONFIG_DRM_PANEL) += drm_panel.o
 
 drm-usb-y   := drm_usb.o
 
-drm_kms_helper-y := drm_crtc_helper.o drm_dp_helper.o drm_probe_helper.o
+drm_kms_helper-y := drm_crtc_helper.o drm_dp_helper.o drm_probe_helper.o \
+		drm_plane_helper.o
 drm_kms_helper-$(CONFIG_DRM_LOAD_EDID_FIRMWARE) += drm_edid_load.o
 drm_kms_helper-$(CONFIG_DRM_KMS_FB_HELPER) += drm_fb_helper.o
 drm_kms_helper-$(CONFIG_DRM_KMS_CMA_HELPER) += drm_fb_cma_helper.o
diff --git a/drivers/gpu/drm/armada/armada_drv.c b/drivers/gpu/drm/armada/armada_drv.c
index 32982da82694..8ab3cd1a8cdb 100644
--- a/drivers/gpu/drm/armada/armada_drv.c
+++ b/drivers/gpu/drm/armada/armada_drv.c
@@ -173,7 +173,7 @@ static int armada_drm_load(struct drm_device *dev, unsigned long flags)
 	if (ret)
 		goto err_kms;
 
-	ret = drm_irq_install(dev);
+	ret = drm_irq_install(dev, platform_get_irq(dev->platformdev, 0));
 	if (ret)
 		goto err_kms;
 
@@ -402,7 +402,7 @@ static struct platform_driver armada_drm_platform_driver = {
 
 static int __init armada_drm_init(void)
 {
-	armada_drm_driver.num_ioctls = DRM_ARRAY_SIZE(armada_ioctls);
+	armada_drm_driver.num_ioctls = ARRAY_SIZE(armada_ioctls);
 	return platform_driver_register(&armada_drm_platform_driver);
 }
 module_init(armada_drm_init);
diff --git a/drivers/gpu/drm/armada/armada_fbdev.c b/drivers/gpu/drm/armada/armada_fbdev.c
index 948cb14c561e..fd166f532ab9 100644
--- a/drivers/gpu/drm/armada/armada_fbdev.c
+++ b/drivers/gpu/drm/armada/armada_fbdev.c
@@ -181,10 +181,8 @@ void armada_fbdev_lastclose(struct drm_device *dev)
 {
 	struct armada_private *priv = dev->dev_private;
 
-	drm_modeset_lock_all(dev);
 	if (priv->fbdev)
-		drm_fb_helper_restore_fbdev_mode(priv->fbdev);
-	drm_modeset_unlock_all(dev);
+		drm_fb_helper_restore_fbdev_mode_unlocked(priv->fbdev);
 }
 
 void armada_fbdev_fini(struct drm_device *dev)
diff --git a/drivers/gpu/drm/armada/armada_gem.c b/drivers/gpu/drm/armada/armada_gem.c
index 887816f43476..bb9b642d8485 100644
--- a/drivers/gpu/drm/armada/armada_gem.c
+++ b/drivers/gpu/drm/armada/armada_gem.c
@@ -433,7 +433,6 @@ armada_gem_prime_map_dma_buf(struct dma_buf_attachment *attach,
 
 	if (dobj->obj.filp) {
 		struct address_space *mapping;
-		gfp_t gfp;
 		int count;
 
 		count = dobj->obj.size / PAGE_SIZE;
@@ -441,12 +440,11 @@ armada_gem_prime_map_dma_buf(struct dma_buf_attachment *attach,
 			goto free_sgt;
 
 		mapping = file_inode(dobj->obj.filp)->i_mapping;
-		gfp = mapping_gfp_mask(mapping);
 
 		for_each_sg(sgt->sgl, sg, count, i) {
 			struct page *page;
 
-			page = shmem_read_mapping_page_gfp(mapping, i, gfp);
+			page = shmem_read_mapping_page(mapping, i);
 			if (IS_ERR(page)) {
 				num = i;
 				goto release;
diff --git a/drivers/gpu/drm/ast/Makefile b/drivers/gpu/drm/ast/Makefile
index 8df4f284ee24..171aa0622b66 100644
--- a/drivers/gpu/drm/ast/Makefile
+++ b/drivers/gpu/drm/ast/Makefile
@@ -4,6 +4,6 @@
 
 ccflags-y := -Iinclude/drm
 
-ast-y := ast_drv.o ast_main.o ast_mode.o ast_fb.o ast_ttm.o ast_post.o
+ast-y := ast_drv.o ast_main.o ast_mode.o ast_fb.o ast_ttm.o ast_post.o ast_dp501.o
 
-obj-$(CONFIG_DRM_AST) := ast.o
\ No newline at end of file
+obj-$(CONFIG_DRM_AST) := ast.o
diff --git a/drivers/gpu/drm/ast/ast_dp501.c b/drivers/gpu/drm/ast/ast_dp501.c
new file mode 100644
index 000000000000..5da4b62285fa
--- /dev/null
+++ b/drivers/gpu/drm/ast/ast_dp501.c
@@ -0,0 +1,410 @@
+
+#include <linux/firmware.h>
+#include <drm/drmP.h>
+#include "ast_drv.h"
+MODULE_FIRMWARE("ast_dp501_fw.bin");
+
+int ast_load_dp501_microcode(struct drm_device *dev)
+{
+	struct ast_private *ast = dev->dev_private;
+	static char *fw_name = "ast_dp501_fw.bin";
+	int err;
+	err = request_firmware(&ast->dp501_fw, fw_name, dev->dev);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+static void send_ack(struct ast_private *ast)
+{
+	u8 sendack;
+	sendack = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0x9b, 0xff);
+	sendack |= 0x80;
+	ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0x9b, 0x00, sendack);
+}
+
+static void send_nack(struct ast_private *ast)
+{
+	u8 sendack;
+	sendack = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0x9b, 0xff);
+	sendack &= ~0x80;
+	ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0x9b, 0x00, sendack);
+}
+
+static bool wait_ack(struct ast_private *ast)
+{
+	u8 waitack;
+	u32 retry = 0;
+	do {
+		waitack = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xd2, 0xff);
+		waitack &= 0x80;
+		udelay(100);
+	} while ((!waitack) && (retry++ < 1000));
+
+	if (retry < 1000)
+		return true;
+	else
+		return false;
+}
+
+static bool wait_nack(struct ast_private *ast)
+{
+	u8 waitack;
+	u32 retry = 0;
+	do {
+		waitack = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xd2, 0xff);
+		waitack &= 0x80;
+		udelay(100);
+	} while ((waitack) && (retry++ < 1000));
+
+	if (retry < 1000)
+		return true;
+	else
+		return false;
+}
+
+static void set_cmd_trigger(struct ast_private *ast)
+{
+	ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0x9b, ~0x40, 0x40);
+}
+
+static void clear_cmd_trigger(struct ast_private *ast)
+{
+	ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0x9b, ~0x40, 0x00);
+}
+
+#if 0
+static bool wait_fw_ready(struct ast_private *ast)
+{
+	u8 waitready;
+	u32 retry = 0;
+	do {
+		waitready = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xd2, 0xff);
+		waitready &= 0x40;
+		udelay(100);
+	} while ((!waitready) && (retry++ < 1000));
+
+	if (retry < 1000)
+		return true;
+	else
+		return false;
+}
+#endif
+
+static bool ast_write_cmd(struct drm_device *dev, u8 data)
+{
+	struct ast_private *ast = dev->dev_private;
+	int retry = 0;
+	if (wait_nack(ast)) {
+		send_nack(ast);
+		ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0x9a, 0x00, data);
+		send_ack(ast);
+		set_cmd_trigger(ast);
+		do {
+			if (wait_ack(ast)) {
+				clear_cmd_trigger(ast);
+				send_nack(ast);
+				return true;
+			}
+		} while (retry++ < 100);
+	}
+	clear_cmd_trigger(ast);
+	send_nack(ast);
+	return false;
+}
+
+static bool ast_write_data(struct drm_device *dev, u8 data)
+{
+	struct ast_private *ast = dev->dev_private;
+
+	if (wait_nack(ast)) {
+		send_nack(ast);
+		ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0x9a, 0x00, data);
+		send_ack(ast);
+		if (wait_ack(ast)) {
+			send_nack(ast);
+			return true;
+		}
+	}
+	send_nack(ast);
+	return false;
+}
+
+#if 0
+static bool ast_read_data(struct drm_device *dev, u8 *data)
+{
+	struct ast_private *ast = dev->dev_private;
+	u8 tmp;
+
+	*data = 0;
+
+	if (wait_ack(ast) == false)
+		return false;
+	tmp = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xd3, 0xff);
+	*data = tmp;
+	if (wait_nack(ast) == false) {
+		send_nack(ast);
+		return false;
+	}
+	send_nack(ast);
+	return true;
+}
+
+static void clear_cmd(struct ast_private *ast)
+{
+	send_nack(ast);
+	ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0x9a, 0x00, 0x00);
+}
+#endif
+
+void ast_set_dp501_video_output(struct drm_device *dev, u8 mode)
+{
+	ast_write_cmd(dev, 0x40);
+	ast_write_data(dev, mode);
+
+	msleep(10);
+}
+
+static u32 get_fw_base(struct ast_private *ast)
+{
+	return ast_mindwm(ast, 0x1e6e2104) & 0x7fffffff;
+}
+
+bool ast_backup_fw(struct drm_device *dev, u8 *addr, u32 size)
+{
+	struct ast_private *ast = dev->dev_private;
+	u32 i, data;
+	u32 boot_address;
+
+	data = ast_mindwm(ast, 0x1e6e2100) & 0x01;
+	if (data) {
+		boot_address = get_fw_base(ast);
+		for (i = 0; i < size; i += 4)
+			*(u32 *)(addr + i) = ast_mindwm(ast, boot_address + i);
+		return true;
+	}
+	return false;
+}
+
+bool ast_launch_m68k(struct drm_device *dev)
+{
+	struct ast_private *ast = dev->dev_private;
+	u32 i, data, len = 0;
+	u32 boot_address;
+	u8 *fw_addr = NULL;
+	u8 jreg;
+
+	data = ast_mindwm(ast, 0x1e6e2100) & 0x01;
+	if (!data) {
+
+		if (ast->dp501_fw_addr) {
+			fw_addr = ast->dp501_fw_addr;
+			len = 32*1024;
+		} else if (ast->dp501_fw) {
+			fw_addr = (u8 *)ast->dp501_fw->data;
+			len = ast->dp501_fw->size;
+		}
+		/* Get BootAddress */
+		ast_moutdwm(ast, 0x1e6e2000, 0x1688a8a8);
+		data = ast_mindwm(ast, 0x1e6e0004);
+		switch (data & 0x03) {
+		case 0:
+			boot_address = 0x44000000;
+			break;
+		default:
+		case 1:
+			boot_address = 0x48000000;
+			break;
+		case 2:
+			boot_address = 0x50000000;
+			break;
+		case 3:
+			boot_address = 0x60000000;
+			break;
+		}
+		boot_address -= 0x200000; /* -2MB */
+
+		/* copy image to buffer */
+		for (i = 0; i < len; i += 4) {
+			data = *(u32 *)(fw_addr + i);
+			ast_moutdwm(ast, boot_address + i, data);
+		}
+
+		/* Init SCU */
+		ast_moutdwm(ast, 0x1e6e2000, 0x1688a8a8);
+
+		/* Launch FW */
+		ast_moutdwm(ast, 0x1e6e2104, 0x80000000 + boot_address);
+		ast_moutdwm(ast, 0x1e6e2100, 1);
+
+		/* Update Scratch */
+		data = ast_mindwm(ast, 0x1e6e2040) & 0xfffff1ff;		/* D[11:9] = 100b: UEFI handling */
+		data |= 0x800;
+		ast_moutdwm(ast, 0x1e6e2040, data);
+
+		jreg = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0x99, 0xfc); /* D[1:0]: Reserved Video Buffer */
+		jreg |= 0x02;
+		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x99, jreg);
+	}
+	return true;
+}
+
+u8 ast_get_dp501_max_clk(struct drm_device *dev)
+{
+	struct ast_private *ast = dev->dev_private;
+	u32 boot_address, offset, data;
+	u8 linkcap[4], linkrate, linklanes, maxclk = 0xff;
+
+	boot_address = get_fw_base(ast);
+
+	/* validate FW version */
+	offset = 0xf000;
+	data = ast_mindwm(ast, boot_address + offset);
+	if ((data & 0xf0) != 0x10) /* version: 1x */
+		return maxclk;
+
+	/* Read Link Capability */
+	offset  = 0xf014;
+	*(u32 *)linkcap = ast_mindwm(ast, boot_address + offset);
+	if (linkcap[2] == 0) {
+		linkrate = linkcap[0];
+		linklanes = linkcap[1];
+		data = (linkrate == 0x0a) ? (90 * linklanes) : (54 * linklanes);
+		if (data > 0xff)
+			data = 0xff;
+		maxclk = (u8)data;
+	}
+	return maxclk;
+}
+
+bool ast_dp501_read_edid(struct drm_device *dev, u8 *ediddata)
+{
+	struct ast_private *ast = dev->dev_private;
+	u32 i, boot_address, offset, data;
+
+	boot_address = get_fw_base(ast);
+
+	/* validate FW version */
+	offset = 0xf000;
+	data = ast_mindwm(ast, boot_address + offset);
+	if ((data & 0xf0) != 0x10)
+		return false;
+
+	/* validate PnP Monitor */
+	offset = 0xf010;
+	data = ast_mindwm(ast, boot_address + offset);
+	if (!(data & 0x01))
+		return false;
+
+	/* Read EDID */
+	offset = 0xf020;
+	for (i = 0; i < 128; i += 4) {
+		data = ast_mindwm(ast, boot_address + offset + i);
+		*(u32 *)(ediddata + i) = data;
+	}
+
+	return true;
+}
+
+static bool ast_init_dvo(struct drm_device *dev)
+{
+	struct ast_private *ast = dev->dev_private;
+	u8 jreg;
+	u32 data;
+	ast_write32(ast, 0xf004, 0x1e6e0000);
+	ast_write32(ast, 0xf000, 0x1);
+	ast_write32(ast, 0x12000, 0x1688a8a8);
+
+	jreg = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xd0, 0xff);
+	if (!(jreg & 0x80)) {
+		/* Init SCU DVO Settings */
+		data = ast_read32(ast, 0x12008);
+		/* delay phase */
+		data &= 0xfffff8ff;
+		data |= 0x00000500;
+		ast_write32(ast, 0x12008, data);
+
+		if (ast->chip == AST2300) {
+			data = ast_read32(ast, 0x12084);
+			/* multi-pins for DVO single-edge */
+			data |= 0xfffe0000;
+			ast_write32(ast, 0x12084, data);
+
+			data = ast_read32(ast, 0x12088);
+			/* multi-pins for DVO single-edge */
+			data |= 0x000fffff;
+			ast_write32(ast, 0x12088, data);
+
+			data = ast_read32(ast, 0x12090);
+			/* multi-pins for DVO single-edge */
+			data &= 0xffffffcf;
+			data |= 0x00000020;
+			ast_write32(ast, 0x12090, data);
+		} else { /* AST2400 */
+			data = ast_read32(ast, 0x12088);
+			/* multi-pins for DVO single-edge */
+			data |= 0x30000000;
+			ast_write32(ast, 0x12088, data);
+
+			data = ast_read32(ast, 0x1208c);
+			/* multi-pins for DVO single-edge */
+			data |= 0x000000cf;
+			ast_write32(ast, 0x1208c, data);
+
+			data = ast_read32(ast, 0x120a4);
+			/* multi-pins for DVO single-edge */
+			data |= 0xffff0000;
+			ast_write32(ast, 0x120a4, data);
+
+			data = ast_read32(ast, 0x120a8);
+			/* multi-pins for DVO single-edge */
+			data |= 0x0000000f;
+			ast_write32(ast, 0x120a8, data);
+
+			data = ast_read32(ast, 0x12094);
+			/* multi-pins for DVO single-edge */
+			data |= 0x00000002;
+			ast_write32(ast, 0x12094, data);
+		}
+	}
+
+	/* Force to DVO */
+	data = ast_read32(ast, 0x1202c);
+	data &= 0xfffbffff;
+	ast_write32(ast, 0x1202c, data);
+
+	/* Init VGA DVO Settings */
+	ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xa3, 0xcf, 0x80);
+	return true;
+}
+
+void ast_init_3rdtx(struct drm_device *dev)
+{
+	struct ast_private *ast = dev->dev_private;
+	u8 jreg;
+	u32 data;
+	if (ast->chip == AST2300 || ast->chip == AST2400) {
+		jreg = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xd1, 0xff);
+		switch (jreg & 0x0e) {
+		case 0x04:
+			ast_init_dvo(dev);
+			break;
+		case 0x08:
+			ast_launch_m68k(dev);
+			break;
+		case 0x0c:
+			ast_init_dvo(dev);
+			break;
+		default:
+			if (ast->tx_chip_type == AST_TX_SIL164)
+				ast_init_dvo(dev);
+			else {
+				ast_write32(ast, 0x12000, 0x1688a8a8);
+				data = ast_read32(ast, 0x1202c);
+				data &= 0xfffcffff;
+				ast_write32(ast, 0, data);
+			}
+		}
+	}
+}
diff --git a/drivers/gpu/drm/ast/ast_drv.c b/drivers/gpu/drm/ast/ast_drv.c
index 5137f15dba19..44074fbcf7ff 100644
--- a/drivers/gpu/drm/ast/ast_drv.c
+++ b/drivers/gpu/drm/ast/ast_drv.c
@@ -94,9 +94,7 @@ static int ast_drm_thaw(struct drm_device *dev)
 	ast_post_gpu(dev);
 
 	drm_mode_config_reset(dev);
-	drm_modeset_lock_all(dev);
 	drm_helper_resume_force_mode(dev);
-	drm_modeset_unlock_all(dev);
 
 	console_lock();
 	ast_fbdev_set_suspend(dev, 0);
@@ -198,7 +196,6 @@ static const struct file_operations ast_fops = {
 
 static struct drm_driver driver = {
 	.driver_features = DRIVER_MODESET | DRIVER_GEM,
-	.dev_priv_size = 0,
 
 	.load = ast_driver_load,
 	.unload = ast_driver_unload,
diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h
index 9833a1b1acc1..5d6a87573c33 100644
--- a/drivers/gpu/drm/ast/ast_drv.h
+++ b/drivers/gpu/drm/ast/ast_drv.h
@@ -61,9 +61,17 @@ enum ast_chip {
 	AST2200,
 	AST2150,
 	AST2300,
+	AST2400,
 	AST1180,
 };
 
+enum ast_tx_chip {
+	AST_TX_NONE,
+	AST_TX_SIL164,
+	AST_TX_ITE66121,
+	AST_TX_DP501,
+};
+
 #define AST_DRAM_512Mx16 0
 #define AST_DRAM_1Gx16   1
 #define AST_DRAM_512Mx32 2
@@ -102,6 +110,12 @@ struct ast_private {
 	 * we have. */
 	struct ttm_bo_kmap_obj cache_kmap;
 	int next_cursor;
+	bool support_wide_screen;
+
+	enum ast_tx_chip tx_chip_type;
+	u8 dp501_maxclk;
+	u8 *dp501_fw_addr;
+	const struct firmware *dp501_fw;	/* dp501 fw */
 };
 
 int ast_driver_load(struct drm_device *dev, unsigned long flags);
@@ -368,4 +382,14 @@ int ast_mmap(struct file *filp, struct vm_area_struct *vma);
 
 /* ast post */
 void ast_post_gpu(struct drm_device *dev);
+u32 ast_mindwm(struct ast_private *ast, u32 r);
+void ast_moutdwm(struct ast_private *ast, u32 r, u32 v);
+/* ast dp501 */
+int ast_load_dp501_microcode(struct drm_device *dev);
+void ast_set_dp501_video_output(struct drm_device *dev, u8 mode);
+bool ast_launch_m68k(struct drm_device *dev);
+bool ast_backup_fw(struct drm_device *dev, u8 *addr, u32 size);
+bool ast_dp501_read_edid(struct drm_device *dev, u8 *ediddata);
+u8 ast_get_dp501_max_clk(struct drm_device *dev);
+void ast_init_3rdtx(struct drm_device *dev);
 #endif
diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c
index 50535fd5a88d..a2cc6be97983 100644
--- a/drivers/gpu/drm/ast/ast_main.c
+++ b/drivers/gpu/drm/ast/ast_main.c
@@ -66,12 +66,16 @@ uint8_t ast_get_index_reg_mask(struct ast_private *ast,
 static int ast_detect_chip(struct drm_device *dev)
 {
 	struct ast_private *ast = dev->dev_private;
+	uint32_t data, jreg;
 
 	if (dev->pdev->device == PCI_CHIP_AST1180) {
 		ast->chip = AST1100;
 		DRM_INFO("AST 1180 detected\n");
 	} else {
-		if (dev->pdev->revision >= 0x20) {
+		if (dev->pdev->revision >= 0x30) {
+			ast->chip = AST2400;
+			DRM_INFO("AST 2400 detected\n");
+		} else if (dev->pdev->revision >= 0x20) {
 			ast->chip = AST2300;
 			DRM_INFO("AST 2300 detected\n");
 		} else if (dev->pdev->revision >= 0x10) {
@@ -104,6 +108,59 @@ static int ast_detect_chip(struct drm_device *dev)
 			DRM_INFO("AST 2000 detected\n");
 		}
 	}
+
+	switch (ast->chip) {
+	case AST1180:
+		ast->support_wide_screen = true;
+		break;
+	case AST2000:
+		ast->support_wide_screen = false;
+		break;
+	default:
+		jreg = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xd0, 0xff);
+		if (!(jreg & 0x80))
+			ast->support_wide_screen = true;
+		else if (jreg & 0x01)
+			ast->support_wide_screen = true;
+		else {
+			ast->support_wide_screen = false;
+			ast_write32(ast, 0xf004, 0x1e6e0000);
+			ast_write32(ast, 0xf000, 0x1);
+			data = ast_read32(ast, 0x1207c);
+			data &= 0x300;
+			if (ast->chip == AST2300 && data == 0x0) /* ast1300 */
+				ast->support_wide_screen = true;
+			if (ast->chip == AST2400 && data == 0x100) /* ast1400 */
+				ast->support_wide_screen = true;
+		}
+		break;
+	}
+
+	ast->tx_chip_type = AST_TX_NONE;
+	jreg = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xa3, 0xff);
+	if (jreg & 0x80)
+		ast->tx_chip_type = AST_TX_SIL164;
+	if ((ast->chip == AST2300) || (ast->chip == AST2400)) {
+		jreg = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xd1, 0xff);
+		switch (jreg) {
+		case 0x04:
+			ast->tx_chip_type = AST_TX_SIL164;
+			break;
+		case 0x08:
+			ast->dp501_fw_addr = kzalloc(32*1024, GFP_KERNEL);
+			if (ast->dp501_fw_addr) {
+				/* backup firmware */
+				if (ast_backup_fw(dev, ast->dp501_fw_addr, 32*1024)) {
+					kfree(ast->dp501_fw_addr);
+					ast->dp501_fw_addr = NULL;
+				}
+			}
+			/* fallthrough */
+		case 0x0c:
+			ast->tx_chip_type = AST_TX_DP501;
+		}
+	}
+
 	return 0;
 }
 
@@ -129,7 +186,7 @@ static int ast_get_dram_info(struct drm_device *dev)
 	else
 		ast->dram_bus_width = 32;
 
-	if (ast->chip == AST2300) {
+	if (ast->chip == AST2300 || ast->chip == AST2400) {
 		switch (data & 0x03) {
 		case 0:
 			ast->dram_type = AST_DRAM_512Mx16;
@@ -257,17 +314,32 @@ static u32 ast_get_vram_info(struct drm_device *dev)
 {
 	struct ast_private *ast = dev->dev_private;
 	u8 jreg;
-
+	u32 vram_size;
 	ast_open_key(ast);
 
+	vram_size = AST_VIDMEM_DEFAULT_SIZE;
 	jreg = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xaa, 0xff);
 	switch (jreg & 3) {
-	case 0: return AST_VIDMEM_SIZE_8M;
-	case 1: return AST_VIDMEM_SIZE_16M;
-	case 2: return AST_VIDMEM_SIZE_32M;
-	case 3: return AST_VIDMEM_SIZE_64M;
+	case 0: vram_size = AST_VIDMEM_SIZE_8M; break;
+	case 1: vram_size = AST_VIDMEM_SIZE_16M; break;
+	case 2: vram_size = AST_VIDMEM_SIZE_32M; break;
+	case 3: vram_size = AST_VIDMEM_SIZE_64M; break;
+	}
+
+	jreg = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0x99, 0xff);
+	switch (jreg & 0x03) {
+	case 1:
+		vram_size -= 0x100000;
+		break;
+	case 2:
+		vram_size -= 0x200000;
+		break;
+	case 3:
+		vram_size -= 0x400000;
+		break;
 	}
-	return AST_VIDMEM_DEFAULT_SIZE;
+
+	return vram_size;
 }
 
 int ast_driver_load(struct drm_device *dev, unsigned long flags)
@@ -316,6 +388,7 @@ int ast_driver_load(struct drm_device *dev, unsigned long flags)
 	if (ast->chip == AST2100 ||
 	    ast->chip == AST2200 ||
 	    ast->chip == AST2300 ||
+	    ast->chip == AST2400 ||
 	    ast->chip == AST1180) {
 		dev->mode_config.max_width = 1920;
 		dev->mode_config.max_height = 2048;
@@ -343,6 +416,7 @@ int ast_driver_unload(struct drm_device *dev)
 {
 	struct ast_private *ast = dev->dev_private;
 
+	kfree(ast->dp501_fw_addr);
 	ast_mode_fini(dev);
 	ast_fbdev_fini(dev);
 	drm_mode_config_cleanup(dev);
@@ -411,16 +485,13 @@ static void ast_bo_unref(struct ast_bo **bo)
 
 	tbo = &((*bo)->bo);
 	ttm_bo_unref(&tbo);
-	if (tbo == NULL)
-		*bo = NULL;
-
+	*bo = NULL;
 }
+
 void ast_gem_free_object(struct drm_gem_object *obj)
 {
 	struct ast_bo *ast_bo = gem_to_ast_bo(obj);
 
-	if (!ast_bo)
-		return;
 	ast_bo_unref(&ast_bo);
 }
 
diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c
index a4afdc8bb578..114aee941d46 100644
--- a/drivers/gpu/drm/ast/ast_mode.c
+++ b/drivers/gpu/drm/ast/ast_mode.c
@@ -115,11 +115,17 @@ static bool ast_get_vbios_mode_info(struct drm_crtc *crtc, struct drm_display_mo
 		else
 			vbios_mode->enh_table = &res_1280x1024[refresh_rate_index];
 		break;
+	case 1360:
+		vbios_mode->enh_table = &res_1360x768[refresh_rate_index];
+		break;
 	case 1440:
 		vbios_mode->enh_table = &res_1440x900[refresh_rate_index];
 		break;
 	case 1600:
-		vbios_mode->enh_table = &res_1600x1200[refresh_rate_index];
+		if (crtc->mode.crtc_vdisplay == 900)
+			vbios_mode->enh_table = &res_1600x900[refresh_rate_index];
+		else
+			vbios_mode->enh_table = &res_1600x1200[refresh_rate_index];
 		break;
 	case 1680:
 		vbios_mode->enh_table = &res_1680x1050[refresh_rate_index];
@@ -175,14 +181,17 @@ static bool ast_get_vbios_mode_info(struct drm_crtc *crtc, struct drm_display_mo
 		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x8d, refresh_rate_index & 0xff);
 		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x8e, mode_id & 0xff);
 
-		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x91, 0xa8);
-		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x92, crtc->primary->fb->bits_per_pixel);
-		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x93, adjusted_mode->clock / 1000);
-		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x94, adjusted_mode->crtc_hdisplay);
-		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x95, adjusted_mode->crtc_hdisplay >> 8);
+		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x91, 0x00);
+		if (vbios_mode->enh_table->flags & NewModeInfo) {
+			ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x91, 0xa8);
+			ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x92, crtc->primary->fb->bits_per_pixel);
+			ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x93, adjusted_mode->clock / 1000);
+			ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x94, adjusted_mode->crtc_hdisplay);
+			ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x95, adjusted_mode->crtc_hdisplay >> 8);
 
-		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x96, adjusted_mode->crtc_vdisplay);
-		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x97, adjusted_mode->crtc_vdisplay >> 8);
+			ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x96, adjusted_mode->crtc_vdisplay);
+			ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0x97, adjusted_mode->crtc_vdisplay >> 8);
+		}
 	}
 
 	return true;
@@ -389,7 +398,7 @@ static void ast_set_ext_reg(struct drm_crtc *crtc, struct drm_display_mode *mode
 	ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xa8, 0xfd, jregA8);
 
 	/* Set Threshold */
-	if (ast->chip == AST2300) {
+	if (ast->chip == AST2300 || ast->chip == AST2400) {
 		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0xa7, 0x78);
 		ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0xa6, 0x60);
 	} else if (ast->chip == AST2100 ||
@@ -451,9 +460,13 @@ static void ast_crtc_dpms(struct drm_crtc *crtc, int mode)
 	case DRM_MODE_DPMS_STANDBY:
 	case DRM_MODE_DPMS_SUSPEND:
 		ast_set_index_reg_mask(ast, AST_IO_SEQ_PORT, 0x1, 0xdf, 0);
+		if (ast->tx_chip_type == AST_TX_DP501)
+			ast_set_dp501_video_output(crtc->dev, 1);
 		ast_crtc_load_lut(crtc);
 		break;
 	case DRM_MODE_DPMS_OFF:
+		if (ast->tx_chip_type == AST_TX_DP501)
+			ast_set_dp501_video_output(crtc->dev, 0);
 		ast_set_index_reg_mask(ast, AST_IO_SEQ_PORT, 0x1, 0xdf, 0x20);
 		break;
 	}
@@ -729,10 +742,24 @@ static int ast_encoder_init(struct drm_device *dev)
 static int ast_get_modes(struct drm_connector *connector)
 {
 	struct ast_connector *ast_connector = to_ast_connector(connector);
+	struct ast_private *ast = connector->dev->dev_private;
 	struct edid *edid;
 	int ret;
-
-	edid = drm_get_edid(connector, &ast_connector->i2c->adapter);
+	bool flags = false;
+	if (ast->tx_chip_type == AST_TX_DP501) {
+		ast->dp501_maxclk = 0xff;
+		edid = kmalloc(128, GFP_KERNEL);
+		if (!edid)
+			return -ENOMEM;
+
+		flags = ast_dp501_read_edid(connector->dev, (u8 *)edid);
+		if (flags)
+			ast->dp501_maxclk = ast_get_dp501_max_clk(connector->dev);
+		else
+			kfree(edid);
+	}
+	if (!flags)
+		edid = drm_get_edid(connector, &ast_connector->i2c->adapter);
 	if (edid) {
 		drm_mode_connector_update_edid_property(&ast_connector->base, edid);
 		ret = drm_add_edid_modes(connector, edid);
@@ -746,7 +773,56 @@ static int ast_get_modes(struct drm_connector *connector)
 static int ast_mode_valid(struct drm_connector *connector,
 			  struct drm_display_mode *mode)
 {
-	return MODE_OK;
+	struct ast_private *ast = connector->dev->dev_private;
+	int flags = MODE_NOMODE;
+	uint32_t jtemp;
+
+	if (ast->support_wide_screen) {
+		if ((mode->hdisplay == 1680) && (mode->vdisplay == 1050))
+			return MODE_OK;
+		if ((mode->hdisplay == 1280) && (mode->vdisplay == 800))
+			return MODE_OK;
+		if ((mode->hdisplay == 1440) && (mode->vdisplay == 900))
+			return MODE_OK;
+		if ((mode->hdisplay == 1360) && (mode->vdisplay == 768))
+			return MODE_OK;
+		if ((mode->hdisplay == 1600) && (mode->vdisplay == 900))
+			return MODE_OK;
+
+		if ((ast->chip == AST2100) || (ast->chip == AST2200) || (ast->chip == AST2300) || (ast->chip == AST2400) || (ast->chip == AST1180)) {
+			if ((mode->hdisplay == 1920) && (mode->vdisplay == 1080))
+				return MODE_OK;
+
+			if ((mode->hdisplay == 1920) && (mode->vdisplay == 1200)) {
+				jtemp = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xd1, 0xff);
+				if (jtemp & 0x01)
+					return MODE_NOMODE;
+				else
+					return MODE_OK;
+			}
+		}
+	}
+	switch (mode->hdisplay) {
+	case 640:
+		if (mode->vdisplay == 480) flags = MODE_OK;
+		break;
+	case 800:
+		if (mode->vdisplay == 600) flags = MODE_OK;
+		break;
+	case 1024:
+		if (mode->vdisplay == 768) flags = MODE_OK;
+		break;
+	case 1280:
+		if (mode->vdisplay == 1024) flags = MODE_OK;
+		break;
+	case 1600:
+		if (mode->vdisplay == 1200) flags = MODE_OK;
+		break;
+	default:
+		return flags;
+	}
+
+	return flags;
 }
 
 static void ast_connector_destroy(struct drm_connector *connector)
diff --git a/drivers/gpu/drm/ast/ast_post.c b/drivers/gpu/drm/ast/ast_post.c
index 635f6ffc27c2..38d437f3a267 100644
--- a/drivers/gpu/drm/ast/ast_post.c
+++ b/drivers/gpu/drm/ast/ast_post.c
@@ -78,7 +78,7 @@ ast_set_def_ext_reg(struct drm_device *dev)
 	for (i = 0x81; i <= 0x8f; i++)
 		ast_set_index_reg(ast, AST_IO_CRTC_PORT, i, 0x00);
 
-	if (ast->chip == AST2300) {
+	if (ast->chip == AST2300 || ast->chip == AST2400) {
 		if (dev->pdev->revision >= 0x20)
 			ext_reg_info = extreginfo_ast2300;
 		else
@@ -102,23 +102,32 @@ ast_set_def_ext_reg(struct drm_device *dev)
 
 	/* Enable RAMDAC for A1 */
 	reg = 0x04;
-	if (ast->chip == AST2300)
+	if (ast->chip == AST2300 || ast->chip == AST2400)
 		reg |= 0x20;
 	ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb6, 0xff, reg);
 }
 
-static inline u32 mindwm(struct ast_private *ast, u32 r)
+u32 ast_mindwm(struct ast_private *ast, u32 r)
 {
+	uint32_t data;
+
 	ast_write32(ast, 0xf004, r & 0xffff0000);
 	ast_write32(ast, 0xf000, 0x1);
 
+	do {
+		data = ast_read32(ast, 0xf004) & 0xffff0000;
+	} while (data != (r & 0xffff0000));
 	return ast_read32(ast, 0x10000 + (r & 0x0000ffff));
 }
 
-static inline void moutdwm(struct ast_private *ast, u32 r, u32 v)
+void ast_moutdwm(struct ast_private *ast, u32 r, u32 v)
 {
+	uint32_t data;
 	ast_write32(ast, 0xf004, r & 0xffff0000);
 	ast_write32(ast, 0xf000, 0x1);
+	do {
+		data = ast_read32(ast, 0xf004) & 0xffff0000;
+	} while (data != (r & 0xffff0000));
 	ast_write32(ast, 0x10000 + (r & 0x0000ffff), v);
 }
 
@@ -154,28 +163,28 @@ static u32 mmctestburst2_ast2150(struct ast_private *ast, u32 datagen)
 {
 	u32 data, timeout;
 
-	moutdwm(ast, 0x1e6e0070, 0x00000000);
-	moutdwm(ast, 0x1e6e0070, 0x00000001 | (datagen << 3));
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000001 | (datagen << 3));
 	timeout = 0;
 	do {
-		data = mindwm(ast, 0x1e6e0070) & 0x40;
+		data = ast_mindwm(ast, 0x1e6e0070) & 0x40;
 		if (++timeout > TIMEOUT_AST2150) {
-			moutdwm(ast, 0x1e6e0070, 0x00000000);
+			ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
 			return 0xffffffff;
 		}
 	} while (!data);
-	moutdwm(ast, 0x1e6e0070, 0x00000000);
-	moutdwm(ast, 0x1e6e0070, 0x00000003 | (datagen << 3));
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000003 | (datagen << 3));
 	timeout = 0;
 	do {
-		data = mindwm(ast, 0x1e6e0070) & 0x40;
+		data = ast_mindwm(ast, 0x1e6e0070) & 0x40;
 		if (++timeout > TIMEOUT_AST2150) {
-			moutdwm(ast, 0x1e6e0070, 0x00000000);
+			ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
 			return 0xffffffff;
 		}
 	} while (!data);
-	data = (mindwm(ast, 0x1e6e0070) & 0x80) >> 7;
-	moutdwm(ast, 0x1e6e0070, 0x00000000);
+	data = (ast_mindwm(ast, 0x1e6e0070) & 0x80) >> 7;
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
 	return data;
 }
 
@@ -184,18 +193,18 @@ static u32 mmctestsingle2_ast2150(struct ast_private *ast, u32 datagen)
 {
 	u32 data, timeout;
 
-	moutdwm(ast, 0x1e6e0070, 0x00000000);
-	moutdwm(ast, 0x1e6e0070, 0x00000005 | (datagen << 3));
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000005 | (datagen << 3));
 	timeout = 0;
 	do {
-		data = mindwm(ast, 0x1e6e0070) & 0x40;
+		data = ast_mindwm(ast, 0x1e6e0070) & 0x40;
 		if (++timeout > TIMEOUT_AST2150) {
-			moutdwm(ast, 0x1e6e0070, 0x00000000);
+			ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
 			return 0xffffffff;
 		}
 	} while (!data);
-	data = (mindwm(ast, 0x1e6e0070) & 0x80) >> 7;
-	moutdwm(ast, 0x1e6e0070, 0x00000000);
+	data = (ast_mindwm(ast, 0x1e6e0070) & 0x80) >> 7;
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
 	return data;
 }
 #endif
@@ -215,7 +224,7 @@ static int cbrscan_ast2150(struct ast_private *ast, int busw)
 	u32 patcnt, loop;
 
 	for (patcnt = 0; patcnt < CBR_PATNUM_AST2150; patcnt++) {
-		moutdwm(ast, 0x1e6e007c, pattern_AST2150[patcnt]);
+		ast_moutdwm(ast, 0x1e6e007c, pattern_AST2150[patcnt]);
 		for (loop = 0; loop < CBR_PASSNUM_AST2150; loop++) {
 			if (cbrtest_ast2150(ast))
 				break;
@@ -237,7 +246,7 @@ cbr_start:
 	passcnt = 0;
 
 	for (dlli = 0; dlli < 100; dlli++) {
-		moutdwm(ast, 0x1e6e0068, dlli | (dlli << 8) | (dlli << 16) | (dlli << 24));
+		ast_moutdwm(ast, 0x1e6e0068, dlli | (dlli << 8) | (dlli << 16) | (dlli << 24));
 		data = cbrscan_ast2150(ast, busw);
 		if (data != 0) {
 			if (data & 0x1) {
@@ -254,7 +263,7 @@ cbr_start:
 		goto cbr_start;
 
 	dlli = dll_min[0] + (((dll_max[0] - dll_min[0]) * 7) >> 4);
-	moutdwm(ast, 0x1e6e0068, dlli | (dlli << 8) | (dlli << 16) | (dlli << 24));
+	ast_moutdwm(ast, 0x1e6e0068, dlli | (dlli << 8) | (dlli << 16) | (dlli << 24));
 }
 
 
@@ -365,10 +374,12 @@ void ast_post_gpu(struct drm_device *dev)
 	ast_open_key(ast);
 	ast_set_def_ext_reg(dev);
 
-	if (ast->chip == AST2300)
+	if (ast->chip == AST2300 || ast->chip == AST2400)
 		ast_init_dram_2300(dev);
 	else
 		ast_init_dram_reg(dev);
+
+	ast_init_3rdtx(dev);
 }
 
 /* AST 2300 DRAM settings */
@@ -403,6 +414,7 @@ struct ast2300_dram_param {
 /*
  * DQSI DLL CBR Setting
  */
+#define CBR_SIZE0            ((1  << 10) - 1)
 #define CBR_SIZE1            ((4  << 10) - 1)
 #define CBR_SIZE2            ((64 << 10) - 1)
 #define CBR_PASSNUM          5
@@ -423,88 +435,84 @@ static const u32 pattern[8] = {
 	0x7C61D253
 };
 
-#if 0 /* unused in DDX, included for completeness */
 static int mmc_test_burst(struct ast_private *ast, u32 datagen)
 {
 	u32 data, timeout;
 
-	moutdwm(ast, 0x1e6e0070, 0x00000000);
-	moutdwm(ast, 0x1e6e0070, 0x000000c1 | (datagen << 3));
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
+	ast_moutdwm(ast, 0x1e6e0070, 0x000000c1 | (datagen << 3));
 	timeout = 0;
 	do {
-		data = mindwm(ast, 0x1e6e0070) & 0x3000;
+		data = ast_mindwm(ast, 0x1e6e0070) & 0x3000;
 		if (data & 0x2000) {
 			return 0;
 		}
 		if (++timeout > TIMEOUT) {
-			moutdwm(ast, 0x1e6e0070, 0x00000000);
+			ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
 			return 0;
 		}
 	} while (!data);
-	moutdwm(ast, 0x1e6e0070, 0x00000000);
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
 	return 1;
 }
-#endif
 
 static int mmc_test_burst2(struct ast_private *ast, u32 datagen)
 {
 	u32 data, timeout;
 
-	moutdwm(ast, 0x1e6e0070, 0x00000000);
-	moutdwm(ast, 0x1e6e0070, 0x00000041 | (datagen << 3));
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000041 | (datagen << 3));
 	timeout = 0;
 	do {
-		data = mindwm(ast, 0x1e6e0070) & 0x1000;
+		data = ast_mindwm(ast, 0x1e6e0070) & 0x1000;
 		if (++timeout > TIMEOUT) {
-			moutdwm(ast, 0x1e6e0070, 0x0);
+			ast_moutdwm(ast, 0x1e6e0070, 0x0);
 			return -1;
 		}
 	} while (!data);
-	data = mindwm(ast, 0x1e6e0078);
+	data = ast_mindwm(ast, 0x1e6e0078);
 	data = (data | (data >> 16)) & 0xffff;
-	moutdwm(ast, 0x1e6e0070, 0x0);
+	ast_moutdwm(ast, 0x1e6e0070, 0x0);
 	return data;
 }
 
-#if 0 /* Unused in DDX here for completeness */
 static int mmc_test_single(struct ast_private *ast, u32 datagen)
 {
 	u32 data, timeout;
 
-	moutdwm(ast, 0x1e6e0070, 0x00000000);
-	moutdwm(ast, 0x1e6e0070, 0x000000c5 | (datagen << 3));
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
+	ast_moutdwm(ast, 0x1e6e0070, 0x000000c5 | (datagen << 3));
 	timeout = 0;
 	do {
-		data = mindwm(ast, 0x1e6e0070) & 0x3000;
+		data = ast_mindwm(ast, 0x1e6e0070) & 0x3000;
 		if (data & 0x2000)
 			return 0;
 		if (++timeout > TIMEOUT) {
-			moutdwm(ast, 0x1e6e0070, 0x0);
+			ast_moutdwm(ast, 0x1e6e0070, 0x0);
 			return 0;
 		}
 	} while (!data);
-	moutdwm(ast, 0x1e6e0070, 0x0);
+	ast_moutdwm(ast, 0x1e6e0070, 0x0);
 	return 1;
 }
-#endif
 
 static int mmc_test_single2(struct ast_private *ast, u32 datagen)
 {
 	u32 data, timeout;
 
-	moutdwm(ast, 0x1e6e0070, 0x00000000);
-	moutdwm(ast, 0x1e6e0070, 0x00000005 | (datagen << 3));
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000000);
+	ast_moutdwm(ast, 0x1e6e0070, 0x00000005 | (datagen << 3));
 	timeout = 0;
 	do {
-		data = mindwm(ast, 0x1e6e0070) & 0x1000;
+		data = ast_mindwm(ast, 0x1e6e0070) & 0x1000;
 		if (++timeout > TIMEOUT) {
-			moutdwm(ast, 0x1e6e0070, 0x0);
+			ast_moutdwm(ast, 0x1e6e0070, 0x0);
 			return -1;
 		}
 	} while (!data);
-	data = mindwm(ast, 0x1e6e0078);
+	data = ast_mindwm(ast, 0x1e6e0078);
 	data = (data | (data >> 16)) & 0xffff;
-	moutdwm(ast, 0x1e6e0070, 0x0);
+	ast_moutdwm(ast, 0x1e6e0070, 0x0);
 	return data;
 }
 
@@ -533,7 +541,7 @@ static int cbr_scan(struct ast_private *ast)
 
 	data2 = 3;
 	for (patcnt = 0; patcnt < CBR_PATNUM; patcnt++) {
-		moutdwm(ast, 0x1e6e007c, pattern[patcnt]);
+		ast_moutdwm(ast, 0x1e6e007c, pattern[patcnt]);
 		for (loop = 0; loop < CBR_PASSNUM2; loop++) {
 			if ((data = cbr_test(ast)) != 0) {
 				data2 &= data;
@@ -568,7 +576,7 @@ static u32 cbr_scan2(struct ast_private *ast)
 
 	data2 = 0xffff;
 	for (patcnt = 0; patcnt < CBR_PATNUM; patcnt++) {
-		moutdwm(ast, 0x1e6e007c, pattern[patcnt]);
+		ast_moutdwm(ast, 0x1e6e007c, pattern[patcnt]);
 		for (loop = 0; loop < CBR_PASSNUM2; loop++) {
 			if ((data = cbr_test2(ast)) != 0) {
 				data2 &= data;
@@ -583,106 +591,35 @@ static u32 cbr_scan2(struct ast_private *ast)
 	return data2;
 }
 
-#if 0 /* unused in DDX - added for completeness */
-static void finetuneDQI(struct ast_private *ast, struct ast2300_dram_param *param)
+static u32 cbr_test3(struct ast_private *ast)
 {
-	u32 gold_sadj[2], dllmin[16], dllmax[16], dlli, data, cnt, mask, passcnt;
-
-	gold_sadj[0] = (mindwm(ast, 0x1E6E0024) >> 16) & 0xffff;
-	gold_sadj[1] = gold_sadj[0] >> 8;
-	gold_sadj[0] = gold_sadj[0] & 0xff;
-	gold_sadj[0] = (gold_sadj[0] + gold_sadj[1]) >> 1;
-	gold_sadj[1] = gold_sadj[0];
-
-	for (cnt = 0; cnt < 16; cnt++) {
-		dllmin[cnt] = 0xff;
-		dllmax[cnt] = 0x0;
-	}
-	passcnt = 0;
-	for (dlli = 0; dlli < 76; dlli++) {
-		moutdwm(ast, 0x1E6E0068, 0x00001400 | (dlli << 16) | (dlli << 24));
-		/* Wait DQSI latch phase calibration */
-		moutdwm(ast, 0x1E6E0074, 0x00000010);
-		moutdwm(ast, 0x1E6E0070, 0x00000003);
-		do {
-			data = mindwm(ast, 0x1E6E0070);
-		} while (!(data & 0x00001000));
-		moutdwm(ast, 0x1E6E0070, 0x00000000);
+	if (!mmc_test_burst(ast, 0))
+		return 0;
+	if (!mmc_test_single(ast, 0))
+		return 0;
+	return 1;
+}
 
-		moutdwm(ast, 0x1E6E0074, CBR_SIZE1);
-		data = cbr_scan2(ast);
-		if (data != 0) {
-			mask = 0x00010001;
-			for (cnt = 0; cnt < 16; cnt++) {
-				if (data & mask) {
-					if (dllmin[cnt] > dlli) {
-						dllmin[cnt] = dlli;
-					}
-					if (dllmax[cnt] < dlli) {
-						dllmax[cnt] = dlli;
-					}
-				}
-				mask <<= 1;
-			}
-			passcnt++;
-		} else if (passcnt >= CBR_THRESHOLD) {
-			break;
-		}
-	}
-	data = 0;
-	for (cnt = 0; cnt < 8; cnt++) {
-		data >>= 3;
-		if ((dllmax[cnt] > dllmin[cnt]) && ((dllmax[cnt] - dllmin[cnt]) >= CBR_THRESHOLD)) {
-			dlli = (dllmin[cnt] + dllmax[cnt]) >> 1;
-			if (gold_sadj[0] >= dlli) {
-				dlli = (gold_sadj[0] - dlli) >> 1;
-				if (dlli > 3) {
-					dlli = 3;
-				}
-			} else {
-				dlli = (dlli - gold_sadj[0]) >> 1;
-				if (dlli > 4) {
-					dlli = 4;
-				}
-				dlli = (8 - dlli) & 0x7;
-			}
-			data |= dlli << 21;
-		}
-	}
-	moutdwm(ast, 0x1E6E0080, data);
+static u32 cbr_scan3(struct ast_private *ast)
+{
+	u32 patcnt, loop;
 
-	data = 0;
-	for (cnt = 8; cnt < 16; cnt++) {
-		data >>= 3;
-		if ((dllmax[cnt] > dllmin[cnt]) && ((dllmax[cnt] - dllmin[cnt]) >= CBR_THRESHOLD)) {
-			dlli = (dllmin[cnt] + dllmax[cnt]) >> 1;
-			if (gold_sadj[1] >= dlli) {
-				dlli = (gold_sadj[1] - dlli) >> 1;
-				if (dlli > 3) {
-					dlli = 3;
-				} else {
-					dlli = (dlli - 1) & 0x7;
-				}
-			} else {
-				dlli = (dlli - gold_sadj[1]) >> 1;
-				dlli += 1;
-				if (dlli > 4) {
-					dlli = 4;
-				}
-				dlli = (8 - dlli) & 0x7;
-			}
-			data |= dlli << 21;
+	for (patcnt = 0; patcnt < CBR_PATNUM; patcnt++) {
+		ast_moutdwm(ast, 0x1e6e007c, pattern[patcnt]);
+		for (loop = 0; loop < 2; loop++) {
+			if (cbr_test3(ast))
+				break;
 		}
+		if (loop == 2)
+			return 0;
 	}
-	moutdwm(ast, 0x1E6E0084, data);
-
-} /* finetuneDQI */
-#endif
+	return 1;
+}
 
-static void finetuneDQI_L(struct ast_private *ast, struct ast2300_dram_param *param)
+static bool finetuneDQI_L(struct ast_private *ast, struct ast2300_dram_param *param)
 {
-	u32 gold_sadj[2], dllmin[16], dllmax[16], dlli, data, cnt, mask, passcnt;
-
+	u32 gold_sadj[2], dllmin[16], dllmax[16], dlli, data, cnt, mask, passcnt, retry = 0;
+	bool status = false;
 FINETUNE_START:
 	for (cnt = 0; cnt < 16; cnt++) {
 		dllmin[cnt] = 0xff;
@@ -690,16 +627,8 @@ FINETUNE_START:
 	}
 	passcnt = 0;
 	for (dlli = 0; dlli < 76; dlli++) {
-		moutdwm(ast, 0x1E6E0068, 0x00001400 | (dlli << 16) | (dlli << 24));
-		/* Wait DQSI latch phase calibration */
-		moutdwm(ast, 0x1E6E0074, 0x00000010);
-		moutdwm(ast, 0x1E6E0070, 0x00000003);
-		do {
-			data = mindwm(ast, 0x1E6E0070);
-		} while (!(data & 0x00001000));
-		moutdwm(ast, 0x1E6E0070, 0x00000000);
-
-		moutdwm(ast, 0x1E6E0074, CBR_SIZE1);
+		ast_moutdwm(ast, 0x1E6E0068, 0x00001400 | (dlli << 16) | (dlli << 24));
+		ast_moutdwm(ast, 0x1E6E0074, CBR_SIZE1);
 		data = cbr_scan2(ast);
 		if (data != 0) {
 			mask = 0x00010001;
@@ -727,9 +656,13 @@ FINETUNE_START:
 			passcnt++;
 		}
 	}
+	if (retry++ > 10)
+		goto FINETUNE_DONE;
 	if (passcnt != 16) {
 		goto FINETUNE_START;
 	}
+	status = true;
+FINETUNE_DONE:
 	gold_sadj[0] = gold_sadj[0] >> 4;
 	gold_sadj[1] = gold_sadj[0];
 
@@ -753,7 +686,7 @@ FINETUNE_START:
 			data |= dlli << 21;
 		}
 	}
-	moutdwm(ast, 0x1E6E0080, data);
+	ast_moutdwm(ast, 0x1E6E0080, data);
 
 	data = 0;
 	for (cnt = 8; cnt < 16; cnt++) {
@@ -778,162 +711,116 @@ FINETUNE_START:
 			data |= dlli << 21;
 		}
 	}
-	moutdwm(ast, 0x1E6E0084, data);
-
+	ast_moutdwm(ast, 0x1E6E0084, data);
+	return status;
 } /* finetuneDQI_L */
 
-static void finetuneDQI_L2(struct ast_private *ast, struct ast2300_dram_param *param)
+static void finetuneDQSI(struct ast_private *ast)
 {
-	u32 gold_sadj[2], dllmin[16], dllmax[16], dlli, data, cnt, mask, passcnt, data2;
+	u32 dlli, dqsip, dqidly;
+	u32 reg_mcr18, reg_mcr0c, passcnt[2], diff;
+	u32 g_dqidly, g_dqsip, g_margin, g_side;
+	u16 pass[32][2][2];
+	char tag[2][76];
+
+	/* Disable DQI CBR */
+	reg_mcr0c  = ast_mindwm(ast, 0x1E6E000C);
+	reg_mcr18  = ast_mindwm(ast, 0x1E6E0018);
+	reg_mcr18 &= 0x0000ffff;
+	ast_moutdwm(ast, 0x1E6E0018, reg_mcr18);
 
-	for (cnt = 0; cnt < 16; cnt++) {
-		dllmin[cnt] = 0xff;
-		dllmax[cnt] = 0x0;
-	}
-	passcnt = 0;
 	for (dlli = 0; dlli < 76; dlli++) {
-		moutdwm(ast, 0x1E6E0068, 0x00001400 | (dlli << 16) | (dlli << 24));
-		/* Wait DQSI latch phase calibration */
-		moutdwm(ast, 0x1E6E0074, 0x00000010);
-		moutdwm(ast, 0x1E6E0070, 0x00000003);
-		do {
-			data = mindwm(ast, 0x1E6E0070);
-		} while (!(data & 0x00001000));
-		moutdwm(ast, 0x1E6E0070, 0x00000000);
-
-		moutdwm(ast, 0x1E6E0074, CBR_SIZE2);
-		data = cbr_scan2(ast);
-		if (data != 0) {
-			mask = 0x00010001;
-			for (cnt = 0; cnt < 16; cnt++) {
-				if (data & mask) {
-					if (dllmin[cnt] > dlli) {
-						dllmin[cnt] = dlli;
-					}
-					if (dllmax[cnt] < dlli) {
-						dllmax[cnt] = dlli;
-					}
-				}
-				mask <<= 1;
-			}
-			passcnt++;
-		} else if (passcnt >= CBR_THRESHOLD2) {
-			break;
-		}
+		tag[0][dlli] = 0x0;
+		tag[1][dlli] = 0x0;
 	}
-	gold_sadj[0] = 0x0;
-	gold_sadj[1] = 0xFF;
-	for (cnt = 0; cnt < 8; cnt++) {
-		if ((dllmax[cnt] > dllmin[cnt]) && ((dllmax[cnt] - dllmin[cnt]) >= CBR_THRESHOLD2)) {
-			if (gold_sadj[0] < dllmin[cnt]) {
-				gold_sadj[0] = dllmin[cnt];
-			}
-			if (gold_sadj[1] > dllmax[cnt]) {
-				gold_sadj[1] = dllmax[cnt];
-			}
-		}
+	for (dqidly = 0; dqidly < 32; dqidly++) {
+		pass[dqidly][0][0] = 0xff;
+		pass[dqidly][0][1] = 0x0;
+		pass[dqidly][1][0] = 0xff;
+		pass[dqidly][1][1] = 0x0;
 	}
-	gold_sadj[0] = (gold_sadj[1] + gold_sadj[0]) >> 1;
-	gold_sadj[1] = mindwm(ast, 0x1E6E0080);
-
-	data = 0;
-	for (cnt = 0; cnt < 8; cnt++) {
-		data >>= 3;
-		data2 = gold_sadj[1] & 0x7;
-		gold_sadj[1] >>= 3;
-		if ((dllmax[cnt] > dllmin[cnt]) && ((dllmax[cnt] - dllmin[cnt]) >= CBR_THRESHOLD2)) {
-			dlli = (dllmin[cnt] + dllmax[cnt]) >> 1;
-			if (gold_sadj[0] >= dlli) {
-				dlli = (gold_sadj[0] - dlli) >> 1;
-				if (dlli > 0) {
-					dlli = 1;
-				}
-				if (data2 != 3) {
-					data2 = (data2 + dlli) & 0x7;
-				}
-			} else {
-				dlli = (dlli - gold_sadj[0]) >> 1;
-				if (dlli > 0) {
-					dlli = 1;
-				}
-				if (data2 != 4) {
-					data2 = (data2 - dlli) & 0x7;
+	for (dqidly = 0; dqidly < 32; dqidly++) {
+		passcnt[0] = passcnt[1] = 0;
+		for (dqsip = 0; dqsip < 2; dqsip++) {
+			ast_moutdwm(ast, 0x1E6E000C, 0);
+			ast_moutdwm(ast, 0x1E6E0018, reg_mcr18 | (dqidly << 16) | (dqsip << 23));
+			ast_moutdwm(ast, 0x1E6E000C, reg_mcr0c);
+			for (dlli = 0; dlli < 76; dlli++) {
+				ast_moutdwm(ast, 0x1E6E0068, 0x00001300 | (dlli << 16) | (dlli << 24));
+				ast_moutdwm(ast, 0x1E6E0070, 0);
+				ast_moutdwm(ast, 0x1E6E0074, CBR_SIZE0);
+				if (cbr_scan3(ast)) {
+					if (dlli == 0)
+						break;
+					passcnt[dqsip]++;
+					tag[dqsip][dlli] = 'P';
+					if (dlli < pass[dqidly][dqsip][0])
+						pass[dqidly][dqsip][0] = (u16) dlli;
+					if (dlli > pass[dqidly][dqsip][1])
+						pass[dqidly][dqsip][1] = (u16) dlli;
+				} else if (passcnt[dqsip] >= 5)
+					break;
+				else {
+					pass[dqidly][dqsip][0] = 0xff;
+					pass[dqidly][dqsip][1] = 0x0;
 				}
 			}
 		}
-		data |= data2 << 21;
-	}
-	moutdwm(ast, 0x1E6E0080, data);
-
-	gold_sadj[0] = 0x0;
-	gold_sadj[1] = 0xFF;
-	for (cnt = 8; cnt < 16; cnt++) {
-		if ((dllmax[cnt] > dllmin[cnt]) && ((dllmax[cnt] - dllmin[cnt]) >= CBR_THRESHOLD2)) {
-			if (gold_sadj[0] < dllmin[cnt]) {
-				gold_sadj[0] = dllmin[cnt];
-			}
-			if (gold_sadj[1] > dllmax[cnt]) {
-				gold_sadj[1] = dllmax[cnt];
-			}
-		}
+		if (passcnt[0] == 0 && passcnt[1] == 0)
+			dqidly++;
 	}
-	gold_sadj[0] = (gold_sadj[1] + gold_sadj[0]) >> 1;
-	gold_sadj[1] = mindwm(ast, 0x1E6E0084);
-
-	data = 0;
-	for (cnt = 8; cnt < 16; cnt++) {
-		data >>= 3;
-		data2 = gold_sadj[1] & 0x7;
-		gold_sadj[1] >>= 3;
-		if ((dllmax[cnt] > dllmin[cnt]) && ((dllmax[cnt] - dllmin[cnt]) >= CBR_THRESHOLD2)) {
-			dlli = (dllmin[cnt] + dllmax[cnt]) >> 1;
-			if (gold_sadj[0] >= dlli) {
-				dlli = (gold_sadj[0] - dlli) >> 1;
-				if (dlli > 0) {
-					dlli = 1;
-				}
-				if (data2 != 3) {
-					data2 = (data2 + dlli) & 0x7;
-				}
-			} else {
-				dlli = (dlli - gold_sadj[0]) >> 1;
-				if (dlli > 0) {
-					dlli = 1;
-				}
-				if (data2 != 4) {
-					data2 = (data2 - dlli) & 0x7;
-				}
+	/* Search margin */
+	g_dqidly = g_dqsip = g_margin = g_side = 0;
+
+	for (dqidly = 0; dqidly < 32; dqidly++) {
+		for (dqsip = 0; dqsip < 2; dqsip++) {
+			if (pass[dqidly][dqsip][0] > pass[dqidly][dqsip][1])
+				continue;
+			diff = pass[dqidly][dqsip][1] - pass[dqidly][dqsip][0];
+			if ((diff+2) < g_margin)
+				continue;
+			passcnt[0] = passcnt[1] = 0;
+			for (dlli = pass[dqidly][dqsip][0]; dlli > 0  && tag[dqsip][dlli] != 0; dlli--, passcnt[0]++);
+			for (dlli = pass[dqidly][dqsip][1]; dlli < 76 && tag[dqsip][dlli] != 0; dlli++, passcnt[1]++);
+			if (passcnt[0] > passcnt[1])
+				passcnt[0] = passcnt[1];
+			passcnt[1] = 0;
+			if (passcnt[0] > g_side)
+				passcnt[1] = passcnt[0] - g_side;
+			if (diff > (g_margin+1) && (passcnt[1] > 0 || passcnt[0] > 8)) {
+				g_margin = diff;
+				g_dqidly = dqidly;
+				g_dqsip  = dqsip;
+				g_side   = passcnt[0];
+			} else if (passcnt[1] > 1 && g_side < 8) {
+				if (diff > g_margin)
+					g_margin = diff;
+				g_dqidly = dqidly;
+				g_dqsip  = dqsip;
+				g_side   = passcnt[0];
 			}
 		}
-		data |= data2 << 21;
 	}
-	moutdwm(ast, 0x1E6E0084, data);
-
-} /* finetuneDQI_L2 */
+	reg_mcr18 = reg_mcr18 | (g_dqidly << 16) | (g_dqsip << 23);
+	ast_moutdwm(ast, 0x1E6E0018, reg_mcr18);
 
-static void cbr_dll2(struct ast_private *ast, struct ast2300_dram_param *param)
+}
+static bool cbr_dll2(struct ast_private *ast, struct ast2300_dram_param *param)
 {
-	u32 dllmin[2], dllmax[2], dlli, data, data2, passcnt;
+	u32 dllmin[2], dllmax[2], dlli, data, passcnt, retry = 0;
+	bool status = false;
 
-
-	finetuneDQI_L(ast, param);
-	finetuneDQI_L2(ast, param);
+	finetuneDQSI(ast);
+	if (finetuneDQI_L(ast, param) == false)
+		return status;
 
 CBR_START2:
 	dllmin[0] = dllmin[1] = 0xff;
 	dllmax[0] = dllmax[1] = 0x0;
 	passcnt = 0;
 	for (dlli = 0; dlli < 76; dlli++) {
-		moutdwm(ast, 0x1E6E0068, 0x00001300 | (dlli << 16) | (dlli << 24));
-		/* Wait DQSI latch phase calibration */
-		moutdwm(ast, 0x1E6E0074, 0x00000010);
-		moutdwm(ast, 0x1E6E0070, 0x00000003);
-		do {
-			data = mindwm(ast, 0x1E6E0070);
-		} while (!(data & 0x00001000));
-		moutdwm(ast, 0x1E6E0070, 0x00000000);
-
-		moutdwm(ast, 0x1E6E0074, CBR_SIZE2);
+		ast_moutdwm(ast, 0x1E6E0068, 0x00001300 | (dlli << 16) | (dlli << 24));
+		ast_moutdwm(ast, 0x1E6E0074, CBR_SIZE2);
 		data = cbr_scan(ast);
 		if (data != 0) {
 			if (data & 0x1) {
@@ -957,44 +844,31 @@ CBR_START2:
 			break;
 		}
 	}
+	if (retry++ > 10)
+		goto CBR_DONE2;
 	if (dllmax[0] == 0 || (dllmax[0]-dllmin[0]) < CBR_THRESHOLD) {
 		goto CBR_START2;
 	}
 	if (dllmax[1] == 0 || (dllmax[1]-dllmin[1]) < CBR_THRESHOLD) {
 		goto CBR_START2;
 	}
+	status = true;
+CBR_DONE2:
 	dlli  = (dllmin[1] + dllmax[1]) >> 1;
 	dlli <<= 8;
 	dlli += (dllmin[0] + dllmax[0]) >> 1;
-	moutdwm(ast, 0x1E6E0068, (mindwm(ast, 0x1E6E0068) & 0xFFFF) | (dlli << 16));
-
-	data  = (mindwm(ast, 0x1E6E0080) >> 24) & 0x1F;
-	data2 = (mindwm(ast, 0x1E6E0018) & 0xff80ffff) | (data << 16);
-	moutdwm(ast, 0x1E6E0018, data2);
-	moutdwm(ast, 0x1E6E0024, 0x8001 | (data << 1) | (param->dll2_finetune_step << 8));
-
-	/* Wait DQSI latch phase calibration */
-	moutdwm(ast, 0x1E6E0074, 0x00000010);
-	moutdwm(ast, 0x1E6E0070, 0x00000003);
-	do {
-		data = mindwm(ast, 0x1E6E0070);
-	} while (!(data & 0x00001000));
-	moutdwm(ast, 0x1E6E0070, 0x00000000);
-	moutdwm(ast, 0x1E6E0070, 0x00000003);
-	do {
-		data = mindwm(ast, 0x1E6E0070);
-	} while (!(data & 0x00001000));
-	moutdwm(ast, 0x1E6E0070, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0068, ast_mindwm(ast, 0x1E720058) | (dlli << 16));
+	return status;
 } /* CBRDLL2 */
 
 static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *param)
 {
 	u32 trap, trap_AC2, trap_MRS;
 
-	moutdwm(ast, 0x1E6E2000, 0x1688A8A8);
+	ast_moutdwm(ast, 0x1E6E2000, 0x1688A8A8);
 
 	/* Ger trap info */
-	trap = (mindwm(ast, 0x1E6E2070) >> 25) & 0x3;
+	trap = (ast_mindwm(ast, 0x1E6E2070) >> 25) & 0x3;
 	trap_AC2  = 0x00020000 + (trap << 16);
 	trap_AC2 |= 0x00300000 + ((trap & 0x2) << 19);
 	trap_MRS  = 0x00000010 + (trap << 4);
@@ -1008,22 +882,35 @@ static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *pa
 
 	switch (param->dram_freq) {
 	case 336:
-		moutdwm(ast, 0x1E6E2020, 0x0190);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0190);
 		param->wodt          = 0;
 		param->reg_AC1       = 0x22202725;
 		param->reg_AC2       = 0xAA007613 | trap_AC2;
 		param->reg_DQSIC     = 0x000000BA;
 		param->reg_MRS       = 0x04001400 | trap_MRS;
 		param->reg_EMRS      = 0x00000000;
-		param->reg_IOZ       = 0x00000034;
+		param->reg_IOZ       = 0x00000023;
 		param->reg_DQIDLY    = 0x00000074;
 		param->reg_FREQ      = 0x00004DC0;
 		param->madj_max      = 96;
 		param->dll2_finetune_step = 3;
+		switch (param->dram_chipid) {
+		default:
+		case AST_DRAM_512Mx16:
+		case AST_DRAM_1Gx16:
+			param->reg_AC2   = 0xAA007613 | trap_AC2;
+			break;
+		case AST_DRAM_2Gx16:
+			param->reg_AC2   = 0xAA00761C | trap_AC2;
+			break;
+		case AST_DRAM_4Gx16:
+			param->reg_AC2   = 0xAA007636 | trap_AC2;
+			break;
+		}
 		break;
 	default:
 	case 396:
-		moutdwm(ast, 0x1E6E2020, 0x03F1);
+		ast_moutdwm(ast, 0x1E6E2020, 0x03F1);
 		param->wodt          = 1;
 		param->reg_AC1       = 0x33302825;
 		param->reg_AC2       = 0xCC009617 | trap_AC2;
@@ -1033,7 +920,7 @@ static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->reg_IOZ       = 0x00000034;
 		param->reg_DRV       = 0x000000FA;
 		param->reg_DQIDLY    = 0x00000089;
-		param->reg_FREQ      = 0x000050C0;
+		param->reg_FREQ      = 0x00005040;
 		param->madj_max      = 96;
 		param->dll2_finetune_step = 4;
 
@@ -1053,14 +940,14 @@ static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		break;
 
 	case 408:
-		moutdwm(ast, 0x1E6E2020, 0x01F0);
+		ast_moutdwm(ast, 0x1E6E2020, 0x01F0);
 		param->wodt          = 1;
 		param->reg_AC1       = 0x33302825;
 		param->reg_AC2       = 0xCC009617 | trap_AC2;
 		param->reg_DQSIC     = 0x000000E2;
 		param->reg_MRS       = 0x04001600 | trap_MRS;
 		param->reg_EMRS      = 0x00000000;
-		param->reg_IOZ       = 0x00000034;
+		param->reg_IOZ       = 0x00000023;
 		param->reg_DRV       = 0x000000FA;
 		param->reg_DQIDLY    = 0x00000089;
 		param->reg_FREQ      = 0x000050C0;
@@ -1083,7 +970,7 @@ static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *pa
 
 		break;
 	case 456:
-		moutdwm(ast, 0x1E6E2020, 0x0230);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0230);
 		param->wodt          = 0;
 		param->reg_AC1       = 0x33302926;
 		param->reg_AC2       = 0xCD44961A;
@@ -1097,7 +984,7 @@ static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->dll2_finetune_step = 4;
 		break;
 	case 504:
-		moutdwm(ast, 0x1E6E2020, 0x0270);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0270);
 		param->wodt          = 1;
 		param->reg_AC1       = 0x33302926;
 		param->reg_AC2       = 0xDE44A61D;
@@ -1111,7 +998,7 @@ static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->dll2_finetune_step = 4;
 		break;
 	case 528:
-		moutdwm(ast, 0x1E6E2020, 0x0290);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0290);
 		param->wodt          = 1;
 		param->rodt          = 1;
 		param->reg_AC1       = 0x33302926;
@@ -1127,7 +1014,7 @@ static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->dll2_finetune_step = 3;
 		break;
 	case 576:
-		moutdwm(ast, 0x1E6E2020, 0x0140);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0140);
 		param->reg_MADJ      = 0x00136868;
 		param->reg_SADJ      = 0x00004534;
 		param->wodt          = 1;
@@ -1145,7 +1032,7 @@ static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->dll2_finetune_step = 3;
 		break;
 	case 600:
-		moutdwm(ast, 0x1E6E2020, 0x02E1);
+		ast_moutdwm(ast, 0x1E6E2020, 0x02E1);
 		param->reg_MADJ      = 0x00136868;
 		param->reg_SADJ      = 0x00004534;
 		param->wodt          = 1;
@@ -1163,7 +1050,7 @@ static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->dll2_finetune_step = 3;
 		break;
 	case 624:
-		moutdwm(ast, 0x1E6E2020, 0x0160);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0160);
 		param->reg_MADJ      = 0x00136868;
 		param->reg_SADJ      = 0x00004534;
 		param->wodt          = 1;
@@ -1196,7 +1083,7 @@ static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *pa
 	case AST_DRAM_4Gx16:
 		param->dram_config = 0x133;
 		break;
-	}; /* switch size */
+	} /* switch size */
 
 	switch (param->vram_size) {
 	default:
@@ -1218,106 +1105,98 @@ static void get_ddr3_info(struct ast_private *ast, struct ast2300_dram_param *pa
 
 static void ddr3_init(struct ast_private *ast, struct ast2300_dram_param *param)
 {
-	u32 data, data2;
+	u32 data, data2, retry = 0;
 
-	moutdwm(ast, 0x1E6E0000, 0xFC600309);
-	moutdwm(ast, 0x1E6E0018, 0x00000100);
-	moutdwm(ast, 0x1E6E0024, 0x00000000);
-	moutdwm(ast, 0x1E6E0034, 0x00000000);
+ddr3_init_start:
+	ast_moutdwm(ast, 0x1E6E0000, 0xFC600309);
+	ast_moutdwm(ast, 0x1E6E0018, 0x00000100);
+	ast_moutdwm(ast, 0x1E6E0024, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0034, 0x00000000);
 	udelay(10);
-	moutdwm(ast, 0x1E6E0064, param->reg_MADJ);
-	moutdwm(ast, 0x1E6E0068, param->reg_SADJ);
+	ast_moutdwm(ast, 0x1E6E0064, param->reg_MADJ);
+	ast_moutdwm(ast, 0x1E6E0068, param->reg_SADJ);
 	udelay(10);
-	moutdwm(ast, 0x1E6E0064, param->reg_MADJ | 0xC0000);
+	ast_moutdwm(ast, 0x1E6E0064, param->reg_MADJ | 0xC0000);
 	udelay(10);
 
-	moutdwm(ast, 0x1E6E0004, param->dram_config);
-	moutdwm(ast, 0x1E6E0008, 0x90040f);
-	moutdwm(ast, 0x1E6E0010, param->reg_AC1);
-	moutdwm(ast, 0x1E6E0014, param->reg_AC2);
-	moutdwm(ast, 0x1E6E0020, param->reg_DQSIC);
-	moutdwm(ast, 0x1E6E0080, 0x00000000);
-	moutdwm(ast, 0x1E6E0084, 0x00000000);
-	moutdwm(ast, 0x1E6E0088, param->reg_DQIDLY);
-	moutdwm(ast, 0x1E6E0018, 0x4040A170);
-	moutdwm(ast, 0x1E6E0018, 0x20402370);
-	moutdwm(ast, 0x1E6E0038, 0x00000000);
-	moutdwm(ast, 0x1E6E0040, 0xFF444444);
-	moutdwm(ast, 0x1E6E0044, 0x22222222);
-	moutdwm(ast, 0x1E6E0048, 0x22222222);
-	moutdwm(ast, 0x1E6E004C, 0x00000002);
-	moutdwm(ast, 0x1E6E0050, 0x80000000);
-	moutdwm(ast, 0x1E6E0050, 0x00000000);
-	moutdwm(ast, 0x1E6E0054, 0);
-	moutdwm(ast, 0x1E6E0060, param->reg_DRV);
-	moutdwm(ast, 0x1E6E006C, param->reg_IOZ);
-	moutdwm(ast, 0x1E6E0070, 0x00000000);
-	moutdwm(ast, 0x1E6E0074, 0x00000000);
-	moutdwm(ast, 0x1E6E0078, 0x00000000);
-	moutdwm(ast, 0x1E6E007C, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0004, param->dram_config);
+	ast_moutdwm(ast, 0x1E6E0008, 0x90040f);
+	ast_moutdwm(ast, 0x1E6E0010, param->reg_AC1);
+	ast_moutdwm(ast, 0x1E6E0014, param->reg_AC2);
+	ast_moutdwm(ast, 0x1E6E0020, param->reg_DQSIC);
+	ast_moutdwm(ast, 0x1E6E0080, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0084, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0088, param->reg_DQIDLY);
+	ast_moutdwm(ast, 0x1E6E0018, 0x4000A170);
+	ast_moutdwm(ast, 0x1E6E0018, 0x00002370);
+	ast_moutdwm(ast, 0x1E6E0038, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0040, 0xFF444444);
+	ast_moutdwm(ast, 0x1E6E0044, 0x22222222);
+	ast_moutdwm(ast, 0x1E6E0048, 0x22222222);
+	ast_moutdwm(ast, 0x1E6E004C, 0x00000002);
+	ast_moutdwm(ast, 0x1E6E0050, 0x80000000);
+	ast_moutdwm(ast, 0x1E6E0050, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0054, 0);
+	ast_moutdwm(ast, 0x1E6E0060, param->reg_DRV);
+	ast_moutdwm(ast, 0x1E6E006C, param->reg_IOZ);
+	ast_moutdwm(ast, 0x1E6E0070, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0074, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0078, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E007C, 0x00000000);
 	/* Wait MCLK2X lock to MCLK */
 	do {
-		data = mindwm(ast, 0x1E6E001C);
+		data = ast_mindwm(ast, 0x1E6E001C);
 	} while (!(data & 0x08000000));
-	moutdwm(ast, 0x1E6E0034, 0x00000001);
-	moutdwm(ast, 0x1E6E000C, 0x00005C04);
-	udelay(10);
-	moutdwm(ast, 0x1E6E000C, 0x00000000);
-	moutdwm(ast, 0x1E6E0034, 0x00000000);
-	data = mindwm(ast, 0x1E6E001C);
+	data = ast_mindwm(ast, 0x1E6E001C);
 	data = (data >> 8) & 0xff;
 	while ((data & 0x08) || ((data & 0x7) < 2) || (data < 4)) {
-		data2 = (mindwm(ast, 0x1E6E0064) & 0xfff3ffff) + 4;
+		data2 = (ast_mindwm(ast, 0x1E6E0064) & 0xfff3ffff) + 4;
 		if ((data2 & 0xff) > param->madj_max) {
 			break;
 		}
-		moutdwm(ast, 0x1E6E0064, data2);
+		ast_moutdwm(ast, 0x1E6E0064, data2);
 		if (data2 & 0x00100000) {
 			data2 = ((data2 & 0xff) >> 3) + 3;
 		} else {
 			data2 = ((data2 & 0xff) >> 2) + 5;
 		}
-		data = mindwm(ast, 0x1E6E0068) & 0xffff00ff;
+		data = ast_mindwm(ast, 0x1E6E0068) & 0xffff00ff;
 		data2 += data & 0xff;
 		data = data | (data2 << 8);
-		moutdwm(ast, 0x1E6E0068, data);
+		ast_moutdwm(ast, 0x1E6E0068, data);
 		udelay(10);
-		moutdwm(ast, 0x1E6E0064, mindwm(ast, 0x1E6E0064) | 0xC0000);
+		ast_moutdwm(ast, 0x1E6E0064, ast_mindwm(ast, 0x1E6E0064) | 0xC0000);
 		udelay(10);
-		data = mindwm(ast, 0x1E6E0018) & 0xfffff1ff;
-		moutdwm(ast, 0x1E6E0018, data);
+		data = ast_mindwm(ast, 0x1E6E0018) & 0xfffff1ff;
+		ast_moutdwm(ast, 0x1E6E0018, data);
 		data = data | 0x200;
-		moutdwm(ast, 0x1E6E0018, data);
+		ast_moutdwm(ast, 0x1E6E0018, data);
 		do {
-			data = mindwm(ast, 0x1E6E001C);
+			data = ast_mindwm(ast, 0x1E6E001C);
 		} while (!(data & 0x08000000));
 
-		moutdwm(ast, 0x1E6E0034, 0x00000001);
-		moutdwm(ast, 0x1E6E000C, 0x00005C04);
-		udelay(10);
-		moutdwm(ast, 0x1E6E000C, 0x00000000);
-		moutdwm(ast, 0x1E6E0034, 0x00000000);
-		data = mindwm(ast, 0x1E6E001C);
+		data = ast_mindwm(ast, 0x1E6E001C);
 		data = (data >> 8) & 0xff;
 	}
-	data = mindwm(ast, 0x1E6E0018) | 0xC00;
-	moutdwm(ast, 0x1E6E0018, data);
+	ast_moutdwm(ast, 0x1E720058, ast_mindwm(ast, 0x1E6E0068) & 0xffff);
+	data = ast_mindwm(ast, 0x1E6E0018) | 0xC00;
+	ast_moutdwm(ast, 0x1E6E0018, data);
 
-	moutdwm(ast, 0x1E6E0034, 0x00000001);
-	moutdwm(ast, 0x1E6E000C, 0x00000040);
+	ast_moutdwm(ast, 0x1E6E0034, 0x00000001);
+	ast_moutdwm(ast, 0x1E6E000C, 0x00000040);
 	udelay(50);
 	/* Mode Register Setting */
-	moutdwm(ast, 0x1E6E002C, param->reg_MRS | 0x100);
-	moutdwm(ast, 0x1E6E0030, param->reg_EMRS);
-	moutdwm(ast, 0x1E6E0028, 0x00000005);
-	moutdwm(ast, 0x1E6E0028, 0x00000007);
-	moutdwm(ast, 0x1E6E0028, 0x00000003);
-	moutdwm(ast, 0x1E6E0028, 0x00000001);
-	moutdwm(ast, 0x1E6E002C, param->reg_MRS);
-	moutdwm(ast, 0x1E6E000C, 0x00005C08);
-	moutdwm(ast, 0x1E6E0028, 0x00000001);
-
-	moutdwm(ast, 0x1E6E000C, 0x7FFF5C01);
+	ast_moutdwm(ast, 0x1E6E002C, param->reg_MRS | 0x100);
+	ast_moutdwm(ast, 0x1E6E0030, param->reg_EMRS);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000005);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000007);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000003);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000001);
+	ast_moutdwm(ast, 0x1E6E002C, param->reg_MRS);
+	ast_moutdwm(ast, 0x1E6E000C, 0x00005C08);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000001);
+
+	ast_moutdwm(ast, 0x1E6E000C, 0x00005C01);
 	data = 0;
 	if (param->wodt) {
 		data = 0x300;
@@ -1325,30 +1204,23 @@ static void ddr3_init(struct ast_private *ast, struct ast2300_dram_param *param)
 	if (param->rodt) {
 		data = data | 0x3000 | ((param->reg_AC2 & 0x60000) >> 3);
 	}
-	moutdwm(ast, 0x1E6E0034, data | 0x3);
+	ast_moutdwm(ast, 0x1E6E0034, data | 0x3);
 
-	/* Wait DQI delay lock */
-	do {
-		data = mindwm(ast, 0x1E6E0080);
-	} while (!(data & 0x40000000));
-	/* Wait DQSI delay lock */
-	do {
-		data = mindwm(ast, 0x1E6E0020);
-	} while (!(data & 0x00000800));
 	/* Calibrate the DQSI delay */
-	cbr_dll2(ast, param);
+	if ((cbr_dll2(ast, param) == false) && (retry++ < 10))
+		goto ddr3_init_start;
 
-	moutdwm(ast, 0x1E6E0120, param->reg_FREQ);
+	ast_moutdwm(ast, 0x1E6E0120, param->reg_FREQ);
 	/* ECC Memory Initialization */
 #ifdef ECC
-	moutdwm(ast, 0x1E6E007C, 0x00000000);
-	moutdwm(ast, 0x1E6E0070, 0x221);
+	ast_moutdwm(ast, 0x1E6E007C, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0070, 0x221);
 	do {
-		data = mindwm(ast, 0x1E6E0070);
+		data = ast_mindwm(ast, 0x1E6E0070);
 	} while (!(data & 0x00001000));
-	moutdwm(ast, 0x1E6E0070, 0x00000000);
-	moutdwm(ast, 0x1E6E0050, 0x80000000);
-	moutdwm(ast, 0x1E6E0050, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0070, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0050, 0x80000000);
+	ast_moutdwm(ast, 0x1E6E0050, 0x00000000);
 #endif
 
 
@@ -1358,10 +1230,10 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 {
 	u32 trap, trap_AC2, trap_MRS;
 
-	moutdwm(ast, 0x1E6E2000, 0x1688A8A8);
+	ast_moutdwm(ast, 0x1E6E2000, 0x1688A8A8);
 
 	/* Ger trap info */
-	trap = (mindwm(ast, 0x1E6E2070) >> 25) & 0x3;
+	trap = (ast_mindwm(ast, 0x1E6E2070) >> 25) & 0x3;
 	trap_AC2  = (trap << 20) | (trap << 16);
 	trap_AC2 += 0x00110000;
 	trap_MRS  = 0x00000040 | (trap << 4);
@@ -1375,7 +1247,7 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 
 	switch (param->dram_freq) {
 	case 264:
-		moutdwm(ast, 0x1E6E2020, 0x0130);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0130);
 		param->wodt          = 0;
 		param->reg_AC1       = 0x11101513;
 		param->reg_AC2       = 0x78117011;
@@ -1390,7 +1262,7 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->dll2_finetune_step = 3;
 		break;
 	case 336:
-		moutdwm(ast, 0x1E6E2020, 0x0190);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0190);
 		param->wodt          = 1;
 		param->reg_AC1       = 0x22202613;
 		param->reg_AC2       = 0xAA009016 | trap_AC2;
@@ -1403,10 +1275,25 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->reg_FREQ      = 0x00004DC0;
 		param->madj_max      = 96;
 		param->dll2_finetune_step = 3;
+		switch (param->dram_chipid) {
+		default:
+		case AST_DRAM_512Mx16:
+			param->reg_AC2   = 0xAA009012 | trap_AC2;
+			break;
+		case AST_DRAM_1Gx16:
+			param->reg_AC2   = 0xAA009016 | trap_AC2;
+			break;
+		case AST_DRAM_2Gx16:
+			param->reg_AC2   = 0xAA009023 | trap_AC2;
+			break;
+		case AST_DRAM_4Gx16:
+			param->reg_AC2   = 0xAA00903B | trap_AC2;
+			break;
+		}
 		break;
 	default:
 	case 396:
-		moutdwm(ast, 0x1E6E2020, 0x03F1);
+		ast_moutdwm(ast, 0x1E6E2020, 0x03F1);
 		param->wodt          = 1;
 		param->rodt          = 0;
 		param->reg_AC1       = 0x33302714;
@@ -1417,7 +1304,7 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->reg_DRV       = 0x000000FA;
 		param->reg_IOZ       = 0x00000034;
 		param->reg_DQIDLY    = 0x00000089;
-		param->reg_FREQ      = 0x000050C0;
+		param->reg_FREQ      = 0x00005040;
 		param->madj_max      = 96;
 		param->dll2_finetune_step = 4;
 
@@ -1440,7 +1327,7 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		break;
 
 	case 408:
-		moutdwm(ast, 0x1E6E2020, 0x01F0);
+		ast_moutdwm(ast, 0x1E6E2020, 0x01F0);
 		param->wodt          = 1;
 		param->rodt          = 0;
 		param->reg_AC1       = 0x33302714;
@@ -1473,7 +1360,7 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 
 		break;
 	case 456:
-		moutdwm(ast, 0x1E6E2020, 0x0230);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0230);
 		param->wodt          = 0;
 		param->reg_AC1       = 0x33302815;
 		param->reg_AC2       = 0xCD44B01E;
@@ -1488,7 +1375,7 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->dll2_finetune_step = 3;
 		break;
 	case 504:
-		moutdwm(ast, 0x1E6E2020, 0x0261);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0261);
 		param->wodt          = 1;
 		param->rodt          = 1;
 		param->reg_AC1       = 0x33302815;
@@ -1504,7 +1391,7 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->dll2_finetune_step = 3;
 		break;
 	case 528:
-		moutdwm(ast, 0x1E6E2020, 0x0120);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0120);
 		param->wodt          = 1;
 		param->rodt          = 1;
 		param->reg_AC1       = 0x33302815;
@@ -1520,7 +1407,7 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->dll2_finetune_step = 3;
 		break;
 	case 552:
-		moutdwm(ast, 0x1E6E2020, 0x02A1);
+		ast_moutdwm(ast, 0x1E6E2020, 0x02A1);
 		param->wodt          = 1;
 		param->rodt          = 1;
 		param->reg_AC1       = 0x43402915;
@@ -1536,7 +1423,7 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 		param->dll2_finetune_step = 3;
 		break;
 	case 576:
-		moutdwm(ast, 0x1E6E2020, 0x0140);
+		ast_moutdwm(ast, 0x1E6E2020, 0x0140);
 		param->wodt          = 1;
 		param->rodt          = 1;
 		param->reg_AC1       = 0x43402915;
@@ -1567,7 +1454,7 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 	case AST_DRAM_4Gx16:
 		param->dram_config = 0x123;
 		break;
-	}; /* switch size */
+	} /* switch size */
 
 	switch (param->vram_size) {
 	default:
@@ -1588,110 +1475,102 @@ static void get_ddr2_info(struct ast_private *ast, struct ast2300_dram_param *pa
 
 static void ddr2_init(struct ast_private *ast, struct ast2300_dram_param *param)
 {
-	u32 data, data2;
-
-	moutdwm(ast, 0x1E6E0000, 0xFC600309);
-	moutdwm(ast, 0x1E6E0018, 0x00000100);
-	moutdwm(ast, 0x1E6E0024, 0x00000000);
-	moutdwm(ast, 0x1E6E0064, param->reg_MADJ);
-	moutdwm(ast, 0x1E6E0068, param->reg_SADJ);
+	u32 data, data2, retry = 0;
+
+ddr2_init_start:
+	ast_moutdwm(ast, 0x1E6E0000, 0xFC600309);
+	ast_moutdwm(ast, 0x1E6E0018, 0x00000100);
+	ast_moutdwm(ast, 0x1E6E0024, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0064, param->reg_MADJ);
+	ast_moutdwm(ast, 0x1E6E0068, param->reg_SADJ);
 	udelay(10);
-	moutdwm(ast, 0x1E6E0064, param->reg_MADJ | 0xC0000);
+	ast_moutdwm(ast, 0x1E6E0064, param->reg_MADJ | 0xC0000);
 	udelay(10);
 
-	moutdwm(ast, 0x1E6E0004, param->dram_config);
-	moutdwm(ast, 0x1E6E0008, 0x90040f);
-	moutdwm(ast, 0x1E6E0010, param->reg_AC1);
-	moutdwm(ast, 0x1E6E0014, param->reg_AC2);
-	moutdwm(ast, 0x1E6E0020, param->reg_DQSIC);
-	moutdwm(ast, 0x1E6E0080, 0x00000000);
-	moutdwm(ast, 0x1E6E0084, 0x00000000);
-	moutdwm(ast, 0x1E6E0088, param->reg_DQIDLY);
-	moutdwm(ast, 0x1E6E0018, 0x4040A130);
-	moutdwm(ast, 0x1E6E0018, 0x20402330);
-	moutdwm(ast, 0x1E6E0038, 0x00000000);
-	moutdwm(ast, 0x1E6E0040, 0xFF808000);
-	moutdwm(ast, 0x1E6E0044, 0x88848466);
-	moutdwm(ast, 0x1E6E0048, 0x44440008);
-	moutdwm(ast, 0x1E6E004C, 0x00000000);
-	moutdwm(ast, 0x1E6E0050, 0x80000000);
-	moutdwm(ast, 0x1E6E0050, 0x00000000);
-	moutdwm(ast, 0x1E6E0054, 0);
-	moutdwm(ast, 0x1E6E0060, param->reg_DRV);
-	moutdwm(ast, 0x1E6E006C, param->reg_IOZ);
-	moutdwm(ast, 0x1E6E0070, 0x00000000);
-	moutdwm(ast, 0x1E6E0074, 0x00000000);
-	moutdwm(ast, 0x1E6E0078, 0x00000000);
-	moutdwm(ast, 0x1E6E007C, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0004, param->dram_config);
+	ast_moutdwm(ast, 0x1E6E0008, 0x90040f);
+	ast_moutdwm(ast, 0x1E6E0010, param->reg_AC1);
+	ast_moutdwm(ast, 0x1E6E0014, param->reg_AC2);
+	ast_moutdwm(ast, 0x1E6E0020, param->reg_DQSIC);
+	ast_moutdwm(ast, 0x1E6E0080, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0084, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0088, param->reg_DQIDLY);
+	ast_moutdwm(ast, 0x1E6E0018, 0x4000A130);
+	ast_moutdwm(ast, 0x1E6E0018, 0x00002330);
+	ast_moutdwm(ast, 0x1E6E0038, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0040, 0xFF808000);
+	ast_moutdwm(ast, 0x1E6E0044, 0x88848466);
+	ast_moutdwm(ast, 0x1E6E0048, 0x44440008);
+	ast_moutdwm(ast, 0x1E6E004C, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0050, 0x80000000);
+	ast_moutdwm(ast, 0x1E6E0050, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0054, 0);
+	ast_moutdwm(ast, 0x1E6E0060, param->reg_DRV);
+	ast_moutdwm(ast, 0x1E6E006C, param->reg_IOZ);
+	ast_moutdwm(ast, 0x1E6E0070, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0074, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0078, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E007C, 0x00000000);
 
 	/* Wait MCLK2X lock to MCLK */
 	do {
-		data = mindwm(ast, 0x1E6E001C);
+		data = ast_mindwm(ast, 0x1E6E001C);
 	} while (!(data & 0x08000000));
-	moutdwm(ast, 0x1E6E0034, 0x00000001);
-	moutdwm(ast, 0x1E6E000C, 0x00005C04);
-	udelay(10);
-	moutdwm(ast, 0x1E6E000C, 0x00000000);
-	moutdwm(ast, 0x1E6E0034, 0x00000000);
-	data = mindwm(ast, 0x1E6E001C);
+	data = ast_mindwm(ast, 0x1E6E001C);
 	data = (data >> 8) & 0xff;
 	while ((data & 0x08) || ((data & 0x7) < 2) || (data < 4)) {
-		data2 = (mindwm(ast, 0x1E6E0064) & 0xfff3ffff) + 4;
+		data2 = (ast_mindwm(ast, 0x1E6E0064) & 0xfff3ffff) + 4;
 		if ((data2 & 0xff) > param->madj_max) {
 			break;
 		}
-		moutdwm(ast, 0x1E6E0064, data2);
+		ast_moutdwm(ast, 0x1E6E0064, data2);
 		if (data2 & 0x00100000) {
 			data2 = ((data2 & 0xff) >> 3) + 3;
 		} else {
 			data2 = ((data2 & 0xff) >> 2) + 5;
 		}
-		data = mindwm(ast, 0x1E6E0068) & 0xffff00ff;
+		data = ast_mindwm(ast, 0x1E6E0068) & 0xffff00ff;
 		data2 += data & 0xff;
 		data = data | (data2 << 8);
-		moutdwm(ast, 0x1E6E0068, data);
+		ast_moutdwm(ast, 0x1E6E0068, data);
 		udelay(10);
-		moutdwm(ast, 0x1E6E0064, mindwm(ast, 0x1E6E0064) | 0xC0000);
+		ast_moutdwm(ast, 0x1E6E0064, ast_mindwm(ast, 0x1E6E0064) | 0xC0000);
 		udelay(10);
-		data = mindwm(ast, 0x1E6E0018) & 0xfffff1ff;
-		moutdwm(ast, 0x1E6E0018, data);
+		data = ast_mindwm(ast, 0x1E6E0018) & 0xfffff1ff;
+		ast_moutdwm(ast, 0x1E6E0018, data);
 		data = data | 0x200;
-		moutdwm(ast, 0x1E6E0018, data);
+		ast_moutdwm(ast, 0x1E6E0018, data);
 		do {
-			data = mindwm(ast, 0x1E6E001C);
+			data = ast_mindwm(ast, 0x1E6E001C);
 		} while (!(data & 0x08000000));
 
-		moutdwm(ast, 0x1E6E0034, 0x00000001);
-		moutdwm(ast, 0x1E6E000C, 0x00005C04);
-		udelay(10);
-		moutdwm(ast, 0x1E6E000C, 0x00000000);
-		moutdwm(ast, 0x1E6E0034, 0x00000000);
-		data = mindwm(ast, 0x1E6E001C);
+		data = ast_mindwm(ast, 0x1E6E001C);
 		data = (data >> 8) & 0xff;
 	}
-	data = mindwm(ast, 0x1E6E0018) | 0xC00;
-	moutdwm(ast, 0x1E6E0018, data);
+	ast_moutdwm(ast, 0x1E720058, ast_mindwm(ast, 0x1E6E0008) & 0xffff);
+	data = ast_mindwm(ast, 0x1E6E0018) | 0xC00;
+	ast_moutdwm(ast, 0x1E6E0018, data);
 
-	moutdwm(ast, 0x1E6E0034, 0x00000001);
-	moutdwm(ast, 0x1E6E000C, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0034, 0x00000001);
+	ast_moutdwm(ast, 0x1E6E000C, 0x00000000);
 	udelay(50);
 	/* Mode Register Setting */
-	moutdwm(ast, 0x1E6E002C, param->reg_MRS | 0x100);
-	moutdwm(ast, 0x1E6E0030, param->reg_EMRS);
-	moutdwm(ast, 0x1E6E0028, 0x00000005);
-	moutdwm(ast, 0x1E6E0028, 0x00000007);
-	moutdwm(ast, 0x1E6E0028, 0x00000003);
-	moutdwm(ast, 0x1E6E0028, 0x00000001);
-
-	moutdwm(ast, 0x1E6E000C, 0x00005C08);
-	moutdwm(ast, 0x1E6E002C, param->reg_MRS);
-	moutdwm(ast, 0x1E6E0028, 0x00000001);
-	moutdwm(ast, 0x1E6E0030, param->reg_EMRS | 0x380);
-	moutdwm(ast, 0x1E6E0028, 0x00000003);
-	moutdwm(ast, 0x1E6E0030, param->reg_EMRS);
-	moutdwm(ast, 0x1E6E0028, 0x00000003);
-
-	moutdwm(ast, 0x1E6E000C, 0x7FFF5C01);
+	ast_moutdwm(ast, 0x1E6E002C, param->reg_MRS | 0x100);
+	ast_moutdwm(ast, 0x1E6E0030, param->reg_EMRS);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000005);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000007);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000003);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000001);
+
+	ast_moutdwm(ast, 0x1E6E000C, 0x00005C08);
+	ast_moutdwm(ast, 0x1E6E002C, param->reg_MRS);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000001);
+	ast_moutdwm(ast, 0x1E6E0030, param->reg_EMRS | 0x380);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000003);
+	ast_moutdwm(ast, 0x1E6E0030, param->reg_EMRS);
+	ast_moutdwm(ast, 0x1E6E0028, 0x00000003);
+
+	ast_moutdwm(ast, 0x1E6E000C, 0x7FFF5C01);
 	data = 0;
 	if (param->wodt) {
 		data = 0x500;
@@ -1699,30 +1578,23 @@ static void ddr2_init(struct ast_private *ast, struct ast2300_dram_param *param)
 	if (param->rodt) {
 		data = data | 0x3000 | ((param->reg_AC2 & 0x60000) >> 3);
 	}
-	moutdwm(ast, 0x1E6E0034, data | 0x3);
-	moutdwm(ast, 0x1E6E0120, param->reg_FREQ);
+	ast_moutdwm(ast, 0x1E6E0034, data | 0x3);
+	ast_moutdwm(ast, 0x1E6E0120, param->reg_FREQ);
 
-	/* Wait DQI delay lock */
-	do {
-		data = mindwm(ast, 0x1E6E0080);
-	} while (!(data & 0x40000000));
-	/* Wait DQSI delay lock */
-	do {
-		data = mindwm(ast, 0x1E6E0020);
-	} while (!(data & 0x00000800));
 	/* Calibrate the DQSI delay */
-	cbr_dll2(ast, param);
+	if ((cbr_dll2(ast, param) == false) && (retry++ < 10))
+		goto ddr2_init_start;
 
 	/* ECC Memory Initialization */
 #ifdef ECC
-	moutdwm(ast, 0x1E6E007C, 0x00000000);
-	moutdwm(ast, 0x1E6E0070, 0x221);
+	ast_moutdwm(ast, 0x1E6E007C, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0070, 0x221);
 	do {
-		data = mindwm(ast, 0x1E6E0070);
+		data = ast_mindwm(ast, 0x1E6E0070);
 	} while (!(data & 0x00001000));
-	moutdwm(ast, 0x1E6E0070, 0x00000000);
-	moutdwm(ast, 0x1E6E0050, 0x80000000);
-	moutdwm(ast, 0x1E6E0050, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0070, 0x00000000);
+	ast_moutdwm(ast, 0x1E6E0050, 0x80000000);
+	ast_moutdwm(ast, 0x1E6E0050, 0x00000000);
 #endif
 
 }
@@ -1768,8 +1640,8 @@ static void ast_init_dram_2300(struct drm_device *dev)
 			ddr2_init(ast, &param);
 		}
 
-		temp = mindwm(ast, 0x1e6e2040);
-		moutdwm(ast, 0x1e6e2040, temp | 0x40);
+		temp = ast_mindwm(ast, 0x1e6e2040);
+		ast_moutdwm(ast, 0x1e6e2040, temp | 0x40);
 	}
 
 	/* wait ready */
diff --git a/drivers/gpu/drm/ast/ast_tables.h b/drivers/gpu/drm/ast/ast_tables.h
index 95fa6aba26bc..4c761dcea972 100644
--- a/drivers/gpu/drm/ast/ast_tables.h
+++ b/drivers/gpu/drm/ast/ast_tables.h
@@ -42,7 +42,7 @@
 #define HBorder                 0x00000020
 #define VBorder                 0x00000010
 #define WideScreenMode		0x00000100
-
+#define NewModeInfo		0x00000200
 
 /* DCLK Index */
 #define VCLK25_175     		0x00
@@ -67,6 +67,11 @@
 #define VCLK106_5   		0x12
 #define VCLK146_25  		0x13
 #define VCLK148_5   		0x14
+#define VCLK71      		0x15
+#define VCLK88_75   		0x16
+#define VCLK119     		0x17
+#define VCLK85_5     		0x18
+#define VCLK97_75     		0x19
 
 static struct ast_vbios_dclk_info dclk_table[] = {
 	{0x2C, 0xE7, 0x03},					/* 00: VCLK25_175	*/
@@ -90,6 +95,10 @@ static struct ast_vbios_dclk_info dclk_table[] = {
 	{0x28, 0x49, 0x80},					/* 12: VCLK106.5        */
 	{0x37, 0x49, 0x80},					/* 13: VCLK146.25       */
 	{0x1f, 0x45, 0x80},					/* 14: VCLK148.5        */
+	{0x47, 0x6c, 0x80},					/* 15: VCLK71       */
+	{0x25, 0x65, 0x80},					/* 16: VCLK88.75    */
+	{0x77, 0x58, 0x80},					/* 17: VCLK119      */
+	{0x32, 0x67, 0x80},				    /* 18: VCLK85_5     */
 };
 
 static struct ast_vbios_stdtable vbios_stdtable[] = {
@@ -225,41 +234,63 @@ static struct ast_vbios_enhtable res_1600x1200[] = {
 	 (SyncPP | Charx8Dot), 0xFF, 1, 0x33 },
 };
 
-static struct ast_vbios_enhtable res_1920x1200[] = {
-	{2080, 1920, 48, 32, 1235, 1200, 3, 6, VCLK154,	/* 60Hz */
-	 (SyncNP | Charx8Dot), 60, 1, 0x34 },
-	{2080, 1920, 48, 32, 1235, 1200, 3, 6, VCLK154,	/* 60Hz */
-	 (SyncNP | Charx8Dot), 0xFF, 1, 0x34 },
+/* 16:9 */
+static struct ast_vbios_enhtable res_1360x768[] = {
+	{1792, 1360, 64,112, 795,  768, 3, 6, VCLK85_5,	         /* 60Hz */
+	 (SyncPP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 0x39 },
+	{1792, 1360, 64,112, 795,  768, 3, 6, VCLK85_5,	         /* end */
+	 (SyncPP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 0xFF, 1, 0x39 },
+};
+
+static struct ast_vbios_enhtable res_1600x900[] = {
+	{1760, 1600, 48, 32, 926,  900, 3, 5, VCLK97_75,	/* 60Hz CVT RB */
+	 (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 0x3A },
+	{1760, 1600, 48, 32, 926,  900, 3, 5, VCLK97_75,	/* end */
+	 (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 0xFF, 1, 0x3A }
 };
 
+static struct ast_vbios_enhtable res_1920x1080[] = {
+	{2200, 1920, 88, 44, 1125, 1080, 4, 5, VCLK148_5,	/* 60Hz */
+	 (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 0x38 },
+	{2200, 1920, 88, 44, 1125, 1080, 4, 5, VCLK148_5,	/* 60Hz */
+	 (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 0xFF, 1, 0x38 },
+};
+
+
 /* 16:10 */
 static struct ast_vbios_enhtable res_1280x800[] = {
+	{1440, 1280, 48, 32,  823,  800, 3, 6, VCLK71,	/* 60Hz RB */
+	 (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 35 },
 	{1680, 1280, 72,128,  831,  800, 3, 6, VCLK83_5,	/* 60Hz */
-	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode), 60, 1, 0x35 },
+	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 0x35 },
 	{1680, 1280, 72,128,  831,  800, 3, 6, VCLK83_5,	/* 60Hz */
-	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode), 0xFF, 1, 0x35 },
+	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 0xFF, 1, 0x35 },
 
 };
 
 static struct ast_vbios_enhtable res_1440x900[] = {
+	{1600, 1440, 48, 32,  926,  900, 3, 6, VCLK88_75,	/* 60Hz RB */
+	 (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 0x36 },
 	{1904, 1440, 80,152,  934,  900, 3, 6, VCLK106_5,	/* 60Hz */
-	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode), 60, 1, 0x36 },
+	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 0x36 },
 	{1904, 1440, 80,152,  934,  900, 3, 6, VCLK106_5,	/* 60Hz */
-	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode), 0xFF, 1, 0x36 },
+	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 0xFF, 1, 0x36 },
 };
 
 static struct ast_vbios_enhtable res_1680x1050[] = {
+	{1840, 1680, 48, 32, 1080, 1050, 3, 6, VCLK119,	/* 60Hz RB */
+	 (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 0x37 },
 	{2240, 1680,104,176, 1089, 1050, 3, 6, VCLK146_25,	/* 60Hz */
-	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode), 60, 1, 0x37 },
+	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 0x37 },
 	{2240, 1680,104,176, 1089, 1050, 3, 6, VCLK146_25,	/* 60Hz */
-	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode), 0xFF, 1, 0x37 },
+	 (SyncPN | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 0xFF, 1, 0x37 },
 };
 
-/* HDTV */
-static struct ast_vbios_enhtable res_1920x1080[] = {
-	{2200, 1920, 88, 44, 1125, 1080, 4, 5, VCLK148_5,	/* 60Hz */
-	 (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode), 60, 1, 0x38 },
-	{2200, 1920, 88, 44, 1125, 1080, 4, 5, VCLK148_5,	/* 60Hz */
-	 (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode), 0xFF, 1, 0x38 },
+static struct ast_vbios_enhtable res_1920x1200[] = {
+	{2080, 1920, 48, 32, 1235, 1200, 3, 6, VCLK154,	/* 60Hz */
+	 (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 60, 1, 0x34 },
+	{2080, 1920, 48, 32, 1235, 1200, 3, 6, VCLK154,	/* 60Hz */
+	 (SyncNP | Charx8Dot | LineCompareOff | WideScreenMode | NewModeInfo), 0xFF, 1, 0x34 },
 };
+
 #endif
diff --git a/drivers/gpu/drm/bochs/bochs_mm.c b/drivers/gpu/drm/bochs/bochs_mm.c
index f488be55d650..b9a695d92792 100644
--- a/drivers/gpu/drm/bochs/bochs_mm.c
+++ b/drivers/gpu/drm/bochs/bochs_mm.c
@@ -434,17 +434,13 @@ static void bochs_bo_unref(struct bochs_bo **bo)
 
 	tbo = &((*bo)->bo);
 	ttm_bo_unref(&tbo);
-	if (tbo == NULL)
-		*bo = NULL;
-
+	*bo = NULL;
 }
 
 void bochs_gem_free_object(struct drm_gem_object *obj)
 {
 	struct bochs_bo *bochs_bo = gem_to_bochs_bo(obj);
 
-	if (!bochs_bo)
-		return;
 	bochs_bo_unref(&bochs_bo);
 }
 
diff --git a/drivers/gpu/drm/bridge/ptn3460.c b/drivers/gpu/drm/bridge/ptn3460.c
index b171901a3553..98fd17ae4916 100644
--- a/drivers/gpu/drm/bridge/ptn3460.c
+++ b/drivers/gpu/drm/bridge/ptn3460.c
@@ -225,12 +225,6 @@ out:
 	return num_modes;
 }
 
-static int ptn3460_mode_valid(struct drm_connector *connector,
-		struct drm_display_mode *mode)
-{
-	return MODE_OK;
-}
-
 struct drm_encoder *ptn3460_best_encoder(struct drm_connector *connector)
 {
 	struct ptn3460_bridge *ptn_bridge;
@@ -242,7 +236,6 @@ struct drm_encoder *ptn3460_best_encoder(struct drm_connector *connector)
 
 struct drm_connector_helper_funcs ptn3460_connector_helper_funcs = {
 	.get_modes = ptn3460_get_modes,
-	.mode_valid = ptn3460_mode_valid,
 	.best_encoder = ptn3460_best_encoder,
 };
 
diff --git a/drivers/gpu/drm/cirrus/cirrus_main.c b/drivers/gpu/drm/cirrus/cirrus_main.c
index 4b0170cf53fd..99c1983f99d2 100644
--- a/drivers/gpu/drm/cirrus/cirrus_main.c
+++ b/drivers/gpu/drm/cirrus/cirrus_main.c
@@ -264,17 +264,13 @@ static void cirrus_bo_unref(struct cirrus_bo **bo)
 
 	tbo = &((*bo)->bo);
 	ttm_bo_unref(&tbo);
-	if (tbo == NULL)
-		*bo = NULL;
-
+	*bo = NULL;
 }
 
 void cirrus_gem_free_object(struct drm_gem_object *obj)
 {
 	struct cirrus_bo *cirrus_bo = gem_to_cirrus_bo(obj);
 
-	if (!cirrus_bo)
-		return;
 	cirrus_bo_unref(&cirrus_bo);
 }
 
diff --git a/drivers/gpu/drm/cirrus/cirrus_mode.c b/drivers/gpu/drm/cirrus/cirrus_mode.c
index f59433b7610c..49332c5fe35b 100644
--- a/drivers/gpu/drm/cirrus/cirrus_mode.c
+++ b/drivers/gpu/drm/cirrus/cirrus_mode.c
@@ -505,13 +505,6 @@ static int cirrus_vga_get_modes(struct drm_connector *connector)
 	return count;
 }
 
-static int cirrus_vga_mode_valid(struct drm_connector *connector,
-				 struct drm_display_mode *mode)
-{
-	/* Any mode we've added is valid */
-	return MODE_OK;
-}
-
 static struct drm_encoder *cirrus_connector_best_encoder(struct drm_connector
 						  *connector)
 {
@@ -546,7 +539,6 @@ static void cirrus_connector_destroy(struct drm_connector *connector)
 
 struct drm_connector_helper_funcs cirrus_vga_connector_helper_funcs = {
 	.get_modes = cirrus_vga_get_modes,
-	.mode_valid = cirrus_vga_mode_valid,
 	.best_encoder = cirrus_connector_best_encoder,
 };
 
diff --git a/drivers/gpu/drm/drm_bufs.c b/drivers/gpu/drm/drm_bufs.c
index edec31fe3fed..68175b54504b 100644
--- a/drivers/gpu/drm/drm_bufs.c
+++ b/drivers/gpu/drm/drm_bufs.c
@@ -363,7 +363,7 @@ static int drm_addmap_core(struct drm_device * dev, resource_size_t offset,
 		list->master = dev->primary->master;
 	*maplist = list;
 	return 0;
-	}
+}
 
 int drm_addmap(struct drm_device * dev, resource_size_t offset,
 	       unsigned int size, enum drm_map_type type,
@@ -656,13 +656,13 @@ int drm_addbufs_agp(struct drm_device * dev, struct drm_buf_desc * request)
 		DRM_DEBUG("zone invalid\n");
 		return -EINVAL;
 	}
-	spin_lock(&dev->count_lock);
+	spin_lock(&dev->buf_lock);
 	if (dev->buf_use) {
-		spin_unlock(&dev->count_lock);
+		spin_unlock(&dev->buf_lock);
 		return -EBUSY;
 	}
 	atomic_inc(&dev->buf_alloc);
-	spin_unlock(&dev->count_lock);
+	spin_unlock(&dev->buf_lock);
 
 	mutex_lock(&dev->struct_mutex);
 	entry = &dma->bufs[order];
@@ -805,13 +805,13 @@ int drm_addbufs_pci(struct drm_device * dev, struct drm_buf_desc * request)
 	page_order = order - PAGE_SHIFT > 0 ? order - PAGE_SHIFT : 0;
 	total = PAGE_SIZE << page_order;
 
-	spin_lock(&dev->count_lock);
+	spin_lock(&dev->buf_lock);
 	if (dev->buf_use) {
-		spin_unlock(&dev->count_lock);
+		spin_unlock(&dev->buf_lock);
 		return -EBUSY;
 	}
 	atomic_inc(&dev->buf_alloc);
-	spin_unlock(&dev->count_lock);
+	spin_unlock(&dev->buf_lock);
 
 	mutex_lock(&dev->struct_mutex);
 	entry = &dma->bufs[order];
@@ -1015,13 +1015,13 @@ static int drm_addbufs_sg(struct drm_device * dev, struct drm_buf_desc * request
 	if (order < DRM_MIN_ORDER || order > DRM_MAX_ORDER)
 		return -EINVAL;
 
-	spin_lock(&dev->count_lock);
+	spin_lock(&dev->buf_lock);
 	if (dev->buf_use) {
-		spin_unlock(&dev->count_lock);
+		spin_unlock(&dev->buf_lock);
 		return -EBUSY;
 	}
 	atomic_inc(&dev->buf_alloc);
-	spin_unlock(&dev->count_lock);
+	spin_unlock(&dev->buf_lock);
 
 	mutex_lock(&dev->struct_mutex);
 	entry = &dma->bufs[order];
@@ -1175,7 +1175,7 @@ int drm_addbufs(struct drm_device *dev, void *data,
  * \param arg pointer to a drm_buf_info structure.
  * \return zero on success or a negative number on failure.
  *
- * Increments drm_device::buf_use while holding the drm_device::count_lock
+ * Increments drm_device::buf_use while holding the drm_device::buf_lock
  * lock, preventing of allocating more buffers after this call. Information
  * about each requested buffer is then copied into user space.
  */
@@ -1196,13 +1196,13 @@ int drm_infobufs(struct drm_device *dev, void *data,
 	if (!dma)
 		return -EINVAL;
 
-	spin_lock(&dev->count_lock);
+	spin_lock(&dev->buf_lock);
 	if (atomic_read(&dev->buf_alloc)) {
-		spin_unlock(&dev->count_lock);
+		spin_unlock(&dev->buf_lock);
 		return -EBUSY;
 	}
 	++dev->buf_use;		/* Can't allocate more after this call */
-	spin_unlock(&dev->count_lock);
+	spin_unlock(&dev->buf_lock);
 
 	for (i = 0, count = 0; i < DRM_MAX_ORDER + 1; i++) {
 		if (dma->bufs[i].buf_count)
@@ -1381,13 +1381,13 @@ int drm_mapbufs(struct drm_device *dev, void *data,
 	if (!dma)
 		return -EINVAL;
 
-	spin_lock(&dev->count_lock);
+	spin_lock(&dev->buf_lock);
 	if (atomic_read(&dev->buf_alloc)) {
-		spin_unlock(&dev->count_lock);
+		spin_unlock(&dev->buf_lock);
 		return -EBUSY;
 	}
 	dev->buf_use++;		/* Can't allocate more after this call */
-	spin_unlock(&dev->count_lock);
+	spin_unlock(&dev->buf_lock);
 
 	if (request->count >= dma->buf_count) {
 		if ((dev->agp && (dma->flags & _DRM_DMA_USE_AGP))
diff --git a/drivers/gpu/drm/drm_cache.c b/drivers/gpu/drm/drm_cache.c
index 534cb89b160d..a6b690626a6b 100644
--- a/drivers/gpu/drm/drm_cache.c
+++ b/drivers/gpu/drm/drm_cache.c
@@ -131,14 +131,14 @@ drm_clflush_sg(struct sg_table *st)
 EXPORT_SYMBOL(drm_clflush_sg);
 
 void
-drm_clflush_virt_range(char *addr, unsigned long length)
+drm_clflush_virt_range(void *addr, unsigned long length)
 {
 #if defined(CONFIG_X86)
 	if (cpu_has_clflush) {
-		char *end = addr + length;
+		void *end = addr + length;
 		mb();
 		for (; addr < end; addr += boot_cpu_data.x86_clflush_size)
-			clflush(addr);
+			clflushopt(addr);
 		clflushopt(end - 1);
 		mb();
 		return;
diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index d8b7099abece..fe94cc10cd35 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -37,6 +37,7 @@
 #include <drm/drm_crtc.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_fourcc.h>
+#include <drm/drm_modeset_lock.h>
 
 #include "drm_crtc_internal.h"
 
@@ -50,12 +51,42 @@
  */
 void drm_modeset_lock_all(struct drm_device *dev)
 {
-	struct drm_crtc *crtc;
+	struct drm_mode_config *config = &dev->mode_config;
+	struct drm_modeset_acquire_ctx *ctx;
+	int ret;
 
-	mutex_lock(&dev->mode_config.mutex);
+	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
+	if (WARN_ON(!ctx))
+		return;
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
-		mutex_lock_nest_lock(&crtc->mutex, &dev->mode_config.mutex);
+	mutex_lock(&config->mutex);
+
+	drm_modeset_acquire_init(ctx, 0);
+
+retry:
+	ret = drm_modeset_lock(&config->connection_mutex, ctx);
+	if (ret)
+		goto fail;
+	ret = drm_modeset_lock_all_crtcs(dev, ctx);
+	if (ret)
+		goto fail;
+
+	WARN_ON(config->acquire_ctx);
+
+	/* now we hold the locks, so now that it is safe, stash the
+	 * ctx for drm_modeset_unlock_all():
+	 */
+	config->acquire_ctx = ctx;
+
+	drm_warn_on_modeset_not_all_locked(dev);
+
+	return;
+
+fail:
+	if (ret == -EDEADLK) {
+		drm_modeset_backoff(ctx);
+		goto retry;
+	}
 }
 EXPORT_SYMBOL(drm_modeset_lock_all);
 
@@ -67,10 +98,17 @@ EXPORT_SYMBOL(drm_modeset_lock_all);
  */
 void drm_modeset_unlock_all(struct drm_device *dev)
 {
-	struct drm_crtc *crtc;
+	struct drm_mode_config *config = &dev->mode_config;
+	struct drm_modeset_acquire_ctx *ctx = config->acquire_ctx;
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
-		mutex_unlock(&crtc->mutex);
+	if (WARN_ON(!ctx))
+		return;
+
+	config->acquire_ctx = NULL;
+	drm_modeset_drop_locks(ctx);
+	drm_modeset_acquire_fini(ctx);
+
+	kfree(ctx);
 
 	mutex_unlock(&dev->mode_config.mutex);
 }
@@ -91,8 +129,9 @@ void drm_warn_on_modeset_not_all_locked(struct drm_device *dev)
 		return;
 
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
-		WARN_ON(!mutex_is_locked(&crtc->mutex));
+		WARN_ON(!drm_modeset_is_locked(&crtc->mutex));
 
+	WARN_ON(!drm_modeset_is_locked(&dev->mode_config.connection_mutex));
 	WARN_ON(!mutex_is_locked(&dev->mode_config.mutex));
 }
 EXPORT_SYMBOL(drm_warn_on_modeset_not_all_locked);
@@ -227,6 +266,7 @@ static const struct drm_prop_enum_list drm_encoder_enum_list[] =
 	{ DRM_MODE_ENCODER_TVDAC, "TV" },
 	{ DRM_MODE_ENCODER_VIRTUAL, "Virtual" },
 	{ DRM_MODE_ENCODER_DSI, "DSI" },
+	{ DRM_MODE_ENCODER_DPMST, "DP MST" },
 };
 
 static const struct drm_prop_enum_list drm_subpixel_enum_list[] =
@@ -256,46 +296,6 @@ void drm_connector_ida_destroy(void)
 }
 
 /**
- * drm_get_encoder_name - return a string for encoder
- * @encoder: encoder to compute name of
- *
- * Note that the buffer used by this function is globally shared and owned by
- * the function itself.
- *
- * FIXME: This isn't really multithreading safe.
- */
-const char *drm_get_encoder_name(const struct drm_encoder *encoder)
-{
-	static char buf[32];
-
-	snprintf(buf, 32, "%s-%d",
-		 drm_encoder_enum_list[encoder->encoder_type].name,
-		 encoder->base.id);
-	return buf;
-}
-EXPORT_SYMBOL(drm_get_encoder_name);
-
-/**
- * drm_get_connector_name - return a string for connector
- * @connector: connector to compute name of
- *
- * Note that the buffer used by this function is globally shared and owned by
- * the function itself.
- *
- * FIXME: This isn't really multithreading safe.
- */
-const char *drm_get_connector_name(const struct drm_connector *connector)
-{
-	static char buf[32];
-
-	snprintf(buf, 32, "%s-%d",
-		 drm_connector_enum_list[connector->connector_type].name,
-		 connector->connector_type_id);
-	return buf;
-}
-EXPORT_SYMBOL(drm_get_connector_name);
-
-/**
  * drm_get_connector_status_name - return a string for connector status
  * @status: connector status to compute name of
  *
@@ -409,6 +409,21 @@ void drm_mode_object_put(struct drm_device *dev,
 	mutex_unlock(&dev->mode_config.idr_mutex);
 }
 
+static struct drm_mode_object *_object_find(struct drm_device *dev,
+		uint32_t id, uint32_t type)
+{
+	struct drm_mode_object *obj = NULL;
+
+	mutex_lock(&dev->mode_config.idr_mutex);
+	obj = idr_find(&dev->mode_config.crtc_idr, id);
+	if (!obj || (type != DRM_MODE_OBJECT_ANY && obj->type != type) ||
+	    (obj->id != id))
+		obj = NULL;
+	mutex_unlock(&dev->mode_config.idr_mutex);
+
+	return obj;
+}
+
 /**
  * drm_mode_object_find - look up a drm object with static lifetime
  * @dev: drm device
@@ -416,7 +431,9 @@ void drm_mode_object_put(struct drm_device *dev,
  * @type: type of the mode object
  *
  * Note that framebuffers cannot be looked up with this functions - since those
- * are reference counted, they need special treatment.
+ * are reference counted, they need special treatment.  Even with
+ * DRM_MODE_OBJECT_ANY (although that will simply return NULL
+ * rather than WARN_ON()).
  */
 struct drm_mode_object *drm_mode_object_find(struct drm_device *dev,
 		uint32_t id, uint32_t type)
@@ -426,13 +443,10 @@ struct drm_mode_object *drm_mode_object_find(struct drm_device *dev,
 	/* Framebuffers are reference counted and need their own lookup
 	 * function.*/
 	WARN_ON(type == DRM_MODE_OBJECT_FB);
-
-	mutex_lock(&dev->mode_config.idr_mutex);
-	obj = idr_find(&dev->mode_config.crtc_idr, id);
-	if (!obj || (obj->type != type) || (obj->id != id))
+	obj = _object_find(dev, id, type);
+	/* don't leak out unref'd fb's */
+	if (obj && (obj->type == DRM_MODE_OBJECT_FB))
 		obj = NULL;
-	mutex_unlock(&dev->mode_config.idr_mutex);
-
 	return obj;
 }
 EXPORT_SYMBOL(drm_mode_object_find);
@@ -538,7 +552,7 @@ EXPORT_SYMBOL(drm_framebuffer_lookup);
  */
 void drm_framebuffer_unreference(struct drm_framebuffer *fb)
 {
-	DRM_DEBUG("FB ID: %d\n", fb->base.id);
+	DRM_DEBUG("%p: FB ID: %d (%d)\n", fb, fb->base.id, atomic_read(&fb->refcount.refcount));
 	kref_put(&fb->refcount, drm_framebuffer_free);
 }
 EXPORT_SYMBOL(drm_framebuffer_unreference);
@@ -551,7 +565,7 @@ EXPORT_SYMBOL(drm_framebuffer_unreference);
  */
 void drm_framebuffer_reference(struct drm_framebuffer *fb)
 {
-	DRM_DEBUG("FB ID: %d\n", fb->base.id);
+	DRM_DEBUG("%p: FB ID: %d (%d)\n", fb, fb->base.id, atomic_read(&fb->refcount.refcount));
 	kref_get(&fb->refcount);
 }
 EXPORT_SYMBOL(drm_framebuffer_reference);
@@ -563,7 +577,7 @@ static void drm_framebuffer_free_bug(struct kref *kref)
 
 static void __drm_framebuffer_unreference(struct drm_framebuffer *fb)
 {
-	DRM_DEBUG("FB ID: %d\n", fb->base.id);
+	DRM_DEBUG("%p: FB ID: %d (%d)\n", fb, fb->base.id, atomic_read(&fb->refcount.refcount));
 	kref_put(&fb->refcount, drm_framebuffer_free_bug);
 }
 
@@ -691,6 +705,8 @@ void drm_framebuffer_remove(struct drm_framebuffer *fb)
 }
 EXPORT_SYMBOL(drm_framebuffer_remove);
 
+DEFINE_WW_CLASS(crtc_ww_class);
+
 /**
  * drm_crtc_init_with_planes - Initialise a new CRTC object with
  *    specified primary and cursor planes.
@@ -710,6 +726,7 @@ int drm_crtc_init_with_planes(struct drm_device *dev, struct drm_crtc *crtc,
 			      void *cursor,
 			      const struct drm_crtc_funcs *funcs)
 {
+	struct drm_mode_config *config = &dev->mode_config;
 	int ret;
 
 	crtc->dev = dev;
@@ -717,8 +734,9 @@ int drm_crtc_init_with_planes(struct drm_device *dev, struct drm_crtc *crtc,
 	crtc->invert_dimensions = false;
 
 	drm_modeset_lock_all(dev);
-	mutex_init(&crtc->mutex);
-	mutex_lock_nest_lock(&crtc->mutex, &dev->mode_config.mutex);
+	drm_modeset_lock_init(&crtc->mutex);
+	/* dropped by _unlock_all(): */
+	drm_modeset_lock(&crtc->mutex, config->acquire_ctx);
 
 	ret = drm_mode_object_get(dev, &crtc->base, DRM_MODE_OBJECT_CRTC);
 	if (ret)
@@ -726,8 +744,8 @@ int drm_crtc_init_with_planes(struct drm_device *dev, struct drm_crtc *crtc,
 
 	crtc->base.properties = &crtc->properties;
 
-	list_add_tail(&crtc->head, &dev->mode_config.crtc_list);
-	dev->mode_config.num_crtc++;
+	list_add_tail(&crtc->head, &config->crtc_list);
+	config->num_crtc++;
 
 	crtc->primary = primary;
 	if (primary)
@@ -755,6 +773,8 @@ void drm_crtc_cleanup(struct drm_crtc *crtc)
 	kfree(crtc->gamma_store);
 	crtc->gamma_store = NULL;
 
+	drm_modeset_lock_fini(&crtc->mutex);
+
 	drm_mode_object_put(dev, &crtc->base);
 	list_del(&crtc->head);
 	dev->mode_config.num_crtc--;
@@ -824,7 +844,7 @@ int drm_connector_init(struct drm_device *dev,
 
 	ret = drm_mode_object_get(dev, &connector->base, DRM_MODE_OBJECT_CONNECTOR);
 	if (ret)
-		goto out;
+		goto out_unlock;
 
 	connector->base.properties = &connector->properties;
 	connector->dev = dev;
@@ -834,9 +854,17 @@ int drm_connector_init(struct drm_device *dev,
 		ida_simple_get(connector_ida, 1, 0, GFP_KERNEL);
 	if (connector->connector_type_id < 0) {
 		ret = connector->connector_type_id;
-		drm_mode_object_put(dev, &connector->base);
-		goto out;
+		goto out_put;
+	}
+	connector->name =
+		kasprintf(GFP_KERNEL, "%s-%d",
+			  drm_connector_enum_list[connector_type].name,
+			  connector->connector_type_id);
+	if (!connector->name) {
+		ret = -ENOMEM;
+		goto out_put;
 	}
+
 	INIT_LIST_HEAD(&connector->probed_modes);
 	INIT_LIST_HEAD(&connector->modes);
 	connector->edid_blob_ptr = NULL;
@@ -853,7 +881,11 @@ int drm_connector_init(struct drm_device *dev,
 	drm_object_attach_property(&connector->base,
 				      dev->mode_config.dpms_property, 0);
 
- out:
+out_put:
+	if (ret)
+		drm_mode_object_put(dev, &connector->base);
+
+out_unlock:
 	drm_modeset_unlock_all(dev);
 
 	return ret;
@@ -881,6 +913,8 @@ void drm_connector_cleanup(struct drm_connector *connector)
 		   connector->connector_type_id);
 
 	drm_mode_object_put(dev, &connector->base);
+	kfree(connector->name);
+	connector->name = NULL;
 	list_del(&connector->head);
 	dev->mode_config.num_connector--;
 }
@@ -982,16 +1016,27 @@ int drm_encoder_init(struct drm_device *dev,
 
 	ret = drm_mode_object_get(dev, &encoder->base, DRM_MODE_OBJECT_ENCODER);
 	if (ret)
-		goto out;
+		goto out_unlock;
 
 	encoder->dev = dev;
 	encoder->encoder_type = encoder_type;
 	encoder->funcs = funcs;
+	encoder->name = kasprintf(GFP_KERNEL, "%s-%d",
+				  drm_encoder_enum_list[encoder_type].name,
+				  encoder->base.id);
+	if (!encoder->name) {
+		ret = -ENOMEM;
+		goto out_put;
+	}
 
 	list_add_tail(&encoder->head, &dev->mode_config.encoder_list);
 	dev->mode_config.num_encoder++;
 
- out:
+out_put:
+	if (ret)
+		drm_mode_object_put(dev, &encoder->base);
+
+out_unlock:
 	drm_modeset_unlock_all(dev);
 
 	return ret;
@@ -1009,6 +1054,8 @@ void drm_encoder_cleanup(struct drm_encoder *encoder)
 	struct drm_device *dev = encoder->dev;
 	drm_modeset_lock_all(dev);
 	drm_mode_object_put(dev, &encoder->base);
+	kfree(encoder->name);
+	encoder->name = NULL;
 	list_del(&encoder->head);
 	dev->mode_config.num_encoder--;
 	drm_modeset_unlock_all(dev);
@@ -1145,16 +1192,19 @@ EXPORT_SYMBOL(drm_plane_cleanup);
  */
 void drm_plane_force_disable(struct drm_plane *plane)
 {
+	struct drm_framebuffer *old_fb = plane->fb;
 	int ret;
 
-	if (!plane->fb)
+	if (!old_fb)
 		return;
 
 	ret = plane->funcs->disable_plane(plane);
-	if (ret)
+	if (ret) {
 		DRM_ERROR("failed to disable plane with busy fb\n");
+		return;
+	}
 	/* disconnect the plane from the fb and crtc: */
-	__drm_framebuffer_unreference(plane->fb);
+	__drm_framebuffer_unreference(old_fb);
 	plane->fb = NULL;
 	plane->crtc = NULL;
 }
@@ -1378,6 +1428,12 @@ static int drm_mode_group_init(struct drm_device *dev, struct drm_mode_group *gr
 	return 0;
 }
 
+void drm_mode_group_destroy(struct drm_mode_group *group)
+{
+	kfree(group->id_list);
+	group->id_list = NULL;
+}
+
 /*
  * NOTE: Driver's shouldn't ever call drm_mode_group_init_legacy_group - it is
  * the drm core's responsibility to set up mode control groups.
@@ -1614,7 +1670,7 @@ int drm_mode_getresources(struct drm_device *dev, void *data,
 					    &dev->mode_config.encoder_list,
 					    head) {
 				DRM_DEBUG_KMS("[ENCODER:%d:%s]\n", encoder->base.id,
-						drm_get_encoder_name(encoder));
+						encoder->name);
 				if (put_user(encoder->base.id, encoder_id +
 					     copied)) {
 					ret = -EFAULT;
@@ -1646,7 +1702,7 @@ int drm_mode_getresources(struct drm_device *dev, void *data,
 					    head) {
 				DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
 					connector->base.id,
-					drm_get_connector_name(connector));
+					connector->name);
 				if (put_user(connector->base.id,
 					     connector_id + copied)) {
 					ret = -EFAULT;
@@ -1695,7 +1751,6 @@ int drm_mode_getcrtc(struct drm_device *dev,
 {
 	struct drm_mode_crtc *crtc_resp = data;
 	struct drm_crtc *crtc;
-	struct drm_mode_object *obj;
 	int ret = 0;
 
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
@@ -1703,13 +1758,11 @@ int drm_mode_getcrtc(struct drm_device *dev,
 
 	drm_modeset_lock_all(dev);
 
-	obj = drm_mode_object_find(dev, crtc_resp->crtc_id,
-				   DRM_MODE_OBJECT_CRTC);
-	if (!obj) {
+	crtc = drm_crtc_find(dev, crtc_resp->crtc_id);
+	if (!crtc) {
 		ret = -ENOENT;
 		goto out;
 	}
-	crtc = obj_to_crtc(obj);
 
 	crtc_resp->x = crtc->x;
 	crtc_resp->y = crtc->y;
@@ -1763,7 +1816,6 @@ int drm_mode_getconnector(struct drm_device *dev, void *data,
 			  struct drm_file *file_priv)
 {
 	struct drm_mode_get_connector *out_resp = data;
-	struct drm_mode_object *obj;
 	struct drm_connector *connector;
 	struct drm_display_mode *mode;
 	int mode_count = 0;
@@ -1787,13 +1839,11 @@ int drm_mode_getconnector(struct drm_device *dev, void *data,
 
 	mutex_lock(&dev->mode_config.mutex);
 
-	obj = drm_mode_object_find(dev, out_resp->connector_id,
-				   DRM_MODE_OBJECT_CONNECTOR);
-	if (!obj) {
+	connector = drm_connector_find(dev, out_resp->connector_id);
+	if (!connector) {
 		ret = -ENOENT;
 		goto out;
 	}
-	connector = obj_to_connector(obj);
 
 	props_count = connector->properties.count;
 
@@ -1821,10 +1871,12 @@ int drm_mode_getconnector(struct drm_device *dev, void *data,
 	out_resp->mm_height = connector->display_info.height_mm;
 	out_resp->subpixel = connector->display_info.subpixel_order;
 	out_resp->connection = connector->status;
+	drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
 	if (connector->encoder)
 		out_resp->encoder_id = connector->encoder->base.id;
 	else
 		out_resp->encoder_id = 0;
+	drm_modeset_unlock(&dev->mode_config.connection_mutex);
 
 	/*
 	 * This ioctl is called twice, once to determine how much space is
@@ -1908,7 +1960,6 @@ int drm_mode_getencoder(struct drm_device *dev, void *data,
 			struct drm_file *file_priv)
 {
 	struct drm_mode_get_encoder *enc_resp = data;
-	struct drm_mode_object *obj;
 	struct drm_encoder *encoder;
 	int ret = 0;
 
@@ -1916,13 +1967,11 @@ int drm_mode_getencoder(struct drm_device *dev, void *data,
 		return -EINVAL;
 
 	drm_modeset_lock_all(dev);
-	obj = drm_mode_object_find(dev, enc_resp->encoder_id,
-				   DRM_MODE_OBJECT_ENCODER);
-	if (!obj) {
+	encoder = drm_encoder_find(dev, enc_resp->encoder_id);
+	if (!encoder) {
 		ret = -ENOENT;
 		goto out;
 	}
-	encoder = obj_to_encoder(obj);
 
 	if (encoder->crtc)
 		enc_resp->crtc_id = encoder->crtc->base.id;
@@ -2020,7 +2069,6 @@ int drm_mode_getplane(struct drm_device *dev, void *data,
 		      struct drm_file *file_priv)
 {
 	struct drm_mode_get_plane *plane_resp = data;
-	struct drm_mode_object *obj;
 	struct drm_plane *plane;
 	uint32_t __user *format_ptr;
 	int ret = 0;
@@ -2029,13 +2077,11 @@ int drm_mode_getplane(struct drm_device *dev, void *data,
 		return -EINVAL;
 
 	drm_modeset_lock_all(dev);
-	obj = drm_mode_object_find(dev, plane_resp->plane_id,
-				   DRM_MODE_OBJECT_PLANE);
-	if (!obj) {
+	plane = drm_plane_find(dev, plane_resp->plane_id);
+	if (!plane) {
 		ret = -ENOENT;
 		goto out;
 	}
-	plane = obj_to_plane(obj);
 
 	if (plane->crtc)
 		plane_resp->crtc_id = plane->crtc->base.id;
@@ -2088,7 +2134,6 @@ int drm_mode_setplane(struct drm_device *dev, void *data,
 		      struct drm_file *file_priv)
 {
 	struct drm_mode_set_plane *plane_req = data;
-	struct drm_mode_object *obj;
 	struct drm_plane *plane;
 	struct drm_crtc *crtc;
 	struct drm_framebuffer *fb = NULL, *old_fb = NULL;
@@ -2103,35 +2148,42 @@ int drm_mode_setplane(struct drm_device *dev, void *data,
 	 * First, find the plane, crtc, and fb objects.  If not available,
 	 * we don't bother to call the driver.
 	 */
-	obj = drm_mode_object_find(dev, plane_req->plane_id,
-				   DRM_MODE_OBJECT_PLANE);
-	if (!obj) {
+	plane = drm_plane_find(dev, plane_req->plane_id);
+	if (!plane) {
 		DRM_DEBUG_KMS("Unknown plane ID %d\n",
 			      plane_req->plane_id);
 		return -ENOENT;
 	}
-	plane = obj_to_plane(obj);
 
 	/* No fb means shut it down */
 	if (!plane_req->fb_id) {
 		drm_modeset_lock_all(dev);
 		old_fb = plane->fb;
-		plane->funcs->disable_plane(plane);
-		plane->crtc = NULL;
-		plane->fb = NULL;
+		ret = plane->funcs->disable_plane(plane);
+		if (!ret) {
+			plane->crtc = NULL;
+			plane->fb = NULL;
+		} else {
+			old_fb = NULL;
+		}
 		drm_modeset_unlock_all(dev);
 		goto out;
 	}
 
-	obj = drm_mode_object_find(dev, plane_req->crtc_id,
-				   DRM_MODE_OBJECT_CRTC);
-	if (!obj) {
+	crtc = drm_crtc_find(dev, plane_req->crtc_id);
+	if (!crtc) {
 		DRM_DEBUG_KMS("Unknown crtc ID %d\n",
 			      plane_req->crtc_id);
 		ret = -ENOENT;
 		goto out;
 	}
-	crtc = obj_to_crtc(obj);
+
+	/* Check whether this plane is usable on this CRTC */
+	if (!(plane->possible_crtcs & drm_crtc_mask(crtc))) {
+		DRM_DEBUG_KMS("Invalid crtc for plane\n");
+		ret = -EINVAL;
+		goto out;
+	}
 
 	fb = drm_framebuffer_lookup(dev, plane_req->fb_id);
 	if (!fb) {
@@ -2187,16 +2239,18 @@ int drm_mode_setplane(struct drm_device *dev, void *data,
 	}
 
 	drm_modeset_lock_all(dev);
+	old_fb = plane->fb;
 	ret = plane->funcs->update_plane(plane, crtc, fb,
 					 plane_req->crtc_x, plane_req->crtc_y,
 					 plane_req->crtc_w, plane_req->crtc_h,
 					 plane_req->src_x, plane_req->src_y,
 					 plane_req->src_w, plane_req->src_h);
 	if (!ret) {
-		old_fb = plane->fb;
 		plane->crtc = crtc;
 		plane->fb = fb;
 		fb = NULL;
+	} else {
+		old_fb = NULL;
 	}
 	drm_modeset_unlock_all(dev);
 
@@ -2239,9 +2293,7 @@ int drm_mode_set_config_internal(struct drm_mode_set *set)
 	ret = crtc->funcs->set_config(set);
 	if (ret == 0) {
 		crtc->primary->crtc = crtc;
-
-		/* crtc->fb must be updated by ->set_config, enforces this. */
-		WARN_ON(fb != crtc->primary->fb);
+		crtc->primary->fb = fb;
 	}
 
 	list_for_each_entry(tmp, &crtc->dev->mode_config.crtc_list, head) {
@@ -2318,7 +2370,6 @@ int drm_mode_setcrtc(struct drm_device *dev, void *data,
 {
 	struct drm_mode_config *config = &dev->mode_config;
 	struct drm_mode_crtc *crtc_req = data;
-	struct drm_mode_object *obj;
 	struct drm_crtc *crtc;
 	struct drm_connector **connector_set = NULL, *connector;
 	struct drm_framebuffer *fb = NULL;
@@ -2336,14 +2387,12 @@ int drm_mode_setcrtc(struct drm_device *dev, void *data,
 		return -ERANGE;
 
 	drm_modeset_lock_all(dev);
-	obj = drm_mode_object_find(dev, crtc_req->crtc_id,
-				   DRM_MODE_OBJECT_CRTC);
-	if (!obj) {
+	crtc = drm_crtc_find(dev, crtc_req->crtc_id);
+	if (!crtc) {
 		DRM_DEBUG_KMS("Unknown CRTC ID %d\n", crtc_req->crtc_id);
 		ret = -ENOENT;
 		goto out;
 	}
-	crtc = obj_to_crtc(obj);
 	DRM_DEBUG_KMS("[CRTC:%d]\n", crtc->base.id);
 
 	if (crtc_req->mode_valid) {
@@ -2426,18 +2475,16 @@ int drm_mode_setcrtc(struct drm_device *dev, void *data,
 				goto out;
 			}
 
-			obj = drm_mode_object_find(dev, out_id,
-						   DRM_MODE_OBJECT_CONNECTOR);
-			if (!obj) {
+			connector = drm_connector_find(dev, out_id);
+			if (!connector) {
 				DRM_DEBUG_KMS("Connector id %d unknown\n",
 						out_id);
 				ret = -ENOENT;
 				goto out;
 			}
-			connector = obj_to_connector(obj);
 			DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
 					connector->base.id,
-					drm_get_connector_name(connector));
+					connector->name);
 
 			connector_set[i] = connector;
 		}
@@ -2466,7 +2513,6 @@ static int drm_mode_cursor_common(struct drm_device *dev,
 				  struct drm_mode_cursor2 *req,
 				  struct drm_file *file_priv)
 {
-	struct drm_mode_object *obj;
 	struct drm_crtc *crtc;
 	int ret = 0;
 
@@ -2476,14 +2522,13 @@ static int drm_mode_cursor_common(struct drm_device *dev,
 	if (!req->flags || (~DRM_MODE_CURSOR_FLAGS & req->flags))
 		return -EINVAL;
 
-	obj = drm_mode_object_find(dev, req->crtc_id, DRM_MODE_OBJECT_CRTC);
-	if (!obj) {
+	crtc = drm_crtc_find(dev, req->crtc_id);
+	if (!crtc) {
 		DRM_DEBUG_KMS("Unknown CRTC ID %d\n", req->crtc_id);
 		return -ENOENT;
 	}
-	crtc = obj_to_crtc(obj);
 
-	mutex_lock(&crtc->mutex);
+	drm_modeset_lock(&crtc->mutex, NULL);
 	if (req->flags & DRM_MODE_CURSOR_BO) {
 		if (!crtc->funcs->cursor_set && !crtc->funcs->cursor_set2) {
 			ret = -ENXIO;
@@ -2507,7 +2552,7 @@ static int drm_mode_cursor_common(struct drm_device *dev,
 		}
 	}
 out:
-	mutex_unlock(&crtc->mutex);
+	drm_modeset_unlock(&crtc->mutex);
 
 	return ret;
 
@@ -3097,6 +3142,8 @@ struct drm_property *drm_property_create(struct drm_device *dev, int flags,
 	if (!property)
 		return NULL;
 
+	property->dev = dev;
+
 	if (num_values) {
 		property->values = kzalloc(sizeof(uint64_t)*num_values, GFP_KERNEL);
 		if (!property->values)
@@ -3117,6 +3164,9 @@ struct drm_property *drm_property_create(struct drm_device *dev, int flags,
 	}
 
 	list_add_tail(&property->head, &dev->mode_config.property_list);
+
+	WARN_ON(!drm_property_type_valid(property));
+
 	return property;
 fail:
 	kfree(property->values);
@@ -3217,6 +3267,22 @@ struct drm_property *drm_property_create_bitmask(struct drm_device *dev,
 }
 EXPORT_SYMBOL(drm_property_create_bitmask);
 
+static struct drm_property *property_create_range(struct drm_device *dev,
+					 int flags, const char *name,
+					 uint64_t min, uint64_t max)
+{
+	struct drm_property *property;
+
+	property = drm_property_create(dev, flags, name, 2);
+	if (!property)
+		return NULL;
+
+	property->values[0] = min;
+	property->values[1] = max;
+
+	return property;
+}
+
 /**
  * drm_property_create - create a new ranged property type
  * @dev: drm device
@@ -3239,20 +3305,36 @@ struct drm_property *drm_property_create_range(struct drm_device *dev, int flags
 					 const char *name,
 					 uint64_t min, uint64_t max)
 {
+	return property_create_range(dev, DRM_MODE_PROP_RANGE | flags,
+			name, min, max);
+}
+EXPORT_SYMBOL(drm_property_create_range);
+
+struct drm_property *drm_property_create_signed_range(struct drm_device *dev,
+					 int flags, const char *name,
+					 int64_t min, int64_t max)
+{
+	return property_create_range(dev, DRM_MODE_PROP_SIGNED_RANGE | flags,
+			name, I642U64(min), I642U64(max));
+}
+EXPORT_SYMBOL(drm_property_create_signed_range);
+
+struct drm_property *drm_property_create_object(struct drm_device *dev,
+					 int flags, const char *name, uint32_t type)
+{
 	struct drm_property *property;
 
-	flags |= DRM_MODE_PROP_RANGE;
+	flags |= DRM_MODE_PROP_OBJECT;
 
-	property = drm_property_create(dev, flags, name, 2);
+	property = drm_property_create(dev, flags, name, 1);
 	if (!property)
 		return NULL;
 
-	property->values[0] = min;
-	property->values[1] = max;
+	property->values[0] = type;
 
 	return property;
 }
-EXPORT_SYMBOL(drm_property_create_range);
+EXPORT_SYMBOL(drm_property_create_object);
 
 /**
  * drm_property_add_enum - add a possible value to an enumeration property
@@ -3274,14 +3356,16 @@ int drm_property_add_enum(struct drm_property *property, int index,
 {
 	struct drm_property_enum *prop_enum;
 
-	if (!(property->flags & (DRM_MODE_PROP_ENUM | DRM_MODE_PROP_BITMASK)))
+	if (!(drm_property_type_is(property, DRM_MODE_PROP_ENUM) ||
+			drm_property_type_is(property, DRM_MODE_PROP_BITMASK)))
 		return -EINVAL;
 
 	/*
 	 * Bitmask enum properties have the additional constraint of values
 	 * from 0 to 63
 	 */
-	if ((property->flags & DRM_MODE_PROP_BITMASK) && (value > 63))
+	if (drm_property_type_is(property, DRM_MODE_PROP_BITMASK) &&
+			(value > 63))
 		return -EINVAL;
 
 	if (!list_empty(&property->enum_blob_list)) {
@@ -3438,7 +3522,6 @@ EXPORT_SYMBOL(drm_object_property_get_value);
 int drm_mode_getproperty_ioctl(struct drm_device *dev,
 			       void *data, struct drm_file *file_priv)
 {
-	struct drm_mode_object *obj;
 	struct drm_mode_get_property *out_resp = data;
 	struct drm_property *property;
 	int enum_count = 0;
@@ -3457,17 +3540,17 @@ int drm_mode_getproperty_ioctl(struct drm_device *dev,
 		return -EINVAL;
 
 	drm_modeset_lock_all(dev);
-	obj = drm_mode_object_find(dev, out_resp->prop_id, DRM_MODE_OBJECT_PROPERTY);
-	if (!obj) {
+	property = drm_property_find(dev, out_resp->prop_id);
+	if (!property) {
 		ret = -ENOENT;
 		goto done;
 	}
-	property = obj_to_property(obj);
 
-	if (property->flags & (DRM_MODE_PROP_ENUM | DRM_MODE_PROP_BITMASK)) {
+	if (drm_property_type_is(property, DRM_MODE_PROP_ENUM) ||
+			drm_property_type_is(property, DRM_MODE_PROP_BITMASK)) {
 		list_for_each_entry(prop_enum, &property->enum_blob_list, head)
 			enum_count++;
-	} else if (property->flags & DRM_MODE_PROP_BLOB) {
+	} else if (drm_property_type_is(property, DRM_MODE_PROP_BLOB)) {
 		list_for_each_entry(prop_blob, &property->enum_blob_list, head)
 			blob_count++;
 	}
@@ -3489,7 +3572,8 @@ int drm_mode_getproperty_ioctl(struct drm_device *dev,
 	}
 	out_resp->count_values = value_count;
 
-	if (property->flags & (DRM_MODE_PROP_ENUM | DRM_MODE_PROP_BITMASK)) {
+	if (drm_property_type_is(property, DRM_MODE_PROP_ENUM) ||
+			drm_property_type_is(property, DRM_MODE_PROP_BITMASK)) {
 		if ((out_resp->count_enum_blobs >= enum_count) && enum_count) {
 			copied = 0;
 			enum_ptr = (struct drm_mode_property_enum __user *)(unsigned long)out_resp->enum_blob_ptr;
@@ -3511,7 +3595,7 @@ int drm_mode_getproperty_ioctl(struct drm_device *dev,
 		out_resp->count_enum_blobs = enum_count;
 	}
 
-	if (property->flags & DRM_MODE_PROP_BLOB) {
+	if (drm_property_type_is(property, DRM_MODE_PROP_BLOB)) {
 		if ((out_resp->count_enum_blobs >= blob_count) && blob_count) {
 			copied = 0;
 			blob_id_ptr = (uint32_t __user *)(unsigned long)out_resp->enum_blob_ptr;
@@ -3590,7 +3674,6 @@ static void drm_property_destroy_blob(struct drm_device *dev,
 int drm_mode_getblob_ioctl(struct drm_device *dev,
 			   void *data, struct drm_file *file_priv)
 {
-	struct drm_mode_object *obj;
 	struct drm_mode_get_blob *out_resp = data;
 	struct drm_property_blob *blob;
 	int ret = 0;
@@ -3600,12 +3683,11 @@ int drm_mode_getblob_ioctl(struct drm_device *dev,
 		return -EINVAL;
 
 	drm_modeset_lock_all(dev);
-	obj = drm_mode_object_find(dev, out_resp->blob_id, DRM_MODE_OBJECT_BLOB);
-	if (!obj) {
+	blob = drm_property_blob_find(dev, out_resp->blob_id);
+	if (!blob) {
 		ret = -ENOENT;
 		goto done;
 	}
-	blob = obj_to_blob(obj);
 
 	if (out_resp->length == blob->length) {
 		blob_ptr = (void __user *)(unsigned long)out_resp->data;
@@ -3667,19 +3749,40 @@ static bool drm_property_change_is_valid(struct drm_property *property,
 {
 	if (property->flags & DRM_MODE_PROP_IMMUTABLE)
 		return false;
-	if (property->flags & DRM_MODE_PROP_RANGE) {
+
+	if (drm_property_type_is(property, DRM_MODE_PROP_RANGE)) {
 		if (value < property->values[0] || value > property->values[1])
 			return false;
 		return true;
-	} else if (property->flags & DRM_MODE_PROP_BITMASK) {
+	} else if (drm_property_type_is(property, DRM_MODE_PROP_SIGNED_RANGE)) {
+		int64_t svalue = U642I64(value);
+		if (svalue < U642I64(property->values[0]) ||
+				svalue > U642I64(property->values[1]))
+			return false;
+		return true;
+	} else if (drm_property_type_is(property, DRM_MODE_PROP_BITMASK)) {
 		int i;
 		uint64_t valid_mask = 0;
 		for (i = 0; i < property->num_values; i++)
 			valid_mask |= (1ULL << property->values[i]);
 		return !(value & ~valid_mask);
-	} else if (property->flags & DRM_MODE_PROP_BLOB) {
+	} else if (drm_property_type_is(property, DRM_MODE_PROP_BLOB)) {
 		/* Only the driver knows */
 		return true;
+	} else if (drm_property_type_is(property, DRM_MODE_PROP_OBJECT)) {
+		struct drm_mode_object *obj;
+		/* a zero value for an object property translates to null: */
+		if (value == 0)
+			return true;
+		/*
+		 * NOTE: use _object_find() directly to bypass restriction on
+		 * looking up refcnt'd objects (ie. fb's).  For a refcnt'd
+		 * object this could race against object finalization, so it
+		 * simply tells us that the object *was* valid.  Which is good
+		 * enough.
+		 */
+		obj = _object_find(property->dev, value, property->values[0]);
+		return obj != NULL;
 	} else {
 		int i;
 		for (i = 0; i < property->num_values; i++)
@@ -3987,7 +4090,6 @@ int drm_mode_gamma_set_ioctl(struct drm_device *dev,
 			     void *data, struct drm_file *file_priv)
 {
 	struct drm_mode_crtc_lut *crtc_lut = data;
-	struct drm_mode_object *obj;
 	struct drm_crtc *crtc;
 	void *r_base, *g_base, *b_base;
 	int size;
@@ -3997,12 +4099,11 @@ int drm_mode_gamma_set_ioctl(struct drm_device *dev,
 		return -EINVAL;
 
 	drm_modeset_lock_all(dev);
-	obj = drm_mode_object_find(dev, crtc_lut->crtc_id, DRM_MODE_OBJECT_CRTC);
-	if (!obj) {
+	crtc = drm_crtc_find(dev, crtc_lut->crtc_id);
+	if (!crtc) {
 		ret = -ENOENT;
 		goto out;
 	}
-	crtc = obj_to_crtc(obj);
 
 	if (crtc->funcs->gamma_set == NULL) {
 		ret = -ENOSYS;
@@ -4061,7 +4162,6 @@ int drm_mode_gamma_get_ioctl(struct drm_device *dev,
 			     void *data, struct drm_file *file_priv)
 {
 	struct drm_mode_crtc_lut *crtc_lut = data;
-	struct drm_mode_object *obj;
 	struct drm_crtc *crtc;
 	void *r_base, *g_base, *b_base;
 	int size;
@@ -4071,12 +4171,11 @@ int drm_mode_gamma_get_ioctl(struct drm_device *dev,
 		return -EINVAL;
 
 	drm_modeset_lock_all(dev);
-	obj = drm_mode_object_find(dev, crtc_lut->crtc_id, DRM_MODE_OBJECT_CRTC);
-	if (!obj) {
+	crtc = drm_crtc_find(dev, crtc_lut->crtc_id);
+	if (!crtc) {
 		ret = -ENOENT;
 		goto out;
 	}
-	crtc = obj_to_crtc(obj);
 
 	/* memcpy into gamma store */
 	if (crtc_lut->gamma_size != crtc->gamma_size) {
@@ -4129,7 +4228,6 @@ int drm_mode_page_flip_ioctl(struct drm_device *dev,
 			     void *data, struct drm_file *file_priv)
 {
 	struct drm_mode_crtc_page_flip *page_flip = data;
-	struct drm_mode_object *obj;
 	struct drm_crtc *crtc;
 	struct drm_framebuffer *fb = NULL, *old_fb = NULL;
 	struct drm_pending_vblank_event *e = NULL;
@@ -4143,12 +4241,11 @@ int drm_mode_page_flip_ioctl(struct drm_device *dev,
 	if ((page_flip->flags & DRM_MODE_PAGE_FLIP_ASYNC) && !dev->mode_config.async_page_flip)
 		return -EINVAL;
 
-	obj = drm_mode_object_find(dev, page_flip->crtc_id, DRM_MODE_OBJECT_CRTC);
-	if (!obj)
+	crtc = drm_crtc_find(dev, page_flip->crtc_id);
+	if (!crtc)
 		return -ENOENT;
-	crtc = obj_to_crtc(obj);
 
-	mutex_lock(&crtc->mutex);
+	drm_modeset_lock(&crtc->mutex, NULL);
 	if (crtc->primary->fb == NULL) {
 		/* The framebuffer is currently unbound, presumably
 		 * due to a hotplug event, that userspace has not
@@ -4232,7 +4329,7 @@ out:
 		drm_framebuffer_unreference(fb);
 	if (old_fb)
 		drm_framebuffer_unreference(old_fb);
-	mutex_unlock(&crtc->mutex);
+	drm_modeset_unlock(&crtc->mutex);
 
 	return ret;
 }
@@ -4597,6 +4694,7 @@ EXPORT_SYMBOL(drm_format_vert_chroma_subsampling);
 void drm_mode_config_init(struct drm_device *dev)
 {
 	mutex_init(&dev->mode_config.mutex);
+	drm_modeset_lock_init(&dev->mode_config.connection_mutex);
 	mutex_init(&dev->mode_config.idr_mutex);
 	mutex_init(&dev->mode_config.fb_lock);
 	INIT_LIST_HEAD(&dev->mode_config.fb_list);
@@ -4696,5 +4794,6 @@ void drm_mode_config_cleanup(struct drm_device *dev)
 	}
 
 	idr_destroy(&dev->mode_config.crtc_idr);
+	drm_modeset_lock_fini(&dev->mode_config.connection_mutex);
 }
 EXPORT_SYMBOL(drm_mode_config_cleanup);
diff --git a/drivers/gpu/drm/drm_crtc_helper.c b/drivers/gpu/drm/drm_crtc_helper.c
index 872ba11c4533..78b37f3febd3 100644
--- a/drivers/gpu/drm/drm_crtc_helper.c
+++ b/drivers/gpu/drm/drm_crtc_helper.c
@@ -93,8 +93,10 @@ bool drm_helper_encoder_in_use(struct drm_encoder *encoder)
 	 * We can expect this mutex to be locked if we are not panicking.
 	 * Locking is currently fubar in the panic handler.
 	 */
-	if (!oops_in_progress)
+	if (!oops_in_progress) {
 		WARN_ON(!mutex_is_locked(&dev->mode_config.mutex));
+		WARN_ON(!drm_modeset_is_locked(&dev->mode_config.connection_mutex));
+	}
 
 	list_for_each_entry(connector, &dev->mode_config.connector_list, head)
 		if (connector->encoder == encoder)
@@ -153,20 +155,14 @@ drm_encoder_disable(struct drm_encoder *encoder)
 static void __drm_helper_disable_unused_functions(struct drm_device *dev)
 {
 	struct drm_encoder *encoder;
-	struct drm_connector *connector;
 	struct drm_crtc *crtc;
 
 	drm_warn_on_modeset_not_all_locked(dev);
 
-	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
-		if (!connector->encoder)
-			continue;
-	}
-
 	list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
 		if (!drm_helper_encoder_in_use(encoder)) {
 			drm_encoder_disable(encoder);
-			/* disconnector encoder from any connector */
+			/* disconnect encoder from any connector */
 			encoder->crtc = NULL;
 		}
 	}
@@ -349,7 +345,7 @@ bool drm_crtc_helper_set_mode(struct drm_crtc *crtc,
 			continue;
 
 		DRM_DEBUG_KMS("[ENCODER:%d:%s] set [MODE:%d:%s]\n",
-			encoder->base.id, drm_get_encoder_name(encoder),
+			encoder->base.id, encoder->name,
 			mode->base.id, mode->name);
 		encoder_funcs = encoder->helper_private;
 		encoder_funcs->mode_set(encoder, mode, adjusted_mode);
@@ -400,8 +396,7 @@ done:
 }
 EXPORT_SYMBOL(drm_crtc_helper_set_mode);
 
-
-static int
+static void
 drm_crtc_helper_disable(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
@@ -430,7 +425,6 @@ drm_crtc_helper_disable(struct drm_crtc *crtc)
 	}
 
 	__drm_helper_disable_unused_functions(dev);
-	return 0;
 }
 
 /**
@@ -481,7 +475,8 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set)
 				(int)set->num_connectors, set->x, set->y);
 	} else {
 		DRM_DEBUG_KMS("[CRTC:%d] [NOFB]\n", set->crtc->base.id);
-		return drm_crtc_helper_disable(set->crtc);
+		drm_crtc_helper_disable(set->crtc);
+		return 0;
 	}
 
 	dev = set->crtc->dev;
@@ -620,11 +615,11 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set)
 		}
 		if (new_crtc) {
 			DRM_DEBUG_KMS("[CONNECTOR:%d:%s] to [CRTC:%d]\n",
-				connector->base.id, drm_get_connector_name(connector),
+				connector->base.id, connector->name,
 				new_crtc->base.id);
 		} else {
 			DRM_DEBUG_KMS("[CONNECTOR:%d:%s] to [NOCRTC]\n",
-				connector->base.id, drm_get_connector_name(connector));
+				connector->base.id, connector->name);
 		}
 	}
 
@@ -650,7 +645,7 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set)
 			DRM_DEBUG_KMS("Setting connector DPMS state to on\n");
 			for (i = 0; i < set->num_connectors; i++) {
 				DRM_DEBUG_KMS("\t[CONNECTOR:%d:%s] set DPMS on\n", set->connectors[i]->base.id,
-					      drm_get_connector_name(set->connectors[i]));
+					      set->connectors[i]->name);
 				set->connectors[i]->funcs->dpms(set->connectors[i], DRM_MODE_DPMS_ON);
 			}
 		}
diff --git a/drivers/gpu/drm/drm_dp_helper.c b/drivers/gpu/drm/drm_dp_helper.c
index 4b6e6f3ba0a1..08e33b8b13a4 100644
--- a/drivers/gpu/drm/drm_dp_helper.c
+++ b/drivers/gpu/drm/drm_dp_helper.c
@@ -206,13 +206,17 @@ i2c_dp_aux_prepare_bus(struct i2c_adapter *adapter)
  * i2c_dp_aux_add_bus() - register an i2c adapter using the aux ch helper
  * @adapter: i2c adapter to register
  *
- * This registers an i2c adapater that uses dp aux channel as it's underlaying
+ * This registers an i2c adapter that uses dp aux channel as it's underlaying
  * transport. The driver needs to fill out the &i2c_algo_dp_aux_data structure
  * and store it in the algo_data member of the @adapter argument. This will be
  * used by the i2c over dp aux algorithm to drive the hardware.
  *
  * RETURNS:
  * 0 on success, -ERRNO on failure.
+ *
+ * IMPORTANT:
+ * This interface is deprecated, please switch to the new dp aux helpers and
+ * drm_dp_aux_register().
  */
 int
 i2c_dp_aux_add_bus(struct i2c_adapter *adapter)
@@ -378,7 +382,10 @@ static int drm_dp_dpcd_access(struct drm_dp_aux *aux, u8 request,
 	 * transactions.
 	 */
 	for (retry = 0; retry < 7; retry++) {
+
+		mutex_lock(&aux->hw_mutex);
 		err = aux->transfer(aux, &msg);
+		mutex_unlock(&aux->hw_mutex);
 		if (err < 0) {
 			if (err == -EBUSY)
 				continue;
@@ -592,7 +599,9 @@ static int drm_dp_i2c_do_msg(struct drm_dp_aux *aux, struct drm_dp_aux_msg *msg)
 	 * before giving up the AUX transaction.
 	 */
 	for (retry = 0; retry < 7; retry++) {
+		mutex_lock(&aux->hw_mutex);
 		err = aux->transfer(aux, msg);
+		mutex_unlock(&aux->hw_mutex);
 		if (err < 0) {
 			if (err == -EBUSY)
 				continue;
@@ -725,13 +734,15 @@ static const struct i2c_algorithm drm_dp_i2c_algo = {
 };
 
 /**
- * drm_dp_aux_register_i2c_bus() - register an I2C adapter for I2C-over-AUX
+ * drm_dp_aux_register() - initialise and register aux channel
  * @aux: DisplayPort AUX channel
  *
  * Returns 0 on success or a negative error code on failure.
  */
-int drm_dp_aux_register_i2c_bus(struct drm_dp_aux *aux)
+int drm_dp_aux_register(struct drm_dp_aux *aux)
 {
+	mutex_init(&aux->hw_mutex);
+
 	aux->ddc.algo = &drm_dp_i2c_algo;
 	aux->ddc.algo_data = aux;
 	aux->ddc.retries = 3;
@@ -746,14 +757,14 @@ int drm_dp_aux_register_i2c_bus(struct drm_dp_aux *aux)
 
 	return i2c_add_adapter(&aux->ddc);
 }
-EXPORT_SYMBOL(drm_dp_aux_register_i2c_bus);
+EXPORT_SYMBOL(drm_dp_aux_register);
 
 /**
- * drm_dp_aux_unregister_i2c_bus() - unregister an I2C-over-AUX adapter
+ * drm_dp_aux_unregister() - unregister an AUX adapter
  * @aux: DisplayPort AUX channel
  */
-void drm_dp_aux_unregister_i2c_bus(struct drm_dp_aux *aux)
+void drm_dp_aux_unregister(struct drm_dp_aux *aux)
 {
 	i2c_del_adapter(&aux->ddc);
 }
-EXPORT_SYMBOL(drm_dp_aux_unregister_i2c_bus);
+EXPORT_SYMBOL(drm_dp_aux_unregister);
diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c
index d4e3f9d9370f..dfa9769b26b5 100644
--- a/drivers/gpu/drm/drm_edid.c
+++ b/drivers/gpu/drm/drm_edid.c
@@ -70,6 +70,8 @@
 #define EDID_QUIRK_FORCE_REDUCED_BLANKING	(1 << 7)
 /* Force 8bpc */
 #define EDID_QUIRK_FORCE_8BPC			(1 << 8)
+/* Force 12bpc */
+#define EDID_QUIRK_FORCE_12BPC			(1 << 9)
 
 struct detailed_mode_closure {
 	struct drm_connector *connector;
@@ -125,6 +127,9 @@ static struct edid_quirk {
 	{ "SAM", 596, EDID_QUIRK_PREFER_LARGE_60 },
 	{ "SAM", 638, EDID_QUIRK_PREFER_LARGE_60 },
 
+	/* Sony PVM-2541A does up to 12 bpc, but only reports max 8 bpc */
+	{ "SNY", 0x2541, EDID_QUIRK_FORCE_12BPC },
+
 	/* ViewSonic VA2026w */
 	{ "VSC", 5020, EDID_QUIRK_FORCE_REDUCED_BLANKING },
 
@@ -984,9 +989,13 @@ static const u8 edid_header[] = {
 	0x00, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x00
 };
 
- /*
- * Sanity check the header of the base EDID block.  Return 8 if the header
- * is perfect, down to 0 if it's totally wrong.
+/**
+ * drm_edid_header_is_valid - sanity check the header of the base EDID block
+ * @raw_edid: pointer to raw base EDID block
+ *
+ * Sanity check the header of the base EDID block.
+ *
+ * Return: 8 if the header is perfect, down to 0 if it's totally wrong.
  */
 int drm_edid_header_is_valid(const u8 *raw_edid)
 {
@@ -1005,9 +1014,16 @@ module_param_named(edid_fixup, edid_fixup, int, 0400);
 MODULE_PARM_DESC(edid_fixup,
 		 "Minimum number of valid EDID header bytes (0-8, default 6)");
 
-/*
- * Sanity check the EDID block (base or extension).  Return 0 if the block
- * doesn't check out, or 1 if it's valid.
+/**
+ * drm_edid_block_valid - Sanity check the EDID block (base or extension)
+ * @raw_edid: pointer to raw EDID block
+ * @block: type of block to validate (0 for base, extension otherwise)
+ * @print_bad_edid: if true, dump bad EDID blocks to the console
+ *
+ * Validate a base or extension EDID block and optionally dump bad blocks to
+ * the console.
+ *
+ * Return: True if the block is valid, false otherwise.
  */
 bool drm_edid_block_valid(u8 *raw_edid, int block, bool print_bad_edid)
 {
@@ -1077,6 +1093,8 @@ EXPORT_SYMBOL(drm_edid_block_valid);
  * @edid: EDID data
  *
  * Sanity-check an entire EDID record (including extensions)
+ *
+ * Return: True if the EDID data is valid, false otherwise.
  */
 bool drm_edid_is_valid(struct edid *edid)
 {
@@ -1096,18 +1114,15 @@ EXPORT_SYMBOL(drm_edid_is_valid);
 
 #define DDC_SEGMENT_ADDR 0x30
 /**
- * Get EDID information via I2C.
- *
- * @adapter : i2c device adaptor
+ * drm_do_probe_ddc_edid() - get EDID information via I2C
+ * @adapter: I2C device adaptor
  * @buf: EDID data buffer to be filled
  * @block: 128 byte EDID block to start fetching from
  * @len: EDID data buffer length to fetch
  *
- * Returns:
- *
- * 0 on success or -1 on failure.
+ * Try to fetch EDID information by calling I2C driver functions.
  *
- * Try to fetch EDID information by calling i2c driver function.
+ * Return: 0 on success or -1 on failure.
  */
 static int
 drm_do_probe_ddc_edid(struct i2c_adapter *adapter, unsigned char *buf,
@@ -1118,7 +1133,8 @@ drm_do_probe_ddc_edid(struct i2c_adapter *adapter, unsigned char *buf,
 	unsigned char xfers = segment ? 3 : 2;
 	int ret, retries = 5;
 
-	/* The core i2c driver will automatically retry the transfer if the
+	/*
+	 * The core I2C driver will automatically retry the transfer if the
 	 * adapter reports EAGAIN. However, we find that bit-banging transfers
 	 * are susceptible to errors under a heavily loaded machine and
 	 * generate spurious NAKs and timeouts. Retrying the transfer
@@ -1144,10 +1160,10 @@ drm_do_probe_ddc_edid(struct i2c_adapter *adapter, unsigned char *buf,
 			}
 		};
 
-	/*
-	 * Avoid sending the segment addr to not upset non-compliant ddc
-	 * monitors.
-	 */
+		/*
+		 * Avoid sending the segment addr to not upset non-compliant
+		 * DDC monitors.
+		 */
 		ret = i2c_transfer(adapter, &msgs[3 - xfers], xfers);
 
 		if (ret == -ENXIO) {
@@ -1216,7 +1232,7 @@ drm_do_get_edid(struct drm_connector *connector, struct i2c_adapter *adapter)
 		if (i == 4 && print_bad_edid) {
 			dev_warn(connector->dev->dev,
 			 "%s: Ignoring invalid EDID block %d.\n",
-			 drm_get_connector_name(connector), j);
+			 connector->name, j);
 
 			connector->bad_edid_counter++;
 		}
@@ -1236,7 +1252,7 @@ drm_do_get_edid(struct drm_connector *connector, struct i2c_adapter *adapter)
 carp:
 	if (print_bad_edid) {
 		dev_warn(connector->dev->dev, "%s: EDID block %d invalid.\n",
-			 drm_get_connector_name(connector), j);
+			 connector->name, j);
 	}
 	connector->bad_edid_counter++;
 
@@ -1246,12 +1262,10 @@ out:
 }
 
 /**
- * Probe DDC presence.
- * @adapter: i2c adapter to probe
+ * drm_probe_ddc() - probe DDC presence
+ * @adapter: I2C adapter to probe
  *
- * Returns:
- *
- * 1 on success
+ * Return: True on success, false on failure.
  */
 bool
 drm_probe_ddc(struct i2c_adapter *adapter)
@@ -1265,12 +1279,12 @@ EXPORT_SYMBOL(drm_probe_ddc);
 /**
  * drm_get_edid - get EDID data, if available
  * @connector: connector we're probing
- * @adapter: i2c adapter to use for DDC
+ * @adapter: I2C adapter to use for DDC
  *
- * Poke the given i2c channel to grab EDID data if possible.  If found,
+ * Poke the given I2C channel to grab EDID data if possible.  If found,
  * attach it to the connector.
  *
- * Return edid data or NULL if we couldn't find any.
+ * Return: Pointer to valid EDID or NULL if we couldn't find any.
  */
 struct edid *drm_get_edid(struct drm_connector *connector,
 			  struct i2c_adapter *adapter)
@@ -1288,7 +1302,7 @@ EXPORT_SYMBOL(drm_get_edid);
  * drm_edid_duplicate - duplicate an EDID and the extensions
  * @edid: EDID to duplicate
  *
- * Return duplicate edid or NULL on allocation failure.
+ * Return: Pointer to duplicated EDID or NULL on allocation failure.
  */
 struct edid *drm_edid_duplicate(const struct edid *edid)
 {
@@ -1411,7 +1425,8 @@ mode_is_rb(const struct drm_display_mode *mode)
  * @rb: Mode reduced-blanking-ness
  *
  * Walk the DMT mode list looking for a match for the given parameters.
- * Return a newly allocated copy of the mode, or NULL if not found.
+ *
+ * Return: A newly allocated copy of the mode, or NULL if not found.
  */
 struct drm_display_mode *drm_mode_find_dmt(struct drm_device *dev,
 					   int hsize, int vsize, int fresh,
@@ -1595,14 +1610,13 @@ bad_std_timing(u8 a, u8 b)
  * @connector: connector of for the EDID block
  * @edid: EDID block to scan
  * @t: standard timing params
- * @revision: standard timing level
  *
  * Take the standard timing params (in this case width, aspect, and refresh)
  * and convert them into a real mode using CVT/GTF/DMT.
  */
 static struct drm_display_mode *
 drm_mode_std(struct drm_connector *connector, struct edid *edid,
-	     struct std_timing *t, int revision)
+	     struct std_timing *t)
 {
 	struct drm_device *dev = connector->dev;
 	struct drm_display_mode *m, *mode = NULL;
@@ -1623,7 +1637,7 @@ drm_mode_std(struct drm_connector *connector, struct edid *edid,
 	vrefresh_rate = vfreq + 60;
 	/* the vdisplay is calculated based on the aspect ratio */
 	if (aspect_ratio == 0) {
-		if (revision < 3)
+		if (edid->revision < 3)
 			vsize = hsize;
 		else
 			vsize = (hsize * 10) / 16;
@@ -2140,7 +2154,7 @@ do_established_modes(struct detailed_timing *timing, void *c)
 
 /**
  * add_established_modes - get est. modes from EDID and add them
- * @connector: connector of for the EDID block
+ * @connector: connector to add mode(s) to
  * @edid: EDID block to scan
  *
  * Each EDID block contains a bitmap of the supported "established modes" list
@@ -2191,8 +2205,7 @@ do_standard_modes(struct detailed_timing *timing, void *c)
 			struct drm_display_mode *newmode;
 
 			std = &data->data.timings[i];
-			newmode = drm_mode_std(connector, edid, std,
-					       edid->revision);
+			newmode = drm_mode_std(connector, edid, std);
 			if (newmode) {
 				drm_mode_probed_add(connector, newmode);
 				closure->modes++;
@@ -2203,7 +2216,7 @@ do_standard_modes(struct detailed_timing *timing, void *c)
 
 /**
  * add_standard_modes - get std. modes from EDID and add them
- * @connector: connector of for the EDID block
+ * @connector: connector to add mode(s) to
  * @edid: EDID block to scan
  *
  * Standard modes can be calculated using the appropriate standard (DMT,
@@ -2221,8 +2234,7 @@ add_standard_modes(struct drm_connector *connector, struct edid *edid)
 		struct drm_display_mode *newmode;
 
 		newmode = drm_mode_std(connector, edid,
-				       &edid->standard_timings[i],
-				       edid->revision);
+				       &edid->standard_timings[i]);
 		if (newmode) {
 			drm_mode_probed_add(connector, newmode);
 			modes++;
@@ -2425,7 +2437,7 @@ cea_mode_alternate_clock(const struct drm_display_mode *cea_mode)
  * drm_match_cea_mode - look for a CEA mode matching given mode
  * @to_match: display mode
  *
- * Returns the CEA Video ID (VIC) of the mode or 0 if it isn't a CEA-861
+ * Return: The CEA Video ID (VIC) of the mode or 0 if it isn't a CEA-861
  * mode.
  */
 u8 drm_match_cea_mode(const struct drm_display_mode *to_match)
@@ -2452,6 +2464,22 @@ u8 drm_match_cea_mode(const struct drm_display_mode *to_match)
 }
 EXPORT_SYMBOL(drm_match_cea_mode);
 
+/**
+ * drm_get_cea_aspect_ratio - get the picture aspect ratio corresponding to
+ * the input VIC from the CEA mode list
+ * @video_code: ID given to each of the CEA modes
+ *
+ * Returns picture aspect ratio
+ */
+enum hdmi_picture_aspect drm_get_cea_aspect_ratio(const u8 video_code)
+{
+	/* return picture aspect ratio for video_code - 1 to access the
+	 * right array element
+	*/
+	return edid_cea_modes[video_code-1].picture_aspect_ratio;
+}
+EXPORT_SYMBOL(drm_get_cea_aspect_ratio);
+
 /*
  * Calculate the alternate clock for HDMI modes (those from the HDMI vendor
  * specific block).
@@ -3023,11 +3051,9 @@ monitor_name(struct detailed_timing *t, void *data)
  * @connector: connector corresponding to the HDMI/DP sink
  * @edid: EDID to parse
  *
- * Fill the ELD (EDID-Like Data) buffer for passing to the audio driver.
- * Some ELD fields are left to the graphics driver caller:
- * - Conn_Type
- * - HDCP
- * - Port_ID
+ * Fill the ELD (EDID-Like Data) buffer for passing to the audio driver. The
+ * Conn_Type, HDCP and Port_ID ELD fields are left for the graphics driver to
+ * fill in.
  */
 void drm_edid_to_eld(struct drm_connector *connector, struct edid *edid)
 {
@@ -3111,9 +3137,10 @@ EXPORT_SYMBOL(drm_edid_to_eld);
  * @sads: pointer that will be set to the extracted SADs
  *
  * Looks for CEA EDID block and extracts SADs (Short Audio Descriptors) from it.
- * Note: returned pointer needs to be kfreed
  *
- * Return number of found SADs or negative number on error.
+ * Note: The returned pointer needs to be freed using kfree().
+ *
+ * Return: The number of found SADs or negative number on error.
  */
 int drm_edid_to_sad(struct edid *edid, struct cea_sad **sads)
 {
@@ -3170,9 +3197,11 @@ EXPORT_SYMBOL(drm_edid_to_sad);
  * @sadb: pointer to the speaker block
  *
  * Looks for CEA EDID block and extracts the Speaker Allocation Data Block from it.
- * Note: returned pointer needs to be kfreed
  *
- * Return number of found Speaker Allocation Blocks or negative number on error.
+ * Note: The returned pointer needs to be freed using kfree().
+ *
+ * Return: The number of found Speaker Allocation Blocks or negative number on
+ * error.
  */
 int drm_edid_to_speaker_allocation(struct edid *edid, u8 **sadb)
 {
@@ -3204,10 +3233,9 @@ int drm_edid_to_speaker_allocation(struct edid *edid, u8 **sadb)
 
 			/* Speaker Allocation Data Block */
 			if (dbl == 3) {
-				*sadb = kmalloc(dbl, GFP_KERNEL);
+				*sadb = kmemdup(&db[1], dbl, GFP_KERNEL);
 				if (!*sadb)
 					return -ENOMEM;
-				memcpy(*sadb, &db[1], dbl);
 				count = dbl;
 				break;
 			}
@@ -3219,9 +3247,12 @@ int drm_edid_to_speaker_allocation(struct edid *edid, u8 **sadb)
 EXPORT_SYMBOL(drm_edid_to_speaker_allocation);
 
 /**
- * drm_av_sync_delay - HDMI/DP sink audio-video sync delay in millisecond
+ * drm_av_sync_delay - compute the HDMI/DP sink audio-video sync delay
  * @connector: connector associated with the HDMI/DP sink
  * @mode: the display mode
+ *
+ * Return: The HDMI/DP sink's audio-video sync delay in milliseconds or 0 if
+ * the sink doesn't support audio or video.
  */
 int drm_av_sync_delay(struct drm_connector *connector,
 		      struct drm_display_mode *mode)
@@ -3263,6 +3294,9 @@ EXPORT_SYMBOL(drm_av_sync_delay);
  *
  * It's possible for one encoder to be associated with multiple HDMI/DP sinks.
  * The policy is now hard coded to simply use the first HDMI/DP sink's ELD.
+ *
+ * Return: The connector associated with the first HDMI/DP sink that has ELD
+ * attached to it.
  */
 struct drm_connector *drm_select_eld(struct drm_encoder *encoder,
 				     struct drm_display_mode *mode)
@@ -3270,6 +3304,8 @@ struct drm_connector *drm_select_eld(struct drm_encoder *encoder,
 	struct drm_connector *connector;
 	struct drm_device *dev = encoder->dev;
 
+	WARN_ON(!mutex_is_locked(&dev->mode_config.mutex));
+
 	list_for_each_entry(connector, &dev->mode_config.connector_list, head)
 		if (connector->encoder == encoder && connector->eld[0])
 			return connector;
@@ -3279,11 +3315,12 @@ struct drm_connector *drm_select_eld(struct drm_encoder *encoder,
 EXPORT_SYMBOL(drm_select_eld);
 
 /**
- * drm_detect_hdmi_monitor - detect whether monitor is hdmi.
+ * drm_detect_hdmi_monitor - detect whether monitor is HDMI
  * @edid: monitor EDID information
  *
  * Parse the CEA extension according to CEA-861-B.
- * Return true if HDMI, false if not or unknown.
+ *
+ * Return: True if the monitor is HDMI, false if not or unknown.
  */
 bool drm_detect_hdmi_monitor(struct edid *edid)
 {
@@ -3321,6 +3358,7 @@ EXPORT_SYMBOL(drm_detect_hdmi_monitor);
  * audio format, assume at least 'basic audio' support, even if 'basic
  * audio' is not defined in EDID.
  *
+ * Return: True if the monitor supports audio, false otherwise.
  */
 bool drm_detect_monitor_audio(struct edid *edid)
 {
@@ -3364,6 +3402,8 @@ EXPORT_SYMBOL(drm_detect_monitor_audio);
  * Check whether the monitor reports the RGB quantization range selection
  * as supported. The AVI infoframe can then be used to inform the monitor
  * which quantization range (full or limited) is used.
+ *
+ * Return: True if the RGB quantization range is selectable, false otherwise.
  */
 bool drm_rgb_quant_range_selectable(struct edid *edid)
 {
@@ -3390,16 +3430,119 @@ bool drm_rgb_quant_range_selectable(struct edid *edid)
 EXPORT_SYMBOL(drm_rgb_quant_range_selectable);
 
 /**
+ * drm_assign_hdmi_deep_color_info - detect whether monitor supports
+ * hdmi deep color modes and update drm_display_info if so.
+ *
+ * @edid: monitor EDID information
+ * @info: Updated with maximum supported deep color bpc and color format
+ *        if deep color supported.
+ *
+ * Parse the CEA extension according to CEA-861-B.
+ * Return true if HDMI deep color supported, false if not or unknown.
+ */
+static bool drm_assign_hdmi_deep_color_info(struct edid *edid,
+                                            struct drm_display_info *info,
+                                            struct drm_connector *connector)
+{
+	u8 *edid_ext, *hdmi;
+	int i;
+	int start_offset, end_offset;
+	unsigned int dc_bpc = 0;
+
+	edid_ext = drm_find_cea_extension(edid);
+	if (!edid_ext)
+		return false;
+
+	if (cea_db_offsets(edid_ext, &start_offset, &end_offset))
+		return false;
+
+	/*
+	 * Because HDMI identifier is in Vendor Specific Block,
+	 * search it from all data blocks of CEA extension.
+	 */
+	for_each_cea_db(edid_ext, i, start_offset, end_offset) {
+		if (cea_db_is_hdmi_vsdb(&edid_ext[i])) {
+			/* HDMI supports at least 8 bpc */
+			info->bpc = 8;
+
+			hdmi = &edid_ext[i];
+			if (cea_db_payload_len(hdmi) < 6)
+				return false;
+
+			if (hdmi[6] & DRM_EDID_HDMI_DC_30) {
+				dc_bpc = 10;
+				info->edid_hdmi_dc_modes |= DRM_EDID_HDMI_DC_30;
+				DRM_DEBUG("%s: HDMI sink does deep color 30.\n",
+						  connector->name);
+			}
+
+			if (hdmi[6] & DRM_EDID_HDMI_DC_36) {
+				dc_bpc = 12;
+				info->edid_hdmi_dc_modes |= DRM_EDID_HDMI_DC_36;
+				DRM_DEBUG("%s: HDMI sink does deep color 36.\n",
+						  connector->name);
+			}
+
+			if (hdmi[6] & DRM_EDID_HDMI_DC_48) {
+				dc_bpc = 16;
+				info->edid_hdmi_dc_modes |= DRM_EDID_HDMI_DC_48;
+				DRM_DEBUG("%s: HDMI sink does deep color 48.\n",
+						  connector->name);
+			}
+
+			if (dc_bpc > 0) {
+				DRM_DEBUG("%s: Assigning HDMI sink color depth as %d bpc.\n",
+						  connector->name, dc_bpc);
+				info->bpc = dc_bpc;
+
+				/*
+				 * Deep color support mandates RGB444 support for all video
+				 * modes and forbids YCRCB422 support for all video modes per
+				 * HDMI 1.3 spec.
+				 */
+				info->color_formats = DRM_COLOR_FORMAT_RGB444;
+
+				/* YCRCB444 is optional according to spec. */
+				if (hdmi[6] & DRM_EDID_HDMI_DC_Y444) {
+					info->color_formats |= DRM_COLOR_FORMAT_YCRCB444;
+					DRM_DEBUG("%s: HDMI sink does YCRCB444 in deep color.\n",
+							  connector->name);
+				}
+
+				/*
+				 * Spec says that if any deep color mode is supported at all,
+				 * then deep color 36 bit must be supported.
+				 */
+				if (!(hdmi[6] & DRM_EDID_HDMI_DC_36)) {
+					DRM_DEBUG("%s: HDMI sink should do DC_36, but does not!\n",
+							  connector->name);
+				}
+
+				return true;
+			}
+			else {
+				DRM_DEBUG("%s: No deep color support on this HDMI sink.\n",
+						  connector->name);
+			}
+		}
+	}
+
+	return false;
+}
+
+/**
  * drm_add_display_info - pull display info out if present
  * @edid: EDID data
  * @info: display info (attached to connector)
+ * @connector: connector whose edid is used to build display info
  *
  * Grab any available display info and stuff it into the drm_display_info
  * structure that's part of the connector.  Useful for tracking bpp and
  * color spaces.
  */
 static void drm_add_display_info(struct edid *edid,
-				 struct drm_display_info *info)
+                                 struct drm_display_info *info,
+                                 struct drm_connector *connector)
 {
 	u8 *edid_ext;
 
@@ -3429,6 +3572,9 @@ static void drm_add_display_info(struct edid *edid,
 			info->color_formats |= DRM_COLOR_FORMAT_YCRCB422;
 	}
 
+	/* HDMI deep color modes supported? Assign to info, if so */
+	drm_assign_hdmi_deep_color_info(edid, info, connector);
+
 	/* Only defined for 1.4 with digital displays */
 	if (edid->revision < 4)
 		return;
@@ -3458,6 +3604,9 @@ static void drm_add_display_info(struct edid *edid,
 		break;
 	}
 
+	DRM_DEBUG("%s: Assigning EDID-1.4 digital sink color depth as %d bpc.\n",
+			  connector->name, info->bpc);
+
 	info->color_formats |= DRM_COLOR_FORMAT_RGB444;
 	if (edid->features & DRM_EDID_FEATURE_RGB_YCRCB444)
 		info->color_formats |= DRM_COLOR_FORMAT_YCRCB444;
@@ -3468,11 +3617,11 @@ static void drm_add_display_info(struct edid *edid,
 /**
  * drm_add_edid_modes - add modes from EDID data, if available
  * @connector: connector we're probing
- * @edid: edid data
+ * @edid: EDID data
  *
  * Add the specified modes to the connector's mode list.
  *
- * Return number of modes added or 0 if we couldn't find any.
+ * Return: The number of modes added or 0 if we couldn't find any.
  */
 int drm_add_edid_modes(struct drm_connector *connector, struct edid *edid)
 {
@@ -3484,7 +3633,7 @@ int drm_add_edid_modes(struct drm_connector *connector, struct edid *edid)
 	}
 	if (!drm_edid_is_valid(edid)) {
 		dev_warn(connector->dev->dev, "%s: EDID invalid.\n",
-			 drm_get_connector_name(connector));
+			 connector->name);
 		return 0;
 	}
 
@@ -3516,11 +3665,14 @@ int drm_add_edid_modes(struct drm_connector *connector, struct edid *edid)
 	if (quirks & (EDID_QUIRK_PREFER_LARGE_60 | EDID_QUIRK_PREFER_LARGE_75))
 		edid_fixup_preferred(connector, quirks);
 
-	drm_add_display_info(edid, &connector->display_info);
+	drm_add_display_info(edid, &connector->display_info, connector);
 
 	if (quirks & EDID_QUIRK_FORCE_8BPC)
 		connector->display_info.bpc = 8;
 
+	if (quirks & EDID_QUIRK_FORCE_12BPC)
+		connector->display_info.bpc = 12;
+
 	return num_modes;
 }
 EXPORT_SYMBOL(drm_add_edid_modes);
@@ -3534,7 +3686,7 @@ EXPORT_SYMBOL(drm_add_edid_modes);
  * Add the specified modes to the connector's mode list. Only when the
  * hdisplay/vdisplay is not beyond the given limit, it will be added.
  *
- * Return number of modes added or 0 if we couldn't find any.
+ * Return: The number of modes added or 0 if we couldn't find any.
  */
 int drm_add_modes_noedid(struct drm_connector *connector,
 			int hdisplay, int vdisplay)
@@ -3573,13 +3725,22 @@ int drm_add_modes_noedid(struct drm_connector *connector,
 }
 EXPORT_SYMBOL(drm_add_modes_noedid);
 
+/**
+ * drm_set_preferred_mode - Sets the preferred mode of a connector
+ * @connector: connector whose mode list should be processed
+ * @hpref: horizontal resolution of preferred mode
+ * @vpref: vertical resolution of preferred mode
+ *
+ * Marks a mode as preferred if it matches the resolution specified by @hpref
+ * and @vpref.
+ */
 void drm_set_preferred_mode(struct drm_connector *connector,
 			   int hpref, int vpref)
 {
 	struct drm_display_mode *mode;
 
 	list_for_each_entry(mode, &connector->probed_modes, head) {
-		if (mode->hdisplay  == hpref &&
+		if (mode->hdisplay == hpref &&
 		    mode->vdisplay == vpref)
 			mode->type |= DRM_MODE_TYPE_PREFERRED;
 	}
@@ -3592,7 +3753,7 @@ EXPORT_SYMBOL(drm_set_preferred_mode);
  * @frame: HDMI AVI infoframe
  * @mode: DRM display mode
  *
- * Returns 0 on success or a negative error code on failure.
+ * Return: 0 on success or a negative error code on failure.
  */
 int
 drm_hdmi_avi_infoframe_from_display_mode(struct hdmi_avi_infoframe *frame,
@@ -3613,6 +3774,12 @@ drm_hdmi_avi_infoframe_from_display_mode(struct hdmi_avi_infoframe *frame,
 	frame->video_code = drm_match_cea_mode(mode);
 
 	frame->picture_aspect = HDMI_PICTURE_ASPECT_NONE;
+
+	/* Populate picture aspect ratio from CEA mode list */
+	if (frame->video_code > 0)
+		frame->picture_aspect = drm_get_cea_aspect_ratio(
+						frame->video_code);
+
 	frame->active_aspect = HDMI_ACTIVE_ASPECT_PICTURE;
 	frame->scan_mode = HDMI_SCAN_MODE_UNDERSCAN;
 
@@ -3657,7 +3824,7 @@ s3d_structure_from_display_mode(const struct drm_display_mode *mode)
  * 4k or stereoscopic 3D mode. So when giving any other mode as input this
  * function will return -EINVAL, error that can be safely ignored.
  *
- * Returns 0 on success or a negative error code on failure.
+ * Return: 0 on success or a negative error code on failure.
  */
 int
 drm_hdmi_vendor_infoframe_from_display_mode(struct hdmi_vendor_infoframe *frame,
diff --git a/drivers/gpu/drm/drm_edid_load.c b/drivers/gpu/drm/drm_edid_load.c
index 1b4c7a5442c5..0a235fe61c9b 100644
--- a/drivers/gpu/drm/drm_edid_load.c
+++ b/drivers/gpu/drm/drm_edid_load.c
@@ -31,8 +31,9 @@ module_param_string(edid_firmware, edid_firmware, sizeof(edid_firmware), 0644);
 MODULE_PARM_DESC(edid_firmware, "Do not probe monitor, use specified EDID blob "
 	"from built-in data or /lib/firmware instead. ");
 
-#define GENERIC_EDIDS 5
+#define GENERIC_EDIDS 6
 static const char *generic_edid_name[GENERIC_EDIDS] = {
+	"edid/800x600.bin",
 	"edid/1024x768.bin",
 	"edid/1280x1024.bin",
 	"edid/1600x1200.bin",
@@ -44,6 +45,24 @@ static const u8 generic_edid[GENERIC_EDIDS][128] = {
 	{
 	0x00, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x00,
 	0x31, 0xd8, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+	0x05, 0x16, 0x01, 0x03, 0x6d, 0x1b, 0x14, 0x78,
+	0xea, 0x5e, 0xc0, 0xa4, 0x59, 0x4a, 0x98, 0x25,
+	0x20, 0x50, 0x54, 0x01, 0x00, 0x00, 0x45, 0x40,
+	0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+	0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0xa0, 0x0f,
+	0x20, 0x00, 0x31, 0x58, 0x1c, 0x20, 0x28, 0x80,
+	0x14, 0x00, 0x15, 0xd0, 0x10, 0x00, 0x00, 0x1e,
+	0x00, 0x00, 0x00, 0xff, 0x00, 0x4c, 0x69, 0x6e,
+	0x75, 0x78, 0x20, 0x23, 0x30, 0x0a, 0x20, 0x20,
+	0x20, 0x20, 0x00, 0x00, 0x00, 0xfd, 0x00, 0x3b,
+	0x3d, 0x24, 0x26, 0x05, 0x00, 0x0a, 0x20, 0x20,
+	0x20, 0x20, 0x20, 0x20, 0x00, 0x00, 0x00, 0xfc,
+	0x00, 0x4c, 0x69, 0x6e, 0x75, 0x78, 0x20, 0x53,
+	0x56, 0x47, 0x41, 0x0a, 0x20, 0x20, 0x00, 0xc2,
+	},
+	{
+	0x00, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x00,
+	0x31, 0xd8, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
 	0x05, 0x16, 0x01, 0x03, 0x6d, 0x23, 0x1a, 0x78,
 	0xea, 0x5e, 0xc0, 0xa4, 0x59, 0x4a, 0x98, 0x25,
 	0x20, 0x50, 0x54, 0x00, 0x08, 0x00, 0x61, 0x40,
@@ -242,7 +261,7 @@ out:
 
 int drm_load_edid_firmware(struct drm_connector *connector)
 {
-	const char *connector_name = drm_get_connector_name(connector);
+	const char *connector_name = connector->name;
 	char *edidname = edid_firmware, *last, *colon;
 	int ret;
 	struct edid *edid;
diff --git a/drivers/gpu/drm/drm_fb_cma_helper.c b/drivers/gpu/drm/drm_fb_cma_helper.c
index 61b5a47ad239..f27c883be391 100644
--- a/drivers/gpu/drm/drm_fb_cma_helper.c
+++ b/drivers/gpu/drm/drm_fb_cma_helper.c
@@ -429,13 +429,8 @@ EXPORT_SYMBOL_GPL(drm_fbdev_cma_fini);
  */
 void drm_fbdev_cma_restore_mode(struct drm_fbdev_cma *fbdev_cma)
 {
-	if (fbdev_cma) {
-		struct drm_device *dev = fbdev_cma->fb_helper.dev;
-
-		drm_modeset_lock_all(dev);
-		drm_fb_helper_restore_fbdev_mode(&fbdev_cma->fb_helper);
-		drm_modeset_unlock_all(dev);
-	}
+	if (fbdev_cma)
+		drm_fb_helper_restore_fbdev_mode_unlocked(&fbdev_cma->fb_helper);
 }
 EXPORT_SYMBOL_GPL(drm_fbdev_cma_restore_mode);
 
diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index 04d3fd3658f3..d5d8cea1a679 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -45,13 +45,13 @@ static LIST_HEAD(kernel_fb_helper_list);
  * DOC: fbdev helpers
  *
  * The fb helper functions are useful to provide an fbdev on top of a drm kernel
- * mode setting driver. They can be used mostly independantely from the crtc
+ * mode setting driver. They can be used mostly independently from the crtc
  * helper functions used by many drivers to implement the kernel mode setting
  * interfaces.
  *
  * Initialization is done as a three-step process with drm_fb_helper_init(),
  * drm_fb_helper_single_add_all_connectors() and drm_fb_helper_initial_config().
- * Drivers with fancier requirements than the default beheviour can override the
+ * Drivers with fancier requirements than the default behaviour can override the
  * second step with their own code.  Teardown is done with drm_fb_helper_fini().
  *
  * At runtime drivers should restore the fbdev console by calling
@@ -59,7 +59,7 @@ static LIST_HEAD(kernel_fb_helper_list);
  * should also notify the fb helper code from updates to the output
  * configuration by calling drm_fb_helper_hotplug_event(). For easier
  * integration with the output polling code in drm_crtc_helper.c the modeset
- * code proves a ->output_poll_changed callback.
+ * code provides a ->output_poll_changed callback.
  *
  * All other functions exported by the fb helper library can be used to
  * implement the fbdev driver interface by the driver.
@@ -120,7 +120,7 @@ static int drm_fb_helper_parse_command_line(struct drm_fb_helper *fb_helper)
 		mode = &fb_helper_conn->cmdline_mode;
 
 		/* do something on return - turn off connector maybe */
-		if (fb_get_options(drm_get_connector_name(connector), &option))
+		if (fb_get_options(connector->name, &option))
 			continue;
 
 		if (drm_mode_parse_command_line_for_connector(option,
@@ -142,12 +142,12 @@ static int drm_fb_helper_parse_command_line(struct drm_fb_helper *fb_helper)
 				}
 
 				DRM_INFO("forcing %s connector %s\n",
-					 drm_get_connector_name(connector), s);
+					 connector->name, s);
 				connector->force = mode->force;
 			}
 
 			DRM_DEBUG_KMS("cmdline mode for connector %s %dx%d@%dHz%s%s%s\n",
-				      drm_get_connector_name(connector),
+				      connector->name,
 				      mode->xres, mode->yres,
 				      mode->refresh_specified ? mode->refresh : 60,
 				      mode->rb ? " reduced blanking" : "",
@@ -273,15 +273,7 @@ int drm_fb_helper_debug_leave(struct fb_info *info)
 }
 EXPORT_SYMBOL(drm_fb_helper_debug_leave);
 
-/**
- * drm_fb_helper_restore_fbdev_mode - restore fbdev configuration
- * @fb_helper: fbcon to restore
- *
- * This should be called from driver's drm ->lastclose callback
- * when implementing an fbcon on top of kms using this helper. This ensures that
- * the user isn't greeted with a black screen when e.g. X dies.
- */
-bool drm_fb_helper_restore_fbdev_mode(struct drm_fb_helper *fb_helper)
+static bool restore_fbdev_mode(struct drm_fb_helper *fb_helper)
 {
 	struct drm_device *dev = fb_helper->dev;
 	struct drm_plane *plane;
@@ -311,7 +303,40 @@ bool drm_fb_helper_restore_fbdev_mode(struct drm_fb_helper *fb_helper)
 	}
 	return error;
 }
-EXPORT_SYMBOL(drm_fb_helper_restore_fbdev_mode);
+/**
+ * drm_fb_helper_restore_fbdev_mode - restore fbdev configuration
+ * @fb_helper: fbcon to restore
+ *
+ * This should be called from driver's drm ->lastclose callback
+ * when implementing an fbcon on top of kms using this helper. This ensures that
+ * the user isn't greeted with a black screen when e.g. X dies.
+ *
+ * Use this variant if you need to bypass locking (panic), or already
+ * hold all modeset locks.  Otherwise use drm_fb_helper_restore_fbdev_mode_unlocked()
+ */
+static bool drm_fb_helper_restore_fbdev_mode(struct drm_fb_helper *fb_helper)
+{
+	return restore_fbdev_mode(fb_helper);
+}
+
+/**
+ * drm_fb_helper_restore_fbdev_mode_unlocked - restore fbdev configuration
+ * @fb_helper: fbcon to restore
+ *
+ * This should be called from driver's drm ->lastclose callback
+ * when implementing an fbcon on top of kms using this helper. This ensures that
+ * the user isn't greeted with a black screen when e.g. X dies.
+ */
+bool drm_fb_helper_restore_fbdev_mode_unlocked(struct drm_fb_helper *fb_helper)
+{
+	struct drm_device *dev = fb_helper->dev;
+	bool ret;
+	drm_modeset_lock_all(dev);
+	ret = restore_fbdev_mode(fb_helper);
+	drm_modeset_unlock_all(dev);
+	return ret;
+}
+EXPORT_SYMBOL(drm_fb_helper_restore_fbdev_mode_unlocked);
 
 /*
  * restore fbcon display for all kms driver's using this helper, used for sysrq
@@ -326,12 +351,25 @@ static bool drm_fb_helper_force_kernel_mode(void)
 		return false;
 
 	list_for_each_entry(helper, &kernel_fb_helper_list, kernel_fb_list) {
-		if (helper->dev->switch_power_state == DRM_SWITCH_POWER_OFF)
+		struct drm_device *dev = helper->dev;
+
+		if (dev->switch_power_state == DRM_SWITCH_POWER_OFF)
+			continue;
+
+		/* NOTE: we use lockless flag below to avoid grabbing other
+		 * modeset locks.  So just trylock the underlying mutex
+		 * directly:
+		 */
+		if (!mutex_trylock(&dev->mode_config.mutex)) {
+			error = true;
 			continue;
+		}
 
 		ret = drm_fb_helper_restore_fbdev_mode(helper);
 		if (ret)
 			error = true;
+
+		mutex_unlock(&dev->mode_config.mutex);
 	}
 	return error;
 }
@@ -811,7 +849,6 @@ EXPORT_SYMBOL(drm_fb_helper_check_var);
 int drm_fb_helper_set_par(struct fb_info *info)
 {
 	struct drm_fb_helper *fb_helper = info->par;
-	struct drm_device *dev = fb_helper->dev;
 	struct fb_var_screeninfo *var = &info->var;
 
 	if (var->pixclock != 0) {
@@ -819,9 +856,7 @@ int drm_fb_helper_set_par(struct fb_info *info)
 		return -EINVAL;
 	}
 
-	drm_modeset_lock_all(dev);
-	drm_fb_helper_restore_fbdev_mode(fb_helper);
-	drm_modeset_unlock_all(dev);
+	drm_fb_helper_restore_fbdev_mode_unlocked(fb_helper);
 
 	if (fb_helper->delayed_hotplug) {
 		fb_helper->delayed_hotplug = false;
diff --git a/drivers/gpu/drm/drm_fops.c b/drivers/gpu/drm/drm_fops.c
index e1eba0b7cd45..021fe5d11df5 100644
--- a/drivers/gpu/drm/drm_fops.c
+++ b/drivers/gpu/drm/drm_fops.c
@@ -43,8 +43,7 @@
 DEFINE_MUTEX(drm_global_mutex);
 EXPORT_SYMBOL(drm_global_mutex);
 
-static int drm_open_helper(struct inode *inode, struct file *filp,
-			   struct drm_minor *minor);
+static int drm_open_helper(struct file *filp, struct drm_minor *minor);
 
 static int drm_setup(struct drm_device * dev)
 {
@@ -95,7 +94,7 @@ int drm_open(struct inode *inode, struct file *filp)
 	/* share address_space across all char-devs of a single device */
 	filp->f_mapping = dev->anon_inode->i_mapping;
 
-	retcode = drm_open_helper(inode, filp, minor);
+	retcode = drm_open_helper(filp, minor);
 	if (retcode)
 		goto err_undo;
 	if (need_setup) {
@@ -171,7 +170,6 @@ static int drm_cpu_valid(void)
 /**
  * Called whenever a process opens /dev/drm.
  *
- * \param inode device inode.
  * \param filp file pointer.
  * \param minor acquired minor-object.
  * \return zero on success or a negative number on failure.
@@ -179,8 +177,7 @@ static int drm_cpu_valid(void)
  * Creates and initializes a drm_file structure for the file private data in \p
  * filp and add it into the double linked list in \p dev.
  */
-static int drm_open_helper(struct inode *inode, struct file *filp,
-			   struct drm_minor *minor)
+static int drm_open_helper(struct file *filp, struct drm_minor *minor)
 {
 	struct drm_device *dev = minor->dev;
 	struct drm_file *priv;
diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
index 9909bef59800..f7d71190aad5 100644
--- a/drivers/gpu/drm/drm_gem.c
+++ b/drivers/gpu/drm/drm_gem.c
@@ -474,21 +474,10 @@ struct page **drm_gem_get_pages(struct drm_gem_object *obj, gfp_t gfpmask)
 			goto fail;
 		pages[i] = p;
 
-		/* There is a hypothetical issue w/ drivers that require
-		 * buffer memory in the low 4GB.. if the pages are un-
-		 * pinned, and swapped out, they can end up swapped back
-		 * in above 4GB.  If pages are already in memory, then
-		 * shmem_read_mapping_page_gfp will ignore the gfpmask,
-		 * even if the already in-memory page disobeys the mask.
-		 *
-		 * It is only a theoretical issue today, because none of
-		 * the devices with this limitation can be populated with
-		 * enough memory to trigger the issue.  But this BUG_ON()
-		 * is here as a reminder in case the problem with
-		 * shmem_read_mapping_page_gfp() isn't solved by the time
-		 * it does become a real issue.
-		 *
-		 * See this thread: http://lkml.org/lkml/2011/7/11/238
+		/* Make sure shmem keeps __GFP_DMA32 allocated pages in the
+		 * correct region during swapin. Note that this requires
+		 * __GFP_DMA32 to be set in mapping_gfp_mask(inode->i_mapping)
+		 * so shmem can relocate pages during swapin if required.
 		 */
 		BUG_ON((gfpmask & __GFP_DMA32) &&
 				(page_to_pfn(p) >= 0x00100000UL));
diff --git a/drivers/gpu/drm/drm_info.c b/drivers/gpu/drm/drm_info.c
index 7473035dd28b..86feedd5e6f6 100644
--- a/drivers/gpu/drm/drm_info.c
+++ b/drivers/gpu/drm/drm_info.c
@@ -47,18 +47,16 @@ int drm_name_info(struct seq_file *m, void *data)
 	struct drm_minor *minor = node->minor;
 	struct drm_device *dev = minor->dev;
 	struct drm_master *master = minor->master;
-	const char *bus_name;
 	if (!master)
 		return 0;
 
-	bus_name = dev->driver->bus->get_name(dev);
 	if (master->unique) {
 		seq_printf(m, "%s %s %s\n",
-			   bus_name,
+			   dev->driver->name,
 			   dev_name(dev->dev), master->unique);
 	} else {
 		seq_printf(m, "%s %s\n",
-			   bus_name, dev_name(dev->dev));
+			   dev->driver->name, dev_name(dev->dev));
 	}
 	return 0;
 }
diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
index 93a42040bedb..69c61f392e66 100644
--- a/drivers/gpu/drm/drm_ioctl.c
+++ b/drivers/gpu/drm/drm_ioctl.c
@@ -72,9 +72,6 @@ static void
 drm_unset_busid(struct drm_device *dev,
 		struct drm_master *master)
 {
-	kfree(dev->devname);
-	dev->devname = NULL;
-
 	kfree(master->unique);
 	master->unique = NULL;
 	master->unique_len = 0;
@@ -93,7 +90,8 @@ drm_unset_busid(struct drm_device *dev,
  * Copies the bus id from userspace into drm_device::unique, and verifies that
  * it matches the device this DRM is attached to (EINVAL otherwise).  Deprecated
  * in interface version 1.1 and will return EBUSY when setversion has requested
- * version 1.1 or greater.
+ * version 1.1 or greater. Also note that KMS is all version 1.1 and later and
+ * UMS was only ever supported on pci devices.
  */
 int drm_setunique(struct drm_device *dev, void *data,
 		  struct drm_file *file_priv)
@@ -108,10 +106,13 @@ int drm_setunique(struct drm_device *dev, void *data,
 	if (!u->unique_len || u->unique_len > 1024)
 		return -EINVAL;
 
-	if (!dev->driver->bus->set_unique)
+	if (drm_core_check_feature(dev, DRIVER_MODESET))
+		return 0;
+
+	if (WARN_ON(!dev->pdev))
 		return -EINVAL;
 
-	ret = dev->driver->bus->set_unique(dev, master, u);
+	ret = drm_pci_set_unique(dev, master, u);
 	if (ret)
 		goto err;
 
@@ -130,13 +131,25 @@ static int drm_set_busid(struct drm_device *dev, struct drm_file *file_priv)
 	if (master->unique != NULL)
 		drm_unset_busid(dev, master);
 
-	ret = dev->driver->bus->set_busid(dev, master);
-	if (ret)
-		goto err;
+	if (dev->driver->bus && dev->driver->bus->set_busid) {
+		ret = dev->driver->bus->set_busid(dev, master);
+		if (ret) {
+			drm_unset_busid(dev, master);
+			return ret;
+		}
+	} else {
+		if (WARN(dev->unique == NULL,
+			 "No drm_bus.set_busid() implementation provided by "
+			 "%ps. Use drm_dev_set_unique() to set the unique "
+			 "name explicitly.", dev->driver))
+			return -EINVAL;
+
+		master->unique = kstrdup(dev->unique, GFP_KERNEL);
+		if (master->unique)
+			master->unique_len = strlen(dev->unique);
+	}
+
 	return 0;
-err:
-	drm_unset_busid(dev, master);
-	return ret;
 }
 
 /**
diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index ec5c3f4cdd01..0de123afdb34 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -1,6 +1,5 @@
-/**
- * \file drm_irq.c
- * IRQ support
+/*
+ * drm_irq.c IRQ and vblank support
  *
  * \author Rickard E. (Rik) Faith <faith@valinux.com>
  * \author Gareth Hughes <gareth@valinux.com>
@@ -56,33 +55,6 @@
  */
 #define DRM_REDUNDANT_VBLIRQ_THRESH_NS 1000000
 
-/**
- * Get interrupt from bus id.
- *
- * \param inode device inode.
- * \param file_priv DRM file private.
- * \param cmd command.
- * \param arg user argument, pointing to a drm_irq_busid structure.
- * \return zero on success or a negative number on failure.
- *
- * Finds the PCI device with the specified bus id and gets its IRQ number.
- * This IOCTL is deprecated, and will now return EINVAL for any busid not equal
- * to that of the device that this DRM instance attached to.
- */
-int drm_irq_by_busid(struct drm_device *dev, void *data,
-		     struct drm_file *file_priv)
-{
-	struct drm_irq_busid *p = data;
-
-	if (!dev->driver->bus->irq_by_busid)
-		return -EINVAL;
-
-	if (!drm_core_check_feature(dev, DRIVER_HAVE_IRQ))
-		return -EINVAL;
-
-	return dev->driver->bus->irq_by_busid(dev, p);
-}
-
 /*
  * Clear vblank timestamp buffer for a crtc.
  */
@@ -167,33 +139,40 @@ static void vblank_disable_and_save(struct drm_device *dev, int crtc)
 
 static void vblank_disable_fn(unsigned long arg)
 {
-	struct drm_device *dev = (struct drm_device *)arg;
+	struct drm_vblank_crtc *vblank = (void *)arg;
+	struct drm_device *dev = vblank->dev;
 	unsigned long irqflags;
-	int i;
+	int crtc = vblank->crtc;
 
 	if (!dev->vblank_disable_allowed)
 		return;
 
-	for (i = 0; i < dev->num_crtcs; i++) {
-		spin_lock_irqsave(&dev->vbl_lock, irqflags);
-		if (atomic_read(&dev->vblank[i].refcount) == 0 &&
-		    dev->vblank[i].enabled) {
-			DRM_DEBUG("disabling vblank on crtc %d\n", i);
-			vblank_disable_and_save(dev, i);
-		}
-		spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
+	spin_lock_irqsave(&dev->vbl_lock, irqflags);
+	if (atomic_read(&vblank->refcount) == 0 && vblank->enabled) {
+		DRM_DEBUG("disabling vblank on crtc %d\n", crtc);
+		vblank_disable_and_save(dev, crtc);
 	}
+	spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
 }
 
+/**
+ * drm_vblank_cleanup - cleanup vblank support
+ * @dev: DRM device
+ *
+ * This function cleans up any resources allocated in drm_vblank_init.
+ */
 void drm_vblank_cleanup(struct drm_device *dev)
 {
+	int crtc;
+
 	/* Bail if the driver didn't call drm_vblank_init() */
 	if (dev->num_crtcs == 0)
 		return;
 
-	del_timer_sync(&dev->vblank_disable_timer);
-
-	vblank_disable_fn((unsigned long)dev);
+	for (crtc = 0; crtc < dev->num_crtcs; crtc++) {
+		del_timer_sync(&dev->vblank[crtc].disable_timer);
+		vblank_disable_fn((unsigned long)&dev->vblank[crtc]);
+	}
 
 	kfree(dev->vblank);
 
@@ -201,12 +180,20 @@ void drm_vblank_cleanup(struct drm_device *dev)
 }
 EXPORT_SYMBOL(drm_vblank_cleanup);
 
+/**
+ * drm_vblank_init - initialize vblank support
+ * @dev: drm_device
+ * @num_crtcs: number of crtcs supported by @dev
+ *
+ * This function initializes vblank support for @num_crtcs display pipelines.
+ *
+ * Returns:
+ * Zero on success or a negative error code on failure.
+ */
 int drm_vblank_init(struct drm_device *dev, int num_crtcs)
 {
 	int i, ret = -ENOMEM;
 
-	setup_timer(&dev->vblank_disable_timer, vblank_disable_fn,
-		    (unsigned long)dev);
 	spin_lock_init(&dev->vbl_lock);
 	spin_lock_init(&dev->vblank_time_lock);
 
@@ -216,8 +203,13 @@ int drm_vblank_init(struct drm_device *dev, int num_crtcs)
 	if (!dev->vblank)
 		goto err;
 
-	for (i = 0; i < num_crtcs; i++)
+	for (i = 0; i < num_crtcs; i++) {
+		dev->vblank[i].dev = dev;
+		dev->vblank[i].crtc = i;
 		init_waitqueue_head(&dev->vblank[i].queue);
+		setup_timer(&dev->vblank[i].disable_timer, vblank_disable_fn,
+			    (unsigned long)&dev->vblank[i]);
+	}
 
 	DRM_INFO("Supports vblank timestamp caching Rev 2 (21.10.2013).\n");
 
@@ -261,42 +253,42 @@ static void drm_irq_vgaarb_nokms(void *cookie, bool state)
 }
 
 /**
- * Install IRQ handler.
- *
- * \param dev DRM device.
+ * drm_irq_install - install IRQ handler
+ * @dev: DRM device
+ * @irq: IRQ number to install the handler for
  *
  * Initializes the IRQ related data. Installs the handler, calling the driver
- * \c irq_preinstall() and \c irq_postinstall() functions
- * before and after the installation.
+ * irq_preinstall() and irq_postinstall() functions before and after the
+ * installation.
+ *
+ * This is the simplified helper interface provided for drivers with no special
+ * needs. Drivers which need to install interrupt handlers for multiple
+ * interrupts must instead set drm_device->irq_enabled to signal the DRM core
+ * that vblank interrupts are available.
+ *
+ * Returns:
+ * Zero on success or a negative error code on failure.
  */
-int drm_irq_install(struct drm_device *dev)
+int drm_irq_install(struct drm_device *dev, int irq)
 {
 	int ret;
 	unsigned long sh_flags = 0;
-	char *irqname;
 
 	if (!drm_core_check_feature(dev, DRIVER_HAVE_IRQ))
 		return -EINVAL;
 
-	if (drm_dev_to_irq(dev) == 0)
+	if (irq == 0)
 		return -EINVAL;
 
-	mutex_lock(&dev->struct_mutex);
-
 	/* Driver must have been initialized */
-	if (!dev->dev_private) {
-		mutex_unlock(&dev->struct_mutex);
+	if (!dev->dev_private)
 		return -EINVAL;
-	}
 
-	if (dev->irq_enabled) {
-		mutex_unlock(&dev->struct_mutex);
+	if (dev->irq_enabled)
 		return -EBUSY;
-	}
 	dev->irq_enabled = true;
-	mutex_unlock(&dev->struct_mutex);
 
-	DRM_DEBUG("irq=%d\n", drm_dev_to_irq(dev));
+	DRM_DEBUG("irq=%d\n", irq);
 
 	/* Before installing handler */
 	if (dev->driver->irq_preinstall)
@@ -306,18 +298,11 @@ int drm_irq_install(struct drm_device *dev)
 	if (drm_core_check_feature(dev, DRIVER_IRQ_SHARED))
 		sh_flags = IRQF_SHARED;
 
-	if (dev->devname)
-		irqname = dev->devname;
-	else
-		irqname = dev->driver->name;
-
-	ret = request_irq(drm_dev_to_irq(dev), dev->driver->irq_handler,
-			  sh_flags, irqname, dev);
+	ret = request_irq(irq, dev->driver->irq_handler,
+			  sh_flags, dev->driver->name, dev);
 
 	if (ret < 0) {
-		mutex_lock(&dev->struct_mutex);
 		dev->irq_enabled = false;
-		mutex_unlock(&dev->struct_mutex);
 		return ret;
 	}
 
@@ -329,12 +314,12 @@ int drm_irq_install(struct drm_device *dev)
 		ret = dev->driver->irq_postinstall(dev);
 
 	if (ret < 0) {
-		mutex_lock(&dev->struct_mutex);
 		dev->irq_enabled = false;
-		mutex_unlock(&dev->struct_mutex);
 		if (!drm_core_check_feature(dev, DRIVER_MODESET))
 			vga_client_register(dev->pdev, NULL, NULL, NULL);
-		free_irq(drm_dev_to_irq(dev), dev);
+		free_irq(irq, dev);
+	} else {
+		dev->irq = irq;
 	}
 
 	return ret;
@@ -342,11 +327,20 @@ int drm_irq_install(struct drm_device *dev)
 EXPORT_SYMBOL(drm_irq_install);
 
 /**
- * Uninstall the IRQ handler.
+ * drm_irq_uninstall - uninstall the IRQ handler
+ * @dev: DRM device
+ *
+ * Calls the driver's irq_uninstall() function and unregisters the IRQ handler.
+ * This should only be called by drivers which used drm_irq_install() to set up
+ * their interrupt handler. Other drivers must only reset
+ * drm_device->irq_enabled to false.
  *
- * \param dev DRM device.
+ * Note that for kernel modesetting drivers it is a bug if this function fails.
+ * The sanity checks are only to catch buggy user modesetting drivers which call
+ * the same function through an ioctl.
  *
- * Calls the driver's \c irq_uninstall() function, and stops the irq.
+ * Returns:
+ * Zero on success or a negative error code on failure.
  */
 int drm_irq_uninstall(struct drm_device *dev)
 {
@@ -357,10 +351,8 @@ int drm_irq_uninstall(struct drm_device *dev)
 	if (!drm_core_check_feature(dev, DRIVER_HAVE_IRQ))
 		return -EINVAL;
 
-	mutex_lock(&dev->struct_mutex);
 	irq_enabled = dev->irq_enabled;
 	dev->irq_enabled = false;
-	mutex_unlock(&dev->struct_mutex);
 
 	/*
 	 * Wake up any waiters so they don't hang.
@@ -379,7 +371,7 @@ int drm_irq_uninstall(struct drm_device *dev)
 	if (!irq_enabled)
 		return -EINVAL;
 
-	DRM_DEBUG("irq=%d\n", drm_dev_to_irq(dev));
+	DRM_DEBUG("irq=%d\n", dev->irq);
 
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		vga_client_register(dev->pdev, NULL, NULL, NULL);
@@ -387,13 +379,13 @@ int drm_irq_uninstall(struct drm_device *dev)
 	if (dev->driver->irq_uninstall)
 		dev->driver->irq_uninstall(dev);
 
-	free_irq(drm_dev_to_irq(dev), dev);
+	free_irq(dev->irq, dev);
 
 	return 0;
 }
 EXPORT_SYMBOL(drm_irq_uninstall);
 
-/**
+/*
  * IRQ control ioctl.
  *
  * \param inode device inode.
@@ -408,43 +400,52 @@ int drm_control(struct drm_device *dev, void *data,
 		struct drm_file *file_priv)
 {
 	struct drm_control *ctl = data;
+	int ret = 0, irq;
 
 	/* if we haven't irq we fallback for compatibility reasons -
 	 * this used to be a separate function in drm_dma.h
 	 */
 
+	if (!drm_core_check_feature(dev, DRIVER_HAVE_IRQ))
+		return 0;
+	if (drm_core_check_feature(dev, DRIVER_MODESET))
+		return 0;
+	/* UMS was only ever support on pci devices. */
+	if (WARN_ON(!dev->pdev))
+		return -EINVAL;
 
 	switch (ctl->func) {
 	case DRM_INST_HANDLER:
-		if (!drm_core_check_feature(dev, DRIVER_HAVE_IRQ))
-			return 0;
-		if (drm_core_check_feature(dev, DRIVER_MODESET))
-			return 0;
+		irq = dev->pdev->irq;
+
 		if (dev->if_version < DRM_IF_VERSION(1, 2) &&
-		    ctl->irq != drm_dev_to_irq(dev))
+		    ctl->irq != irq)
 			return -EINVAL;
-		return drm_irq_install(dev);
+		mutex_lock(&dev->struct_mutex);
+		ret = drm_irq_install(dev, irq);
+		mutex_unlock(&dev->struct_mutex);
+
+		return ret;
 	case DRM_UNINST_HANDLER:
-		if (!drm_core_check_feature(dev, DRIVER_HAVE_IRQ))
-			return 0;
-		if (drm_core_check_feature(dev, DRIVER_MODESET))
-			return 0;
-		return drm_irq_uninstall(dev);
+		mutex_lock(&dev->struct_mutex);
+		ret = drm_irq_uninstall(dev);
+		mutex_unlock(&dev->struct_mutex);
+
+		return ret;
 	default:
 		return -EINVAL;
 	}
 }
 
 /**
- * drm_calc_timestamping_constants - Calculate vblank timestamp constants
- *
- * @crtc drm_crtc whose timestamp constants should be updated.
- * @mode display mode containing the scanout timings
+ * drm_calc_timestamping_constants - calculate vblank timestamp constants
+ * @crtc: drm_crtc whose timestamp constants should be updated.
+ * @mode: display mode containing the scanout timings
  *
  * Calculate and store various constants which are later
  * needed by vblank and swap-completion timestamping, e.g,
  * by drm_calc_vbltimestamp_from_scanoutpos(). They are
- * derived from crtc's true scanout timing, so they take
+ * derived from CRTC's true scanout timing, so they take
  * things like panel scaling or other adjustments into account.
  */
 void drm_calc_timestamping_constants(struct drm_crtc *crtc,
@@ -489,11 +490,22 @@ void drm_calc_timestamping_constants(struct drm_crtc *crtc,
 EXPORT_SYMBOL(drm_calc_timestamping_constants);
 
 /**
- * drm_calc_vbltimestamp_from_scanoutpos - helper routine for kms
- * drivers. Implements calculation of exact vblank timestamps from
- * given drm_display_mode timings and current video scanout position
- * of a crtc. This can be called from within get_vblank_timestamp()
- * implementation of a kms driver to implement the actual timestamping.
+ * drm_calc_vbltimestamp_from_scanoutpos - precise vblank timestamp helper
+ * @dev: DRM device
+ * @crtc: Which CRTC's vblank timestamp to retrieve
+ * @max_error: Desired maximum allowable error in timestamps (nanosecs)
+ *             On return contains true maximum error of timestamp
+ * @vblank_time: Pointer to struct timeval which should receive the timestamp
+ * @flags: Flags to pass to driver:
+ *         0 = Default,
+ *         DRM_CALLED_FROM_VBLIRQ = If function is called from vbl IRQ handler
+ * @refcrtc: CRTC which defines scanout timing
+ * @mode: mode which defines the scanout timings
+ *
+ * Implements calculation of exact vblank timestamps from given drm_display_mode
+ * timings and current video scanout position of a CRTC. This can be called from
+ * within get_vblank_timestamp() implementation of a kms driver to implement the
+ * actual timestamping.
  *
  * Should return timestamps conforming to the OML_sync_control OpenML
  * extension specification. The timestamp corresponds to the end of
@@ -508,21 +520,11 @@ EXPORT_SYMBOL(drm_calc_timestamping_constants);
  * returns as no operation if a doublescan or interlaced video mode is
  * active. Higher level code is expected to handle this.
  *
- * @dev: DRM device.
- * @crtc: Which crtc's vblank timestamp to retrieve.
- * @max_error: Desired maximum allowable error in timestamps (nanosecs).
- *             On return contains true maximum error of timestamp.
- * @vblank_time: Pointer to struct timeval which should receive the timestamp.
- * @flags: Flags to pass to driver:
- *         0 = Default.
- *         DRM_CALLED_FROM_VBLIRQ = If function is called from vbl irq handler.
- * @refcrtc: drm_crtc* of crtc which defines scanout timing.
- * @mode: mode which defines the scanout timings
- *
- * Returns negative value on error, failure or if not supported in current
+ * Returns:
+ * Negative value on error, failure or if not supported in current
  * video mode:
  *
- * -EINVAL   - Invalid crtc.
+ * -EINVAL   - Invalid CRTC.
  * -EAGAIN   - Temporary unavailable, e.g., called before initial modeset.
  * -ENOTSUPP - Function not supported in current display mode.
  * -EIO      - Failed, e.g., due to failed scanout position query.
@@ -671,23 +673,23 @@ static struct timeval get_drm_timestamp(void)
 
 /**
  * drm_get_last_vbltimestamp - retrieve raw timestamp for the most recent
- * vblank interval.
- *
+ * 			       vblank interval
  * @dev: DRM device
- * @crtc: which crtc's vblank timestamp to retrieve
+ * @crtc: which CRTC's vblank timestamp to retrieve
  * @tvblank: Pointer to target struct timeval which should receive the timestamp
  * @flags: Flags to pass to driver:
- *         0 = Default.
- *         DRM_CALLED_FROM_VBLIRQ = If function is called from vbl irq handler.
+ *         0 = Default,
+ *         DRM_CALLED_FROM_VBLIRQ = If function is called from vbl IRQ handler
  *
  * Fetches the system timestamp corresponding to the time of the most recent
- * vblank interval on specified crtc. May call into kms-driver to
+ * vblank interval on specified CRTC. May call into kms-driver to
  * compute the timestamp with a high-precision GPU specific method.
  *
  * Returns zero if timestamp originates from uncorrected do_gettimeofday()
  * call, i.e., it isn't very precisely locked to the true vblank.
  *
- * Returns non-zero if timestamp is considered to be very precise.
+ * Returns:
+ * Non-zero if timestamp is considered to be very precise, zero otherwise.
  */
 u32 drm_get_last_vbltimestamp(struct drm_device *dev, int crtc,
 			      struct timeval *tvblank, unsigned flags)
@@ -722,6 +724,9 @@ EXPORT_SYMBOL(drm_get_last_vbltimestamp);
  * Fetches the "cooked" vblank count value that represents the number of
  * vblank events since the system was booted, including lost events due to
  * modesetting activity.
+ *
+ * Returns:
+ * The software vblank counter.
  */
 u32 drm_vblank_count(struct drm_device *dev, int crtc)
 {
@@ -740,8 +745,7 @@ EXPORT_SYMBOL(drm_vblank_count);
  * Fetches the "cooked" vblank count value that represents the number of
  * vblank events since the system was booted, including lost events due to
  * modesetting activity. Returns corresponding system timestamp of the time
- * of the vblank interval that corresponds to the current value vblank counter
- * value.
+ * of the vblank interval that corresponds to the current vblank counter value.
  */
 u32 drm_vblank_count_and_time(struct drm_device *dev, int crtc,
 			      struct timeval *vblanktime)
@@ -870,6 +874,42 @@ static void drm_update_vblank_count(struct drm_device *dev, int crtc)
 }
 
 /**
+ * drm_vblank_enable - enable the vblank interrupt on a CRTC
+ * @dev: DRM device
+ * @crtc: CRTC in question
+ */
+static int drm_vblank_enable(struct drm_device *dev, int crtc)
+{
+	int ret = 0;
+
+	assert_spin_locked(&dev->vbl_lock);
+
+	spin_lock(&dev->vblank_time_lock);
+
+	if (!dev->vblank[crtc].enabled) {
+		/*
+		 * Enable vblank irqs under vblank_time_lock protection.
+		 * All vblank count & timestamp updates are held off
+		 * until we are done reinitializing master counter and
+		 * timestamps. Filtercode in drm_handle_vblank() will
+		 * prevent double-accounting of same vblank interval.
+		 */
+		ret = dev->driver->enable_vblank(dev, crtc);
+		DRM_DEBUG("enabling vblank on crtc %d, ret: %d\n", crtc, ret);
+		if (ret)
+			atomic_dec(&dev->vblank[crtc].refcount);
+		else {
+			dev->vblank[crtc].enabled = true;
+			drm_update_vblank_count(dev, crtc);
+		}
+	}
+
+	spin_unlock(&dev->vblank_time_lock);
+
+	return ret;
+}
+
+/**
  * drm_vblank_get - get a reference count on vblank events
  * @dev: DRM device
  * @crtc: which CRTC to own
@@ -877,36 +917,20 @@ static void drm_update_vblank_count(struct drm_device *dev, int crtc)
  * Acquire a reference count on vblank events to avoid having them disabled
  * while in use.
  *
- * RETURNS
+ * This is the legacy version of drm_crtc_vblank_get().
+ *
+ * Returns:
  * Zero on success, nonzero on failure.
  */
 int drm_vblank_get(struct drm_device *dev, int crtc)
 {
-	unsigned long irqflags, irqflags2;
+	unsigned long irqflags;
 	int ret = 0;
 
 	spin_lock_irqsave(&dev->vbl_lock, irqflags);
 	/* Going from 0->1 means we have to enable interrupts again */
 	if (atomic_add_return(1, &dev->vblank[crtc].refcount) == 1) {
-		spin_lock_irqsave(&dev->vblank_time_lock, irqflags2);
-		if (!dev->vblank[crtc].enabled) {
-			/* Enable vblank irqs under vblank_time_lock protection.
-			 * All vblank count & timestamp updates are held off
-			 * until we are done reinitializing master counter and
-			 * timestamps. Filtercode in drm_handle_vblank() will
-			 * prevent double-accounting of same vblank interval.
-			 */
-			ret = dev->driver->enable_vblank(dev, crtc);
-			DRM_DEBUG("enabling vblank on crtc %d, ret: %d\n",
-				  crtc, ret);
-			if (ret)
-				atomic_dec(&dev->vblank[crtc].refcount);
-			else {
-				dev->vblank[crtc].enabled = true;
-				drm_update_vblank_count(dev, crtc);
-			}
-		}
-		spin_unlock_irqrestore(&dev->vblank_time_lock, irqflags2);
+		ret = drm_vblank_enable(dev, crtc);
 	} else {
 		if (!dev->vblank[crtc].enabled) {
 			atomic_dec(&dev->vblank[crtc].refcount);
@@ -920,12 +944,32 @@ int drm_vblank_get(struct drm_device *dev, int crtc)
 EXPORT_SYMBOL(drm_vblank_get);
 
 /**
+ * drm_crtc_vblank_get - get a reference count on vblank events
+ * @crtc: which CRTC to own
+ *
+ * Acquire a reference count on vblank events to avoid having them disabled
+ * while in use.
+ *
+ * This is the native kms version of drm_vblank_off().
+ *
+ * Returns:
+ * Zero on success, nonzero on failure.
+ */
+int drm_crtc_vblank_get(struct drm_crtc *crtc)
+{
+	return drm_vblank_get(crtc->dev, drm_crtc_index(crtc));
+}
+EXPORT_SYMBOL(drm_crtc_vblank_get);
+
+/**
  * drm_vblank_put - give up ownership of vblank events
  * @dev: DRM device
  * @crtc: which counter to give up
  *
  * Release ownership of a given vblank counter, turning off interrupts
  * if possible. Disable interrupts after drm_vblank_offdelay milliseconds.
+ *
+ * This is the legacy version of drm_crtc_vblank_put().
  */
 void drm_vblank_put(struct drm_device *dev, int crtc)
 {
@@ -934,17 +978,39 @@ void drm_vblank_put(struct drm_device *dev, int crtc)
 	/* Last user schedules interrupt disable */
 	if (atomic_dec_and_test(&dev->vblank[crtc].refcount) &&
 	    (drm_vblank_offdelay > 0))
-		mod_timer(&dev->vblank_disable_timer,
+		mod_timer(&dev->vblank[crtc].disable_timer,
 			  jiffies + ((drm_vblank_offdelay * HZ)/1000));
 }
 EXPORT_SYMBOL(drm_vblank_put);
 
 /**
+ * drm_crtc_vblank_put - give up ownership of vblank events
+ * @crtc: which counter to give up
+ *
+ * Release ownership of a given vblank counter, turning off interrupts
+ * if possible. Disable interrupts after drm_vblank_offdelay milliseconds.
+ *
+ * This is the native kms version of drm_vblank_put().
+ */
+void drm_crtc_vblank_put(struct drm_crtc *crtc)
+{
+	drm_vblank_put(crtc->dev, drm_crtc_index(crtc));
+}
+EXPORT_SYMBOL(drm_crtc_vblank_put);
+
+/**
  * drm_vblank_off - disable vblank events on a CRTC
  * @dev: DRM device
  * @crtc: CRTC in question
  *
- * Caller must hold event lock.
+ * Drivers can use this function to shut down the vblank interrupt handling when
+ * disabling a crtc. This function ensures that the latest vblank frame count is
+ * stored so that drm_vblank_on() can restore it again.
+ *
+ * Drivers must use this function when the hardware vblank counter can get
+ * reset, e.g. when suspending.
+ *
+ * This is the legacy version of drm_crtc_vblank_off().
  */
 void drm_vblank_off(struct drm_device *dev, int crtc)
 {
@@ -978,12 +1044,87 @@ void drm_vblank_off(struct drm_device *dev, int crtc)
 EXPORT_SYMBOL(drm_vblank_off);
 
 /**
+ * drm_crtc_vblank_off - disable vblank events on a CRTC
+ * @crtc: CRTC in question
+ *
+ * Drivers can use this function to shut down the vblank interrupt handling when
+ * disabling a crtc. This function ensures that the latest vblank frame count is
+ * stored so that drm_vblank_on can restore it again.
+ *
+ * Drivers must use this function when the hardware vblank counter can get
+ * reset, e.g. when suspending.
+ *
+ * This is the native kms version of drm_vblank_off().
+ */
+void drm_crtc_vblank_off(struct drm_crtc *crtc)
+{
+	drm_vblank_off(crtc->dev, drm_crtc_index(crtc));
+}
+EXPORT_SYMBOL(drm_crtc_vblank_off);
+
+/**
+ * drm_vblank_on - enable vblank events on a CRTC
+ * @dev: DRM device
+ * @crtc: CRTC in question
+ *
+ * This functions restores the vblank interrupt state captured with
+ * drm_vblank_off() again. Note that calls to drm_vblank_on() and
+ * drm_vblank_off() can be unbalanced and so can also be unconditionaly called
+ * in driver load code to reflect the current hardware state of the crtc.
+ *
+ * This is the legacy version of drm_crtc_vblank_on().
+ */
+void drm_vblank_on(struct drm_device *dev, int crtc)
+{
+	unsigned long irqflags;
+
+	spin_lock_irqsave(&dev->vbl_lock, irqflags);
+	/* re-enable interrupts if there's are users left */
+	if (atomic_read(&dev->vblank[crtc].refcount) != 0)
+		WARN_ON(drm_vblank_enable(dev, crtc));
+	spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
+}
+EXPORT_SYMBOL(drm_vblank_on);
+
+/**
+ * drm_crtc_vblank_on - enable vblank events on a CRTC
+ * @crtc: CRTC in question
+ *
+ * This functions restores the vblank interrupt state captured with
+ * drm_vblank_off() again. Note that calls to drm_vblank_on() and
+ * drm_vblank_off() can be unbalanced and so can also be unconditionaly called
+ * in driver load code to reflect the current hardware state of the crtc.
+ *
+ * This is the native kms version of drm_vblank_on().
+ */
+void drm_crtc_vblank_on(struct drm_crtc *crtc)
+{
+	drm_vblank_on(crtc->dev, drm_crtc_index(crtc));
+}
+EXPORT_SYMBOL(drm_crtc_vblank_on);
+
+/**
  * drm_vblank_pre_modeset - account for vblanks across mode sets
  * @dev: DRM device
  * @crtc: CRTC in question
  *
  * Account for vblank events across mode setting events, which will likely
  * reset the hardware frame counter.
+ *
+ * This is done by grabbing a temporary vblank reference to ensure that the
+ * vblank interrupt keeps running across the modeset sequence. With this the
+ * software-side vblank frame counting will ensure that there are no jumps or
+ * discontinuities.
+ *
+ * Unfortunately this approach is racy and also doesn't work when the vblank
+ * interrupt stops running, e.g. across system suspend resume. It is therefore
+ * highly recommended that drivers use the newer drm_vblank_off() and
+ * drm_vblank_on() instead. drm_vblank_pre_modeset() only works correctly when
+ * using "cooked" software vblank frame counters and not relying on any hardware
+ * counters.
+ *
+ * Drivers must call drm_vblank_post_modeset() when re-enabling the same crtc
+ * again.
  */
 void drm_vblank_pre_modeset(struct drm_device *dev, int crtc)
 {
@@ -1005,6 +1146,14 @@ void drm_vblank_pre_modeset(struct drm_device *dev, int crtc)
 }
 EXPORT_SYMBOL(drm_vblank_pre_modeset);
 
+/**
+ * drm_vblank_post_modeset - undo drm_vblank_pre_modeset changes
+ * @dev: DRM device
+ * @crtc: CRTC in question
+ *
+ * This function again drops the temporary vblank reference acquired in
+ * drm_vblank_pre_modeset.
+ */
 void drm_vblank_post_modeset(struct drm_device *dev, int crtc)
 {
 	unsigned long irqflags;
@@ -1026,7 +1175,7 @@ void drm_vblank_post_modeset(struct drm_device *dev, int crtc)
 }
 EXPORT_SYMBOL(drm_vblank_post_modeset);
 
-/**
+/*
  * drm_modeset_ctl - handle vblank event counter changes across mode switch
  * @DRM_IOCTL_ARGS: standard ioctl arguments
  *
@@ -1139,7 +1288,7 @@ err_put:
 	return ret;
 }
 
-/**
+/*
  * Wait for VBLANK.
  *
  * \param inode device inode.
@@ -1150,7 +1299,7 @@ err_put:
  *
  * This function enables the vblank interrupt on the pipe requested, then
  * sleeps waiting for the requested sequence number to occur, and drops
- * the vblank interrupt refcount afterwards. (vblank irq disable follows that
+ * the vblank interrupt refcount afterwards. (vblank IRQ disable follows that
  * after a timeout with no further vblank waits scheduled).
  */
 int drm_wait_vblank(struct drm_device *dev, void *data,
@@ -1160,9 +1309,8 @@ int drm_wait_vblank(struct drm_device *dev, void *data,
 	int ret;
 	unsigned int flags, seq, crtc, high_crtc;
 
-	if (drm_core_check_feature(dev, DRIVER_HAVE_IRQ))
-		if ((!drm_dev_to_irq(dev)) || (!dev->irq_enabled))
-			return -EINVAL;
+	if (!dev->irq_enabled)
+		return -EINVAL;
 
 	if (vblwait->request.type & _DRM_VBLANK_SIGNAL)
 		return -EINVAL;
@@ -1222,6 +1370,7 @@ int drm_wait_vblank(struct drm_device *dev, void *data,
 	DRM_WAIT_ON(ret, dev->vblank[crtc].queue, 3 * HZ,
 		    (((drm_vblank_count(dev, crtc) -
 		       vblwait->request.sequence) <= (1 << 23)) ||
+		     !dev->vblank[crtc].enabled ||
 		     !dev->irq_enabled));
 
 	if (ret != -EINTR) {
diff --git a/drivers/gpu/drm/drm_mipi_dsi.c b/drivers/gpu/drm/drm_mipi_dsi.c
index 09821f46d768..e633df2f68d8 100644
--- a/drivers/gpu/drm/drm_mipi_dsi.c
+++ b/drivers/gpu/drm/drm_mipi_dsi.c
@@ -282,6 +282,14 @@ static int mipi_dsi_drv_remove(struct device *dev)
 	return drv->remove(dsi);
 }
 
+static void mipi_dsi_drv_shutdown(struct device *dev)
+{
+	struct mipi_dsi_driver *drv = to_mipi_dsi_driver(dev->driver);
+	struct mipi_dsi_device *dsi = to_mipi_dsi_device(dev);
+
+	drv->shutdown(dsi);
+}
+
 /**
  * mipi_dsi_driver_register - register a driver for DSI devices
  * @drv: DSI driver structure
@@ -293,6 +301,8 @@ int mipi_dsi_driver_register(struct mipi_dsi_driver *drv)
 		drv->driver.probe = mipi_dsi_drv_probe;
 	if (drv->remove)
 		drv->driver.remove = mipi_dsi_drv_remove;
+	if (drv->shutdown)
+		drv->driver.shutdown = mipi_dsi_drv_shutdown;
 
 	return driver_register(&drv->driver);
 }
diff --git a/drivers/gpu/drm/drm_modes.c b/drivers/gpu/drm/drm_modes.c
index 8b410576fce4..bedf1894e17e 100644
--- a/drivers/gpu/drm/drm_modes.c
+++ b/drivers/gpu/drm/drm_modes.c
@@ -1013,6 +1013,7 @@ EXPORT_SYMBOL(drm_mode_sort);
 /**
  * drm_mode_connector_list_update - update the mode list for the connector
  * @connector: the connector to update
+ * @merge_type_bits: whether to merge or overright type bits.
  *
  * This moves the modes from the @connector probed_modes list
  * to the actual mode list. It compares the probed mode against the current
@@ -1021,7 +1022,8 @@ EXPORT_SYMBOL(drm_mode_sort);
  * This is just a helper functions doesn't validate any modes itself and also
  * doesn't prune any invalid modes. Callers need to do that themselves.
  */
-void drm_mode_connector_list_update(struct drm_connector *connector)
+void drm_mode_connector_list_update(struct drm_connector *connector,
+				    bool merge_type_bits)
 {
 	struct drm_display_mode *mode;
 	struct drm_display_mode *pmode, *pt;
@@ -1039,7 +1041,10 @@ void drm_mode_connector_list_update(struct drm_connector *connector)
 				/* if equal delete the probed mode */
 				mode->status = pmode->status;
 				/* Merge type bits together */
-				mode->type |= pmode->type;
+				if (merge_type_bits)
+					mode->type |= pmode->type;
+				else
+					mode->type = pmode->type;
 				list_del(&pmode->head);
 				drm_mode_destroy(connector->dev, pmode);
 				break;
diff --git a/drivers/gpu/drm/drm_modeset_lock.c b/drivers/gpu/drm/drm_modeset_lock.c
new file mode 100644
index 000000000000..7c2497dea1e9
--- /dev/null
+++ b/drivers/gpu/drm/drm_modeset_lock.c
@@ -0,0 +1,247 @@
+/*
+ * Copyright (C) 2014 Red Hat
+ * Author: Rob Clark <robdclark@gmail.com>
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#include <drm/drmP.h>
+#include <drm/drm_crtc.h>
+#include <drm/drm_modeset_lock.h>
+
+/**
+ * DOC: kms locking
+ *
+ * As KMS moves toward more fine grained locking, and atomic ioctl where
+ * userspace can indirectly control locking order, it becomes necessary
+ * to use ww_mutex and acquire-contexts to avoid deadlocks.  But because
+ * the locking is more distributed around the driver code, we want a bit
+ * of extra utility/tracking out of our acquire-ctx.  This is provided
+ * by drm_modeset_lock / drm_modeset_acquire_ctx.
+ *
+ * For basic principles of ww_mutex, see: Documentation/ww-mutex-design.txt
+ *
+ * The basic usage pattern is to:
+ *
+ *     drm_modeset_acquire_init(&ctx)
+ *   retry:
+ *     foreach (lock in random_ordered_set_of_locks) {
+ *       ret = drm_modeset_lock(lock, &ctx)
+ *       if (ret == -EDEADLK) {
+ *          drm_modeset_backoff(&ctx);
+ *          goto retry;
+ *       }
+ *     }
+ *
+ *     ... do stuff ...
+ *
+ *     drm_modeset_drop_locks(&ctx);
+ *     drm_modeset_acquire_fini(&ctx);
+ */
+
+
+/**
+ * drm_modeset_acquire_init - initialize acquire context
+ * @ctx: the acquire context
+ * @flags: for future
+ */
+void drm_modeset_acquire_init(struct drm_modeset_acquire_ctx *ctx,
+		uint32_t flags)
+{
+	ww_acquire_init(&ctx->ww_ctx, &crtc_ww_class);
+	INIT_LIST_HEAD(&ctx->locked);
+}
+EXPORT_SYMBOL(drm_modeset_acquire_init);
+
+/**
+ * drm_modeset_acquire_fini - cleanup acquire context
+ * @ctx: the acquire context
+ */
+void drm_modeset_acquire_fini(struct drm_modeset_acquire_ctx *ctx)
+{
+	ww_acquire_fini(&ctx->ww_ctx);
+}
+EXPORT_SYMBOL(drm_modeset_acquire_fini);
+
+/**
+ * drm_modeset_drop_locks - drop all locks
+ * @ctx: the acquire context
+ *
+ * Drop all locks currently held against this acquire context.
+ */
+void drm_modeset_drop_locks(struct drm_modeset_acquire_ctx *ctx)
+{
+	WARN_ON(ctx->contended);
+	while (!list_empty(&ctx->locked)) {
+		struct drm_modeset_lock *lock;
+
+		lock = list_first_entry(&ctx->locked,
+				struct drm_modeset_lock, head);
+
+		drm_modeset_unlock(lock);
+	}
+}
+EXPORT_SYMBOL(drm_modeset_drop_locks);
+
+static inline int modeset_lock(struct drm_modeset_lock *lock,
+		struct drm_modeset_acquire_ctx *ctx,
+		bool interruptible, bool slow)
+{
+	int ret;
+
+	WARN_ON(ctx->contended);
+
+	if (interruptible && slow) {
+		ret = ww_mutex_lock_slow_interruptible(&lock->mutex, &ctx->ww_ctx);
+	} else if (interruptible) {
+		ret = ww_mutex_lock_interruptible(&lock->mutex, &ctx->ww_ctx);
+	} else if (slow) {
+		ww_mutex_lock_slow(&lock->mutex, &ctx->ww_ctx);
+		ret = 0;
+	} else {
+		ret = ww_mutex_lock(&lock->mutex, &ctx->ww_ctx);
+	}
+	if (!ret) {
+		WARN_ON(!list_empty(&lock->head));
+		list_add(&lock->head, &ctx->locked);
+	} else if (ret == -EALREADY) {
+		/* we already hold the lock.. this is fine.  For atomic
+		 * we will need to be able to drm_modeset_lock() things
+		 * without having to keep track of what is already locked
+		 * or not.
+		 */
+		ret = 0;
+	} else if (ret == -EDEADLK) {
+		ctx->contended = lock;
+	}
+
+	return ret;
+}
+
+static int modeset_backoff(struct drm_modeset_acquire_ctx *ctx,
+		bool interruptible)
+{
+	struct drm_modeset_lock *contended = ctx->contended;
+
+	ctx->contended = NULL;
+
+	if (WARN_ON(!contended))
+		return 0;
+
+	drm_modeset_drop_locks(ctx);
+
+	return modeset_lock(contended, ctx, interruptible, true);
+}
+
+/**
+ * drm_modeset_backoff - deadlock avoidance backoff
+ * @ctx: the acquire context
+ *
+ * If deadlock is detected (ie. drm_modeset_lock() returns -EDEADLK),
+ * you must call this function to drop all currently held locks and
+ * block until the contended lock becomes available.
+ */
+void drm_modeset_backoff(struct drm_modeset_acquire_ctx *ctx)
+{
+	modeset_backoff(ctx, false);
+}
+EXPORT_SYMBOL(drm_modeset_backoff);
+
+/**
+ * drm_modeset_backoff_interruptible - deadlock avoidance backoff
+ * @ctx: the acquire context
+ *
+ * Interruptible version of drm_modeset_backoff()
+ */
+int drm_modeset_backoff_interruptible(struct drm_modeset_acquire_ctx *ctx)
+{
+	return modeset_backoff(ctx, true);
+}
+EXPORT_SYMBOL(drm_modeset_backoff_interruptible);
+
+/**
+ * drm_modeset_lock - take modeset lock
+ * @lock: lock to take
+ * @ctx: acquire ctx
+ *
+ * If ctx is not NULL, then its ww acquire context is used and the
+ * lock will be tracked by the context and can be released by calling
+ * drm_modeset_drop_locks().  If -EDEADLK is returned, this means a
+ * deadlock scenario has been detected and it is an error to attempt
+ * to take any more locks without first calling drm_modeset_backoff().
+ */
+int drm_modeset_lock(struct drm_modeset_lock *lock,
+		struct drm_modeset_acquire_ctx *ctx)
+{
+	if (ctx)
+		return modeset_lock(lock, ctx, false, false);
+
+	ww_mutex_lock(&lock->mutex, NULL);
+	return 0;
+}
+EXPORT_SYMBOL(drm_modeset_lock);
+
+/**
+ * drm_modeset_lock_interruptible - take modeset lock
+ * @lock: lock to take
+ * @ctx: acquire ctx
+ *
+ * Interruptible version of drm_modeset_lock()
+ */
+int drm_modeset_lock_interruptible(struct drm_modeset_lock *lock,
+		struct drm_modeset_acquire_ctx *ctx)
+{
+	if (ctx)
+		return modeset_lock(lock, ctx, true, false);
+
+	return ww_mutex_lock_interruptible(&lock->mutex, NULL);
+}
+EXPORT_SYMBOL(drm_modeset_lock_interruptible);
+
+/**
+ * drm_modeset_unlock - drop modeset lock
+ * @lock: lock to release
+ */
+void drm_modeset_unlock(struct drm_modeset_lock *lock)
+{
+	list_del_init(&lock->head);
+	ww_mutex_unlock(&lock->mutex);
+}
+EXPORT_SYMBOL(drm_modeset_unlock);
+
+/* Temporary.. until we have sufficiently fine grained locking, there
+ * are a couple scenarios where it is convenient to grab all crtc locks.
+ * It is planned to remove this:
+ */
+int drm_modeset_lock_all_crtcs(struct drm_device *dev,
+		struct drm_modeset_acquire_ctx *ctx)
+{
+	struct drm_mode_config *config = &dev->mode_config;
+	struct drm_crtc *crtc;
+	int ret = 0;
+
+	list_for_each_entry(crtc, &config->crtc_list, head) {
+		ret = drm_modeset_lock(&crtc->mutex, ctx);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(drm_modeset_lock_all_crtcs);
diff --git a/drivers/gpu/drm/drm_pci.c b/drivers/gpu/drm/drm_pci.c
index 9c696a5ad74d..020cfd934854 100644
--- a/drivers/gpu/drm/drm_pci.c
+++ b/drivers/gpu/drm/drm_pci.c
@@ -1,17 +1,3 @@
-/* drm_pci.h -- PCI DMA memory management wrappers for DRM -*- linux-c -*- */
-/**
- * \file drm_pci.c
- * \brief Functions and ioctls to manage PCI memory
- *
- * \warning These interfaces aren't stable yet.
- *
- * \todo Implement the remaining ioctl's for the PCI pools.
- * \todo The wrappers here are so thin that they would be better off inlined..
- *
- * \author José Fonseca <jrfonseca@tungstengraphics.com>
- * \author Leif Delgass <ldelgass@retinalburn.net>
- */
-
 /*
  * Copyright 2003 José Fonseca.
  * Copyright 2003 Leif Delgass.
@@ -42,12 +28,14 @@
 #include <linux/export.h>
 #include <drm/drmP.h>
 
-/**********************************************************************/
-/** \name PCI memory */
-/*@{*/
-
 /**
- * \brief Allocate a PCI consistent memory block, for DMA.
+ * drm_pci_alloc - Allocate a PCI consistent memory block, for DMA.
+ * @dev: DRM device
+ * @size: size of block to allocate
+ * @align: alignment of block
+ *
+ * Return: A handle to the allocated memory block on success or NULL on
+ * failure.
  */
 drm_dma_handle_t *drm_pci_alloc(struct drm_device * dev, size_t size, size_t align)
 {
@@ -88,8 +76,8 @@ drm_dma_handle_t *drm_pci_alloc(struct drm_device * dev, size_t size, size_t ali
 
 EXPORT_SYMBOL(drm_pci_alloc);
 
-/**
- * \brief Free a PCI consistent memory block without freeing its descriptor.
+/*
+ * Free a PCI consistent memory block without freeing its descriptor.
  *
  * This function is for internal use in the Linux-specific DRM core code.
  */
@@ -111,7 +99,9 @@ void __drm_pci_free(struct drm_device * dev, drm_dma_handle_t * dmah)
 }
 
 /**
- * \brief Free a PCI consistent memory block
+ * drm_pci_free - Free a PCI consistent memory block
+ * @dev: DRM device
+ * @dmah: handle to memory block
  */
 void drm_pci_free(struct drm_device * dev, drm_dma_handle_t * dmah)
 {
@@ -137,21 +127,9 @@ static int drm_get_pci_domain(struct drm_device *dev)
 	return pci_domain_nr(dev->pdev->bus);
 }
 
-static int drm_pci_get_irq(struct drm_device *dev)
-{
-	return dev->pdev->irq;
-}
-
-static const char *drm_pci_get_name(struct drm_device *dev)
-{
-	struct pci_driver *pdriver = dev->driver->kdriver.pci;
-	return pdriver->name;
-}
-
 static int drm_pci_set_busid(struct drm_device *dev, struct drm_master *master)
 {
 	int len, ret;
-	struct pci_driver *pdriver = dev->driver->kdriver.pci;
 	master->unique_len = 40;
 	master->unique_size = master->unique_len;
 	master->unique = kmalloc(master->unique_size, GFP_KERNEL);
@@ -173,29 +151,16 @@ static int drm_pci_set_busid(struct drm_device *dev, struct drm_master *master)
 	} else
 		master->unique_len = len;
 
-	dev->devname =
-		kmalloc(strlen(pdriver->name) +
-			master->unique_len + 2, GFP_KERNEL);
-
-	if (dev->devname == NULL) {
-		ret = -ENOMEM;
-		goto err;
-	}
-
-	sprintf(dev->devname, "%s@%s", pdriver->name,
-		master->unique);
-
 	return 0;
 err:
 	return ret;
 }
 
-static int drm_pci_set_unique(struct drm_device *dev,
-			      struct drm_master *master,
-			      struct drm_unique *u)
+int drm_pci_set_unique(struct drm_device *dev,
+		       struct drm_master *master,
+		       struct drm_unique *u)
 {
 	int domain, bus, slot, func, ret;
-	const char *bus_name;
 
 	master->unique_len = u->unique_len;
 	master->unique_size = u->unique_len + 1;
@@ -212,17 +177,6 @@ static int drm_pci_set_unique(struct drm_device *dev,
 
 	master->unique[master->unique_len] = '\0';
 
-	bus_name = dev->driver->bus->get_name(dev);
-	dev->devname = kmalloc(strlen(bus_name) +
-			       strlen(master->unique) + 2, GFP_KERNEL);
-	if (!dev->devname) {
-		ret = -ENOMEM;
-		goto err;
-	}
-
-	sprintf(dev->devname, "%s@%s", bus_name,
-		master->unique);
-
 	/* Return error if the busid submitted doesn't match the device's actual
 	 * busid.
 	 */
@@ -247,7 +201,6 @@ err:
 	return ret;
 }
 
-
 static int drm_pci_irq_by_busid(struct drm_device *dev, struct drm_irq_busid *p)
 {
 	if ((p->busnum >> 8) != drm_get_pci_domain(dev) ||
@@ -262,6 +215,36 @@ static int drm_pci_irq_by_busid(struct drm_device *dev, struct drm_irq_busid *p)
 	return 0;
 }
 
+/**
+ * drm_irq_by_busid - Get interrupt from bus ID
+ * @dev: DRM device
+ * @data: IOCTL parameter pointing to a drm_irq_busid structure
+ * @file_priv: DRM file private.
+ *
+ * Finds the PCI device with the specified bus id and gets its IRQ number.
+ * This IOCTL is deprecated, and will now return EINVAL for any busid not equal
+ * to that of the device that this DRM instance attached to.
+ *
+ * Return: 0 on success or a negative error code on failure.
+ */
+int drm_irq_by_busid(struct drm_device *dev, void *data,
+		     struct drm_file *file_priv)
+{
+	struct drm_irq_busid *p = data;
+
+	if (drm_core_check_feature(dev, DRIVER_MODESET))
+		return -EINVAL;
+
+	/* UMS was only ever support on PCI devices. */
+	if (WARN_ON(!dev->pdev))
+		return -EINVAL;
+
+	if (!drm_core_check_feature(dev, DRIVER_HAVE_IRQ))
+		return -EINVAL;
+
+	return drm_pci_irq_by_busid(dev, p);
+}
+
 static void drm_pci_agp_init(struct drm_device *dev)
 {
 	if (drm_core_check_feature(dev, DRIVER_USE_AGP)) {
@@ -287,24 +270,20 @@ void drm_pci_agp_destroy(struct drm_device *dev)
 }
 
 static struct drm_bus drm_pci_bus = {
-	.bus_type = DRIVER_BUS_PCI,
-	.get_irq = drm_pci_get_irq,
-	.get_name = drm_pci_get_name,
 	.set_busid = drm_pci_set_busid,
-	.set_unique = drm_pci_set_unique,
-	.irq_by_busid = drm_pci_irq_by_busid,
 };
 
 /**
- * Register.
- *
- * \param pdev - PCI device structure
- * \param ent entry from the PCI ID table with device type flags
- * \return zero on success or a negative number on failure.
+ * drm_get_pci_dev - Register a PCI device with the DRM subsystem
+ * @pdev: PCI device
+ * @ent: entry from the PCI ID table that matches @pdev
+ * @driver: DRM device driver
  *
  * Attempt to gets inter module "drm" information. If we are first
  * then register the character device and inter module information.
  * Try and register, if we fail to register, backout previous work.
+ *
+ * Return: 0 on success or a negative error code on failure.
  */
 int drm_get_pci_dev(struct pci_dev *pdev, const struct pci_device_id *ent,
 		    struct drm_driver *driver)
@@ -357,15 +336,14 @@ err_free:
 EXPORT_SYMBOL(drm_get_pci_dev);
 
 /**
- * PCI device initialization. Called direct from modules at load time.
- *
- * \return zero on success or a negative number on failure.
+ * drm_pci_init - Register matching PCI devices with the DRM subsystem
+ * @driver: DRM device driver
+ * @pdriver: PCI device driver
  *
- * Initializes a drm_device structures,registering the
- * stubs and initializing the AGP device.
+ * Initializes a drm_device structures, registering the stubs and initializing
+ * the AGP device.
  *
- * Expands the \c DRIVER_PREINIT and \c DRIVER_POST_INIT macros before and
- * after the initialization for driver customization.
+ * Return: 0 on success or a negative error code on failure.
  */
 int drm_pci_init(struct drm_driver *driver, struct pci_driver *pdriver)
 {
@@ -375,7 +353,6 @@ int drm_pci_init(struct drm_driver *driver, struct pci_driver *pdriver)
 
 	DRM_DEBUG("\n");
 
-	driver->kdriver.pci = pdriver;
 	driver->bus = &drm_pci_bus;
 
 	if (driver->driver_features & DRIVER_MODESET)
@@ -453,11 +430,31 @@ int drm_pci_init(struct drm_driver *driver, struct pci_driver *pdriver)
 }
 
 void drm_pci_agp_destroy(struct drm_device *dev) {}
+
+int drm_irq_by_busid(struct drm_device *dev, void *data,
+		     struct drm_file *file_priv)
+{
+	return -EINVAL;
+}
+
+int drm_pci_set_unique(struct drm_device *dev,
+		       struct drm_master *master,
+		       struct drm_unique *u)
+{
+	return -EINVAL;
+}
 #endif
 
 EXPORT_SYMBOL(drm_pci_init);
 
-/*@}*/
+/**
+ * drm_pci_exit - Unregister matching PCI devices from the DRM subsystem
+ * @driver: DRM device driver
+ * @pdriver: PCI device driver
+ *
+ * Unregisters one or more devices matched by a PCI driver from the DRM
+ * subsystem.
+ */
 void drm_pci_exit(struct drm_driver *driver, struct pci_driver *pdriver)
 {
 	struct drm_device *dev, *tmp;
diff --git a/drivers/gpu/drm/drm_plane_helper.c b/drivers/gpu/drm/drm_plane_helper.c
index d2b1c03b3d71..6d133149cc74 100644
--- a/drivers/gpu/drm/drm_plane_helper.c
+++ b/drivers/gpu/drm/drm_plane_helper.c
@@ -25,7 +25,9 @@
 
 #include <linux/list.h>
 #include <drm/drmP.h>
+#include <drm/drm_plane_helper.h>
 #include <drm/drm_rect.h>
+#include <drm/drm_plane_helper.h>
 
 #define SUBPIXEL_MASK 0xffff
 
@@ -36,9 +38,9 @@
  * creating the primary plane.  However drivers that still call
  * drm_plane_init() will use this minimal format list as the default.
  */
-const static uint32_t safe_modeset_formats[] = {
-       DRM_FORMAT_XRGB8888,
-       DRM_FORMAT_ARGB8888,
+static const uint32_t safe_modeset_formats[] = {
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_ARGB8888,
 };
 
 /*
@@ -54,6 +56,13 @@ static int get_connectors_for_crtc(struct drm_crtc *crtc,
 	struct drm_connector *connector;
 	int count = 0;
 
+	/*
+	 * Note: Once we change the plane hooks to more fine-grained locking we
+	 * need to grab the connection_mutex here to be able to make these
+	 * checks.
+	 */
+	WARN_ON(!drm_modeset_is_locked(&dev->mode_config.connection_mutex));
+
 	list_for_each_entry(connector, &dev->mode_config.connector_list, head)
 		if (connector->encoder && connector->encoder->crtc == crtc) {
 			if (connector_list != NULL && count < num_connectors)
@@ -66,6 +75,79 @@ static int get_connectors_for_crtc(struct drm_crtc *crtc,
 }
 
 /**
+ * drm_plane_helper_check_update() - Check plane update for validity
+ * @plane: plane object to update
+ * @crtc: owning CRTC of owning plane
+ * @fb: framebuffer to flip onto plane
+ * @src: source coordinates in 16.16 fixed point
+ * @dest: integer destination coordinates
+ * @clip: integer clipping coordinates
+ * @min_scale: minimum @src:@dest scaling factor in 16.16 fixed point
+ * @max_scale: maximum @src:@dest scaling factor in 16.16 fixed point
+ * @can_position: is it legal to position the plane such that it
+ *                doesn't cover the entire crtc?  This will generally
+ *                only be false for primary planes.
+ * @can_update_disabled: can the plane be updated while the crtc
+ *                       is disabled?
+ * @visible: output parameter indicating whether plane is still visible after
+ *           clipping
+ *
+ * Checks that a desired plane update is valid.  Drivers that provide
+ * their own plane handling rather than helper-provided implementations may
+ * still wish to call this function to avoid duplication of error checking
+ * code.
+ *
+ * RETURNS:
+ * Zero if update appears valid, error code on failure
+ */
+int drm_plane_helper_check_update(struct drm_plane *plane,
+				    struct drm_crtc *crtc,
+				    struct drm_framebuffer *fb,
+				    struct drm_rect *src,
+				    struct drm_rect *dest,
+				    const struct drm_rect *clip,
+				    int min_scale,
+				    int max_scale,
+				    bool can_position,
+				    bool can_update_disabled,
+				    bool *visible)
+{
+	int hscale, vscale;
+
+	if (!crtc->enabled && !can_update_disabled) {
+		DRM_DEBUG_KMS("Cannot update plane of a disabled CRTC.\n");
+		return -EINVAL;
+	}
+
+	/* Check scaling */
+	hscale = drm_rect_calc_hscale(src, dest, min_scale, max_scale);
+	vscale = drm_rect_calc_vscale(src, dest, min_scale, max_scale);
+	if (hscale < 0 || vscale < 0) {
+		DRM_DEBUG_KMS("Invalid scaling of plane\n");
+		return -ERANGE;
+	}
+
+	*visible = drm_rect_clip_scaled(src, dest, clip, hscale, vscale);
+	if (!*visible)
+		/*
+		 * Plane isn't visible; some drivers can handle this
+		 * so we just return success here.  Drivers that can't
+		 * (including those that use the primary plane helper's
+		 * update function) will return an error from their
+		 * update_plane handler.
+		 */
+		return 0;
+
+	if (!can_position && !drm_rect_equals(dest, clip)) {
+		DRM_DEBUG_KMS("Plane must cover entire CRTC\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(drm_plane_helper_check_update);
+
+/**
  * drm_primary_helper_update() - Helper for primary plane update
  * @plane: plane object to update
  * @crtc: owning CRTC of owning plane
@@ -113,56 +195,42 @@ int drm_primary_helper_update(struct drm_plane *plane, struct drm_crtc *crtc,
 		.x = src_x >> 16,
 		.y = src_y >> 16,
 	};
+	struct drm_rect src = {
+		.x1 = src_x,
+		.y1 = src_y,
+		.x2 = src_x + src_w,
+		.y2 = src_y + src_h,
+	};
 	struct drm_rect dest = {
 		.x1 = crtc_x,
 		.y1 = crtc_y,
 		.x2 = crtc_x + crtc_w,
 		.y2 = crtc_y + crtc_h,
 	};
-	struct drm_rect clip = {
+	const struct drm_rect clip = {
 		.x2 = crtc->mode.hdisplay,
 		.y2 = crtc->mode.vdisplay,
 	};
 	struct drm_connector **connector_list;
-	struct drm_framebuffer *tmpfb;
 	int num_connectors, ret;
+	bool visible;
 
-	if (!crtc->enabled) {
-		DRM_DEBUG_KMS("Cannot update primary plane of a disabled CRTC.\n");
-		return -EINVAL;
-	}
-
-	/* Disallow subpixel positioning */
-	if ((src_x | src_y | src_w | src_h) & SUBPIXEL_MASK) {
-		DRM_DEBUG_KMS("Primary plane does not support subpixel positioning\n");
-		return -EINVAL;
-	}
-
-	/* Primary planes are locked to their owning CRTC */
-	if (plane->possible_crtcs != drm_crtc_mask(crtc)) {
-		DRM_DEBUG_KMS("Cannot change primary plane CRTC\n");
-		return -EINVAL;
-	}
-
-	/* Disallow scaling */
-	if (crtc_w != src_w || crtc_h != src_h) {
-		DRM_DEBUG_KMS("Can't scale primary plane\n");
-		return -EINVAL;
-	}
-
-	/* Make sure primary plane covers entire CRTC */
-	drm_rect_intersect(&dest, &clip);
-	if (dest.x1 != 0 || dest.y1 != 0 ||
-	    dest.x2 != crtc->mode.hdisplay || dest.y2 != crtc->mode.vdisplay) {
-		DRM_DEBUG_KMS("Primary plane must cover entire CRTC\n");
-		return -EINVAL;
-	}
-
-	/* Framebuffer must be big enough to cover entire plane */
-	ret = drm_crtc_check_viewport(crtc, crtc_x, crtc_y, &crtc->mode, fb);
+	ret = drm_plane_helper_check_update(plane, crtc, fb,
+					    &src, &dest, &clip,
+					    DRM_PLANE_HELPER_NO_SCALING,
+					    DRM_PLANE_HELPER_NO_SCALING,
+					    false, false, &visible);
 	if (ret)
 		return ret;
 
+	if (!visible)
+		/*
+		 * Primary plane isn't visible.  Note that unless a driver
+		 * provides their own disable function, this will just
+		 * wind up returning -EINVAL to userspace.
+		 */
+		return plane->funcs->disable_plane(plane);
+
 	/* Find current connectors for CRTC */
 	num_connectors = get_connectors_for_crtc(crtc, NULL, 0);
 	BUG_ON(num_connectors == 0);
@@ -176,21 +244,14 @@ int drm_primary_helper_update(struct drm_plane *plane, struct drm_crtc *crtc,
 	set.num_connectors = num_connectors;
 
 	/*
-	 * set_config() adjusts crtc->primary->fb; however the DRM setplane
-	 * code that called us expects to handle the framebuffer update and
-	 * reference counting; save and restore the current fb before
-	 * calling it.
-	 *
-	 * N.B., we call set_config() directly here rather than using
+	 * We call set_config() directly here rather than using
 	 * drm_mode_set_config_internal.  We're reprogramming the same
 	 * connectors that were already in use, so we shouldn't need the extra
 	 * cross-CRTC fb refcounting to accomodate stealing connectors.
 	 * drm_mode_setplane() already handles the basic refcounting for the
 	 * framebuffers involved in this operation.
 	 */
-	tmpfb = plane->fb;
 	ret = crtc->funcs->set_config(&set);
-	plane->fb = tmpfb;
 
 	kfree(connector_list);
 	return ret;
@@ -232,7 +293,6 @@ EXPORT_SYMBOL(drm_primary_helper_disable);
  */
 void drm_primary_helper_destroy(struct drm_plane *plane)
 {
-	plane->funcs->disable_plane(plane);
 	drm_plane_cleanup(plane);
 	kfree(plane);
 }
diff --git a/drivers/gpu/drm/drm_platform.c b/drivers/gpu/drm/drm_platform.c
index 319ff5385601..d5b76f148c12 100644
--- a/drivers/gpu/drm/drm_platform.c
+++ b/drivers/gpu/drm/drm_platform.c
@@ -68,16 +68,6 @@ err_free:
 	return ret;
 }
 
-static int drm_platform_get_irq(struct drm_device *dev)
-{
-	return platform_get_irq(dev->platformdev, 0);
-}
-
-static const char *drm_platform_get_name(struct drm_device *dev)
-{
-	return dev->platformdev->name;
-}
-
 static int drm_platform_set_busid(struct drm_device *dev, struct drm_master *master)
 {
 	int len, ret, id;
@@ -106,46 +96,30 @@ static int drm_platform_set_busid(struct drm_device *dev, struct drm_master *mas
 		goto err;
 	}
 
-	dev->devname =
-		kmalloc(strlen(dev->platformdev->name) +
-			master->unique_len + 2, GFP_KERNEL);
-
-	if (dev->devname == NULL) {
-		ret = -ENOMEM;
-		goto err;
-	}
-
-	sprintf(dev->devname, "%s@%s", dev->platformdev->name,
-		master->unique);
 	return 0;
 err:
 	return ret;
 }
 
 static struct drm_bus drm_platform_bus = {
-	.bus_type = DRIVER_BUS_PLATFORM,
-	.get_irq = drm_platform_get_irq,
-	.get_name = drm_platform_get_name,
 	.set_busid = drm_platform_set_busid,
 };
 
 /**
- * Platform device initialization. Called direct from modules.
+ * drm_platform_init - Register a platform device with the DRM subsystem
+ * @driver: DRM device driver
+ * @platform_device: platform device to register
  *
- * \return zero on success or a negative number on failure.
- *
- * Initializes a drm_device structures,registering the
- * stubs
+ * Registers the specified DRM device driver and platform device with the DRM
+ * subsystem, initializing a drm_device structure and calling the driver's
+ * .load() function.
  *
- * Expands the \c DRIVER_PREINIT and \c DRIVER_POST_INIT macros before and
- * after the initialization for driver customization.
+ * Return: 0 on success or a negative error code on failure.
  */
-
 int drm_platform_init(struct drm_driver *driver, struct platform_device *platform_device)
 {
 	DRM_DEBUG("\n");
 
-	driver->kdriver.platform_device = platform_device;
 	driver->bus = &drm_platform_bus;
 	return drm_get_platform_dev(platform_device, driver);
 }
diff --git a/drivers/gpu/drm/drm_probe_helper.c b/drivers/gpu/drm/drm_probe_helper.c
index e70f54d4a581..d22676b89cbb 100644
--- a/drivers/gpu/drm/drm_probe_helper.c
+++ b/drivers/gpu/drm/drm_probe_helper.c
@@ -82,26 +82,8 @@ static void drm_mode_validate_flag(struct drm_connector *connector,
 	return;
 }
 
-/**
- * drm_helper_probe_single_connector_modes - get complete set of display modes
- * @connector: connector to probe
- * @maxX: max width for modes
- * @maxY: max height for modes
- *
- * Based on the helper callbacks implemented by @connector try to detect all
- * valid modes.  Modes will first be added to the connector's probed_modes list,
- * then culled (based on validity and the @maxX, @maxY parameters) and put into
- * the normal modes list.
- *
- * Intended to be use as a generic implementation of the ->fill_modes()
- * @connector vfunc for drivers that use the crtc helpers for output mode
- * filtering and detection.
- *
- * Returns:
- * The number of modes found on @connector.
- */
-int drm_helper_probe_single_connector_modes(struct drm_connector *connector,
-					    uint32_t maxX, uint32_t maxY)
+static int drm_helper_probe_single_connector_modes_merge_bits(struct drm_connector *connector,
+							      uint32_t maxX, uint32_t maxY, bool merge_type_bits)
 {
 	struct drm_device *dev = connector->dev;
 	struct drm_display_mode *mode;
@@ -114,7 +96,7 @@ int drm_helper_probe_single_connector_modes(struct drm_connector *connector,
 	WARN_ON(!mutex_is_locked(&dev->mode_config.mutex));
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n", connector->base.id,
-			drm_get_connector_name(connector));
+			connector->name);
 	/* set all modes to the unverified state */
 	list_for_each_entry(mode, &connector->modes, head)
 		mode->status = MODE_UNVERIFIED;
@@ -138,7 +120,7 @@ int drm_helper_probe_single_connector_modes(struct drm_connector *connector,
 
 	if (connector->status == connector_status_disconnected) {
 		DRM_DEBUG_KMS("[CONNECTOR:%d:%s] disconnected\n",
-			connector->base.id, drm_get_connector_name(connector));
+			connector->base.id, connector->name);
 		drm_mode_connector_update_edid_property(connector, NULL);
 		verbose_prune = false;
 		goto prune;
@@ -155,7 +137,7 @@ int drm_helper_probe_single_connector_modes(struct drm_connector *connector,
 	if (count == 0)
 		goto prune;
 
-	drm_mode_connector_list_update(connector);
+	drm_mode_connector_list_update(connector, merge_type_bits);
 
 	if (maxX && maxY)
 		drm_mode_validate_size(dev, &connector->modes, maxX, maxY);
@@ -169,7 +151,7 @@ int drm_helper_probe_single_connector_modes(struct drm_connector *connector,
 	drm_mode_validate_flag(connector, mode_flags);
 
 	list_for_each_entry(mode, &connector->modes, head) {
-		if (mode->status == MODE_OK)
+		if (mode->status == MODE_OK && connector_funcs->mode_valid)
 			mode->status = connector_funcs->mode_valid(connector,
 								   mode);
 	}
@@ -186,7 +168,7 @@ prune:
 	drm_mode_sort(&connector->modes);
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s] probed modes :\n", connector->base.id,
-			drm_get_connector_name(connector));
+			connector->name);
 	list_for_each_entry(mode, &connector->modes, head) {
 		drm_mode_set_crtcinfo(mode, CRTC_INTERLACE_HALVE_V);
 		drm_mode_debug_printmodeline(mode);
@@ -194,9 +176,49 @@ prune:
 
 	return count;
 }
+
+/**
+ * drm_helper_probe_single_connector_modes - get complete set of display modes
+ * @connector: connector to probe
+ * @maxX: max width for modes
+ * @maxY: max height for modes
+ *
+ * Based on the helper callbacks implemented by @connector try to detect all
+ * valid modes.  Modes will first be added to the connector's probed_modes list,
+ * then culled (based on validity and the @maxX, @maxY parameters) and put into
+ * the normal modes list.
+ *
+ * Intended to be use as a generic implementation of the ->fill_modes()
+ * @connector vfunc for drivers that use the crtc helpers for output mode
+ * filtering and detection.
+ *
+ * Returns:
+ * The number of modes found on @connector.
+ */
+int drm_helper_probe_single_connector_modes(struct drm_connector *connector,
+					    uint32_t maxX, uint32_t maxY)
+{
+	return drm_helper_probe_single_connector_modes_merge_bits(connector, maxX, maxY, true);
+}
 EXPORT_SYMBOL(drm_helper_probe_single_connector_modes);
 
 /**
+ * drm_helper_probe_single_connector_modes_nomerge - get complete set of display modes
+ * @connector: connector to probe
+ * @maxX: max width for modes
+ * @maxY: max height for modes
+ *
+ * This operates like drm_hehlper_probe_single_connector_modes except it
+ * replaces the mode bits instead of merging them for preferred modes.
+ */
+int drm_helper_probe_single_connector_modes_nomerge(struct drm_connector *connector,
+					    uint32_t maxX, uint32_t maxY)
+{
+	return drm_helper_probe_single_connector_modes_merge_bits(connector, maxX, maxY, false);
+}
+EXPORT_SYMBOL(drm_helper_probe_single_connector_modes_nomerge);
+
+/**
  * drm_kms_helper_hotplug_event - fire off KMS hotplug events
  * @dev: drm_device whose connector state changed
  *
@@ -264,7 +286,7 @@ static void output_poll_execute(struct work_struct *work)
 			DRM_DEBUG_KMS("[CONNECTOR:%d:%s] "
 				      "status updated from %s to %s\n",
 				      connector->base.id,
-				      drm_get_connector_name(connector),
+				      connector->name,
 				      old, new);
 
 			changed = true;
@@ -409,7 +431,7 @@ bool drm_helper_hpd_irq_event(struct drm_device *dev)
 		connector->status = connector->funcs->detect(connector, false);
 		DRM_DEBUG_KMS("[CONNECTOR:%d:%s] status updated from %s to %s\n",
 			      connector->base.id,
-			      drm_get_connector_name(connector),
+			      connector->name,
 			      drm_get_connector_status_name(old_status),
 			      drm_get_connector_status_name(connector->status));
 		if (old_status != connector->status)
diff --git a/drivers/gpu/drm/drm_stub.c b/drivers/gpu/drm/drm_stub.c
index 4c24c3ac1efa..14d16464000a 100644
--- a/drivers/gpu/drm/drm_stub.c
+++ b/drivers/gpu/drm/drm_stub.c
@@ -1,16 +1,11 @@
-/**
- * \file drm_stub.h
- * Stub support
- *
- * \author Rickard E. (Rik) Faith <faith@valinux.com>
- */
-
 /*
  * Created: Fri Jan 19 10:48:35 2001 by faith@acm.org
  *
  * Copyright 2001 VA Linux Systems, Inc., Sunnyvale, California.
  * All Rights Reserved.
  *
+ * Author Rickard E. (Rik) Faith <faith@valinux.com>
+ *
  * Permission is hereby granted, free of charge, to any person obtaining a
  * copy of this software and associated documentation files (the "Software"),
  * to deal in the Software without restriction, including without limitation
@@ -128,7 +123,10 @@ struct drm_master *drm_master_create(struct drm_minor *minor)
 	kref_init(&master->refcount);
 	spin_lock_init(&master->lock.spinlock);
 	init_waitqueue_head(&master->lock.lock_queue);
-	drm_ht_create(&master->magiclist, DRM_MAGIC_HASH_ORDER);
+	if (drm_ht_create(&master->magiclist, DRM_MAGIC_HASH_ORDER)) {
+		kfree(master);
+		return NULL;
+	}
 	INIT_LIST_HEAD(&master->magicfree);
 	master->minor = minor;
 
@@ -166,9 +164,6 @@ static void drm_master_destroy(struct kref *kref)
 		master->unique_len = 0;
 	}
 
-	kfree(dev->devname);
-	dev->devname = NULL;
-
 	list_for_each_entry_safe(pt, next, &master->magicfree, head) {
 		list_del(&pt->head);
 		drm_ht_remove_item(&master->magiclist, &pt->hash_item);
@@ -294,6 +289,7 @@ static void drm_minor_free(struct drm_device *dev, unsigned int type)
 
 	slot = drm_minor_get_slot(dev, type);
 	if (*slot) {
+		drm_mode_group_destroy(&(*slot)->mode_group);
 		kfree(*slot);
 		*slot = NULL;
 	}
@@ -424,11 +420,15 @@ void drm_minor_release(struct drm_minor *minor)
 }
 
 /**
- * Called via drm_exit() at module unload time or when pci device is
- * unplugged.
+ * drm_put_dev - Unregister and release a DRM device
+ * @dev: DRM device
  *
- * Cleans up all DRM device, calling drm_lastclose().
+ * Called at module unload time or when a PCI device is unplugged.
  *
+ * Use of this function is discouraged. It will eventually go away completely.
+ * Please use drm_dev_unregister() and drm_dev_unref() explicitly instead.
+ *
+ * Cleans up all DRM device, calling drm_lastclose().
  */
 void drm_put_dev(struct drm_device *dev)
 {
@@ -535,7 +535,7 @@ static void drm_fs_inode_free(struct inode *inode)
 }
 
 /**
- * drm_dev_alloc - Allocate new drm device
+ * drm_dev_alloc - Allocate new DRM device
  * @driver: DRM driver to allocate device for
  * @parent: Parent device object
  *
@@ -569,7 +569,7 @@ struct drm_device *drm_dev_alloc(struct drm_driver *driver,
 	INIT_LIST_HEAD(&dev->maplist);
 	INIT_LIST_HEAD(&dev->vblank_event_list);
 
-	spin_lock_init(&dev->count_lock);
+	spin_lock_init(&dev->buf_lock);
 	spin_lock_init(&dev->event_lock);
 	mutex_init(&dev->struct_mutex);
 	mutex_init(&dev->ctxlist_mutex);
@@ -648,9 +648,8 @@ static void drm_dev_release(struct kref *ref)
 	drm_minor_free(dev, DRM_MINOR_RENDER);
 	drm_minor_free(dev, DRM_MINOR_CONTROL);
 
-	kfree(dev->devname);
-
 	mutex_destroy(&dev->master_mutex);
+	kfree(dev->unique);
 	kfree(dev);
 }
 
@@ -690,6 +689,7 @@ EXPORT_SYMBOL(drm_dev_unref);
 /**
  * drm_dev_register - Register DRM device
  * @dev: Device to register
+ * @flags: Flags passed to the driver's .load() function
  *
  * Register the DRM device @dev with the system, advertise device to user-space
  * and start normal device operation. @dev must be allocated via drm_dev_alloc()
@@ -778,3 +778,28 @@ void drm_dev_unregister(struct drm_device *dev)
 	drm_minor_unregister(dev, DRM_MINOR_CONTROL);
 }
 EXPORT_SYMBOL(drm_dev_unregister);
+
+/**
+ * drm_dev_set_unique - Set the unique name of a DRM device
+ * @dev: device of which to set the unique name
+ * @fmt: format string for unique name
+ *
+ * Sets the unique name of a DRM device using the specified format string and
+ * a variable list of arguments. Drivers can use this at driver probe time if
+ * the unique name of the devices they drive is static.
+ *
+ * Return: 0 on success or a negative error code on failure.
+ */
+int drm_dev_set_unique(struct drm_device *dev, const char *fmt, ...)
+{
+	va_list ap;
+
+	kfree(dev->unique);
+
+	va_start(ap, fmt);
+	dev->unique = kvasprintf(GFP_KERNEL, fmt, ap);
+	va_end(ap);
+
+	return dev->unique ? 0 : -ENOMEM;
+}
+EXPORT_SYMBOL(drm_dev_set_unique);
diff --git a/drivers/gpu/drm/drm_sysfs.c b/drivers/gpu/drm/drm_sysfs.c
index c22c3097c3e8..369b26278e76 100644
--- a/drivers/gpu/drm/drm_sysfs.c
+++ b/drivers/gpu/drm/drm_sysfs.c
@@ -380,9 +380,9 @@ int drm_sysfs_connector_add(struct drm_connector *connector)
 
 	connector->kdev = device_create(drm_class, dev->primary->kdev,
 					0, connector, "card%d-%s",
-					dev->primary->index, drm_get_connector_name(connector));
+					dev->primary->index, connector->name);
 	DRM_DEBUG("adding \"%s\" to sysfs\n",
-		  drm_get_connector_name(connector));
+		  connector->name);
 
 	if (IS_ERR(connector->kdev)) {
 		DRM_ERROR("failed to register connector device: %ld\n", PTR_ERR(connector->kdev));
@@ -460,7 +460,7 @@ void drm_sysfs_connector_remove(struct drm_connector *connector)
 	if (!connector->kdev)
 		return;
 	DRM_DEBUG("removing \"%s\" from sysfs\n",
-		  drm_get_connector_name(connector));
+		  connector->name);
 
 	for (i = 0; i < ARRAY_SIZE(connector_attrs); i++)
 		device_remove_file(connector->kdev, &connector_attrs[i]);
diff --git a/drivers/gpu/drm/drm_usb.c b/drivers/gpu/drm/drm_usb.c
index c3406aad2944..f2fe94aab901 100644
--- a/drivers/gpu/drm/drm_usb.c
+++ b/drivers/gpu/drm/drm_usb.c
@@ -36,16 +36,6 @@ err_free:
 }
 EXPORT_SYMBOL(drm_get_usb_dev);
 
-static int drm_usb_get_irq(struct drm_device *dev)
-{
-	return 0;
-}
-
-static const char *drm_usb_get_name(struct drm_device *dev)
-{
-	return "USB";
-}
-
 static int drm_usb_set_busid(struct drm_device *dev,
 			       struct drm_master *master)
 {
@@ -53,18 +43,24 @@ static int drm_usb_set_busid(struct drm_device *dev,
 }
 
 static struct drm_bus drm_usb_bus = {
-	.bus_type = DRIVER_BUS_USB,
-	.get_irq = drm_usb_get_irq,
-	.get_name = drm_usb_get_name,
 	.set_busid = drm_usb_set_busid,
 };
-    
+
+/**
+ * drm_usb_init - Register matching USB devices with the DRM subsystem
+ * @driver: DRM device driver
+ * @udriver: USB device driver
+ *
+ * Registers one or more devices matched by a USB driver with the DRM
+ * subsystem.
+ *
+ * Return: 0 on success or a negative error code on failure.
+ */
 int drm_usb_init(struct drm_driver *driver, struct usb_driver *udriver)
 {
 	int res;
 	DRM_DEBUG("\n");
 
-	driver->kdriver.usb = udriver;
 	driver->bus = &drm_usb_bus;
 
 	res = usb_register(udriver);
@@ -72,6 +68,14 @@ int drm_usb_init(struct drm_driver *driver, struct usb_driver *udriver)
 }
 EXPORT_SYMBOL(drm_usb_init);
 
+/**
+ * drm_usb_exit - Unregister matching USB devices from the DRM subsystem
+ * @driver: DRM device driver
+ * @udriver: USB device driver
+ *
+ * Unregisters one or more devices matched by a USB driver from the DRM
+ * subsystem.
+ */
 void drm_usb_exit(struct drm_driver *driver,
 		  struct usb_driver *udriver)
 {
diff --git a/drivers/gpu/drm/exynos/Kconfig b/drivers/gpu/drm/exynos/Kconfig
index 5bf5bca94f56..178d2a9672a8 100644
--- a/drivers/gpu/drm/exynos/Kconfig
+++ b/drivers/gpu/drm/exynos/Kconfig
@@ -26,14 +26,14 @@ config DRM_EXYNOS_DMABUF
 
 config DRM_EXYNOS_FIMD
 	bool "Exynos DRM FIMD"
-	depends on DRM_EXYNOS && !FB_S3C && !ARCH_MULTIPLATFORM
+	depends on DRM_EXYNOS && !FB_S3C
 	select FB_MODE_HELPERS
 	help
 	  Choose this option if you want to use Exynos FIMD for DRM.
 
 config DRM_EXYNOS_DPI
 	bool "EXYNOS DRM parallel output support"
-	depends on DRM_EXYNOS
+	depends on DRM_EXYNOS_FIMD
 	select DRM_PANEL
 	default n
 	help
@@ -41,7 +41,7 @@ config DRM_EXYNOS_DPI
 
 config DRM_EXYNOS_DSI
 	bool "EXYNOS DRM MIPI-DSI driver support"
-	depends on DRM_EXYNOS
+	depends on DRM_EXYNOS_FIMD
 	select DRM_MIPI_DSI
 	select DRM_PANEL
 	default n
@@ -50,7 +50,7 @@ config DRM_EXYNOS_DSI
 
 config DRM_EXYNOS_DP
 	bool "EXYNOS DRM DP driver support"
-	depends on DRM_EXYNOS && ARCH_EXYNOS
+	depends on DRM_EXYNOS_FIMD && ARCH_EXYNOS && (DRM_PTN3460=n || DRM_PTN3460=y || DRM_PTN3460=DRM_EXYNOS)
 	default DRM_EXYNOS
 	help
 	  This enables support for DP device.
diff --git a/drivers/gpu/drm/exynos/exynos_ddc.c b/drivers/gpu/drm/exynos/exynos_ddc.c
deleted file mode 100644
index 6a8c84e7c839..000000000000
--- a/drivers/gpu/drm/exynos/exynos_ddc.c
+++ /dev/null
@@ -1,63 +0,0 @@
-/*
- * Copyright (C) 2011 Samsung Electronics Co.Ltd
- * Authors:
- *	Seung-Woo Kim <sw0312.kim@samsung.com>
- *	Inki Dae <inki.dae@samsung.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- */
-
-#include <drm/drmP.h>
-
-#include <linux/kernel.h>
-#include <linux/i2c.h>
-#include <linux/of.h>
-
-#include "exynos_drm_drv.h"
-#include "exynos_hdmi.h"
-
-static int s5p_ddc_probe(struct i2c_client *client,
-			const struct i2c_device_id *dev_id)
-{
-	hdmi_attach_ddc_client(client);
-
-	dev_info(&client->adapter->dev,
-		"attached %s into i2c adapter successfully\n",
-		client->name);
-
-	return 0;
-}
-
-static int s5p_ddc_remove(struct i2c_client *client)
-{
-	dev_info(&client->adapter->dev,
-		"detached %s from i2c adapter successfully\n",
-		client->name);
-
-	return 0;
-}
-
-static struct of_device_id hdmiddc_match_types[] = {
-	{
-		.compatible = "samsung,exynos5-hdmiddc",
-	}, {
-		.compatible = "samsung,exynos4210-hdmiddc",
-	}, {
-		/* end node */
-	}
-};
-
-struct i2c_driver ddc_driver = {
-	.driver = {
-		.name = "exynos-hdmiddc",
-		.owner = THIS_MODULE,
-		.of_match_table = hdmiddc_match_types,
-	},
-	.probe		= s5p_ddc_probe,
-	.remove		= s5p_ddc_remove,
-	.command		= NULL,
-};
diff --git a/drivers/gpu/drm/exynos/exynos_dp_core.c b/drivers/gpu/drm/exynos/exynos_dp_core.c
index aed533bbfd31..a8ffc8c1477b 100644
--- a/drivers/gpu/drm/exynos/exynos_dp_core.c
+++ b/drivers/gpu/drm/exynos/exynos_dp_core.c
@@ -18,6 +18,9 @@
 #include <linux/interrupt.h>
 #include <linux/delay.h>
 #include <linux/of.h>
+#include <linux/of_gpio.h>
+#include <linux/gpio.h>
+#include <linux/component.h>
 #include <linux/phy/phy.h>
 #include <video/of_display_timing.h>
 #include <video/of_videomode.h>
@@ -141,15 +144,15 @@ static int exynos_dp_read_edid(struct exynos_dp_device *dp)
 			return -EIO;
 		}
 
-		exynos_dp_read_byte_from_dpcd(dp, DPCD_ADDR_TEST_REQUEST,
+		exynos_dp_read_byte_from_dpcd(dp, DP_TEST_REQUEST,
 					&test_vector);
-		if (test_vector & DPCD_TEST_EDID_READ) {
+		if (test_vector & DP_TEST_LINK_EDID_READ) {
 			exynos_dp_write_byte_to_dpcd(dp,
-				DPCD_ADDR_TEST_EDID_CHECKSUM,
+				DP_TEST_EDID_CHECKSUM,
 				edid[EDID_BLOCK_LENGTH + EDID_CHECKSUM]);
 			exynos_dp_write_byte_to_dpcd(dp,
-				DPCD_ADDR_TEST_RESPONSE,
-				DPCD_TEST_EDID_CHECKSUM_WRITE);
+				DP_TEST_RESPONSE,
+				DP_TEST_EDID_CHECKSUM_WRITE);
 		}
 	} else {
 		dev_info(dp->dev, "EDID data does not include any extensions.\n");
@@ -171,15 +174,15 @@ static int exynos_dp_read_edid(struct exynos_dp_device *dp)
 		}
 
 		exynos_dp_read_byte_from_dpcd(dp,
-			DPCD_ADDR_TEST_REQUEST,
+			DP_TEST_REQUEST,
 			&test_vector);
-		if (test_vector & DPCD_TEST_EDID_READ) {
+		if (test_vector & DP_TEST_LINK_EDID_READ) {
 			exynos_dp_write_byte_to_dpcd(dp,
-				DPCD_ADDR_TEST_EDID_CHECKSUM,
+				DP_TEST_EDID_CHECKSUM,
 				edid[EDID_CHECKSUM]);
 			exynos_dp_write_byte_to_dpcd(dp,
-				DPCD_ADDR_TEST_RESPONSE,
-				DPCD_TEST_EDID_CHECKSUM_WRITE);
+				DP_TEST_RESPONSE,
+				DP_TEST_EDID_CHECKSUM_WRITE);
 		}
 	}
 
@@ -193,8 +196,8 @@ static int exynos_dp_handle_edid(struct exynos_dp_device *dp)
 	int i;
 	int retval;
 
-	/* Read DPCD DPCD_ADDR_DPCD_REV~RECEIVE_PORT1_CAP_1 */
-	retval = exynos_dp_read_bytes_from_dpcd(dp, DPCD_ADDR_DPCD_REV,
+	/* Read DPCD DP_DPCD_REV~RECEIVE_PORT1_CAP_1 */
+	retval = exynos_dp_read_bytes_from_dpcd(dp, DP_DPCD_REV,
 				12, buf);
 	if (retval)
 		return retval;
@@ -214,14 +217,14 @@ static void exynos_dp_enable_rx_to_enhanced_mode(struct exynos_dp_device *dp,
 {
 	u8 data;
 
-	exynos_dp_read_byte_from_dpcd(dp, DPCD_ADDR_LANE_COUNT_SET, &data);
+	exynos_dp_read_byte_from_dpcd(dp, DP_LANE_COUNT_SET, &data);
 
 	if (enable)
-		exynos_dp_write_byte_to_dpcd(dp, DPCD_ADDR_LANE_COUNT_SET,
-			DPCD_ENHANCED_FRAME_EN |
+		exynos_dp_write_byte_to_dpcd(dp, DP_LANE_COUNT_SET,
+			DP_LANE_COUNT_ENHANCED_FRAME_EN |
 			DPCD_LANE_COUNT_SET(data));
 	else
-		exynos_dp_write_byte_to_dpcd(dp, DPCD_ADDR_LANE_COUNT_SET,
+		exynos_dp_write_byte_to_dpcd(dp, DP_LANE_COUNT_SET,
 			DPCD_LANE_COUNT_SET(data));
 }
 
@@ -230,7 +233,7 @@ static int exynos_dp_is_enhanced_mode_available(struct exynos_dp_device *dp)
 	u8 data;
 	int retval;
 
-	exynos_dp_read_byte_from_dpcd(dp, DPCD_ADDR_MAX_LANE_COUNT, &data);
+	exynos_dp_read_byte_from_dpcd(dp, DP_MAX_LANE_COUNT, &data);
 	retval = DPCD_ENHANCED_FRAME_CAP(data);
 
 	return retval;
@@ -250,8 +253,8 @@ static void exynos_dp_training_pattern_dis(struct exynos_dp_device *dp)
 	exynos_dp_set_training_pattern(dp, DP_NONE);
 
 	exynos_dp_write_byte_to_dpcd(dp,
-		DPCD_ADDR_TRAINING_PATTERN_SET,
-		DPCD_TRAINING_PATTERN_DISABLED);
+		DP_TRAINING_PATTERN_SET,
+		DP_TRAINING_PATTERN_DISABLE);
 }
 
 static void exynos_dp_set_lane_lane_pre_emphasis(struct exynos_dp_device *dp,
@@ -295,7 +298,7 @@ static int exynos_dp_link_start(struct exynos_dp_device *dp)
 	/* Setup RX configuration */
 	buf[0] = dp->link_train.link_rate;
 	buf[1] = dp->link_train.lane_count;
-	retval = exynos_dp_write_bytes_to_dpcd(dp, DPCD_ADDR_LINK_BW_SET,
+	retval = exynos_dp_write_bytes_to_dpcd(dp, DP_LINK_BW_SET,
 				2, buf);
 	if (retval)
 		return retval;
@@ -322,16 +325,16 @@ static int exynos_dp_link_start(struct exynos_dp_device *dp)
 
 	/* Set RX training pattern */
 	retval = exynos_dp_write_byte_to_dpcd(dp,
-			DPCD_ADDR_TRAINING_PATTERN_SET,
-			DPCD_SCRAMBLING_DISABLED | DPCD_TRAINING_PATTERN_1);
+			DP_TRAINING_PATTERN_SET,
+			DP_LINK_SCRAMBLING_DISABLE | DP_TRAINING_PATTERN_1);
 	if (retval)
 		return retval;
 
 	for (lane = 0; lane < lane_count; lane++)
-		buf[lane] = DPCD_PRE_EMPHASIS_PATTERN2_LEVEL0 |
-			    DPCD_VOLTAGE_SWING_PATTERN1_LEVEL0;
+		buf[lane] = DP_TRAIN_PRE_EMPHASIS_0 |
+			    DP_TRAIN_VOLTAGE_SWING_400;
 
-	retval = exynos_dp_write_bytes_to_dpcd(dp, DPCD_ADDR_TRAINING_LANE0_SET,
+	retval = exynos_dp_write_bytes_to_dpcd(dp, DP_TRAINING_LANE0_SET,
 			lane_count, buf);
 
 	return retval;
@@ -352,7 +355,7 @@ static int exynos_dp_clock_recovery_ok(u8 link_status[2], int lane_count)
 
 	for (lane = 0; lane < lane_count; lane++) {
 		lane_status = exynos_dp_get_lane_status(link_status, lane);
-		if ((lane_status & DPCD_LANE_CR_DONE) == 0)
+		if ((lane_status & DP_LANE_CR_DONE) == 0)
 			return -EINVAL;
 	}
 	return 0;
@@ -364,13 +367,13 @@ static int exynos_dp_channel_eq_ok(u8 link_status[2], u8 link_align,
 	int lane;
 	u8 lane_status;
 
-	if ((link_align & DPCD_INTERLANE_ALIGN_DONE) == 0)
+	if ((link_align & DP_INTERLANE_ALIGN_DONE) == 0)
 		return -EINVAL;
 
 	for (lane = 0; lane < lane_count; lane++) {
 		lane_status = exynos_dp_get_lane_status(link_status, lane);
-		lane_status &= DPCD_CHANNEL_EQ_BITS;
-		if (lane_status != DPCD_CHANNEL_EQ_BITS)
+		lane_status &= DP_CHANNEL_EQ_BITS;
+		if (lane_status != DP_CHANNEL_EQ_BITS)
 			return -EINVAL;
 	}
 
@@ -468,9 +471,9 @@ static void exynos_dp_get_adjust_training_lane(struct exynos_dp_device *dp,
 				DPCD_PRE_EMPHASIS_SET(pre_emphasis);
 
 		if (voltage_swing == VOLTAGE_LEVEL_3)
-			training_lane |= DPCD_MAX_SWING_REACHED;
+			training_lane |= DP_TRAIN_MAX_SWING_REACHED;
 		if (pre_emphasis == PRE_EMPHASIS_LEVEL_3)
-			training_lane |= DPCD_MAX_PRE_EMPHASIS_REACHED;
+			training_lane |= DP_TRAIN_MAX_PRE_EMPHASIS_REACHED;
 
 		dp->link_train.training_lane[lane] = training_lane;
 	}
@@ -487,12 +490,12 @@ static int exynos_dp_process_clock_recovery(struct exynos_dp_device *dp)
 	lane_count = dp->link_train.lane_count;
 
 	retval =  exynos_dp_read_bytes_from_dpcd(dp,
-			DPCD_ADDR_LANE0_1_STATUS, 2, link_status);
+			DP_LANE0_1_STATUS, 2, link_status);
 	if (retval)
 		return retval;
 
 	retval =  exynos_dp_read_bytes_from_dpcd(dp,
-			DPCD_ADDR_ADJUST_REQUEST_LANE0_1, 2, adjust_request);
+			DP_ADJUST_REQUEST_LANE0_1, 2, adjust_request);
 	if (retval)
 		return retval;
 
@@ -501,9 +504,9 @@ static int exynos_dp_process_clock_recovery(struct exynos_dp_device *dp)
 		exynos_dp_set_training_pattern(dp, TRAINING_PTN2);
 
 		retval = exynos_dp_write_byte_to_dpcd(dp,
-				DPCD_ADDR_TRAINING_PATTERN_SET,
-				DPCD_SCRAMBLING_DISABLED |
-				DPCD_TRAINING_PATTERN_2);
+				DP_TRAINING_PATTERN_SET,
+				DP_LINK_SCRAMBLING_DISABLE |
+				DP_TRAINING_PATTERN_2);
 		if (retval)
 			return retval;
 
@@ -543,7 +546,7 @@ static int exynos_dp_process_clock_recovery(struct exynos_dp_device *dp)
 			dp->link_train.training_lane[lane], lane);
 
 	retval = exynos_dp_write_bytes_to_dpcd(dp,
-			DPCD_ADDR_TRAINING_LANE0_SET, lane_count,
+			DP_TRAINING_LANE0_SET, lane_count,
 			dp->link_train.training_lane);
 	if (retval)
 		return retval;
@@ -562,7 +565,7 @@ static int exynos_dp_process_equalizer_training(struct exynos_dp_device *dp)
 	lane_count = dp->link_train.lane_count;
 
 	retval = exynos_dp_read_bytes_from_dpcd(dp,
-			DPCD_ADDR_LANE0_1_STATUS, 2, link_status);
+			DP_LANE0_1_STATUS, 2, link_status);
 	if (retval)
 		return retval;
 
@@ -572,12 +575,12 @@ static int exynos_dp_process_equalizer_training(struct exynos_dp_device *dp)
 	}
 
 	retval = exynos_dp_read_bytes_from_dpcd(dp,
-			DPCD_ADDR_ADJUST_REQUEST_LANE0_1, 2, adjust_request);
+			DP_ADJUST_REQUEST_LANE0_1, 2, adjust_request);
 	if (retval)
 		return retval;
 
 	retval = exynos_dp_read_byte_from_dpcd(dp,
-			DPCD_ADDR_LANE_ALIGN_STATUS_UPDATED, &link_align);
+			DP_LANE_ALIGN_STATUS_UPDATED, &link_align);
 	if (retval)
 		return retval;
 
@@ -619,7 +622,7 @@ static int exynos_dp_process_equalizer_training(struct exynos_dp_device *dp)
 		exynos_dp_set_lane_link_training(dp,
 			dp->link_train.training_lane[lane], lane);
 
-	retval = exynos_dp_write_bytes_to_dpcd(dp, DPCD_ADDR_TRAINING_LANE0_SET,
+	retval = exynos_dp_write_bytes_to_dpcd(dp, DP_TRAINING_LANE0_SET,
 			lane_count, dp->link_train.training_lane);
 
 	return retval;
@@ -634,7 +637,7 @@ static void exynos_dp_get_max_rx_bandwidth(struct exynos_dp_device *dp,
 	 * For DP rev.1.1, Maximum link rate of Main Link lanes
 	 * 0x06 = 1.62 Gbps, 0x0a = 2.7 Gbps
 	 */
-	exynos_dp_read_byte_from_dpcd(dp, DPCD_ADDR_MAX_LINK_RATE, &data);
+	exynos_dp_read_byte_from_dpcd(dp, DP_MAX_LINK_RATE, &data);
 	*bandwidth = data;
 }
 
@@ -647,7 +650,7 @@ static void exynos_dp_get_max_rx_lane_count(struct exynos_dp_device *dp,
 	 * For DP rev.1.1, Maximum number of Main Link lanes
 	 * 0x01 = 1 lane, 0x02 = 2 lanes, 0x04 = 4 lanes
 	 */
-	exynos_dp_read_byte_from_dpcd(dp, DPCD_ADDR_MAX_LANE_COUNT, &data);
+	exynos_dp_read_byte_from_dpcd(dp, DP_MAX_LANE_COUNT, &data);
 	*lane_count = DPCD_MAX_LANE_COUNT(data);
 }
 
@@ -819,20 +822,20 @@ static void exynos_dp_enable_scramble(struct exynos_dp_device *dp, bool enable)
 		exynos_dp_enable_scrambling(dp);
 
 		exynos_dp_read_byte_from_dpcd(dp,
-			DPCD_ADDR_TRAINING_PATTERN_SET,
+			DP_TRAINING_PATTERN_SET,
 			&data);
 		exynos_dp_write_byte_to_dpcd(dp,
-			DPCD_ADDR_TRAINING_PATTERN_SET,
-			(u8)(data & ~DPCD_SCRAMBLING_DISABLED));
+			DP_TRAINING_PATTERN_SET,
+			(u8)(data & ~DP_LINK_SCRAMBLING_DISABLE));
 	} else {
 		exynos_dp_disable_scrambling(dp);
 
 		exynos_dp_read_byte_from_dpcd(dp,
-			DPCD_ADDR_TRAINING_PATTERN_SET,
+			DP_TRAINING_PATTERN_SET,
 			&data);
 		exynos_dp_write_byte_to_dpcd(dp,
-			DPCD_ADDR_TRAINING_PATTERN_SET,
-			(u8)(data | DPCD_SCRAMBLING_DISABLED));
+			DP_TRAINING_PATTERN_SET,
+			(u8)(data | DP_LINK_SCRAMBLING_DISABLE));
 	}
 }
 
@@ -949,12 +952,6 @@ static int exynos_dp_get_modes(struct drm_connector *connector)
 	return 1;
 }
 
-static int exynos_dp_mode_valid(struct drm_connector *connector,
-			struct drm_display_mode *mode)
-{
-	return MODE_OK;
-}
-
 static struct drm_encoder *exynos_dp_best_encoder(
 			struct drm_connector *connector)
 {
@@ -965,20 +962,9 @@ static struct drm_encoder *exynos_dp_best_encoder(
 
 static struct drm_connector_helper_funcs exynos_dp_connector_helper_funcs = {
 	.get_modes = exynos_dp_get_modes,
-	.mode_valid = exynos_dp_mode_valid,
 	.best_encoder = exynos_dp_best_encoder,
 };
 
-static int exynos_dp_initialize(struct exynos_drm_display *display,
-				struct drm_device *drm_dev)
-{
-	struct exynos_dp_device *dp = display->ctx;
-
-	dp->drm_dev = drm_dev;
-
-	return 0;
-}
-
 static bool find_bridge(const char *compat, struct bridge_init *bridge)
 {
 	bridge->client = NULL;
@@ -1101,12 +1087,11 @@ static void exynos_dp_dpms(struct exynos_drm_display *display, int mode)
 		break;
 	default:
 		break;
-	};
+	}
 	dp->dpms_mode = mode;
 }
 
 static struct exynos_drm_display_ops exynos_dp_display_ops = {
-	.initialize = exynos_dp_initialize,
 	.create_connector = exynos_dp_create_connector,
 	.dpms = exynos_dp_dpms,
 };
@@ -1123,10 +1108,8 @@ static struct video_info *exynos_dp_dt_parse_pdata(struct device *dev)
 
 	dp_video_config = devm_kzalloc(dev,
 				sizeof(*dp_video_config), GFP_KERNEL);
-	if (!dp_video_config) {
-		dev_err(dev, "memory allocation for video config failed\n");
+	if (!dp_video_config)
 		return ERR_PTR(-ENOMEM);
-	}
 
 	dp_video_config->h_sync_polarity =
 		of_property_read_bool(dp_node, "hsync-active-high");
@@ -1185,10 +1168,7 @@ static int exynos_dp_dt_parse_phydata(struct exynos_dp_device *dp)
 	dp_phy_node = of_find_node_by_name(dp_phy_node, "dptx-phy");
 	if (!dp_phy_node) {
 		dp->phy = devm_phy_get(dp->dev, "dp");
-		if (IS_ERR(dp->phy))
-			return PTR_ERR(dp->phy);
-		else
-			return 0;
+		return PTR_ERR_OR_ZERO(dp->phy);
 	}
 
 	if (of_property_read_u32(dp_phy_node, "reg", &phy_base)) {
@@ -1230,19 +1210,20 @@ static int exynos_dp_dt_parse_panel(struct exynos_dp_device *dp)
 	return 0;
 }
 
-static int exynos_dp_probe(struct platform_device *pdev)
+static int exynos_dp_bind(struct device *dev, struct device *master, void *data)
 {
+	struct platform_device *pdev = to_platform_device(dev);
+	struct drm_device *drm_dev = data;
 	struct resource *res;
 	struct exynos_dp_device *dp;
+	unsigned int irq_flags;
 
 	int ret = 0;
 
 	dp = devm_kzalloc(&pdev->dev, sizeof(struct exynos_dp_device),
 				GFP_KERNEL);
-	if (!dp) {
-		dev_err(&pdev->dev, "no memory for device data\n");
+	if (!dp)
 		return -ENOMEM;
-	}
 
 	dp->dev = &pdev->dev;
 	dp->dpms_mode = DRM_MODE_DPMS_OFF;
@@ -1273,7 +1254,30 @@ static int exynos_dp_probe(struct platform_device *pdev)
 	if (IS_ERR(dp->reg_base))
 		return PTR_ERR(dp->reg_base);
 
-	dp->irq = platform_get_irq(pdev, 0);
+	dp->hpd_gpio = of_get_named_gpio(dev->of_node, "samsung,hpd-gpio", 0);
+
+	if (gpio_is_valid(dp->hpd_gpio)) {
+		/*
+		 * Set up the hotplug GPIO from the device tree as an interrupt.
+		 * Simply specifying a different interrupt in the device tree
+		 * doesn't work since we handle hotplug rather differently when
+		 * using a GPIO.  We also need the actual GPIO specifier so
+		 * that we can get the current state of the GPIO.
+		 */
+		ret = devm_gpio_request_one(&pdev->dev, dp->hpd_gpio, GPIOF_IN,
+					    "hpd_gpio");
+		if (ret) {
+			dev_err(&pdev->dev, "failed to get hpd gpio\n");
+			return ret;
+		}
+		dp->irq = gpio_to_irq(dp->hpd_gpio);
+		irq_flags = IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING;
+	} else {
+		dp->hpd_gpio = -ENODEV;
+		dp->irq = platform_get_irq(pdev, 0);
+		irq_flags = 0;
+	}
+
 	if (dp->irq == -ENXIO) {
 		dev_err(&pdev->dev, "failed to get irq\n");
 		return -ENODEV;
@@ -1285,28 +1289,61 @@ static int exynos_dp_probe(struct platform_device *pdev)
 
 	exynos_dp_init_dp(dp);
 
-	ret = devm_request_irq(&pdev->dev, dp->irq, exynos_dp_irq_handler, 0,
-				"exynos-dp", dp);
+	ret = devm_request_irq(&pdev->dev, dp->irq, exynos_dp_irq_handler,
+			irq_flags, "exynos-dp", dp);
 	if (ret) {
 		dev_err(&pdev->dev, "failed to request irq\n");
 		return ret;
 	}
 	disable_irq(dp->irq);
 
+	dp->drm_dev = drm_dev;
 	exynos_dp_display.ctx = dp;
 
 	platform_set_drvdata(pdev, &exynos_dp_display);
-	exynos_drm_display_register(&exynos_dp_display);
 
-	return 0;
+	return exynos_drm_create_enc_conn(drm_dev, &exynos_dp_display);
 }
 
-static int exynos_dp_remove(struct platform_device *pdev)
+static void exynos_dp_unbind(struct device *dev, struct device *master,
+				void *data)
 {
-	struct exynos_drm_display *display = platform_get_drvdata(pdev);
+	struct exynos_drm_display *display = dev_get_drvdata(dev);
+	struct exynos_dp_device *dp = display->ctx;
+	struct drm_encoder *encoder = dp->encoder;
 
 	exynos_dp_dpms(display, DRM_MODE_DPMS_OFF);
-	exynos_drm_display_unregister(&exynos_dp_display);
+
+	encoder->funcs->destroy(encoder);
+	drm_connector_cleanup(&dp->connector);
+}
+
+static const struct component_ops exynos_dp_ops = {
+	.bind	= exynos_dp_bind,
+	.unbind	= exynos_dp_unbind,
+};
+
+static int exynos_dp_probe(struct platform_device *pdev)
+{
+	int ret;
+
+	ret = exynos_drm_component_add(&pdev->dev, EXYNOS_DEVICE_TYPE_CONNECTOR,
+					exynos_dp_display.type);
+	if (ret)
+		return ret;
+
+	ret = component_add(&pdev->dev, &exynos_dp_ops);
+	if (ret)
+		exynos_drm_component_del(&pdev->dev,
+						EXYNOS_DEVICE_TYPE_CONNECTOR);
+
+	return ret;
+}
+
+static int exynos_dp_remove(struct platform_device *pdev)
+{
+	component_del(&pdev->dev, &exynos_dp_ops);
+	exynos_drm_component_del(&pdev->dev, EXYNOS_DEVICE_TYPE_CONNECTOR);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/exynos/exynos_dp_core.h b/drivers/gpu/drm/exynos/exynos_dp_core.h
index d6a900d4ee40..02cc4f9ab903 100644
--- a/drivers/gpu/drm/exynos/exynos_dp_core.h
+++ b/drivers/gpu/drm/exynos/exynos_dp_core.h
@@ -14,6 +14,7 @@
 #define _EXYNOS_DP_CORE_H
 
 #include <drm/drm_crtc.h>
+#include <drm/drm_dp_helper.h>
 #include <drm/exynos_drm.h>
 
 #define DP_TIMEOUT_LOOP_COUNT 100
@@ -159,6 +160,7 @@ struct exynos_dp_device {
 	struct work_struct	hotplug_work;
 	struct phy		*phy;
 	int			dpms_mode;
+	int			hpd_gpio;
 
 	struct exynos_drm_panel_info panel;
 };
@@ -261,69 +263,17 @@ void exynos_dp_disable_scrambling(struct exynos_dp_device *dp);
 #define EDID_EXTENSION_FLAG			0x7e
 #define EDID_CHECKSUM				0x7f
 
-/* Definition for DPCD Register */
-#define DPCD_ADDR_DPCD_REV			0x0000
-#define DPCD_ADDR_MAX_LINK_RATE			0x0001
-#define DPCD_ADDR_MAX_LANE_COUNT		0x0002
-#define DPCD_ADDR_LINK_BW_SET			0x0100
-#define DPCD_ADDR_LANE_COUNT_SET		0x0101
-#define DPCD_ADDR_TRAINING_PATTERN_SET		0x0102
-#define DPCD_ADDR_TRAINING_LANE0_SET		0x0103
-#define DPCD_ADDR_LANE0_1_STATUS		0x0202
-#define DPCD_ADDR_LANE_ALIGN_STATUS_UPDATED	0x0204
-#define DPCD_ADDR_ADJUST_REQUEST_LANE0_1	0x0206
-#define DPCD_ADDR_ADJUST_REQUEST_LANE2_3	0x0207
-#define DPCD_ADDR_TEST_REQUEST			0x0218
-#define DPCD_ADDR_TEST_RESPONSE			0x0260
-#define DPCD_ADDR_TEST_EDID_CHECKSUM		0x0261
-#define DPCD_ADDR_SINK_POWER_STATE		0x0600
-
-/* DPCD_ADDR_MAX_LANE_COUNT */
+/* DP_MAX_LANE_COUNT */
 #define DPCD_ENHANCED_FRAME_CAP(x)		(((x) >> 7) & 0x1)
 #define DPCD_MAX_LANE_COUNT(x)			((x) & 0x1f)
 
-/* DPCD_ADDR_LANE_COUNT_SET */
-#define DPCD_ENHANCED_FRAME_EN			(0x1 << 7)
+/* DP_LANE_COUNT_SET */
 #define DPCD_LANE_COUNT_SET(x)			((x) & 0x1f)
 
-/* DPCD_ADDR_TRAINING_PATTERN_SET */
-#define DPCD_SCRAMBLING_DISABLED		(0x1 << 5)
-#define DPCD_SCRAMBLING_ENABLED			(0x0 << 5)
-#define DPCD_TRAINING_PATTERN_2			(0x2 << 0)
-#define DPCD_TRAINING_PATTERN_1			(0x1 << 0)
-#define DPCD_TRAINING_PATTERN_DISABLED		(0x0 << 0)
-
-/* DPCD_ADDR_TRAINING_LANE0_SET */
-#define DPCD_MAX_PRE_EMPHASIS_REACHED		(0x1 << 5)
+/* DP_TRAINING_LANE0_SET */
 #define DPCD_PRE_EMPHASIS_SET(x)		(((x) & 0x3) << 3)
 #define DPCD_PRE_EMPHASIS_GET(x)		(((x) >> 3) & 0x3)
-#define DPCD_PRE_EMPHASIS_PATTERN2_LEVEL0	(0x0 << 3)
-#define DPCD_MAX_SWING_REACHED			(0x1 << 2)
 #define DPCD_VOLTAGE_SWING_SET(x)		(((x) & 0x3) << 0)
 #define DPCD_VOLTAGE_SWING_GET(x)		(((x) >> 0) & 0x3)
-#define DPCD_VOLTAGE_SWING_PATTERN1_LEVEL0	(0x0 << 0)
-
-/* DPCD_ADDR_LANE0_1_STATUS */
-#define DPCD_LANE_SYMBOL_LOCKED			(0x1 << 2)
-#define DPCD_LANE_CHANNEL_EQ_DONE		(0x1 << 1)
-#define DPCD_LANE_CR_DONE			(0x1 << 0)
-#define DPCD_CHANNEL_EQ_BITS			(DPCD_LANE_CR_DONE|	\
-						 DPCD_LANE_CHANNEL_EQ_DONE|\
-						 DPCD_LANE_SYMBOL_LOCKED)
-
-/* DPCD_ADDR_LANE_ALIGN__STATUS_UPDATED */
-#define DPCD_LINK_STATUS_UPDATED		(0x1 << 7)
-#define DPCD_DOWNSTREAM_PORT_STATUS_CHANGED	(0x1 << 6)
-#define DPCD_INTERLANE_ALIGN_DONE		(0x1 << 0)
-
-/* DPCD_ADDR_TEST_REQUEST */
-#define DPCD_TEST_EDID_READ			(0x1 << 2)
-
-/* DPCD_ADDR_TEST_RESPONSE */
-#define DPCD_TEST_EDID_CHECKSUM_WRITE		(0x1 << 2)
-
-/* DPCD_ADDR_SINK_POWER_STATE */
-#define DPCD_SET_POWER_STATE_D0			(0x1 << 0)
-#define DPCD_SET_POWER_STATE_D4			(0x2 << 0)
 
 #endif /* _EXYNOS_DP_CORE_H */
diff --git a/drivers/gpu/drm/exynos/exynos_dp_reg.c b/drivers/gpu/drm/exynos/exynos_dp_reg.c
index b70da5052ff0..c1f87a2a9284 100644
--- a/drivers/gpu/drm/exynos/exynos_dp_reg.c
+++ b/drivers/gpu/drm/exynos/exynos_dp_reg.c
@@ -13,6 +13,7 @@
 #include <linux/device.h>
 #include <linux/io.h>
 #include <linux/delay.h>
+#include <linux/gpio.h>
 
 #include "exynos_dp_core.h"
 #include "exynos_dp_reg.h"
@@ -326,6 +327,9 @@ void exynos_dp_clear_hotplug_interrupts(struct exynos_dp_device *dp)
 {
 	u32 reg;
 
+	if (gpio_is_valid(dp->hpd_gpio))
+		return;
+
 	reg = HOTPLUG_CHG | HPD_LOST | PLUG;
 	writel(reg, dp->reg_base + EXYNOS_DP_COMMON_INT_STA_4);
 
@@ -337,6 +341,9 @@ void exynos_dp_init_hpd(struct exynos_dp_device *dp)
 {
 	u32 reg;
 
+	if (gpio_is_valid(dp->hpd_gpio))
+		return;
+
 	exynos_dp_clear_hotplug_interrupts(dp);
 
 	reg = readl(dp->reg_base + EXYNOS_DP_SYS_CTL_3);
@@ -348,19 +355,27 @@ enum dp_irq_type exynos_dp_get_irq_type(struct exynos_dp_device *dp)
 {
 	u32 reg;
 
-	/* Parse hotplug interrupt status register */
-	reg = readl(dp->reg_base + EXYNOS_DP_COMMON_INT_STA_4);
+	if (gpio_is_valid(dp->hpd_gpio)) {
+		reg = gpio_get_value(dp->hpd_gpio);
+		if (reg)
+			return DP_IRQ_TYPE_HP_CABLE_IN;
+		else
+			return DP_IRQ_TYPE_HP_CABLE_OUT;
+	} else {
+		/* Parse hotplug interrupt status register */
+		reg = readl(dp->reg_base + EXYNOS_DP_COMMON_INT_STA_4);
 
-	if (reg & PLUG)
-		return DP_IRQ_TYPE_HP_CABLE_IN;
+		if (reg & PLUG)
+			return DP_IRQ_TYPE_HP_CABLE_IN;
 
-	if (reg & HPD_LOST)
-		return DP_IRQ_TYPE_HP_CABLE_OUT;
+		if (reg & HPD_LOST)
+			return DP_IRQ_TYPE_HP_CABLE_OUT;
 
-	if (reg & HOTPLUG_CHG)
-		return DP_IRQ_TYPE_HP_CHANGE;
+		if (reg & HOTPLUG_CHG)
+			return DP_IRQ_TYPE_HP_CHANGE;
 
-	return DP_IRQ_TYPE_UNKNOWN;
+		return DP_IRQ_TYPE_UNKNOWN;
+	}
 }
 
 void exynos_dp_reset_aux(struct exynos_dp_device *dp)
@@ -386,7 +401,7 @@ void exynos_dp_init_aux(struct exynos_dp_device *dp)
 	/* Disable AUX transaction H/W retry */
 	reg = AUX_BIT_PERIOD_EXPECTED_DELAY(3) | AUX_HW_RETRY_COUNT_SEL(0)|
 		AUX_HW_RETRY_INTERVAL_600_MICROSECONDS;
-	writel(reg, dp->reg_base + EXYNOS_DP_AUX_HW_RETRY_CTL) ;
+	writel(reg, dp->reg_base + EXYNOS_DP_AUX_HW_RETRY_CTL);
 
 	/* Receive AUX Channel DEFER commands equal to DEFFER_COUNT*64 */
 	reg = DEFER_CTRL_EN | DEFER_COUNT(1);
@@ -402,9 +417,14 @@ int exynos_dp_get_plug_in_status(struct exynos_dp_device *dp)
 {
 	u32 reg;
 
-	reg = readl(dp->reg_base + EXYNOS_DP_SYS_CTL_3);
-	if (reg & HPD_STATUS)
-		return 0;
+	if (gpio_is_valid(dp->hpd_gpio)) {
+		if (gpio_get_value(dp->hpd_gpio))
+			return 0;
+	} else {
+		reg = readl(dp->reg_base + EXYNOS_DP_SYS_CTL_3);
+		if (reg & HPD_STATUS)
+			return 0;
+	}
 
 	return -EINVAL;
 }
diff --git a/drivers/gpu/drm/exynos/exynos_drm_core.c b/drivers/gpu/drm/exynos/exynos_drm_core.c
index 0e9e06ce36b8..4c9f972eaa07 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_core.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_core.c
@@ -19,21 +19,19 @@
 #include "exynos_drm_fbdev.h"
 
 static LIST_HEAD(exynos_drm_subdrv_list);
-static LIST_HEAD(exynos_drm_manager_list);
-static LIST_HEAD(exynos_drm_display_list);
 
-static int exynos_drm_create_enc_conn(struct drm_device *dev,
+int exynos_drm_create_enc_conn(struct drm_device *dev,
 					struct exynos_drm_display *display)
 {
 	struct drm_encoder *encoder;
-	struct exynos_drm_manager *manager;
 	int ret;
 	unsigned long possible_crtcs = 0;
 
-	/* Find possible crtcs for this display */
-	list_for_each_entry(manager, &exynos_drm_manager_list, list)
-		if (manager->type == display->type)
-			possible_crtcs |= 1 << manager->pipe;
+	ret = exynos_drm_crtc_get_pipe_from_type(dev, display->type);
+	if (ret < 0)
+		return ret;
+
+	possible_crtcs |= 1 << ret;
 
 	/* create and initialize a encoder for this sub driver. */
 	encoder = exynos_drm_encoder_create(dev, display, possible_crtcs);
@@ -57,127 +55,29 @@ err_destroy_encoder:
 	return ret;
 }
 
-static int exynos_drm_subdrv_probe(struct drm_device *dev,
-					struct exynos_drm_subdrv *subdrv)
-{
-	if (subdrv->probe) {
-		int ret;
-
-		subdrv->drm_dev = dev;
-
-		/*
-		 * this probe callback would be called by sub driver
-		 * after setting of all resources to this sub driver,
-		 * such as clock, irq and register map are done or by load()
-		 * of exynos drm driver.
-		 *
-		 * P.S. note that this driver is considered for modularization.
-		 */
-		ret = subdrv->probe(dev, subdrv->dev);
-		if (ret)
-			return ret;
-	}
-
-	return 0;
-}
-
-static void exynos_drm_subdrv_remove(struct drm_device *dev,
-				      struct exynos_drm_subdrv *subdrv)
-{
-	if (subdrv->remove)
-		subdrv->remove(dev, subdrv->dev);
-}
-
-int exynos_drm_initialize_managers(struct drm_device *dev)
+int exynos_drm_subdrv_register(struct exynos_drm_subdrv *subdrv)
 {
-	struct exynos_drm_manager *manager, *n;
-	int ret, pipe = 0;
-
-	list_for_each_entry(manager, &exynos_drm_manager_list, list) {
-		if (manager->ops->initialize) {
-			ret = manager->ops->initialize(manager, dev, pipe);
-			if (ret) {
-				DRM_ERROR("Mgr init [%d] failed with %d\n",
-						manager->type, ret);
-				goto err;
-			}
-		}
+	if (!subdrv)
+		return -EINVAL;
 
-		manager->drm_dev = dev;
-		manager->pipe = pipe++;
+	list_add_tail(&subdrv->list, &exynos_drm_subdrv_list);
 
-		ret = exynos_drm_crtc_create(manager);
-		if (ret) {
-			DRM_ERROR("CRTC create [%d] failed with %d\n",
-					manager->type, ret);
-			goto err;
-		}
-	}
 	return 0;
-
-err:
-	list_for_each_entry_safe(manager, n, &exynos_drm_manager_list, list) {
-		if (pipe-- > 0)
-			exynos_drm_manager_unregister(manager);
-		else
-			list_del(&manager->list);
-	}
-	return ret;
-}
-
-void exynos_drm_remove_managers(struct drm_device *dev)
-{
-	struct exynos_drm_manager *manager, *n;
-
-	list_for_each_entry_safe(manager, n, &exynos_drm_manager_list, list)
-		exynos_drm_manager_unregister(manager);
 }
+EXPORT_SYMBOL_GPL(exynos_drm_subdrv_register);
 
-int exynos_drm_initialize_displays(struct drm_device *dev)
+int exynos_drm_subdrv_unregister(struct exynos_drm_subdrv *subdrv)
 {
-	struct exynos_drm_display *display, *n;
-	int ret, initialized = 0;
-
-	list_for_each_entry(display, &exynos_drm_display_list, list) {
-		if (display->ops->initialize) {
-			ret = display->ops->initialize(display, dev);
-			if (ret) {
-				DRM_ERROR("Display init [%d] failed with %d\n",
-						display->type, ret);
-				goto err;
-			}
-		}
+	if (!subdrv)
+		return -EINVAL;
 
-		initialized++;
+	list_del(&subdrv->list);
 
-		ret = exynos_drm_create_enc_conn(dev, display);
-		if (ret) {
-			DRM_ERROR("Encoder create [%d] failed with %d\n",
-					display->type, ret);
-			goto err;
-		}
-	}
 	return 0;
-
-err:
-	list_for_each_entry_safe(display, n, &exynos_drm_display_list, list) {
-		if (initialized-- > 0)
-			exynos_drm_display_unregister(display);
-		else
-			list_del(&display->list);
-	}
-	return ret;
-}
-
-void exynos_drm_remove_displays(struct drm_device *dev)
-{
-	struct exynos_drm_display *display, *n;
-
-	list_for_each_entry_safe(display, n, &exynos_drm_display_list, list)
-		exynos_drm_display_unregister(display);
 }
+EXPORT_SYMBOL_GPL(exynos_drm_subdrv_unregister);
 
-int exynos_drm_device_register(struct drm_device *dev)
+int exynos_drm_device_subdrv_probe(struct drm_device *dev)
 {
 	struct exynos_drm_subdrv *subdrv, *n;
 	int err;
@@ -186,19 +86,28 @@ int exynos_drm_device_register(struct drm_device *dev)
 		return -EINVAL;
 
 	list_for_each_entry_safe(subdrv, n, &exynos_drm_subdrv_list, list) {
-		err = exynos_drm_subdrv_probe(dev, subdrv);
-		if (err) {
-			DRM_DEBUG("exynos drm subdrv probe failed.\n");
-			list_del(&subdrv->list);
-			continue;
+		if (subdrv->probe) {
+			subdrv->drm_dev = dev;
+
+			/*
+			 * this probe callback would be called by sub driver
+			 * after setting of all resources to this sub driver,
+			 * such as clock, irq and register map are done.
+			 */
+			err = subdrv->probe(dev, subdrv->dev);
+			if (err) {
+				DRM_DEBUG("exynos drm subdrv probe failed.\n");
+				list_del(&subdrv->list);
+				continue;
+			}
 		}
 	}
 
 	return 0;
 }
-EXPORT_SYMBOL_GPL(exynos_drm_device_register);
+EXPORT_SYMBOL_GPL(exynos_drm_device_subdrv_probe);
 
-int exynos_drm_device_unregister(struct drm_device *dev)
+int exynos_drm_device_subdrv_remove(struct drm_device *dev)
 {
 	struct exynos_drm_subdrv *subdrv;
 
@@ -208,66 +117,13 @@ int exynos_drm_device_unregister(struct drm_device *dev)
 	}
 
 	list_for_each_entry(subdrv, &exynos_drm_subdrv_list, list) {
-		exynos_drm_subdrv_remove(dev, subdrv);
+		if (subdrv->remove)
+			subdrv->remove(dev, subdrv->dev);
 	}
 
 	return 0;
 }
-EXPORT_SYMBOL_GPL(exynos_drm_device_unregister);
-
-int exynos_drm_manager_register(struct exynos_drm_manager *manager)
-{
-	BUG_ON(!manager->ops);
-	list_add_tail(&manager->list, &exynos_drm_manager_list);
-	return 0;
-}
-
-int exynos_drm_manager_unregister(struct exynos_drm_manager *manager)
-{
-	if (manager->ops->remove)
-		manager->ops->remove(manager);
-
-	list_del(&manager->list);
-	return 0;
-}
-
-int exynos_drm_display_register(struct exynos_drm_display *display)
-{
-	BUG_ON(!display->ops);
-	list_add_tail(&display->list, &exynos_drm_display_list);
-	return 0;
-}
-
-int exynos_drm_display_unregister(struct exynos_drm_display *display)
-{
-	if (display->ops->remove)
-		display->ops->remove(display);
-
-	list_del(&display->list);
-	return 0;
-}
-
-int exynos_drm_subdrv_register(struct exynos_drm_subdrv *subdrv)
-{
-	if (!subdrv)
-		return -EINVAL;
-
-	list_add_tail(&subdrv->list, &exynos_drm_subdrv_list);
-
-	return 0;
-}
-EXPORT_SYMBOL_GPL(exynos_drm_subdrv_register);
-
-int exynos_drm_subdrv_unregister(struct exynos_drm_subdrv *subdrv)
-{
-	if (!subdrv)
-		return -EINVAL;
-
-	list_del(&subdrv->list);
-
-	return 0;
-}
-EXPORT_SYMBOL_GPL(exynos_drm_subdrv_unregister);
+EXPORT_SYMBOL_GPL(exynos_drm_device_subdrv_remove);
 
 int exynos_drm_subdrv_open(struct drm_device *dev, struct drm_file *file)
 {
diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.c b/drivers/gpu/drm/exynos/exynos_drm_crtc.c
index 1ef5ab9c9d51..95c9435d0266 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_crtc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.c
@@ -368,6 +368,7 @@ int exynos_drm_crtc_create(struct exynos_drm_manager *manager)
 		return -ENOMEM;
 	}
 
+	manager->crtc = &exynos_crtc->drm_crtc;
 	crtc = &exynos_crtc->drm_crtc;
 
 	private->crtc[manager->pipe] = crtc;
@@ -491,3 +492,19 @@ void exynos_drm_crtc_complete_scanout(struct drm_framebuffer *fb)
 			manager->ops->wait_for_vblank(manager);
 	}
 }
+
+int exynos_drm_crtc_get_pipe_from_type(struct drm_device *drm_dev,
+					unsigned int out_type)
+{
+	struct drm_crtc *crtc;
+
+	list_for_each_entry(crtc, &drm_dev->mode_config.crtc_list, head) {
+		struct exynos_drm_crtc *exynos_crtc;
+
+		exynos_crtc = to_exynos_crtc(crtc);
+		if (exynos_crtc->manager->type == out_type)
+			return exynos_crtc->manager->pipe;
+	}
+
+	return -EPERM;
+}
diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.h b/drivers/gpu/drm/exynos/exynos_drm_crtc.h
index c27b66cc5d24..9f74b10a8a01 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_crtc.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.h
@@ -32,4 +32,8 @@ void exynos_drm_crtc_plane_commit(struct drm_crtc *crtc, int zpos);
 void exynos_drm_crtc_plane_enable(struct drm_crtc *crtc, int zpos);
 void exynos_drm_crtc_plane_disable(struct drm_crtc *crtc, int zpos);
 
+/* This function gets pipe value to crtc device matched with out_type. */
+int exynos_drm_crtc_get_pipe_from_type(struct drm_device *drm_dev,
+					unsigned int out_type);
+
 #endif
diff --git a/drivers/gpu/drm/exynos/exynos_drm_dpi.c b/drivers/gpu/drm/exynos/exynos_drm_dpi.c
index 2b09c7c0bfcc..482127f633c5 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_dpi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_dpi.c
@@ -40,20 +40,10 @@ exynos_dpi_detect(struct drm_connector *connector, bool force)
 {
 	struct exynos_dpi *ctx = connector_to_dpi(connector);
 
-	/* panels supported only by boot-loader are always connected */
-	if (!ctx->panel_node)
-		return connector_status_connected;
-
-	if (!ctx->panel) {
-		ctx->panel = of_drm_find_panel(ctx->panel_node);
-		if (ctx->panel)
-			drm_panel_attach(ctx->panel, &ctx->connector);
-	}
-
-	if (ctx->panel)
-		return connector_status_connected;
+	if (!ctx->panel->connector)
+		drm_panel_attach(ctx->panel, &ctx->connector);
 
-	return connector_status_disconnected;
+	return connector_status_connected;
 }
 
 static void exynos_dpi_connector_destroy(struct drm_connector *connector)
@@ -94,12 +84,6 @@ static int exynos_dpi_get_modes(struct drm_connector *connector)
 	return 0;
 }
 
-static int exynos_dpi_mode_valid(struct drm_connector *connector,
-				 struct drm_display_mode *mode)
-{
-	return MODE_OK;
-}
-
 static struct drm_encoder *
 exynos_dpi_best_encoder(struct drm_connector *connector)
 {
@@ -110,7 +94,6 @@ exynos_dpi_best_encoder(struct drm_connector *connector)
 
 static struct drm_connector_helper_funcs exynos_dpi_connector_helper_funcs = {
 	.get_modes = exynos_dpi_get_modes,
-	.mode_valid = exynos_dpi_mode_valid,
 	.best_encoder = exynos_dpi_best_encoder,
 };
 
@@ -123,10 +106,7 @@ static int exynos_dpi_create_connector(struct exynos_drm_display *display,
 
 	ctx->encoder = encoder;
 
-	if (ctx->panel_node)
-		connector->polled = DRM_CONNECTOR_POLL_CONNECT;
-	else
-		connector->polled = DRM_CONNECTOR_POLL_HPD;
+	connector->polled = DRM_CONNECTOR_POLL_HPD;
 
 	ret = drm_connector_init(encoder->dev, connector,
 				 &exynos_dpi_connector_funcs,
@@ -172,7 +152,7 @@ static void exynos_dpi_dpms(struct exynos_drm_display *display, int mode)
 		break;
 	default:
 		break;
-	};
+	}
 	ctx->dpms_mode = mode;
 }
 
@@ -294,8 +274,10 @@ static int exynos_dpi_parse_dt(struct exynos_dpi *ctx)
 			return -ENOMEM;
 
 		ret = of_get_videomode(dn, vm, 0);
-		if (ret < 0)
+		if (ret < 0) {
+			devm_kfree(dev, vm);
 			return ret;
+		}
 
 		ctx->vm = vm;
 
@@ -308,32 +290,58 @@ static int exynos_dpi_parse_dt(struct exynos_dpi *ctx)
 	return 0;
 }
 
-int exynos_dpi_probe(struct device *dev)
+struct exynos_drm_display *exynos_dpi_probe(struct device *dev)
 {
 	struct exynos_dpi *ctx;
 	int ret;
 
+	ret = exynos_drm_component_add(dev,
+					EXYNOS_DEVICE_TYPE_CONNECTOR,
+					exynos_dpi_display.type);
+	if (ret)
+		return ERR_PTR(ret);
+
 	ctx = devm_kzalloc(dev, sizeof(*ctx), GFP_KERNEL);
 	if (!ctx)
-		return -ENOMEM;
+		goto err_del_component;
 
 	ctx->dev = dev;
 	exynos_dpi_display.ctx = ctx;
 	ctx->dpms_mode = DRM_MODE_DPMS_OFF;
 
 	ret = exynos_dpi_parse_dt(ctx);
-	if (ret < 0)
-		return ret;
+	if (ret < 0) {
+		devm_kfree(dev, ctx);
+		goto err_del_component;
+	}
 
-	exynos_drm_display_register(&exynos_dpi_display);
+	if (ctx->panel_node) {
+		ctx->panel = of_drm_find_panel(ctx->panel_node);
+		if (!ctx->panel) {
+			exynos_drm_component_del(dev,
+						EXYNOS_DEVICE_TYPE_CONNECTOR);
+			return ERR_PTR(-EPROBE_DEFER);
+		}
+	}
 
-	return 0;
+	return &exynos_dpi_display;
+
+err_del_component:
+	exynos_drm_component_del(dev, EXYNOS_DEVICE_TYPE_CONNECTOR);
+
+	return NULL;
 }
 
 int exynos_dpi_remove(struct device *dev)
 {
+	struct drm_encoder *encoder = exynos_dpi_display.encoder;
+	struct exynos_dpi *ctx = exynos_dpi_display.ctx;
+
 	exynos_dpi_dpms(&exynos_dpi_display, DRM_MODE_DPMS_OFF);
-	exynos_drm_display_unregister(&exynos_dpi_display);
+	encoder->funcs->destroy(encoder);
+	drm_connector_cleanup(&ctx->connector);
+
+	exynos_drm_component_del(dev, EXYNOS_DEVICE_TYPE_CONNECTOR);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c
index 2d27ba23a6a8..d91f27777537 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c
@@ -16,6 +16,7 @@
 #include <drm/drm_crtc_helper.h>
 
 #include <linux/anon_inodes.h>
+#include <linux/component.h>
 
 #include <drm/exynos_drm.h>
 
@@ -40,9 +41,19 @@
 
 #define VBLANK_OFF_DELAY	50000
 
-/* platform device pointer for eynos drm device. */
 static struct platform_device *exynos_drm_pdev;
 
+static DEFINE_MUTEX(drm_component_lock);
+static LIST_HEAD(drm_component_list);
+
+struct component_dev {
+	struct list_head list;
+	struct device *crtc_dev;
+	struct device *conn_dev;
+	enum exynos_drm_output_type out_type;
+	unsigned int dev_type_flag;
+};
+
 static int exynos_drm_load(struct drm_device *dev, unsigned long flags)
 {
 	struct exynos_drm_private *private;
@@ -73,38 +84,21 @@ static int exynos_drm_load(struct drm_device *dev, unsigned long flags)
 
 	exynos_drm_mode_config_init(dev);
 
-	ret = exynos_drm_initialize_managers(dev);
-	if (ret)
-		goto err_mode_config_cleanup;
-
 	for (nr = 0; nr < MAX_PLANE; nr++) {
 		struct drm_plane *plane;
 		unsigned long possible_crtcs = (1 << MAX_CRTC) - 1;
 
 		plane = exynos_plane_init(dev, possible_crtcs, false);
 		if (!plane)
-			goto err_manager_cleanup;
+			goto err_mode_config_cleanup;
 	}
 
-	ret = exynos_drm_initialize_displays(dev);
-	if (ret)
-		goto err_manager_cleanup;
-
 	/* init kms poll for handling hpd */
 	drm_kms_helper_poll_init(dev);
 
 	ret = drm_vblank_init(dev, MAX_CRTC);
 	if (ret)
-		goto err_display_cleanup;
-
-	/*
-	 * probe sub drivers such as display controller and hdmi driver,
-	 * that were registered at probe() of platform driver
-	 * to the sub driver and create encoder and connector for them.
-	 */
-	ret = exynos_drm_device_register(dev);
-	if (ret)
-		goto err_vblank;
+		goto err_mode_config_cleanup;
 
 	/* setup possible_clones. */
 	exynos_drm_encoder_setup(dev);
@@ -113,17 +107,25 @@ static int exynos_drm_load(struct drm_device *dev, unsigned long flags)
 
 	platform_set_drvdata(dev->platformdev, dev);
 
+	/* Try to bind all sub drivers. */
+	ret = component_bind_all(dev->dev, dev);
+	if (ret)
+		goto err_cleanup_vblank;
+
+	/* Probe non kms sub drivers and virtual display driver. */
+	ret = exynos_drm_device_subdrv_probe(dev);
+	if (ret)
+		goto err_unbind_all;
+
 	/* force connectors detection */
 	drm_helper_hpd_irq_event(dev);
 
 	return 0;
 
-err_vblank:
+err_unbind_all:
+	component_unbind_all(dev->dev, dev);
+err_cleanup_vblank:
 	drm_vblank_cleanup(dev);
-err_display_cleanup:
-	exynos_drm_remove_displays(dev);
-err_manager_cleanup:
-	exynos_drm_remove_managers(dev);
 err_mode_config_cleanup:
 	drm_mode_config_cleanup(dev);
 	drm_release_iommu_mapping(dev);
@@ -135,17 +137,17 @@ err_free_private:
 
 static int exynos_drm_unload(struct drm_device *dev)
 {
+	exynos_drm_device_subdrv_remove(dev);
+
 	exynos_drm_fbdev_fini(dev);
-	exynos_drm_device_unregister(dev);
 	drm_vblank_cleanup(dev);
 	drm_kms_helper_poll_fini(dev);
-	exynos_drm_remove_displays(dev);
-	exynos_drm_remove_managers(dev);
 	drm_mode_config_cleanup(dev);
 
 	drm_release_iommu_mapping(dev);
 	kfree(dev->dev_private);
 
+	component_unbind_all(dev->dev, dev);
 	dev->dev_private = NULL;
 
 	return 0;
@@ -183,9 +185,9 @@ static int exynos_drm_resume(struct drm_device *dev)
 		if (connector->funcs->dpms)
 			connector->funcs->dpms(connector, connector->dpms);
 	}
+	drm_modeset_unlock_all(dev);
 
 	drm_helper_resume_force_mode(dev);
-	drm_modeset_unlock_all(dev);
 
 	return 0;
 }
@@ -323,8 +325,7 @@ static const struct file_operations exynos_drm_driver_fops = {
 };
 
 static struct drm_driver exynos_drm_driver = {
-	.driver_features	= DRIVER_HAVE_IRQ | DRIVER_MODESET |
-					DRIVER_GEM | DRIVER_PRIME,
+	.driver_features	= DRIVER_MODESET | DRIVER_GEM | DRIVER_PRIME,
 	.load			= exynos_drm_load,
 	.unload			= exynos_drm_unload,
 	.suspend		= exynos_drm_suspend,
@@ -355,27 +356,6 @@ static struct drm_driver exynos_drm_driver = {
 	.minor	= DRIVER_MINOR,
 };
 
-static int exynos_drm_platform_probe(struct platform_device *pdev)
-{
-	int ret;
-
-	ret = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32));
-	if (ret)
-		return ret;
-
-	pm_runtime_enable(&pdev->dev);
-	pm_runtime_get_sync(&pdev->dev);
-
-	return drm_platform_init(&exynos_drm_driver, pdev);
-}
-
-static int exynos_drm_platform_remove(struct platform_device *pdev)
-{
-	drm_put_dev(platform_get_drvdata(pdev));
-
-	return 0;
-}
-
 #ifdef CONFIG_PM_SLEEP
 static int exynos_drm_sys_suspend(struct device *dev)
 {
@@ -400,196 +380,319 @@ static int exynos_drm_sys_resume(struct device *dev)
 }
 #endif
 
-#ifdef CONFIG_PM_RUNTIME
-static int exynos_drm_runtime_suspend(struct device *dev)
+static const struct dev_pm_ops exynos_drm_pm_ops = {
+	SET_SYSTEM_SLEEP_PM_OPS(exynos_drm_sys_suspend, exynos_drm_sys_resume)
+};
+
+int exynos_drm_component_add(struct device *dev,
+				enum exynos_drm_device_type dev_type,
+				enum exynos_drm_output_type out_type)
 {
-	struct drm_device *drm_dev = dev_get_drvdata(dev);
-	pm_message_t message;
+	struct component_dev *cdev;
 
-	if (pm_runtime_suspended(dev))
-		return 0;
+	if (dev_type != EXYNOS_DEVICE_TYPE_CRTC &&
+			dev_type != EXYNOS_DEVICE_TYPE_CONNECTOR) {
+		DRM_ERROR("invalid device type.\n");
+		return -EINVAL;
+	}
 
-	message.event = PM_EVENT_SUSPEND;
-	return exynos_drm_suspend(drm_dev, message);
+	mutex_lock(&drm_component_lock);
+
+	/*
+	 * Make sure to check if there is a component which has two device
+	 * objects, for connector and for encoder/connector.
+	 * It should make sure that crtc and encoder/connector drivers are
+	 * ready before exynos drm core binds them.
+	 */
+	list_for_each_entry(cdev, &drm_component_list, list) {
+		if (cdev->out_type == out_type) {
+			/*
+			 * If crtc and encoder/connector device objects are
+			 * added already just return.
+			 */
+			if (cdev->dev_type_flag == (EXYNOS_DEVICE_TYPE_CRTC |
+						EXYNOS_DEVICE_TYPE_CONNECTOR)) {
+				mutex_unlock(&drm_component_lock);
+				return 0;
+			}
+
+			if (dev_type == EXYNOS_DEVICE_TYPE_CRTC) {
+				cdev->crtc_dev = dev;
+				cdev->dev_type_flag |= dev_type;
+			}
+
+			if (dev_type == EXYNOS_DEVICE_TYPE_CONNECTOR) {
+				cdev->conn_dev = dev;
+				cdev->dev_type_flag |= dev_type;
+			}
+
+			mutex_unlock(&drm_component_lock);
+			return 0;
+		}
+	}
+
+	mutex_unlock(&drm_component_lock);
+
+	cdev = kzalloc(sizeof(*cdev), GFP_KERNEL);
+	if (!cdev)
+		return -ENOMEM;
+
+	if (dev_type == EXYNOS_DEVICE_TYPE_CRTC)
+		cdev->crtc_dev = dev;
+	if (dev_type == EXYNOS_DEVICE_TYPE_CONNECTOR)
+		cdev->conn_dev = dev;
+
+	cdev->out_type = out_type;
+	cdev->dev_type_flag = dev_type;
+
+	mutex_lock(&drm_component_lock);
+	list_add_tail(&cdev->list, &drm_component_list);
+	mutex_unlock(&drm_component_lock);
+
+	return 0;
 }
 
-static int exynos_drm_runtime_resume(struct device *dev)
+void exynos_drm_component_del(struct device *dev,
+				enum exynos_drm_device_type dev_type)
 {
-	struct drm_device *drm_dev = dev_get_drvdata(dev);
+	struct component_dev *cdev, *next;
 
-	if (!pm_runtime_suspended(dev))
-		return 0;
+	mutex_lock(&drm_component_lock);
 
-	return exynos_drm_resume(drm_dev);
+	list_for_each_entry_safe(cdev, next, &drm_component_list, list) {
+		if (dev_type == EXYNOS_DEVICE_TYPE_CRTC) {
+			if (cdev->crtc_dev == dev) {
+				cdev->crtc_dev = NULL;
+				cdev->dev_type_flag &= ~dev_type;
+			}
+		}
+
+		if (dev_type == EXYNOS_DEVICE_TYPE_CONNECTOR) {
+			if (cdev->conn_dev == dev) {
+				cdev->conn_dev = NULL;
+				cdev->dev_type_flag &= ~dev_type;
+			}
+		}
+
+		/*
+		 * Release cdev object only in case that both of crtc and
+		 * encoder/connector device objects are NULL.
+		 */
+		if (!cdev->crtc_dev && !cdev->conn_dev) {
+			list_del(&cdev->list);
+			kfree(cdev);
+		}
+
+		break;
+	}
+
+	mutex_unlock(&drm_component_lock);
 }
-#endif
 
-static const struct dev_pm_ops exynos_drm_pm_ops = {
-	SET_SYSTEM_SLEEP_PM_OPS(exynos_drm_sys_suspend, exynos_drm_sys_resume)
-	SET_RUNTIME_PM_OPS(exynos_drm_runtime_suspend,
-			exynos_drm_runtime_resume, NULL)
-};
+static int compare_of(struct device *dev, void *data)
+{
+	return dev == (struct device *)data;
+}
 
-static struct platform_driver exynos_drm_platform_driver = {
-	.probe		= exynos_drm_platform_probe,
-	.remove		= exynos_drm_platform_remove,
-	.driver		= {
-		.owner	= THIS_MODULE,
-		.name	= "exynos-drm",
-		.pm	= &exynos_drm_pm_ops,
-	},
+static int exynos_drm_add_components(struct device *dev, struct master *m)
+{
+	struct component_dev *cdev;
+	unsigned int attach_cnt = 0;
+
+	mutex_lock(&drm_component_lock);
+
+	list_for_each_entry(cdev, &drm_component_list, list) {
+		int ret;
+
+		/*
+		 * Add components to master only in case that crtc and
+		 * encoder/connector device objects exist.
+		 */
+		if (!cdev->crtc_dev || !cdev->conn_dev)
+			continue;
+
+		attach_cnt++;
+
+		mutex_unlock(&drm_component_lock);
+
+		/*
+		 * fimd and dpi modules have same device object so add
+		 * only crtc device object in this case.
+		 *
+		 * TODO. if dpi module follows driver-model driver then
+		 * below codes can be removed.
+		 */
+		if (cdev->crtc_dev == cdev->conn_dev) {
+			ret = component_master_add_child(m, compare_of,
+					cdev->crtc_dev);
+			if (ret < 0)
+				return ret;
+
+			goto out_lock;
+		}
+
+		/*
+		 * Do not chage below call order.
+		 * crtc device first should be added to master because
+		 * connector/encoder need pipe number of crtc when they
+		 * are created.
+		 */
+		ret = component_master_add_child(m, compare_of, cdev->crtc_dev);
+		ret |= component_master_add_child(m, compare_of,
+							cdev->conn_dev);
+		if (ret < 0)
+			return ret;
+
+out_lock:
+		mutex_lock(&drm_component_lock);
+	}
+
+	mutex_unlock(&drm_component_lock);
+
+	return attach_cnt ? 0 : -ENODEV;
+}
+
+static int exynos_drm_bind(struct device *dev)
+{
+	return drm_platform_init(&exynos_drm_driver, to_platform_device(dev));
+}
+
+static void exynos_drm_unbind(struct device *dev)
+{
+	drm_put_dev(dev_get_drvdata(dev));
+}
+
+static const struct component_master_ops exynos_drm_ops = {
+	.add_components = exynos_drm_add_components,
+	.bind		= exynos_drm_bind,
+	.unbind		= exynos_drm_unbind,
 };
 
-static int __init exynos_drm_init(void)
+static int exynos_drm_platform_probe(struct platform_device *pdev)
 {
 	int ret;
 
+	pdev->dev.coherent_dma_mask = DMA_BIT_MASK(32);
+	exynos_drm_driver.num_ioctls = ARRAY_SIZE(exynos_ioctls);
+
+#ifdef CONFIG_DRM_EXYNOS_FIMD
+	ret = platform_driver_register(&fimd_driver);
+	if (ret < 0)
+		return ret;
+#endif
+
 #ifdef CONFIG_DRM_EXYNOS_DP
 	ret = platform_driver_register(&dp_driver);
 	if (ret < 0)
-		goto out_dp;
+		goto err_unregister_fimd_drv;
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_DSI
 	ret = platform_driver_register(&dsi_driver);
 	if (ret < 0)
-		goto out_dsi;
-#endif
-
-#ifdef CONFIG_DRM_EXYNOS_FIMD
-	ret = platform_driver_register(&fimd_driver);
-	if (ret < 0)
-		goto out_fimd;
+		goto err_unregister_dp_drv;
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_HDMI
-	ret = platform_driver_register(&hdmi_driver);
-	if (ret < 0)
-		goto out_hdmi;
 	ret = platform_driver_register(&mixer_driver);
 	if (ret < 0)
-		goto out_mixer;
-#endif
-
-#ifdef CONFIG_DRM_EXYNOS_VIDI
-	ret = platform_driver_register(&vidi_driver);
+		goto err_unregister_dsi_drv;
+	ret = platform_driver_register(&hdmi_driver);
 	if (ret < 0)
-		goto out_vidi;
+		goto err_unregister_mixer_drv;
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_G2D
 	ret = platform_driver_register(&g2d_driver);
 	if (ret < 0)
-		goto out_g2d;
+		goto err_unregister_hdmi_drv;
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_FIMC
 	ret = platform_driver_register(&fimc_driver);
 	if (ret < 0)
-		goto out_fimc;
+		goto err_unregister_g2d_drv;
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_ROTATOR
 	ret = platform_driver_register(&rotator_driver);
 	if (ret < 0)
-		goto out_rotator;
+		goto err_unregister_fimc_drv;
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_GSC
 	ret = platform_driver_register(&gsc_driver);
 	if (ret < 0)
-		goto out_gsc;
+		goto err_unregister_rotator_drv;
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_IPP
 	ret = platform_driver_register(&ipp_driver);
 	if (ret < 0)
-		goto out_ipp;
+		goto err_unregister_gsc_drv;
 
 	ret = exynos_platform_device_ipp_register();
 	if (ret < 0)
-		goto out_ipp_dev;
+		goto err_unregister_ipp_drv;
 #endif
 
-	ret = platform_driver_register(&exynos_drm_platform_driver);
+	ret = component_master_add(&pdev->dev, &exynos_drm_ops);
 	if (ret < 0)
-		goto out_drm;
-
-	exynos_drm_pdev = platform_device_register_simple("exynos-drm", -1,
-				NULL, 0);
-	if (IS_ERR(exynos_drm_pdev)) {
-		ret = PTR_ERR(exynos_drm_pdev);
-		goto out;
-	}
+		DRM_DEBUG_KMS("re-tried by last sub driver probed later.\n");
 
 	return 0;
 
-out:
-	platform_driver_unregister(&exynos_drm_platform_driver);
-
-out_drm:
 #ifdef CONFIG_DRM_EXYNOS_IPP
-	exynos_platform_device_ipp_unregister();
-out_ipp_dev:
+err_unregister_ipp_drv:
 	platform_driver_unregister(&ipp_driver);
-out_ipp:
+err_unregister_gsc_drv:
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_GSC
 	platform_driver_unregister(&gsc_driver);
-out_gsc:
+err_unregister_rotator_drv:
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_ROTATOR
 	platform_driver_unregister(&rotator_driver);
-out_rotator:
+err_unregister_fimc_drv:
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_FIMC
 	platform_driver_unregister(&fimc_driver);
-out_fimc:
+err_unregister_g2d_drv:
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_G2D
 	platform_driver_unregister(&g2d_driver);
-out_g2d:
-#endif
-
-#ifdef CONFIG_DRM_EXYNOS_VIDI
-	platform_driver_unregister(&vidi_driver);
-out_vidi:
+err_unregister_hdmi_drv:
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_HDMI
-	platform_driver_unregister(&mixer_driver);
-out_mixer:
 	platform_driver_unregister(&hdmi_driver);
-out_hdmi:
-#endif
-
-#ifdef CONFIG_DRM_EXYNOS_FIMD
-	platform_driver_unregister(&fimd_driver);
-out_fimd:
+err_unregister_mixer_drv:
+	platform_driver_unregister(&mixer_driver);
+err_unregister_dsi_drv:
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_DSI
 	platform_driver_unregister(&dsi_driver);
-out_dsi:
+err_unregister_dp_drv:
 #endif
 
 #ifdef CONFIG_DRM_EXYNOS_DP
 	platform_driver_unregister(&dp_driver);
-out_dp:
+err_unregister_fimd_drv:
+#endif
+
+#ifdef CONFIG_DRM_EXYNOS_FIMD
+	platform_driver_unregister(&fimd_driver);
 #endif
 	return ret;
 }
 
-static void __exit exynos_drm_exit(void)
+static int exynos_drm_platform_remove(struct platform_device *pdev)
 {
-	platform_device_unregister(exynos_drm_pdev);
-
-	platform_driver_unregister(&exynos_drm_platform_driver);
-
 #ifdef CONFIG_DRM_EXYNOS_IPP
 	exynos_platform_device_ipp_unregister();
 	platform_driver_unregister(&ipp_driver);
@@ -616,10 +719,6 @@ static void __exit exynos_drm_exit(void)
 	platform_driver_unregister(&hdmi_driver);
 #endif
 
-#ifdef CONFIG_DRM_EXYNOS_VIDI
-	platform_driver_unregister(&vidi_driver);
-#endif
-
 #ifdef CONFIG_DRM_EXYNOS_FIMD
 	platform_driver_unregister(&fimd_driver);
 #endif
@@ -631,6 +730,59 @@ static void __exit exynos_drm_exit(void)
 #ifdef CONFIG_DRM_EXYNOS_DP
 	platform_driver_unregister(&dp_driver);
 #endif
+	component_master_del(&pdev->dev, &exynos_drm_ops);
+	return 0;
+}
+
+static struct platform_driver exynos_drm_platform_driver = {
+	.probe	= exynos_drm_platform_probe,
+	.remove	= exynos_drm_platform_remove,
+	.driver	= {
+		.owner	= THIS_MODULE,
+		.name	= "exynos-drm",
+		.pm	= &exynos_drm_pm_ops,
+	},
+};
+
+static int exynos_drm_init(void)
+{
+	int ret;
+
+	exynos_drm_pdev = platform_device_register_simple("exynos-drm", -1,
+								NULL, 0);
+	if (IS_ERR(exynos_drm_pdev))
+		return PTR_ERR(exynos_drm_pdev);
+
+#ifdef CONFIG_DRM_EXYNOS_VIDI
+	ret = exynos_drm_probe_vidi();
+	if (ret < 0)
+		goto err_unregister_pd;
+#endif
+
+	ret = platform_driver_register(&exynos_drm_platform_driver);
+	if (ret)
+		goto err_remove_vidi;
+
+	return 0;
+
+err_unregister_pd:
+	platform_device_unregister(exynos_drm_pdev);
+
+err_remove_vidi:
+#ifdef CONFIG_DRM_EXYNOS_VIDI
+	exynos_drm_remove_vidi();
+#endif
+
+	return ret;
+}
+
+static void exynos_drm_exit(void)
+{
+#ifdef CONFIG_DRM_EXYNOS_VIDI
+	exynos_drm_remove_vidi();
+#endif
+	platform_device_unregister(exynos_drm_pdev);
+	platform_driver_unregister(&exynos_drm_platform_driver);
 }
 
 module_init(exynos_drm_init);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.h b/drivers/gpu/drm/exynos/exynos_drm_drv.h
index ce3e6a30deaa..36535f398848 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.h
@@ -42,6 +42,13 @@ struct drm_connector;
 
 extern unsigned int drm_vblank_offdelay;
 
+/* This enumerates device type. */
+enum exynos_drm_device_type {
+	EXYNOS_DEVICE_TYPE_NONE,
+	EXYNOS_DEVICE_TYPE_CRTC,
+	EXYNOS_DEVICE_TYPE_CONNECTOR,
+};
+
 /* this enumerates display type. */
 enum exynos_drm_output_type {
 	EXYNOS_DISPLAY_TYPE_NONE,
@@ -122,7 +129,6 @@ struct exynos_drm_overlay {
  * Exynos DRM Display Structure.
  *	- this structure is common to analog tv, digital tv and lcd panel.
  *
- * @initialize: initializes the display with drm_dev
  * @remove: cleans up the display for removal
  * @mode_fixup: fix mode data comparing to hw specific display mode.
  * @mode_set: convert drm_display_mode to hw specific display mode and
@@ -133,8 +139,6 @@ struct exynos_drm_overlay {
  */
 struct exynos_drm_display;
 struct exynos_drm_display_ops {
-	int (*initialize)(struct exynos_drm_display *display,
-				struct drm_device *drm_dev);
 	int (*create_connector)(struct exynos_drm_display *display,
 				struct drm_encoder *encoder);
 	void (*remove)(struct exynos_drm_display *display);
@@ -172,8 +176,6 @@ struct exynos_drm_display {
 /*
  * Exynos drm manager ops
  *
- * @initialize: initializes the manager with drm_dev
- * @remove: cleans up the manager for removal
  * @dpms: control device power.
  * @mode_fixup: fix mode data before applying it
  * @mode_set: set the given mode to the manager
@@ -189,9 +191,6 @@ struct exynos_drm_display {
  */
 struct exynos_drm_manager;
 struct exynos_drm_manager_ops {
-	int (*initialize)(struct exynos_drm_manager *mgr,
-				struct drm_device *drm_dev, int pipe);
-	void (*remove)(struct exynos_drm_manager *mgr);
 	void (*dpms)(struct exynos_drm_manager *mgr, int mode);
 	bool (*mode_fixup)(struct exynos_drm_manager *mgr,
 				const struct drm_display_mode *mode,
@@ -215,6 +214,7 @@ struct exynos_drm_manager_ops {
  * @list: the list entry for this manager
  * @type: one of EXYNOS_DISPLAY_TYPE_LCD and HDMI.
  * @drm_dev: pointer to the drm device
+ * @crtc: crtc object.
  * @pipe: the pipe number for this crtc/manager
  * @ops: pointer to callbacks for exynos drm specific functionality
  * @ctx: A pointer to the manager's implementation specific context
@@ -223,6 +223,7 @@ struct exynos_drm_manager {
 	struct list_head list;
 	enum exynos_drm_output_type type;
 	struct drm_device *drm_dev;
+	struct drm_crtc *crtc;
 	int pipe;
 	struct exynos_drm_manager_ops *ops;
 	void *ctx;
@@ -254,6 +255,7 @@ struct drm_exynos_file_private {
  *	otherwise default one.
  * @da_space_size: size of device address space.
  *	if 0 then default value is used for it.
+ * @pipe: the pipe number for this crtc/manager.
  */
 struct exynos_drm_private {
 	struct drm_fb_helper *fb_helper;
@@ -271,6 +273,8 @@ struct exynos_drm_private {
 
 	unsigned long da_start;
 	unsigned long da_space_size;
+
+	unsigned int pipe;
 };
 
 /*
@@ -281,11 +285,11 @@ struct exynos_drm_private {
  * @drm_dev: pointer to drm_device and this pointer would be set
  *	when sub driver calls exynos_drm_subdrv_register().
  * @manager: subdrv has its own manager to control a hardware appropriately
- *	and we can access a hardware drawing on this manager.
+ *     and we can access a hardware drawing on this manager.
  * @probe: this callback would be called by exynos drm driver after
- *	subdrv is registered to it.
+ *     subdrv is registered to it.
  * @remove: this callback is used to release resources created
- *	by probe callback.
+ *     by probe callback.
  * @open: this would be called with drm device file open.
  * @close: this would be called with drm device file close.
  */
@@ -302,39 +306,14 @@ struct exynos_drm_subdrv {
 			struct drm_file *file);
 };
 
-/*
- * this function calls a probe callback registered to sub driver list and
- * create its own encoder and connector and then set drm_device object
- * to global one.
- */
-int exynos_drm_device_register(struct drm_device *dev);
-/*
- * this function calls a remove callback registered to sub driver list and
- * destroy its own encoder and connetor.
- */
-int exynos_drm_device_unregister(struct drm_device *dev);
-
-int exynos_drm_initialize_managers(struct drm_device *dev);
-void exynos_drm_remove_managers(struct drm_device *dev);
-int exynos_drm_initialize_displays(struct drm_device *dev);
-void exynos_drm_remove_displays(struct drm_device *dev);
-
-int exynos_drm_manager_register(struct exynos_drm_manager *manager);
-int exynos_drm_manager_unregister(struct exynos_drm_manager *manager);
-int exynos_drm_display_register(struct exynos_drm_display *display);
-int exynos_drm_display_unregister(struct exynos_drm_display *display);
-
-/*
- * this function would be called by sub drivers such as display controller
- * or hdmi driver to register this sub driver object to exynos drm driver
- * and when a sub driver is registered to exynos drm driver a probe callback
- * of the sub driver is called and creates its own encoder and connector.
- */
+ /* This function would be called by non kms drivers such as g2d and ipp. */
 int exynos_drm_subdrv_register(struct exynos_drm_subdrv *drm_subdrv);
 
 /* this function removes subdrv list from exynos drm driver */
 int exynos_drm_subdrv_unregister(struct exynos_drm_subdrv *drm_subdrv);
 
+int exynos_drm_device_subdrv_probe(struct drm_device *dev);
+int exynos_drm_device_subdrv_remove(struct drm_device *dev);
 int exynos_drm_subdrv_open(struct drm_device *dev, struct drm_file *file);
 void exynos_drm_subdrv_close(struct drm_device *dev, struct drm_file *file);
 
@@ -360,18 +339,40 @@ int exynos_platform_device_ipp_register(void);
 void exynos_platform_device_ipp_unregister(void);
 
 #ifdef CONFIG_DRM_EXYNOS_DPI
-int exynos_dpi_probe(struct device *dev);
+struct exynos_drm_display * exynos_dpi_probe(struct device *dev);
 int exynos_dpi_remove(struct device *dev);
 #else
-static inline int exynos_dpi_probe(struct device *dev) { return 0; }
+static inline struct exynos_drm_display *
+exynos_dpi_probe(struct device *dev) { return 0; }
 static inline int exynos_dpi_remove(struct device *dev) { return 0; }
 #endif
 
+/*
+ * this function registers exynos drm vidi platform device/driver.
+ */
+int exynos_drm_probe_vidi(void);
+
+/*
+ * this function unregister exynos drm vidi platform device/driver.
+ */
+void exynos_drm_remove_vidi(void);
+
+/* This function creates a encoder and a connector, and initializes them. */
+int exynos_drm_create_enc_conn(struct drm_device *dev,
+				struct exynos_drm_display *display);
+
+int exynos_drm_component_add(struct device *dev,
+				enum exynos_drm_device_type dev_type,
+				enum exynos_drm_output_type out_type);
+
+void exynos_drm_component_del(struct device *dev,
+				enum exynos_drm_device_type dev_type);
+
+extern struct platform_driver fimd_driver;
 extern struct platform_driver dp_driver;
 extern struct platform_driver dsi_driver;
-extern struct platform_driver fimd_driver;
-extern struct platform_driver hdmi_driver;
 extern struct platform_driver mixer_driver;
+extern struct platform_driver hdmi_driver;
 extern struct platform_driver exynos_drm_common_hdmi_driver;
 extern struct platform_driver vidi_driver;
 extern struct platform_driver g2d_driver;
diff --git a/drivers/gpu/drm/exynos/exynos_drm_dsi.c b/drivers/gpu/drm/exynos/exynos_drm_dsi.c
index 4ac438187568..6302aa64f6c1 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_dsi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_dsi.c
@@ -19,6 +19,7 @@
 #include <linux/irq.h>
 #include <linux/phy/phy.h>
 #include <linux/regulator/consumer.h>
+#include <linux/component.h>
 
 #include <video/mipi_display.h>
 #include <video/videomode.h>
@@ -1378,16 +1379,60 @@ end:
 	return ret;
 }
 
+static int exynos_dsi_bind(struct device *dev, struct device *master,
+				void *data)
+{
+	struct drm_device *drm_dev = data;
+	struct exynos_dsi *dsi;
+	int ret;
+
+	ret = exynos_drm_create_enc_conn(drm_dev, &exynos_dsi_display);
+	if (ret) {
+		DRM_ERROR("Encoder create [%d] failed with %d\n",
+				exynos_dsi_display.type, ret);
+		return ret;
+	}
+
+	dsi = exynos_dsi_display.ctx;
+
+	return mipi_dsi_host_register(&dsi->dsi_host);
+}
+
+static void exynos_dsi_unbind(struct device *dev, struct device *master,
+				void *data)
+{
+	struct exynos_dsi *dsi = exynos_dsi_display.ctx;
+	struct drm_encoder *encoder = dsi->encoder;
+
+	exynos_dsi_dpms(&exynos_dsi_display, DRM_MODE_DPMS_OFF);
+
+	mipi_dsi_host_unregister(&dsi->dsi_host);
+
+	encoder->funcs->destroy(encoder);
+	drm_connector_cleanup(&dsi->connector);
+}
+
+static const struct component_ops exynos_dsi_component_ops = {
+	.bind	= exynos_dsi_bind,
+	.unbind	= exynos_dsi_unbind,
+};
+
 static int exynos_dsi_probe(struct platform_device *pdev)
 {
 	struct resource *res;
 	struct exynos_dsi *dsi;
 	int ret;
 
+	ret = exynos_drm_component_add(&pdev->dev, EXYNOS_DEVICE_TYPE_CONNECTOR,
+					exynos_dsi_display.type);
+	if (ret)
+		return ret;
+
 	dsi = devm_kzalloc(&pdev->dev, sizeof(*dsi), GFP_KERNEL);
 	if (!dsi) {
 		dev_err(&pdev->dev, "failed to allocate dsi object.\n");
-		return -ENOMEM;
+		ret = -ENOMEM;
+		goto err_del_component;
 	}
 
 	init_completion(&dsi->completed);
@@ -1401,7 +1446,7 @@ static int exynos_dsi_probe(struct platform_device *pdev)
 
 	ret = exynos_dsi_parse_dt(dsi);
 	if (ret)
-		return ret;
+		goto err_del_component;
 
 	dsi->supplies[0].supply = "vddcore";
 	dsi->supplies[1].supply = "vddio";
@@ -1415,32 +1460,37 @@ static int exynos_dsi_probe(struct platform_device *pdev)
 	dsi->pll_clk = devm_clk_get(&pdev->dev, "pll_clk");
 	if (IS_ERR(dsi->pll_clk)) {
 		dev_info(&pdev->dev, "failed to get dsi pll input clock\n");
-		return -EPROBE_DEFER;
+		ret = PTR_ERR(dsi->pll_clk);
+		goto err_del_component;
 	}
 
 	dsi->bus_clk = devm_clk_get(&pdev->dev, "bus_clk");
 	if (IS_ERR(dsi->bus_clk)) {
 		dev_info(&pdev->dev, "failed to get dsi bus clock\n");
-		return -EPROBE_DEFER;
+		ret = PTR_ERR(dsi->bus_clk);
+		goto err_del_component;
 	}
 
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	dsi->reg_base = devm_ioremap_resource(&pdev->dev, res);
 	if (IS_ERR(dsi->reg_base)) {
 		dev_err(&pdev->dev, "failed to remap io region\n");
-		return PTR_ERR(dsi->reg_base);
+		ret = PTR_ERR(dsi->reg_base);
+		goto err_del_component;
 	}
 
 	dsi->phy = devm_phy_get(&pdev->dev, "dsim");
 	if (IS_ERR(dsi->phy)) {
 		dev_info(&pdev->dev, "failed to get dsim phy\n");
-		return -EPROBE_DEFER;
+		ret = PTR_ERR(dsi->phy);
+		goto err_del_component;
 	}
 
 	dsi->irq = platform_get_irq(pdev, 0);
 	if (dsi->irq < 0) {
 		dev_err(&pdev->dev, "failed to request dsi irq resource\n");
-		return dsi->irq;
+		ret = dsi->irq;
+		goto err_del_component;
 	}
 
 	irq_set_status_flags(dsi->irq, IRQ_NOAUTOEN);
@@ -1449,58 +1499,31 @@ static int exynos_dsi_probe(struct platform_device *pdev)
 					dev_name(&pdev->dev), dsi);
 	if (ret) {
 		dev_err(&pdev->dev, "failed to request dsi irq\n");
-		return ret;
+		goto err_del_component;
 	}
 
 	exynos_dsi_display.ctx = dsi;
 
 	platform_set_drvdata(pdev, &exynos_dsi_display);
-	exynos_drm_display_register(&exynos_dsi_display);
-
-	return mipi_dsi_host_register(&dsi->dsi_host);
-}
-
-static int exynos_dsi_remove(struct platform_device *pdev)
-{
-	struct exynos_dsi *dsi = exynos_dsi_display.ctx;
-
-	exynos_dsi_dpms(&exynos_dsi_display, DRM_MODE_DPMS_OFF);
-
-	exynos_drm_display_unregister(&exynos_dsi_display);
-	mipi_dsi_host_unregister(&dsi->dsi_host);
-
-	return 0;
-}
 
-#if CONFIG_PM_SLEEP
-static int exynos_dsi_resume(struct device *dev)
-{
-	struct exynos_dsi *dsi = exynos_dsi_display.ctx;
+	ret = component_add(&pdev->dev, &exynos_dsi_component_ops);
+	if (ret)
+		goto err_del_component;
 
-	if (dsi->state & DSIM_STATE_ENABLED) {
-		dsi->state &= ~DSIM_STATE_ENABLED;
-		exynos_dsi_enable(dsi);
-	}
+	return ret;
 
-	return 0;
+err_del_component:
+	exynos_drm_component_del(&pdev->dev, EXYNOS_DEVICE_TYPE_CONNECTOR);
+	return ret;
 }
 
-static int exynos_dsi_suspend(struct device *dev)
+static int exynos_dsi_remove(struct platform_device *pdev)
 {
-	struct exynos_dsi *dsi = exynos_dsi_display.ctx;
-
-	if (dsi->state & DSIM_STATE_ENABLED) {
-		exynos_dsi_disable(dsi);
-		dsi->state |= DSIM_STATE_ENABLED;
-	}
+	component_del(&pdev->dev, &exynos_dsi_component_ops);
+	exynos_drm_component_del(&pdev->dev, EXYNOS_DEVICE_TYPE_CONNECTOR);
 
 	return 0;
 }
-#endif
-
-static const struct dev_pm_ops exynos_dsi_pm_ops = {
-	SET_SYSTEM_SLEEP_PM_OPS(exynos_dsi_suspend, exynos_dsi_resume)
-};
 
 static struct of_device_id exynos_dsi_of_match[] = {
 	{ .compatible = "samsung,exynos4210-mipi-dsi" },
@@ -1513,7 +1536,6 @@ struct platform_driver dsi_driver = {
 	.driver = {
 		   .name = "exynos-dsi",
 		   .owner = THIS_MODULE,
-		   .pm = &exynos_dsi_pm_ops,
 		   .of_match_table = exynos_dsi_of_match,
 	},
 };
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
index addbf7536da4..d771b467cf0c 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
@@ -121,16 +121,8 @@ static int exynos_drm_fbdev_update(struct drm_fb_helper *helper,
 	offset = fbi->var.xoffset * (fb->bits_per_pixel >> 3);
 	offset += fbi->var.yoffset * fb->pitches[0];
 
-	dev->mode_config.fb_base = (resource_size_t)buffer->dma_addr;
 	fbi->screen_base = buffer->kvaddr + offset;
-	if (is_drm_iommu_supported(dev))
-		fbi->fix.smem_start = (unsigned long)
-			(page_to_phys(sg_page(buffer->sgt->sgl)) + offset);
-	else
-		fbi->fix.smem_start = (unsigned long)buffer->dma_addr;
-
 	fbi->screen_size = size;
-	fbi->fix.smem_len = size;
 
 	return 0;
 }
@@ -237,7 +229,7 @@ static struct drm_fb_helper_funcs exynos_drm_fb_helper_funcs = {
 	.fb_probe =	exynos_drm_fbdev_create,
 };
 
-bool exynos_drm_fbdev_is_anything_connected(struct drm_device *dev)
+static bool exynos_drm_fbdev_is_anything_connected(struct drm_device *dev)
 {
 	struct drm_connector *connector;
 	bool ret = false;
@@ -375,7 +367,5 @@ void exynos_drm_fbdev_restore_mode(struct drm_device *dev)
 	if (!private || !private->fb_helper)
 		return;
 
-	drm_modeset_lock_all(dev);
-	drm_fb_helper_restore_fbdev_mode(private->fb_helper);
-	drm_modeset_unlock_all(dev);
+	drm_fb_helper_restore_fbdev_mode_unlocked(private->fb_helper);
 }
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimc.c b/drivers/gpu/drm/exynos/exynos_drm_fimc.c
index 30d76b2ff9c2..831dde9034c6 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fimc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fimc.c
@@ -18,6 +18,7 @@
 #include <linux/clk.h>
 #include <linux/pm_runtime.h>
 #include <linux/of.h>
+#include <linux/spinlock.h>
 
 #include <drm/drmP.h>
 #include <drm/exynos_drm.h>
@@ -57,7 +58,6 @@
 #define FIMC_SHFACTOR	10
 #define FIMC_BUF_STOP	1
 #define FIMC_BUF_START	2
-#define FIMC_REG_SZ		32
 #define FIMC_WIDTH_ITU_709	1280
 #define FIMC_REFRESH_MAX	60
 #define FIMC_REFRESH_MIN	12
@@ -69,9 +69,6 @@
 #define get_fimc_context(dev)	platform_get_drvdata(to_platform_device(dev))
 #define get_ctx_from_ippdrv(ippdrv)	container_of(ippdrv,\
 					struct fimc_context, ippdrv);
-#define fimc_read(offset)		readl(ctx->regs + (offset))
-#define fimc_write(cfg, offset)	writel(cfg, ctx->regs + (offset))
-
 enum fimc_wb {
 	FIMC_WB_NONE,
 	FIMC_WB_A,
@@ -161,7 +158,7 @@ struct fimc_context {
 	struct exynos_drm_ippdrv	ippdrv;
 	struct resource	*regs_res;
 	void __iomem	*regs;
-	struct mutex	lock;
+	spinlock_t	lock;
 	struct clk	*clocks[FIMC_CLKS_MAX];
 	u32		clk_frequency;
 	struct regmap	*sysreg;
@@ -172,39 +169,53 @@ struct fimc_context {
 	bool	suspended;
 };
 
+static u32 fimc_read(struct fimc_context *ctx, u32 reg)
+{
+	return readl(ctx->regs + reg);
+}
+
+static void fimc_write(struct fimc_context *ctx, u32 val, u32 reg)
+{
+	writel(val, ctx->regs + reg);
+}
+
+static void fimc_set_bits(struct fimc_context *ctx, u32 reg, u32 bits)
+{
+	void __iomem *r = ctx->regs + reg;
+
+	writel(readl(r) | bits, r);
+}
+
+static void fimc_clear_bits(struct fimc_context *ctx, u32 reg, u32 bits)
+{
+	void __iomem *r = ctx->regs + reg;
+
+	writel(readl(r) & ~bits, r);
+}
+
 static void fimc_sw_reset(struct fimc_context *ctx)
 {
 	u32 cfg;
 
 	/* stop dma operation */
-	cfg = fimc_read(EXYNOS_CISTATUS);
-	if (EXYNOS_CISTATUS_GET_ENVID_STATUS(cfg)) {
-		cfg = fimc_read(EXYNOS_MSCTRL);
-		cfg &= ~EXYNOS_MSCTRL_ENVID;
-		fimc_write(cfg, EXYNOS_MSCTRL);
-	}
+	cfg = fimc_read(ctx, EXYNOS_CISTATUS);
+	if (EXYNOS_CISTATUS_GET_ENVID_STATUS(cfg))
+		fimc_clear_bits(ctx, EXYNOS_MSCTRL, EXYNOS_MSCTRL_ENVID);
 
-	cfg = fimc_read(EXYNOS_CISRCFMT);
-	cfg |= EXYNOS_CISRCFMT_ITU601_8BIT;
-	fimc_write(cfg, EXYNOS_CISRCFMT);
+	fimc_set_bits(ctx, EXYNOS_CISRCFMT, EXYNOS_CISRCFMT_ITU601_8BIT);
 
 	/* disable image capture */
-	cfg = fimc_read(EXYNOS_CIIMGCPT);
-	cfg &= ~(EXYNOS_CIIMGCPT_IMGCPTEN_SC | EXYNOS_CIIMGCPT_IMGCPTEN);
-	fimc_write(cfg, EXYNOS_CIIMGCPT);
+	fimc_clear_bits(ctx, EXYNOS_CIIMGCPT,
+		EXYNOS_CIIMGCPT_IMGCPTEN_SC | EXYNOS_CIIMGCPT_IMGCPTEN);
 
 	/* s/w reset */
-	cfg = fimc_read(EXYNOS_CIGCTRL);
-	cfg |= (EXYNOS_CIGCTRL_SWRST);
-	fimc_write(cfg, EXYNOS_CIGCTRL);
+	fimc_set_bits(ctx, EXYNOS_CIGCTRL, EXYNOS_CIGCTRL_SWRST);
 
 	/* s/w reset complete */
-	cfg = fimc_read(EXYNOS_CIGCTRL);
-	cfg &= ~EXYNOS_CIGCTRL_SWRST;
-	fimc_write(cfg, EXYNOS_CIGCTRL);
+	fimc_clear_bits(ctx, EXYNOS_CIGCTRL, EXYNOS_CIGCTRL_SWRST);
 
 	/* reset sequence */
-	fimc_write(0x0, EXYNOS_CIFCNTSEQ);
+	fimc_write(ctx, 0x0, EXYNOS_CIFCNTSEQ);
 }
 
 static int fimc_set_camblk_fimd0_wb(struct fimc_context *ctx)
@@ -220,7 +231,7 @@ static void fimc_set_type_ctrl(struct fimc_context *ctx, enum fimc_wb wb)
 
 	DRM_DEBUG_KMS("wb[%d]\n", wb);
 
-	cfg = fimc_read(EXYNOS_CIGCTRL);
+	cfg = fimc_read(ctx, EXYNOS_CIGCTRL);
 	cfg &= ~(EXYNOS_CIGCTRL_TESTPATTERN_MASK |
 		EXYNOS_CIGCTRL_SELCAM_ITU_MASK |
 		EXYNOS_CIGCTRL_SELCAM_MIPI_MASK |
@@ -246,7 +257,7 @@ static void fimc_set_type_ctrl(struct fimc_context *ctx, enum fimc_wb wb)
 		break;
 	}
 
-	fimc_write(cfg, EXYNOS_CIGCTRL);
+	fimc_write(ctx, cfg, EXYNOS_CIGCTRL);
 }
 
 static void fimc_set_polarity(struct fimc_context *ctx,
@@ -259,7 +270,7 @@ static void fimc_set_polarity(struct fimc_context *ctx,
 	DRM_DEBUG_KMS("inv_href[%d]inv_hsync[%d]\n",
 		pol->inv_href, pol->inv_hsync);
 
-	cfg = fimc_read(EXYNOS_CIGCTRL);
+	cfg = fimc_read(ctx, EXYNOS_CIGCTRL);
 	cfg &= ~(EXYNOS_CIGCTRL_INVPOLPCLK | EXYNOS_CIGCTRL_INVPOLVSYNC |
 		 EXYNOS_CIGCTRL_INVPOLHREF | EXYNOS_CIGCTRL_INVPOLHSYNC);
 
@@ -272,7 +283,7 @@ static void fimc_set_polarity(struct fimc_context *ctx,
 	if (pol->inv_hsync)
 		cfg |= EXYNOS_CIGCTRL_INVPOLHSYNC;
 
-	fimc_write(cfg, EXYNOS_CIGCTRL);
+	fimc_write(ctx, cfg, EXYNOS_CIGCTRL);
 }
 
 static void fimc_handle_jpeg(struct fimc_context *ctx, bool enable)
@@ -281,70 +292,54 @@ static void fimc_handle_jpeg(struct fimc_context *ctx, bool enable)
 
 	DRM_DEBUG_KMS("enable[%d]\n", enable);
 
-	cfg = fimc_read(EXYNOS_CIGCTRL);
+	cfg = fimc_read(ctx, EXYNOS_CIGCTRL);
 	if (enable)
 		cfg |= EXYNOS_CIGCTRL_CAM_JPEG;
 	else
 		cfg &= ~EXYNOS_CIGCTRL_CAM_JPEG;
 
-	fimc_write(cfg, EXYNOS_CIGCTRL);
+	fimc_write(ctx, cfg, EXYNOS_CIGCTRL);
 }
 
-static void fimc_handle_irq(struct fimc_context *ctx, bool enable,
-		bool overflow, bool level)
+static void fimc_mask_irq(struct fimc_context *ctx, bool enable)
 {
 	u32 cfg;
 
-	DRM_DEBUG_KMS("enable[%d]overflow[%d]level[%d]\n",
-			enable, overflow, level);
+	DRM_DEBUG_KMS("enable[%d]\n", enable);
 
-	cfg = fimc_read(EXYNOS_CIGCTRL);
+	cfg = fimc_read(ctx, EXYNOS_CIGCTRL);
 	if (enable) {
-		cfg &= ~(EXYNOS_CIGCTRL_IRQ_OVFEN | EXYNOS_CIGCTRL_IRQ_LEVEL);
-		cfg |= EXYNOS_CIGCTRL_IRQ_ENABLE;
-		if (overflow)
-			cfg |= EXYNOS_CIGCTRL_IRQ_OVFEN;
-		if (level)
-			cfg |= EXYNOS_CIGCTRL_IRQ_LEVEL;
+		cfg &= ~EXYNOS_CIGCTRL_IRQ_OVFEN;
+		cfg |= EXYNOS_CIGCTRL_IRQ_ENABLE | EXYNOS_CIGCTRL_IRQ_LEVEL;
 	} else
-		cfg &= ~(EXYNOS_CIGCTRL_IRQ_OVFEN | EXYNOS_CIGCTRL_IRQ_ENABLE);
-
-	fimc_write(cfg, EXYNOS_CIGCTRL);
+		cfg &= ~EXYNOS_CIGCTRL_IRQ_ENABLE;
+	fimc_write(ctx, cfg, EXYNOS_CIGCTRL);
 }
 
 static void fimc_clear_irq(struct fimc_context *ctx)
 {
-	u32 cfg;
-
-	cfg = fimc_read(EXYNOS_CIGCTRL);
-	cfg |= EXYNOS_CIGCTRL_IRQ_CLR;
-	fimc_write(cfg, EXYNOS_CIGCTRL);
+	fimc_set_bits(ctx, EXYNOS_CIGCTRL, EXYNOS_CIGCTRL_IRQ_CLR);
 }
 
 static bool fimc_check_ovf(struct fimc_context *ctx)
 {
 	struct exynos_drm_ippdrv *ippdrv = &ctx->ippdrv;
-	u32 cfg, status, flag;
+	u32 status, flag;
 
-	status = fimc_read(EXYNOS_CISTATUS);
+	status = fimc_read(ctx, EXYNOS_CISTATUS);
 	flag = EXYNOS_CISTATUS_OVFIY | EXYNOS_CISTATUS_OVFICB |
 		EXYNOS_CISTATUS_OVFICR;
 
 	DRM_DEBUG_KMS("flag[0x%x]\n", flag);
 
 	if (status & flag) {
-		cfg = fimc_read(EXYNOS_CIWDOFST);
-		cfg |= (EXYNOS_CIWDOFST_CLROVFIY | EXYNOS_CIWDOFST_CLROVFICB |
+		fimc_set_bits(ctx, EXYNOS_CIWDOFST,
+			EXYNOS_CIWDOFST_CLROVFIY | EXYNOS_CIWDOFST_CLROVFICB |
 			EXYNOS_CIWDOFST_CLROVFICR);
-
-		fimc_write(cfg, EXYNOS_CIWDOFST);
-
-		cfg = fimc_read(EXYNOS_CIWDOFST);
-		cfg &= ~(EXYNOS_CIWDOFST_CLROVFIY | EXYNOS_CIWDOFST_CLROVFICB |
+		fimc_clear_bits(ctx, EXYNOS_CIWDOFST,
+			EXYNOS_CIWDOFST_CLROVFIY | EXYNOS_CIWDOFST_CLROVFICB |
 			EXYNOS_CIWDOFST_CLROVFICR);
 
-		fimc_write(cfg, EXYNOS_CIWDOFST);
-
 		dev_err(ippdrv->dev, "occurred overflow at %d, status 0x%x.\n",
 			ctx->id, status);
 		return true;
@@ -357,7 +352,7 @@ static bool fimc_check_frame_end(struct fimc_context *ctx)
 {
 	u32 cfg;
 
-	cfg = fimc_read(EXYNOS_CISTATUS);
+	cfg = fimc_read(ctx, EXYNOS_CISTATUS);
 
 	DRM_DEBUG_KMS("cfg[0x%x]\n", cfg);
 
@@ -365,7 +360,7 @@ static bool fimc_check_frame_end(struct fimc_context *ctx)
 		return false;
 
 	cfg &= ~(EXYNOS_CISTATUS_FRAMEEND);
-	fimc_write(cfg, EXYNOS_CISTATUS);
+	fimc_write(ctx, cfg, EXYNOS_CISTATUS);
 
 	return true;
 }
@@ -375,7 +370,7 @@ static int fimc_get_buf_id(struct fimc_context *ctx)
 	u32 cfg;
 	int frame_cnt, buf_id;
 
-	cfg = fimc_read(EXYNOS_CISTATUS2);
+	cfg = fimc_read(ctx, EXYNOS_CISTATUS2);
 	frame_cnt = EXYNOS_CISTATUS2_GET_FRAMECOUNT_BEFORE(cfg);
 
 	if (frame_cnt == 0)
@@ -402,13 +397,13 @@ static void fimc_handle_lastend(struct fimc_context *ctx, bool enable)
 
 	DRM_DEBUG_KMS("enable[%d]\n", enable);
 
-	cfg = fimc_read(EXYNOS_CIOCTRL);
+	cfg = fimc_read(ctx, EXYNOS_CIOCTRL);
 	if (enable)
 		cfg |= EXYNOS_CIOCTRL_LASTENDEN;
 	else
 		cfg &= ~EXYNOS_CIOCTRL_LASTENDEN;
 
-	fimc_write(cfg, EXYNOS_CIOCTRL);
+	fimc_write(ctx, cfg, EXYNOS_CIOCTRL);
 }
 
 
@@ -420,18 +415,18 @@ static int fimc_src_set_fmt_order(struct fimc_context *ctx, u32 fmt)
 	DRM_DEBUG_KMS("fmt[0x%x]\n", fmt);
 
 	/* RGB */
-	cfg = fimc_read(EXYNOS_CISCCTRL);
+	cfg = fimc_read(ctx, EXYNOS_CISCCTRL);
 	cfg &= ~EXYNOS_CISCCTRL_INRGB_FMT_RGB_MASK;
 
 	switch (fmt) {
 	case DRM_FORMAT_RGB565:
 		cfg |= EXYNOS_CISCCTRL_INRGB_FMT_RGB565;
-		fimc_write(cfg, EXYNOS_CISCCTRL);
+		fimc_write(ctx, cfg, EXYNOS_CISCCTRL);
 		return 0;
 	case DRM_FORMAT_RGB888:
 	case DRM_FORMAT_XRGB8888:
 		cfg |= EXYNOS_CISCCTRL_INRGB_FMT_RGB888;
-		fimc_write(cfg, EXYNOS_CISCCTRL);
+		fimc_write(ctx, cfg, EXYNOS_CISCCTRL);
 		return 0;
 	default:
 		/* bypass */
@@ -439,7 +434,7 @@ static int fimc_src_set_fmt_order(struct fimc_context *ctx, u32 fmt)
 	}
 
 	/* YUV */
-	cfg = fimc_read(EXYNOS_MSCTRL);
+	cfg = fimc_read(ctx, EXYNOS_MSCTRL);
 	cfg &= ~(EXYNOS_MSCTRL_ORDER2P_SHIFT_MASK |
 		EXYNOS_MSCTRL_C_INT_IN_2PLANE |
 		EXYNOS_MSCTRL_ORDER422_YCBYCR);
@@ -479,7 +474,7 @@ static int fimc_src_set_fmt_order(struct fimc_context *ctx, u32 fmt)
 		return -EINVAL;
 	}
 
-	fimc_write(cfg, EXYNOS_MSCTRL);
+	fimc_write(ctx, cfg, EXYNOS_MSCTRL);
 
 	return 0;
 }
@@ -492,7 +487,7 @@ static int fimc_src_set_fmt(struct device *dev, u32 fmt)
 
 	DRM_DEBUG_KMS("fmt[0x%x]\n", fmt);
 
-	cfg = fimc_read(EXYNOS_MSCTRL);
+	cfg = fimc_read(ctx, EXYNOS_MSCTRL);
 	cfg &= ~EXYNOS_MSCTRL_INFORMAT_RGB;
 
 	switch (fmt) {
@@ -527,9 +522,9 @@ static int fimc_src_set_fmt(struct device *dev, u32 fmt)
 		return -EINVAL;
 	}
 
-	fimc_write(cfg, EXYNOS_MSCTRL);
+	fimc_write(ctx, cfg, EXYNOS_MSCTRL);
 
-	cfg = fimc_read(EXYNOS_CIDMAPARAM);
+	cfg = fimc_read(ctx, EXYNOS_CIDMAPARAM);
 	cfg &= ~EXYNOS_CIDMAPARAM_R_MODE_MASK;
 
 	if (fmt == DRM_FORMAT_NV12MT)
@@ -537,7 +532,7 @@ static int fimc_src_set_fmt(struct device *dev, u32 fmt)
 	else
 		cfg |= EXYNOS_CIDMAPARAM_R_MODE_LINEAR;
 
-	fimc_write(cfg, EXYNOS_CIDMAPARAM);
+	fimc_write(ctx, cfg, EXYNOS_CIDMAPARAM);
 
 	return fimc_src_set_fmt_order(ctx, fmt);
 }
@@ -552,11 +547,11 @@ static int fimc_src_set_transf(struct device *dev,
 
 	DRM_DEBUG_KMS("degree[%d]flip[0x%x]\n", degree, flip);
 
-	cfg1 = fimc_read(EXYNOS_MSCTRL);
+	cfg1 = fimc_read(ctx, EXYNOS_MSCTRL);
 	cfg1 &= ~(EXYNOS_MSCTRL_FLIP_X_MIRROR |
 		EXYNOS_MSCTRL_FLIP_Y_MIRROR);
 
-	cfg2 = fimc_read(EXYNOS_CITRGFMT);
+	cfg2 = fimc_read(ctx, EXYNOS_CITRGFMT);
 	cfg2 &= ~EXYNOS_CITRGFMT_INROT90_CLOCKWISE;
 
 	switch (degree) {
@@ -595,8 +590,8 @@ static int fimc_src_set_transf(struct device *dev,
 		return -EINVAL;
 	}
 
-	fimc_write(cfg1, EXYNOS_MSCTRL);
-	fimc_write(cfg2, EXYNOS_CITRGFMT);
+	fimc_write(ctx, cfg1, EXYNOS_MSCTRL);
+	fimc_write(ctx, cfg2, EXYNOS_CITRGFMT);
 	*swap = (cfg2 & EXYNOS_CITRGFMT_INROT90_CLOCKWISE) ? 1 : 0;
 
 	return 0;
@@ -621,17 +616,17 @@ static int fimc_set_window(struct fimc_context *ctx,
 	 * set window offset 1, 2 size
 	 * check figure 43-21 in user manual
 	 */
-	cfg = fimc_read(EXYNOS_CIWDOFST);
+	cfg = fimc_read(ctx, EXYNOS_CIWDOFST);
 	cfg &= ~(EXYNOS_CIWDOFST_WINHOROFST_MASK |
 		EXYNOS_CIWDOFST_WINVEROFST_MASK);
 	cfg |= (EXYNOS_CIWDOFST_WINHOROFST(h1) |
 		EXYNOS_CIWDOFST_WINVEROFST(v1));
 	cfg |= EXYNOS_CIWDOFST_WINOFSEN;
-	fimc_write(cfg, EXYNOS_CIWDOFST);
+	fimc_write(ctx, cfg, EXYNOS_CIWDOFST);
 
 	cfg = (EXYNOS_CIWDOFST2_WINHOROFST2(h2) |
 		EXYNOS_CIWDOFST2_WINVEROFST2(v2));
-	fimc_write(cfg, EXYNOS_CIWDOFST2);
+	fimc_write(ctx, cfg, EXYNOS_CIWDOFST2);
 
 	return 0;
 }
@@ -651,7 +646,7 @@ static int fimc_src_set_size(struct device *dev, int swap,
 	cfg = (EXYNOS_ORGISIZE_HORIZONTAL(img_sz.hsize) |
 		EXYNOS_ORGISIZE_VERTICAL(img_sz.vsize));
 
-	fimc_write(cfg, EXYNOS_ORGISIZE);
+	fimc_write(ctx, cfg, EXYNOS_ORGISIZE);
 
 	DRM_DEBUG_KMS("x[%d]y[%d]w[%d]h[%d]\n", pos->x, pos->y, pos->w, pos->h);
 
@@ -663,12 +658,12 @@ static int fimc_src_set_size(struct device *dev, int swap,
 	}
 
 	/* set input DMA image size */
-	cfg = fimc_read(EXYNOS_CIREAL_ISIZE);
+	cfg = fimc_read(ctx, EXYNOS_CIREAL_ISIZE);
 	cfg &= ~(EXYNOS_CIREAL_ISIZE_HEIGHT_MASK |
 		EXYNOS_CIREAL_ISIZE_WIDTH_MASK);
 	cfg |= (EXYNOS_CIREAL_ISIZE_WIDTH(img_pos.w) |
 		EXYNOS_CIREAL_ISIZE_HEIGHT(img_pos.h));
-	fimc_write(cfg, EXYNOS_CIREAL_ISIZE);
+	fimc_write(ctx, cfg, EXYNOS_CIREAL_ISIZE);
 
 	/*
 	 * set input FIFO image size
@@ -677,18 +672,18 @@ static int fimc_src_set_size(struct device *dev, int swap,
 	cfg = (EXYNOS_CISRCFMT_ITU601_8BIT |
 		EXYNOS_CISRCFMT_SOURCEHSIZE(img_sz.hsize) |
 		EXYNOS_CISRCFMT_SOURCEVSIZE(img_sz.vsize));
-	fimc_write(cfg, EXYNOS_CISRCFMT);
+	fimc_write(ctx, cfg, EXYNOS_CISRCFMT);
 
 	/* offset Y(RGB), Cb, Cr */
 	cfg = (EXYNOS_CIIYOFF_HORIZONTAL(img_pos.x) |
 		EXYNOS_CIIYOFF_VERTICAL(img_pos.y));
-	fimc_write(cfg, EXYNOS_CIIYOFF);
+	fimc_write(ctx, cfg, EXYNOS_CIIYOFF);
 	cfg = (EXYNOS_CIICBOFF_HORIZONTAL(img_pos.x) |
 		EXYNOS_CIICBOFF_VERTICAL(img_pos.y));
-	fimc_write(cfg, EXYNOS_CIICBOFF);
+	fimc_write(ctx, cfg, EXYNOS_CIICBOFF);
 	cfg = (EXYNOS_CIICROFF_HORIZONTAL(img_pos.x) |
 		EXYNOS_CIICROFF_VERTICAL(img_pos.y));
-	fimc_write(cfg, EXYNOS_CIICROFF);
+	fimc_write(ctx, cfg, EXYNOS_CIICROFF);
 
 	return fimc_set_window(ctx, &img_pos, &img_sz);
 }
@@ -722,25 +717,25 @@ static int fimc_src_set_addr(struct device *dev,
 	switch (buf_type) {
 	case IPP_BUF_ENQUEUE:
 		config = &property->config[EXYNOS_DRM_OPS_SRC];
-		fimc_write(buf_info->base[EXYNOS_DRM_PLANAR_Y],
+		fimc_write(ctx, buf_info->base[EXYNOS_DRM_PLANAR_Y],
 			EXYNOS_CIIYSA(buf_id));
 
 		if (config->fmt == DRM_FORMAT_YVU420) {
-			fimc_write(buf_info->base[EXYNOS_DRM_PLANAR_CR],
+			fimc_write(ctx, buf_info->base[EXYNOS_DRM_PLANAR_CR],
 				EXYNOS_CIICBSA(buf_id));
-			fimc_write(buf_info->base[EXYNOS_DRM_PLANAR_CB],
+			fimc_write(ctx, buf_info->base[EXYNOS_DRM_PLANAR_CB],
 				EXYNOS_CIICRSA(buf_id));
 		} else {
-			fimc_write(buf_info->base[EXYNOS_DRM_PLANAR_CB],
+			fimc_write(ctx, buf_info->base[EXYNOS_DRM_PLANAR_CB],
 				EXYNOS_CIICBSA(buf_id));
-			fimc_write(buf_info->base[EXYNOS_DRM_PLANAR_CR],
+			fimc_write(ctx, buf_info->base[EXYNOS_DRM_PLANAR_CR],
 				EXYNOS_CIICRSA(buf_id));
 		}
 		break;
 	case IPP_BUF_DEQUEUE:
-		fimc_write(0x0, EXYNOS_CIIYSA(buf_id));
-		fimc_write(0x0, EXYNOS_CIICBSA(buf_id));
-		fimc_write(0x0, EXYNOS_CIICRSA(buf_id));
+		fimc_write(ctx, 0x0, EXYNOS_CIIYSA(buf_id));
+		fimc_write(ctx, 0x0, EXYNOS_CIICBSA(buf_id));
+		fimc_write(ctx, 0x0, EXYNOS_CIICRSA(buf_id));
 		break;
 	default:
 		/* bypass */
@@ -765,22 +760,22 @@ static int fimc_dst_set_fmt_order(struct fimc_context *ctx, u32 fmt)
 	DRM_DEBUG_KMS("fmt[0x%x]\n", fmt);
 
 	/* RGB */
-	cfg = fimc_read(EXYNOS_CISCCTRL);
+	cfg = fimc_read(ctx, EXYNOS_CISCCTRL);
 	cfg &= ~EXYNOS_CISCCTRL_OUTRGB_FMT_RGB_MASK;
 
 	switch (fmt) {
 	case DRM_FORMAT_RGB565:
 		cfg |= EXYNOS_CISCCTRL_OUTRGB_FMT_RGB565;
-		fimc_write(cfg, EXYNOS_CISCCTRL);
+		fimc_write(ctx, cfg, EXYNOS_CISCCTRL);
 		return 0;
 	case DRM_FORMAT_RGB888:
 		cfg |= EXYNOS_CISCCTRL_OUTRGB_FMT_RGB888;
-		fimc_write(cfg, EXYNOS_CISCCTRL);
+		fimc_write(ctx, cfg, EXYNOS_CISCCTRL);
 		return 0;
 	case DRM_FORMAT_XRGB8888:
 		cfg |= (EXYNOS_CISCCTRL_OUTRGB_FMT_RGB888 |
 			EXYNOS_CISCCTRL_EXTRGB_EXTENSION);
-		fimc_write(cfg, EXYNOS_CISCCTRL);
+		fimc_write(ctx, cfg, EXYNOS_CISCCTRL);
 		break;
 	default:
 		/* bypass */
@@ -788,7 +783,7 @@ static int fimc_dst_set_fmt_order(struct fimc_context *ctx, u32 fmt)
 	}
 
 	/* YUV */
-	cfg = fimc_read(EXYNOS_CIOCTRL);
+	cfg = fimc_read(ctx, EXYNOS_CIOCTRL);
 	cfg &= ~(EXYNOS_CIOCTRL_ORDER2P_MASK |
 		EXYNOS_CIOCTRL_ORDER422_MASK |
 		EXYNOS_CIOCTRL_YCBCR_PLANE_MASK);
@@ -830,7 +825,7 @@ static int fimc_dst_set_fmt_order(struct fimc_context *ctx, u32 fmt)
 		return -EINVAL;
 	}
 
-	fimc_write(cfg, EXYNOS_CIOCTRL);
+	fimc_write(ctx, cfg, EXYNOS_CIOCTRL);
 
 	return 0;
 }
@@ -843,16 +838,16 @@ static int fimc_dst_set_fmt(struct device *dev, u32 fmt)
 
 	DRM_DEBUG_KMS("fmt[0x%x]\n", fmt);
 
-	cfg = fimc_read(EXYNOS_CIEXTEN);
+	cfg = fimc_read(ctx, EXYNOS_CIEXTEN);
 
 	if (fmt == DRM_FORMAT_AYUV) {
 		cfg |= EXYNOS_CIEXTEN_YUV444_OUT;
-		fimc_write(cfg, EXYNOS_CIEXTEN);
+		fimc_write(ctx, cfg, EXYNOS_CIEXTEN);
 	} else {
 		cfg &= ~EXYNOS_CIEXTEN_YUV444_OUT;
-		fimc_write(cfg, EXYNOS_CIEXTEN);
+		fimc_write(ctx, cfg, EXYNOS_CIEXTEN);
 
-		cfg = fimc_read(EXYNOS_CITRGFMT);
+		cfg = fimc_read(ctx, EXYNOS_CITRGFMT);
 		cfg &= ~EXYNOS_CITRGFMT_OUTFORMAT_MASK;
 
 		switch (fmt) {
@@ -885,10 +880,10 @@ static int fimc_dst_set_fmt(struct device *dev, u32 fmt)
 			return -EINVAL;
 		}
 
-		fimc_write(cfg, EXYNOS_CITRGFMT);
+		fimc_write(ctx, cfg, EXYNOS_CITRGFMT);
 	}
 
-	cfg = fimc_read(EXYNOS_CIDMAPARAM);
+	cfg = fimc_read(ctx, EXYNOS_CIDMAPARAM);
 	cfg &= ~EXYNOS_CIDMAPARAM_W_MODE_MASK;
 
 	if (fmt == DRM_FORMAT_NV12MT)
@@ -896,7 +891,7 @@ static int fimc_dst_set_fmt(struct device *dev, u32 fmt)
 	else
 		cfg |= EXYNOS_CIDMAPARAM_W_MODE_LINEAR;
 
-	fimc_write(cfg, EXYNOS_CIDMAPARAM);
+	fimc_write(ctx, cfg, EXYNOS_CIDMAPARAM);
 
 	return fimc_dst_set_fmt_order(ctx, fmt);
 }
@@ -911,7 +906,7 @@ static int fimc_dst_set_transf(struct device *dev,
 
 	DRM_DEBUG_KMS("degree[%d]flip[0x%x]\n", degree, flip);
 
-	cfg = fimc_read(EXYNOS_CITRGFMT);
+	cfg = fimc_read(ctx, EXYNOS_CITRGFMT);
 	cfg &= ~EXYNOS_CITRGFMT_FLIP_MASK;
 	cfg &= ~EXYNOS_CITRGFMT_OUTROT90_CLOCKWISE;
 
@@ -951,53 +946,23 @@ static int fimc_dst_set_transf(struct device *dev,
 		return -EINVAL;
 	}
 
-	fimc_write(cfg, EXYNOS_CITRGFMT);
+	fimc_write(ctx, cfg, EXYNOS_CITRGFMT);
 	*swap = (cfg & EXYNOS_CITRGFMT_OUTROT90_CLOCKWISE) ? 1 : 0;
 
 	return 0;
 }
 
-static int fimc_get_ratio_shift(u32 src, u32 dst, u32 *ratio, u32 *shift)
-{
-	DRM_DEBUG_KMS("src[%d]dst[%d]\n", src, dst);
-
-	if (src >= dst * 64) {
-		DRM_ERROR("failed to make ratio and shift.\n");
-		return -EINVAL;
-	} else if (src >= dst * 32) {
-		*ratio = 32;
-		*shift = 5;
-	} else if (src >= dst * 16) {
-		*ratio = 16;
-		*shift = 4;
-	} else if (src >= dst * 8) {
-		*ratio = 8;
-		*shift = 3;
-	} else if (src >= dst * 4) {
-		*ratio = 4;
-		*shift = 2;
-	} else if (src >= dst * 2) {
-		*ratio = 2;
-		*shift = 1;
-	} else {
-		*ratio = 1;
-		*shift = 0;
-	}
-
-	return 0;
-}
-
 static int fimc_set_prescaler(struct fimc_context *ctx, struct fimc_scaler *sc,
 		struct drm_exynos_pos *src, struct drm_exynos_pos *dst)
 {
 	struct exynos_drm_ippdrv *ippdrv = &ctx->ippdrv;
 	u32 cfg, cfg_ext, shfactor;
 	u32 pre_dst_width, pre_dst_height;
-	u32 pre_hratio, hfactor, pre_vratio, vfactor;
+	u32 hfactor, vfactor;
 	int ret = 0;
 	u32 src_w, src_h, dst_w, dst_h;
 
-	cfg_ext = fimc_read(EXYNOS_CITRGFMT);
+	cfg_ext = fimc_read(ctx, EXYNOS_CITRGFMT);
 	if (cfg_ext & EXYNOS_CITRGFMT_INROT90_CLOCKWISE) {
 		src_w = src->h;
 		src_h = src->w;
@@ -1014,24 +979,24 @@ static int fimc_set_prescaler(struct fimc_context *ctx, struct fimc_scaler *sc,
 		dst_h = dst->h;
 	}
 
-	ret = fimc_get_ratio_shift(src_w, dst_w, &pre_hratio, &hfactor);
-	if (ret) {
+	/* fimc_ippdrv_check_property assures that dividers are not null */
+	hfactor = fls(src_w / dst_w / 2);
+	if (hfactor > FIMC_SHFACTOR / 2) {
 		dev_err(ippdrv->dev, "failed to get ratio horizontal.\n");
-		return ret;
+		return -EINVAL;
 	}
 
-	ret = fimc_get_ratio_shift(src_h, dst_h, &pre_vratio, &vfactor);
-	if (ret) {
+	vfactor = fls(src_h / dst_h / 2);
+	if (vfactor > FIMC_SHFACTOR / 2) {
 		dev_err(ippdrv->dev, "failed to get ratio vertical.\n");
-		return ret;
+		return -EINVAL;
 	}
 
-	pre_dst_width = src_w / pre_hratio;
-	pre_dst_height = src_h / pre_vratio;
+	pre_dst_width = src_w >> hfactor;
+	pre_dst_height = src_h >> vfactor;
 	DRM_DEBUG_KMS("pre_dst_width[%d]pre_dst_height[%d]\n",
 		pre_dst_width, pre_dst_height);
-	DRM_DEBUG_KMS("pre_hratio[%d]hfactor[%d]pre_vratio[%d]vfactor[%d]\n",
-		pre_hratio, hfactor, pre_vratio, vfactor);
+	DRM_DEBUG_KMS("hfactor[%d]vfactor[%d]\n", hfactor, vfactor);
 
 	sc->hratio = (src_w << 14) / (dst_w << hfactor);
 	sc->vratio = (src_h << 14) / (dst_h << vfactor);
@@ -1044,13 +1009,13 @@ static int fimc_set_prescaler(struct fimc_context *ctx, struct fimc_scaler *sc,
 	DRM_DEBUG_KMS("shfactor[%d]\n", shfactor);
 
 	cfg = (EXYNOS_CISCPRERATIO_SHFACTOR(shfactor) |
-		EXYNOS_CISCPRERATIO_PREHORRATIO(pre_hratio) |
-		EXYNOS_CISCPRERATIO_PREVERRATIO(pre_vratio));
-	fimc_write(cfg, EXYNOS_CISCPRERATIO);
+		EXYNOS_CISCPRERATIO_PREHORRATIO(1 << hfactor) |
+		EXYNOS_CISCPRERATIO_PREVERRATIO(1 << vfactor));
+	fimc_write(ctx, cfg, EXYNOS_CISCPRERATIO);
 
 	cfg = (EXYNOS_CISCPREDST_PREDSTWIDTH(pre_dst_width) |
 		EXYNOS_CISCPREDST_PREDSTHEIGHT(pre_dst_height));
-	fimc_write(cfg, EXYNOS_CISCPREDST);
+	fimc_write(ctx, cfg, EXYNOS_CISCPREDST);
 
 	return ret;
 }
@@ -1064,7 +1029,7 @@ static void fimc_set_scaler(struct fimc_context *ctx, struct fimc_scaler *sc)
 	DRM_DEBUG_KMS("hratio[%d]vratio[%d]\n",
 		sc->hratio, sc->vratio);
 
-	cfg = fimc_read(EXYNOS_CISCCTRL);
+	cfg = fimc_read(ctx, EXYNOS_CISCCTRL);
 	cfg &= ~(EXYNOS_CISCCTRL_SCALERBYPASS |
 		EXYNOS_CISCCTRL_SCALEUP_H | EXYNOS_CISCCTRL_SCALEUP_V |
 		EXYNOS_CISCCTRL_MAIN_V_RATIO_MASK |
@@ -1084,14 +1049,14 @@ static void fimc_set_scaler(struct fimc_context *ctx, struct fimc_scaler *sc)
 
 	cfg |= (EXYNOS_CISCCTRL_MAINHORRATIO((sc->hratio >> 6)) |
 		EXYNOS_CISCCTRL_MAINVERRATIO((sc->vratio >> 6)));
-	fimc_write(cfg, EXYNOS_CISCCTRL);
+	fimc_write(ctx, cfg, EXYNOS_CISCCTRL);
 
-	cfg_ext = fimc_read(EXYNOS_CIEXTEN);
+	cfg_ext = fimc_read(ctx, EXYNOS_CIEXTEN);
 	cfg_ext &= ~EXYNOS_CIEXTEN_MAINHORRATIO_EXT_MASK;
 	cfg_ext &= ~EXYNOS_CIEXTEN_MAINVERRATIO_EXT_MASK;
 	cfg_ext |= (EXYNOS_CIEXTEN_MAINHORRATIO_EXT(sc->hratio) |
 		EXYNOS_CIEXTEN_MAINVERRATIO_EXT(sc->vratio));
-	fimc_write(cfg_ext, EXYNOS_CIEXTEN);
+	fimc_write(ctx, cfg_ext, EXYNOS_CIEXTEN);
 }
 
 static int fimc_dst_set_size(struct device *dev, int swap,
@@ -1109,12 +1074,12 @@ static int fimc_dst_set_size(struct device *dev, int swap,
 	cfg = (EXYNOS_ORGOSIZE_HORIZONTAL(img_sz.hsize) |
 		EXYNOS_ORGOSIZE_VERTICAL(img_sz.vsize));
 
-	fimc_write(cfg, EXYNOS_ORGOSIZE);
+	fimc_write(ctx, cfg, EXYNOS_ORGOSIZE);
 
 	DRM_DEBUG_KMS("x[%d]y[%d]w[%d]h[%d]\n", pos->x, pos->y, pos->w, pos->h);
 
 	/* CSC ITU */
-	cfg = fimc_read(EXYNOS_CIGCTRL);
+	cfg = fimc_read(ctx, EXYNOS_CIGCTRL);
 	cfg &= ~EXYNOS_CIGCTRL_CSC_MASK;
 
 	if (sz->hsize >= FIMC_WIDTH_ITU_709)
@@ -1122,7 +1087,7 @@ static int fimc_dst_set_size(struct device *dev, int swap,
 	else
 		cfg |= EXYNOS_CIGCTRL_CSC_ITU601;
 
-	fimc_write(cfg, EXYNOS_CIGCTRL);
+	fimc_write(ctx, cfg, EXYNOS_CIGCTRL);
 
 	if (swap) {
 		img_pos.w = pos->h;
@@ -1132,41 +1097,38 @@ static int fimc_dst_set_size(struct device *dev, int swap,
 	}
 
 	/* target image size */
-	cfg = fimc_read(EXYNOS_CITRGFMT);
+	cfg = fimc_read(ctx, EXYNOS_CITRGFMT);
 	cfg &= ~(EXYNOS_CITRGFMT_TARGETH_MASK |
 		EXYNOS_CITRGFMT_TARGETV_MASK);
 	cfg |= (EXYNOS_CITRGFMT_TARGETHSIZE(img_pos.w) |
 		EXYNOS_CITRGFMT_TARGETVSIZE(img_pos.h));
-	fimc_write(cfg, EXYNOS_CITRGFMT);
+	fimc_write(ctx, cfg, EXYNOS_CITRGFMT);
 
 	/* target area */
 	cfg = EXYNOS_CITAREA_TARGET_AREA(img_pos.w * img_pos.h);
-	fimc_write(cfg, EXYNOS_CITAREA);
+	fimc_write(ctx, cfg, EXYNOS_CITAREA);
 
 	/* offset Y(RGB), Cb, Cr */
 	cfg = (EXYNOS_CIOYOFF_HORIZONTAL(img_pos.x) |
 		EXYNOS_CIOYOFF_VERTICAL(img_pos.y));
-	fimc_write(cfg, EXYNOS_CIOYOFF);
+	fimc_write(ctx, cfg, EXYNOS_CIOYOFF);
 	cfg = (EXYNOS_CIOCBOFF_HORIZONTAL(img_pos.x) |
 		EXYNOS_CIOCBOFF_VERTICAL(img_pos.y));
-	fimc_write(cfg, EXYNOS_CIOCBOFF);
+	fimc_write(ctx, cfg, EXYNOS_CIOCBOFF);
 	cfg = (EXYNOS_CIOCROFF_HORIZONTAL(img_pos.x) |
 		EXYNOS_CIOCROFF_VERTICAL(img_pos.y));
-	fimc_write(cfg, EXYNOS_CIOCROFF);
+	fimc_write(ctx, cfg, EXYNOS_CIOCROFF);
 
 	return 0;
 }
 
-static int fimc_dst_get_buf_seq(struct fimc_context *ctx)
+static int fimc_dst_get_buf_count(struct fimc_context *ctx)
 {
-	u32 cfg, i, buf_num = 0;
-	u32 mask = 0x00000001;
+	u32 cfg, buf_num;
 
-	cfg = fimc_read(EXYNOS_CIFCNTSEQ);
+	cfg = fimc_read(ctx, EXYNOS_CIFCNTSEQ);
 
-	for (i = 0; i < FIMC_REG_SZ; i++)
-		if (cfg & (mask << i))
-			buf_num++;
+	buf_num = hweight32(cfg);
 
 	DRM_DEBUG_KMS("buf_num[%d]\n", buf_num);
 
@@ -1181,13 +1143,14 @@ static int fimc_dst_set_buf_seq(struct fimc_context *ctx, u32 buf_id,
 	u32 cfg;
 	u32 mask = 0x00000001 << buf_id;
 	int ret = 0;
+	unsigned long flags;
 
 	DRM_DEBUG_KMS("buf_id[%d]buf_type[%d]\n", buf_id, buf_type);
 
-	mutex_lock(&ctx->lock);
+	spin_lock_irqsave(&ctx->lock, flags);
 
 	/* mask register set */
-	cfg = fimc_read(EXYNOS_CIFCNTSEQ);
+	cfg = fimc_read(ctx, EXYNOS_CIFCNTSEQ);
 
 	switch (buf_type) {
 	case IPP_BUF_ENQUEUE:
@@ -1205,20 +1168,20 @@ static int fimc_dst_set_buf_seq(struct fimc_context *ctx, u32 buf_id,
 	/* sequence id */
 	cfg &= ~mask;
 	cfg |= (enable << buf_id);
-	fimc_write(cfg, EXYNOS_CIFCNTSEQ);
+	fimc_write(ctx, cfg, EXYNOS_CIFCNTSEQ);
 
 	/* interrupt enable */
 	if (buf_type == IPP_BUF_ENQUEUE &&
-	    fimc_dst_get_buf_seq(ctx) >= FIMC_BUF_START)
-		fimc_handle_irq(ctx, true, false, true);
+	    fimc_dst_get_buf_count(ctx) >= FIMC_BUF_START)
+		fimc_mask_irq(ctx, true);
 
 	/* interrupt disable */
 	if (buf_type == IPP_BUF_DEQUEUE &&
-	    fimc_dst_get_buf_seq(ctx) <= FIMC_BUF_STOP)
-		fimc_handle_irq(ctx, false, false, true);
+	    fimc_dst_get_buf_count(ctx) <= FIMC_BUF_STOP)
+		fimc_mask_irq(ctx, false);
 
 err_unlock:
-	mutex_unlock(&ctx->lock);
+	spin_unlock_irqrestore(&ctx->lock, flags);
 	return ret;
 }
 
@@ -1252,25 +1215,25 @@ static int fimc_dst_set_addr(struct device *dev,
 	case IPP_BUF_ENQUEUE:
 		config = &property->config[EXYNOS_DRM_OPS_DST];
 
-		fimc_write(buf_info->base[EXYNOS_DRM_PLANAR_Y],
+		fimc_write(ctx, buf_info->base[EXYNOS_DRM_PLANAR_Y],
 			EXYNOS_CIOYSA(buf_id));
 
 		if (config->fmt == DRM_FORMAT_YVU420) {
-			fimc_write(buf_info->base[EXYNOS_DRM_PLANAR_CR],
+			fimc_write(ctx, buf_info->base[EXYNOS_DRM_PLANAR_CR],
 				EXYNOS_CIOCBSA(buf_id));
-			fimc_write(buf_info->base[EXYNOS_DRM_PLANAR_CB],
+			fimc_write(ctx, buf_info->base[EXYNOS_DRM_PLANAR_CB],
 				EXYNOS_CIOCRSA(buf_id));
 		} else {
-			fimc_write(buf_info->base[EXYNOS_DRM_PLANAR_CB],
+			fimc_write(ctx, buf_info->base[EXYNOS_DRM_PLANAR_CB],
 				EXYNOS_CIOCBSA(buf_id));
-			fimc_write(buf_info->base[EXYNOS_DRM_PLANAR_CR],
+			fimc_write(ctx, buf_info->base[EXYNOS_DRM_PLANAR_CR],
 				EXYNOS_CIOCRSA(buf_id));
 		}
 		break;
 	case IPP_BUF_DEQUEUE:
-		fimc_write(0x0, EXYNOS_CIOYSA(buf_id));
-		fimc_write(0x0, EXYNOS_CIOCBSA(buf_id));
-		fimc_write(0x0, EXYNOS_CIOCRSA(buf_id));
+		fimc_write(ctx, 0x0, EXYNOS_CIOYSA(buf_id));
+		fimc_write(ctx, 0x0, EXYNOS_CIOCBSA(buf_id));
+		fimc_write(ctx, 0x0, EXYNOS_CIOCRSA(buf_id));
 		break;
 	default:
 		/* bypass */
@@ -1342,11 +1305,7 @@ static irqreturn_t fimc_irq_handler(int irq, void *dev_id)
 
 static int fimc_init_prop_list(struct exynos_drm_ippdrv *ippdrv)
 {
-	struct drm_exynos_ipp_prop_list *prop_list;
-
-	prop_list = devm_kzalloc(ippdrv->dev, sizeof(*prop_list), GFP_KERNEL);
-	if (!prop_list)
-		return -ENOMEM;
+	struct drm_exynos_ipp_prop_list *prop_list = &ippdrv->prop_list;
 
 	prop_list->version = 1;
 	prop_list->writeback = 1;
@@ -1371,8 +1330,6 @@ static int fimc_init_prop_list(struct exynos_drm_ippdrv *ippdrv)
 	prop_list->scale_min.hsize = FIMC_SCALE_MIN;
 	prop_list->scale_min.vsize = FIMC_SCALE_MIN;
 
-	ippdrv->prop_list = prop_list;
-
 	return 0;
 }
 
@@ -1395,7 +1352,7 @@ static int fimc_ippdrv_check_property(struct device *dev,
 {
 	struct fimc_context *ctx = get_fimc_context(dev);
 	struct exynos_drm_ippdrv *ippdrv = &ctx->ippdrv;
-	struct drm_exynos_ipp_prop_list *pp = ippdrv->prop_list;
+	struct drm_exynos_ipp_prop_list *pp = &ippdrv->prop_list;
 	struct drm_exynos_ipp_config *config;
 	struct drm_exynos_pos *pos;
 	struct drm_exynos_sz *sz;
@@ -1508,15 +1465,15 @@ static void fimc_clear_addr(struct fimc_context *ctx)
 	int i;
 
 	for (i = 0; i < FIMC_MAX_SRC; i++) {
-		fimc_write(0, EXYNOS_CIIYSA(i));
-		fimc_write(0, EXYNOS_CIICBSA(i));
-		fimc_write(0, EXYNOS_CIICRSA(i));
+		fimc_write(ctx, 0, EXYNOS_CIIYSA(i));
+		fimc_write(ctx, 0, EXYNOS_CIICBSA(i));
+		fimc_write(ctx, 0, EXYNOS_CIICRSA(i));
 	}
 
 	for (i = 0; i < FIMC_MAX_DST; i++) {
-		fimc_write(0, EXYNOS_CIOYSA(i));
-		fimc_write(0, EXYNOS_CIOCBSA(i));
-		fimc_write(0, EXYNOS_CIOCRSA(i));
+		fimc_write(ctx, 0, EXYNOS_CIOYSA(i));
+		fimc_write(ctx, 0, EXYNOS_CIOCBSA(i));
+		fimc_write(ctx, 0, EXYNOS_CIOCRSA(i));
 	}
 }
 
@@ -1556,7 +1513,7 @@ static int fimc_ippdrv_start(struct device *dev, enum drm_exynos_ipp_cmd cmd)
 
 	property = &c_node->property;
 
-	fimc_handle_irq(ctx, true, false, true);
+	fimc_mask_irq(ctx, true);
 
 	for_each_ipp_ops(i) {
 		config = &property->config[i];
@@ -1582,10 +1539,10 @@ static int fimc_ippdrv_start(struct device *dev, enum drm_exynos_ipp_cmd cmd)
 		fimc_handle_lastend(ctx, false);
 
 		/* setup dma */
-		cfg0 = fimc_read(EXYNOS_MSCTRL);
+		cfg0 = fimc_read(ctx, EXYNOS_MSCTRL);
 		cfg0 &= ~EXYNOS_MSCTRL_INPUT_MASK;
 		cfg0 |= EXYNOS_MSCTRL_INPUT_MEMORY;
-		fimc_write(cfg0, EXYNOS_MSCTRL);
+		fimc_write(ctx, cfg0, EXYNOS_MSCTRL);
 		break;
 	case IPP_CMD_WB:
 		fimc_set_type_ctrl(ctx, FIMC_WB_A);
@@ -1610,41 +1567,33 @@ static int fimc_ippdrv_start(struct device *dev, enum drm_exynos_ipp_cmd cmd)
 	}
 
 	/* Reset status */
-	fimc_write(0x0, EXYNOS_CISTATUS);
+	fimc_write(ctx, 0x0, EXYNOS_CISTATUS);
 
-	cfg0 = fimc_read(EXYNOS_CIIMGCPT);
+	cfg0 = fimc_read(ctx, EXYNOS_CIIMGCPT);
 	cfg0 &= ~EXYNOS_CIIMGCPT_IMGCPTEN_SC;
 	cfg0 |= EXYNOS_CIIMGCPT_IMGCPTEN_SC;
 
 	/* Scaler */
-	cfg1 = fimc_read(EXYNOS_CISCCTRL);
+	cfg1 = fimc_read(ctx, EXYNOS_CISCCTRL);
 	cfg1 &= ~EXYNOS_CISCCTRL_SCAN_MASK;
 	cfg1 |= (EXYNOS_CISCCTRL_PROGRESSIVE |
 		EXYNOS_CISCCTRL_SCALERSTART);
 
-	fimc_write(cfg1, EXYNOS_CISCCTRL);
+	fimc_write(ctx, cfg1, EXYNOS_CISCCTRL);
 
 	/* Enable image capture*/
 	cfg0 |= EXYNOS_CIIMGCPT_IMGCPTEN;
-	fimc_write(cfg0, EXYNOS_CIIMGCPT);
+	fimc_write(ctx, cfg0, EXYNOS_CIIMGCPT);
 
 	/* Disable frame end irq */
-	cfg0 = fimc_read(EXYNOS_CIGCTRL);
-	cfg0 &= ~EXYNOS_CIGCTRL_IRQ_END_DISABLE;
-	fimc_write(cfg0, EXYNOS_CIGCTRL);
+	fimc_clear_bits(ctx, EXYNOS_CIGCTRL, EXYNOS_CIGCTRL_IRQ_END_DISABLE);
 
-	cfg0 = fimc_read(EXYNOS_CIOCTRL);
-	cfg0 &= ~EXYNOS_CIOCTRL_WEAVE_MASK;
-	fimc_write(cfg0, EXYNOS_CIOCTRL);
+	fimc_clear_bits(ctx, EXYNOS_CIOCTRL, EXYNOS_CIOCTRL_WEAVE_MASK);
 
 	if (cmd == IPP_CMD_M2M) {
-		cfg0 = fimc_read(EXYNOS_MSCTRL);
-		cfg0 |= EXYNOS_MSCTRL_ENVID;
-		fimc_write(cfg0, EXYNOS_MSCTRL);
+		fimc_set_bits(ctx, EXYNOS_MSCTRL, EXYNOS_MSCTRL_ENVID);
 
-		cfg0 = fimc_read(EXYNOS_MSCTRL);
-		cfg0 |= EXYNOS_MSCTRL_ENVID;
-		fimc_write(cfg0, EXYNOS_MSCTRL);
+		fimc_set_bits(ctx, EXYNOS_MSCTRL, EXYNOS_MSCTRL_ENVID);
 	}
 
 	return 0;
@@ -1661,10 +1610,10 @@ static void fimc_ippdrv_stop(struct device *dev, enum drm_exynos_ipp_cmd cmd)
 	switch (cmd) {
 	case IPP_CMD_M2M:
 		/* Source clear */
-		cfg = fimc_read(EXYNOS_MSCTRL);
+		cfg = fimc_read(ctx, EXYNOS_MSCTRL);
 		cfg &= ~EXYNOS_MSCTRL_INPUT_MASK;
 		cfg &= ~EXYNOS_MSCTRL_ENVID;
-		fimc_write(cfg, EXYNOS_MSCTRL);
+		fimc_write(ctx, cfg, EXYNOS_MSCTRL);
 		break;
 	case IPP_CMD_WB:
 		exynos_drm_ippnb_send_event(IPP_SET_WRITEBACK, (void *)&set_wb);
@@ -1675,25 +1624,20 @@ static void fimc_ippdrv_stop(struct device *dev, enum drm_exynos_ipp_cmd cmd)
 		break;
 	}
 
-	fimc_handle_irq(ctx, false, false, true);
+	fimc_mask_irq(ctx, false);
 
 	/* reset sequence */
-	fimc_write(0x0, EXYNOS_CIFCNTSEQ);
+	fimc_write(ctx, 0x0, EXYNOS_CIFCNTSEQ);
 
 	/* Scaler disable */
-	cfg = fimc_read(EXYNOS_CISCCTRL);
-	cfg &= ~EXYNOS_CISCCTRL_SCALERSTART;
-	fimc_write(cfg, EXYNOS_CISCCTRL);
+	fimc_clear_bits(ctx, EXYNOS_CISCCTRL, EXYNOS_CISCCTRL_SCALERSTART);
 
 	/* Disable image capture */
-	cfg = fimc_read(EXYNOS_CIIMGCPT);
-	cfg &= ~(EXYNOS_CIIMGCPT_IMGCPTEN_SC | EXYNOS_CIIMGCPT_IMGCPTEN);
-	fimc_write(cfg, EXYNOS_CIIMGCPT);
+	fimc_clear_bits(ctx, EXYNOS_CIIMGCPT,
+		EXYNOS_CIIMGCPT_IMGCPTEN_SC | EXYNOS_CIIMGCPT_IMGCPTEN);
 
 	/* Enable frame end irq */
-	cfg = fimc_read(EXYNOS_CIGCTRL);
-	cfg |= EXYNOS_CIGCTRL_IRQ_END_DISABLE;
-	fimc_write(cfg, EXYNOS_CIGCTRL);
+	fimc_set_bits(ctx, EXYNOS_CIGCTRL, EXYNOS_CIGCTRL_IRQ_END_DISABLE);
 }
 
 static void fimc_put_clocks(struct fimc_context *ctx)
@@ -1848,7 +1792,7 @@ static int fimc_probe(struct platform_device *pdev)
 
 	DRM_DEBUG_KMS("id[%d]ippdrv[0x%x]\n", ctx->id, (int)ippdrv);
 
-	mutex_init(&ctx->lock);
+	spin_lock_init(&ctx->lock);
 	platform_set_drvdata(pdev, ctx);
 
 	pm_runtime_set_active(dev);
@@ -1879,7 +1823,6 @@ static int fimc_remove(struct platform_device *pdev)
 	struct exynos_drm_ippdrv *ippdrv = &ctx->ippdrv;
 
 	exynos_drm_ippdrv_unregister(ippdrv);
-	mutex_destroy(&ctx->lock);
 
 	fimc_put_clocks(ctx);
 	pm_runtime_set_suspended(dev);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
index 40fd6ccfcd6f..bb45ab2e7384 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
@@ -19,6 +19,7 @@
 #include <linux/of.h>
 #include <linux/of_device.h>
 #include <linux/pm_runtime.h>
+#include <linux/component.h>
 
 #include <video/of_display_timing.h>
 #include <video/of_videomode.h>
@@ -38,6 +39,7 @@
  */
 
 #define FIMD_DEFAULT_FRAMERATE 60
+#define MIN_FB_WIDTH_FOR_16WORD_BURST 128
 
 /* position control register for hardware window 0, 2 ~ 4.*/
 #define VIDOSD_A(win)		(VIDOSD_BASE + 0x00 + (win) * 16)
@@ -122,6 +124,7 @@ struct fimd_context {
 
 	struct exynos_drm_panel_info panel;
 	struct fimd_driver_data *driver_data;
+	struct exynos_drm_display *display;
 };
 
 static const struct of_device_id fimd_driver_dt_match[] = {
@@ -143,13 +146,57 @@ static inline struct fimd_driver_data *drm_fimd_get_driver_data(
 	return (struct fimd_driver_data *)of_id->data;
 }
 
+static void fimd_wait_for_vblank(struct exynos_drm_manager *mgr)
+{
+	struct fimd_context *ctx = mgr->ctx;
+
+	if (ctx->suspended)
+		return;
+
+	atomic_set(&ctx->wait_vsync_event, 1);
+
+	/*
+	 * wait for FIMD to signal VSYNC interrupt or return after
+	 * timeout which is set to 50ms (refresh rate of 20).
+	 */
+	if (!wait_event_timeout(ctx->wait_vsync_queue,
+				!atomic_read(&ctx->wait_vsync_event),
+				HZ/20))
+		DRM_DEBUG_KMS("vblank wait timed out.\n");
+}
+
+
+static void fimd_clear_channel(struct exynos_drm_manager *mgr)
+{
+	struct fimd_context *ctx = mgr->ctx;
+	int win, ch_enabled = 0;
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	/* Check if any channel is enabled. */
+	for (win = 0; win < WINDOWS_NR; win++) {
+		u32 val = readl(ctx->regs + SHADOWCON);
+		if (val & SHADOWCON_CHx_ENABLE(win)) {
+			val &= ~SHADOWCON_CHx_ENABLE(win);
+			writel(val, ctx->regs + SHADOWCON);
+			ch_enabled = 1;
+		}
+	}
+
+	/* Wait for vsync, as disable channel takes effect at next vsync */
+	if (ch_enabled)
+		fimd_wait_for_vblank(mgr);
+}
+
 static int fimd_mgr_initialize(struct exynos_drm_manager *mgr,
-			struct drm_device *drm_dev, int pipe)
+			struct drm_device *drm_dev)
 {
 	struct fimd_context *ctx = mgr->ctx;
+	struct exynos_drm_private *priv;
+	priv = drm_dev->dev_private;
 
-	ctx->drm_dev = drm_dev;
-	ctx->pipe = pipe;
+	mgr->drm_dev = ctx->drm_dev = drm_dev;
+	mgr->pipe = ctx->pipe = priv->pipe++;
 
 	/*
 	 * enable drm irq mode.
@@ -169,8 +216,14 @@ static int fimd_mgr_initialize(struct exynos_drm_manager *mgr,
 	drm_dev->vblank_disable_allowed = true;
 
 	/* attach this sub driver to iommu mapping if supported. */
-	if (is_drm_iommu_supported(ctx->drm_dev))
+	if (is_drm_iommu_supported(ctx->drm_dev)) {
+		/*
+		 * If any channel is already active, iommu will throw
+		 * a PAGE FAULT when enabled. So clear any channel if enabled.
+		 */
+		fimd_clear_channel(mgr);
 		drm_iommu_attach_device(ctx->drm_dev, ctx->dev);
+	}
 
 	return 0;
 }
@@ -324,25 +377,6 @@ static void fimd_disable_vblank(struct exynos_drm_manager *mgr)
 	}
 }
 
-static void fimd_wait_for_vblank(struct exynos_drm_manager *mgr)
-{
-	struct fimd_context *ctx = mgr->ctx;
-
-	if (ctx->suspended)
-		return;
-
-	atomic_set(&ctx->wait_vsync_event, 1);
-
-	/*
-	 * wait for FIMD to signal VSYNC interrupt or return after
-	 * timeout which is set to 50ms (refresh rate of 20).
-	 */
-	if (!wait_event_timeout(ctx->wait_vsync_queue,
-				!atomic_read(&ctx->wait_vsync_event),
-				HZ/20))
-		DRM_DEBUG_KMS("vblank wait timed out.\n");
-}
-
 static void fimd_win_mode_set(struct exynos_drm_manager *mgr,
 			struct exynos_drm_overlay *overlay)
 {
@@ -446,6 +480,19 @@ static void fimd_win_set_pixfmt(struct fimd_context *ctx, unsigned int win)
 
 	DRM_DEBUG_KMS("bpp = %d\n", win_data->bpp);
 
+	/*
+	 * In case of exynos, setting dma-burst to 16Word causes permanent
+	 * tearing for very small buffers, e.g. cursor buffer. Burst Mode
+	 * switching which is based on overlay size is not recommended as
+	 * overlay size varies alot towards the end of the screen and rapid
+	 * movement causes unstable DMA which results into iommu crash/tear.
+	 */
+
+	if (win_data->fb_width < MIN_FB_WIDTH_FOR_16WORD_BURST) {
+		val &= ~WINCONx_BURSTLEN_MASK;
+		val |= WINCONx_BURSTLEN_4WORD;
+	}
+
 	writel(val, ctx->regs + WINCON(win));
 }
 
@@ -656,19 +703,6 @@ static void fimd_win_disable(struct exynos_drm_manager *mgr, int zpos)
 	win_data->enabled = false;
 }
 
-static void fimd_clear_win(struct fimd_context *ctx, int win)
-{
-	writel(0, ctx->regs + WINCON(win));
-	writel(0, ctx->regs + VIDOSD_A(win));
-	writel(0, ctx->regs + VIDOSD_B(win));
-	writel(0, ctx->regs + VIDOSD_C(win));
-
-	if (win == 1 || win == 2)
-		writel(0, ctx->regs + VIDOSD_D(win));
-
-	fimd_shadow_protect_win(ctx, win, false);
-}
-
 static void fimd_window_suspend(struct exynos_drm_manager *mgr)
 {
 	struct fimd_context *ctx = mgr->ctx;
@@ -803,8 +837,6 @@ static void fimd_dpms(struct exynos_drm_manager *mgr, int mode)
 }
 
 static struct exynos_drm_manager_ops fimd_manager_ops = {
-	.initialize = fimd_mgr_initialize,
-	.remove = fimd_mgr_remove,
 	.dpms = fimd_dpms,
 	.mode_fixup = fimd_mode_fixup,
 	.mode_set = fimd_mode_set,
@@ -849,20 +881,64 @@ out:
 	return IRQ_HANDLED;
 }
 
+static int fimd_bind(struct device *dev, struct device *master, void *data)
+{
+	struct fimd_context *ctx = fimd_manager.ctx;
+	struct drm_device *drm_dev = data;
+
+	fimd_mgr_initialize(&fimd_manager, drm_dev);
+	exynos_drm_crtc_create(&fimd_manager);
+	if (ctx->display)
+		exynos_drm_create_enc_conn(drm_dev, ctx->display);
+
+	return 0;
+
+}
+
+static void fimd_unbind(struct device *dev, struct device *master,
+			void *data)
+{
+	struct exynos_drm_manager *mgr = dev_get_drvdata(dev);
+	struct fimd_context *ctx = fimd_manager.ctx;
+	struct drm_crtc *crtc = mgr->crtc;
+
+	fimd_dpms(mgr, DRM_MODE_DPMS_OFF);
+
+	if (ctx->display)
+		exynos_dpi_remove(dev);
+
+	fimd_mgr_remove(mgr);
+
+	crtc->funcs->destroy(crtc);
+}
+
+static const struct component_ops fimd_component_ops = {
+	.bind	= fimd_bind,
+	.unbind = fimd_unbind,
+};
+
 static int fimd_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
 	struct fimd_context *ctx;
 	struct resource *res;
-	int win;
 	int ret = -EINVAL;
 
-	if (!dev->of_node)
-		return -ENODEV;
+	ret = exynos_drm_component_add(&pdev->dev, EXYNOS_DEVICE_TYPE_CRTC,
+					fimd_manager.type);
+	if (ret)
+		return ret;
+
+	if (!dev->of_node) {
+		ret = -ENODEV;
+		goto err_del_component;
+	}
 
 	ctx = devm_kzalloc(dev, sizeof(*ctx), GFP_KERNEL);
-	if (!ctx)
-		return -ENOMEM;
+	if (!ctx) {
+		ret = -ENOMEM;
+		goto err_del_component;
+	}
 
 	ctx->dev = dev;
 	ctx->suspended = true;
@@ -875,32 +951,37 @@ static int fimd_probe(struct platform_device *pdev)
 	ctx->bus_clk = devm_clk_get(dev, "fimd");
 	if (IS_ERR(ctx->bus_clk)) {
 		dev_err(dev, "failed to get bus clock\n");
-		return PTR_ERR(ctx->bus_clk);
+		ret = PTR_ERR(ctx->bus_clk);
+		goto err_del_component;
 	}
 
 	ctx->lcd_clk = devm_clk_get(dev, "sclk_fimd");
 	if (IS_ERR(ctx->lcd_clk)) {
 		dev_err(dev, "failed to get lcd clock\n");
-		return PTR_ERR(ctx->lcd_clk);
+		ret = PTR_ERR(ctx->lcd_clk);
+		goto err_del_component;
 	}
 
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 
 	ctx->regs = devm_ioremap_resource(dev, res);
-	if (IS_ERR(ctx->regs))
-		return PTR_ERR(ctx->regs);
+	if (IS_ERR(ctx->regs)) {
+		ret = PTR_ERR(ctx->regs);
+		goto err_del_component;
+	}
 
 	res = platform_get_resource_byname(pdev, IORESOURCE_IRQ, "vsync");
 	if (!res) {
 		dev_err(dev, "irq request failed.\n");
-		return -ENXIO;
+		ret = -ENXIO;
+		goto err_del_component;
 	}
 
 	ret = devm_request_irq(dev, res->start, fimd_irq_handler,
 							0, "drm_fimd", ctx);
 	if (ret) {
 		dev_err(dev, "irq request failed.\n");
-		return ret;
+		goto err_del_component;
 	}
 
 	ctx->driver_data = drm_fimd_get_driver_data(pdev);
@@ -910,30 +991,34 @@ static int fimd_probe(struct platform_device *pdev)
 	platform_set_drvdata(pdev, &fimd_manager);
 
 	fimd_manager.ctx = ctx;
-	exynos_drm_manager_register(&fimd_manager);
 
-	exynos_dpi_probe(ctx->dev);
+	ctx->display = exynos_dpi_probe(dev);
+	if (IS_ERR(ctx->display))
+		return PTR_ERR(ctx->display);
 
-	pm_runtime_enable(dev);
+	pm_runtime_enable(&pdev->dev);
 
-	for (win = 0; win < WINDOWS_NR; win++)
-		fimd_clear_win(ctx, win);
+	ret = component_add(&pdev->dev, &fimd_component_ops);
+	if (ret)
+		goto err_disable_pm_runtime;
 
-	return 0;
+	return ret;
+
+err_disable_pm_runtime:
+	pm_runtime_disable(&pdev->dev);
+
+err_del_component:
+	exynos_drm_component_del(&pdev->dev, EXYNOS_DEVICE_TYPE_CRTC);
+	return ret;
 }
 
 static int fimd_remove(struct platform_device *pdev)
 {
-	struct exynos_drm_manager *mgr = platform_get_drvdata(pdev);
-
-	exynos_dpi_remove(&pdev->dev);
-
-	exynos_drm_manager_unregister(&fimd_manager);
-
-	fimd_dpms(mgr, DRM_MODE_DPMS_OFF);
-
 	pm_runtime_disable(&pdev->dev);
 
+	component_del(&pdev->dev, &fimd_component_ops);
+	exynos_drm_component_del(&pdev->dev, EXYNOS_DEVICE_TYPE_CRTC);
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c
index 42d2904d88c7..163a054922cb 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gem.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c
@@ -612,22 +612,20 @@ int exynos_drm_gem_dumb_create(struct drm_file *file_priv,
 	args->pitch = args->width * ((args->bpp + 7) / 8);
 	args->size = args->pitch * args->height;
 
-	exynos_gem_obj = exynos_drm_gem_create(dev, EXYNOS_BO_CONTIG |
-						EXYNOS_BO_WC, args->size);
-	/*
-	 * If physically contiguous memory allocation fails and if IOMMU is
-	 * supported then try to get buffer from non physically contiguous
-	 * memory area.
-	 */
-	if (IS_ERR(exynos_gem_obj) && is_drm_iommu_supported(dev)) {
-		dev_warn(dev->dev, "contiguous FB allocation failed, falling back to non-contiguous\n");
+	if (is_drm_iommu_supported(dev)) {
+		exynos_gem_obj = exynos_drm_gem_create(dev,
+			EXYNOS_BO_NONCONTIG | EXYNOS_BO_WC,
+			args->size);
+	} else {
 		exynos_gem_obj = exynos_drm_gem_create(dev,
-					EXYNOS_BO_NONCONTIG | EXYNOS_BO_WC,
-					args->size);
+			EXYNOS_BO_CONTIG | EXYNOS_BO_WC,
+			args->size);
 	}
 
-	if (IS_ERR(exynos_gem_obj))
+	if (IS_ERR(exynos_gem_obj)) {
+		dev_warn(dev->dev, "FB allocation failed.\n");
 		return PTR_ERR(exynos_gem_obj);
+	}
 
 	ret = exynos_drm_gem_handle_create(&exynos_gem_obj->base, file_priv,
 			&args->handle);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gsc.c b/drivers/gpu/drm/exynos/exynos_drm_gsc.c
index fa75059a6104..9e3ff1672965 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gsc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_gsc.c
@@ -1335,11 +1335,7 @@ static irqreturn_t gsc_irq_handler(int irq, void *dev_id)
 
 static int gsc_init_prop_list(struct exynos_drm_ippdrv *ippdrv)
 {
-	struct drm_exynos_ipp_prop_list *prop_list;
-
-	prop_list = devm_kzalloc(ippdrv->dev, sizeof(*prop_list), GFP_KERNEL);
-	if (!prop_list)
-		return -ENOMEM;
+	struct drm_exynos_ipp_prop_list *prop_list = &ippdrv->prop_list;
 
 	prop_list->version = 1;
 	prop_list->writeback = 1;
@@ -1363,8 +1359,6 @@ static int gsc_init_prop_list(struct exynos_drm_ippdrv *ippdrv)
 	prop_list->scale_min.hsize = GSC_SCALE_MIN;
 	prop_list->scale_min.vsize = GSC_SCALE_MIN;
 
-	ippdrv->prop_list = prop_list;
-
 	return 0;
 }
 
@@ -1387,7 +1381,7 @@ static int gsc_ippdrv_check_property(struct device *dev,
 {
 	struct gsc_context *ctx = get_gsc_context(dev);
 	struct exynos_drm_ippdrv *ippdrv = &ctx->ippdrv;
-	struct drm_exynos_ipp_prop_list *pp = ippdrv->prop_list;
+	struct drm_exynos_ipp_prop_list *pp = &ippdrv->prop_list;
 	struct drm_exynos_ipp_config *config;
 	struct drm_exynos_pos *pos;
 	struct drm_exynos_sz *sz;
diff --git a/drivers/gpu/drm/exynos/exynos_drm_ipp.c b/drivers/gpu/drm/exynos/exynos_drm_ipp.c
index 3d78144387ac..a1888e128f1d 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_ipp.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_ipp.c
@@ -167,6 +167,13 @@ static int ipp_create_id(struct idr *id_idr, struct mutex *lock, void *obj,
 	return 0;
 }
 
+static void ipp_remove_id(struct idr *id_idr, struct mutex *lock, u32 id)
+{
+	mutex_lock(lock);
+	idr_remove(id_idr, id);
+	mutex_unlock(lock);
+}
+
 static void *ipp_find_obj(struct idr *id_idr, struct mutex *lock, u32 id)
 {
 	void *obj;
@@ -276,11 +283,6 @@ static struct exynos_drm_ippdrv *ipp_find_drv_by_handle(u32 prop_id)
 
 	DRM_DEBUG_KMS("prop_id[%d]\n", prop_id);
 
-	if (list_empty(&exynos_drm_ippdrv_list)) {
-		DRM_DEBUG_KMS("ippdrv_list is empty.\n");
-		return ERR_PTR(-ENODEV);
-	}
-
 	/*
 	 * This case is search ipp driver by prop_id handle.
 	 * sometimes, ipp subsystem find driver by prop_id.
@@ -289,11 +291,14 @@ static struct exynos_drm_ippdrv *ipp_find_drv_by_handle(u32 prop_id)
 	list_for_each_entry(ippdrv, &exynos_drm_ippdrv_list, drv_list) {
 		DRM_DEBUG_KMS("count[%d]ippdrv[0x%x]\n", count++, (int)ippdrv);
 
-		if (!list_empty(&ippdrv->cmd_list)) {
-			list_for_each_entry(c_node, &ippdrv->cmd_list, list)
-				if (c_node->property.prop_id == prop_id)
-					return ippdrv;
+		mutex_lock(&ippdrv->cmd_lock);
+		list_for_each_entry(c_node, &ippdrv->cmd_list, list) {
+			if (c_node->property.prop_id == prop_id) {
+				mutex_unlock(&ippdrv->cmd_lock);
+				return ippdrv;
+			}
 		}
+		mutex_unlock(&ippdrv->cmd_lock);
 	}
 
 	return ERR_PTR(-ENODEV);
@@ -325,6 +330,7 @@ int exynos_drm_ipp_get_property(struct drm_device *drm_dev, void *data,
 	if (!prop_list->ipp_id) {
 		list_for_each_entry(ippdrv, &exynos_drm_ippdrv_list, drv_list)
 			count++;
+
 		/*
 		 * Supports ippdrv list count for user application.
 		 * First step user application getting ippdrv count.
@@ -346,7 +352,7 @@ int exynos_drm_ipp_get_property(struct drm_device *drm_dev, void *data,
 			return PTR_ERR(ippdrv);
 		}
 
-		prop_list = ippdrv->prop_list;
+		*prop_list = ippdrv->prop_list;
 	}
 
 	return 0;
@@ -386,9 +392,11 @@ static int ipp_find_and_set_property(struct drm_exynos_ipp_property *property)
 	 * when we find this command no using prop_id.
 	 * return property information set in this command node.
 	 */
+	mutex_lock(&ippdrv->cmd_lock);
 	list_for_each_entry(c_node, &ippdrv->cmd_list, list) {
 		if ((c_node->property.prop_id == prop_id) &&
 		    (c_node->state == IPP_STATE_STOP)) {
+			mutex_unlock(&ippdrv->cmd_lock);
 			DRM_DEBUG_KMS("found cmd[%d]ippdrv[0x%x]\n",
 				property->cmd, (int)ippdrv);
 
@@ -396,6 +404,7 @@ static int ipp_find_and_set_property(struct drm_exynos_ipp_property *property)
 			return 0;
 		}
 	}
+	mutex_unlock(&ippdrv->cmd_lock);
 
 	DRM_ERROR("failed to search property.\n");
 
@@ -499,7 +508,7 @@ int exynos_drm_ipp_set_property(struct drm_device *drm_dev, void *data,
 	c_node->start_work = ipp_create_cmd_work();
 	if (IS_ERR(c_node->start_work)) {
 		DRM_ERROR("failed to create start work.\n");
-		goto err_clear;
+		goto err_remove_id;
 	}
 
 	c_node->stop_work = ipp_create_cmd_work();
@@ -514,7 +523,7 @@ int exynos_drm_ipp_set_property(struct drm_device *drm_dev, void *data,
 		goto err_free_stop;
 	}
 
-	mutex_init(&c_node->cmd_lock);
+	mutex_init(&c_node->lock);
 	mutex_init(&c_node->mem_lock);
 	mutex_init(&c_node->event_lock);
 
@@ -526,7 +535,9 @@ int exynos_drm_ipp_set_property(struct drm_device *drm_dev, void *data,
 
 	INIT_LIST_HEAD(&c_node->event_list);
 	list_splice_init(&priv->event_list, &c_node->event_list);
+	mutex_lock(&ippdrv->cmd_lock);
 	list_add_tail(&c_node->list, &ippdrv->cmd_list);
+	mutex_unlock(&ippdrv->cmd_lock);
 
 	/* make dedicated state without m2m */
 	if (!ipp_is_m2m_cmd(property->cmd))
@@ -538,18 +549,24 @@ err_free_stop:
 	kfree(c_node->stop_work);
 err_free_start:
 	kfree(c_node->start_work);
+err_remove_id:
+	ipp_remove_id(&ctx->prop_idr, &ctx->prop_lock, property->prop_id);
 err_clear:
 	kfree(c_node);
 	return ret;
 }
 
-static void ipp_clean_cmd_node(struct drm_exynos_ipp_cmd_node *c_node)
+static void ipp_clean_cmd_node(struct ipp_context *ctx,
+				struct drm_exynos_ipp_cmd_node *c_node)
 {
 	/* delete list */
 	list_del(&c_node->list);
 
+	ipp_remove_id(&ctx->prop_idr, &ctx->prop_lock,
+			c_node->property.prop_id);
+
 	/* destroy mutex */
-	mutex_destroy(&c_node->cmd_lock);
+	mutex_destroy(&c_node->lock);
 	mutex_destroy(&c_node->mem_lock);
 	mutex_destroy(&c_node->event_lock);
 
@@ -567,17 +584,10 @@ static int ipp_check_mem_list(struct drm_exynos_ipp_cmd_node *c_node)
 	struct list_head *head;
 	int ret, i, count[EXYNOS_DRM_OPS_MAX] = { 0, };
 
-	mutex_lock(&c_node->mem_lock);
-
 	for_each_ipp_ops(i) {
 		/* source/destination memory list */
 		head = &c_node->mem_list[i];
 
-		if (list_empty(head)) {
-			DRM_DEBUG_KMS("%s memory empty.\n", i ? "dst" : "src");
-			continue;
-		}
-
 		/* find memory node entry */
 		list_for_each_entry(m_node, head, list) {
 			DRM_DEBUG_KMS("%s,count[%d]m_node[0x%x]\n",
@@ -602,8 +612,6 @@ static int ipp_check_mem_list(struct drm_exynos_ipp_cmd_node *c_node)
 		ret = max(count[EXYNOS_DRM_OPS_SRC],
 			count[EXYNOS_DRM_OPS_DST]);
 
-	mutex_unlock(&c_node->mem_lock);
-
 	return ret;
 }
 
@@ -646,16 +654,13 @@ static int ipp_set_mem_node(struct exynos_drm_ippdrv *ippdrv,
 		return -EFAULT;
 	}
 
-	mutex_lock(&c_node->mem_lock);
-
 	DRM_DEBUG_KMS("ops_id[%d]\n", m_node->ops_id);
 
 	/* get operations callback */
 	ops = ippdrv->ops[m_node->ops_id];
 	if (!ops) {
 		DRM_ERROR("not support ops.\n");
-		ret = -EFAULT;
-		goto err_unlock;
+		return -EFAULT;
 	}
 
 	/* set address and enable irq */
@@ -664,12 +669,10 @@ static int ipp_set_mem_node(struct exynos_drm_ippdrv *ippdrv,
 			m_node->buf_id, IPP_BUF_ENQUEUE);
 		if (ret) {
 			DRM_ERROR("failed to set addr.\n");
-			goto err_unlock;
+			return ret;
 		}
 	}
 
-err_unlock:
-	mutex_unlock(&c_node->mem_lock);
 	return ret;
 }
 
@@ -684,11 +687,9 @@ static struct drm_exynos_ipp_mem_node
 	void *addr;
 	int i;
 
-	mutex_lock(&c_node->mem_lock);
-
 	m_node = kzalloc(sizeof(*m_node), GFP_KERNEL);
 	if (!m_node)
-		goto err_unlock;
+		return ERR_PTR(-ENOMEM);
 
 	/* clear base address for error handling */
 	memset(&buf_info, 0x0, sizeof(buf_info));
@@ -722,15 +723,14 @@ static struct drm_exynos_ipp_mem_node
 
 	m_node->filp = file;
 	m_node->buf_info = buf_info;
+	mutex_lock(&c_node->mem_lock);
 	list_add_tail(&m_node->list, &c_node->mem_list[qbuf->ops_id]);
-
 	mutex_unlock(&c_node->mem_lock);
+
 	return m_node;
 
 err_clear:
 	kfree(m_node);
-err_unlock:
-	mutex_unlock(&c_node->mem_lock);
 	return ERR_PTR(-EFAULT);
 }
 
@@ -747,13 +747,6 @@ static int ipp_put_mem_node(struct drm_device *drm_dev,
 		return -EFAULT;
 	}
 
-	if (list_empty(&m_node->list)) {
-		DRM_ERROR("empty memory node.\n");
-		return -ENOMEM;
-	}
-
-	mutex_lock(&c_node->mem_lock);
-
 	DRM_DEBUG_KMS("ops_id[%d]\n", m_node->ops_id);
 
 	/* put gem buffer */
@@ -768,8 +761,6 @@ static int ipp_put_mem_node(struct drm_device *drm_dev,
 	list_del(&m_node->list);
 	kfree(m_node);
 
-	mutex_unlock(&c_node->mem_lock);
-
 	return 0;
 }
 
@@ -805,7 +796,9 @@ static int ipp_get_event(struct drm_device *drm_dev,
 	e->base.event = &e->event.base;
 	e->base.file_priv = file;
 	e->base.destroy = ipp_free_event;
+	mutex_lock(&c_node->event_lock);
 	list_add_tail(&e->base.link, &c_node->event_list);
+	mutex_unlock(&c_node->event_lock);
 
 	return 0;
 }
@@ -816,11 +809,7 @@ static void ipp_put_event(struct drm_exynos_ipp_cmd_node *c_node,
 	struct drm_exynos_ipp_send_event *e, *te;
 	int count = 0;
 
-	if (list_empty(&c_node->event_list)) {
-		DRM_DEBUG_KMS("event_list is empty.\n");
-		return;
-	}
-
+	mutex_lock(&c_node->event_lock);
 	list_for_each_entry_safe(e, te, &c_node->event_list, base.link) {
 		DRM_DEBUG_KMS("count[%d]e[0x%x]\n", count++, (int)e);
 
@@ -841,9 +830,13 @@ static void ipp_put_event(struct drm_exynos_ipp_cmd_node *c_node,
 			/* delete list */
 			list_del(&e->base.link);
 			kfree(e);
-			return;
+			goto out_unlock;
 		}
 	}
+
+out_unlock:
+	mutex_unlock(&c_node->event_lock);
+	return;
 }
 
 static void ipp_handle_cmd_work(struct device *dev,
@@ -887,7 +880,9 @@ static int ipp_queue_buf_with_run(struct device *dev,
 		return 0;
 	}
 
+	mutex_lock(&c_node->mem_lock);
 	if (!ipp_check_mem_list(c_node)) {
+		mutex_unlock(&c_node->mem_lock);
 		DRM_DEBUG_KMS("empty memory.\n");
 		return 0;
 	}
@@ -904,10 +899,12 @@ static int ipp_queue_buf_with_run(struct device *dev,
 	} else {
 		ret = ipp_set_mem_node(ippdrv, c_node, m_node);
 		if (ret) {
+			mutex_unlock(&c_node->mem_lock);
 			DRM_ERROR("failed to set m node.\n");
 			return ret;
 		}
 	}
+	mutex_unlock(&c_node->mem_lock);
 
 	return 0;
 }
@@ -918,15 +915,15 @@ static void ipp_clean_queue_buf(struct drm_device *drm_dev,
 {
 	struct drm_exynos_ipp_mem_node *m_node, *tm_node;
 
-	if (!list_empty(&c_node->mem_list[qbuf->ops_id])) {
-		/* delete list */
-		list_for_each_entry_safe(m_node, tm_node,
-			&c_node->mem_list[qbuf->ops_id], list) {
-			if (m_node->buf_id == qbuf->buf_id &&
-			    m_node->ops_id == qbuf->ops_id)
-				ipp_put_mem_node(drm_dev, c_node, m_node);
-		}
+	/* delete list */
+	mutex_lock(&c_node->mem_lock);
+	list_for_each_entry_safe(m_node, tm_node,
+		&c_node->mem_list[qbuf->ops_id], list) {
+		if (m_node->buf_id == qbuf->buf_id &&
+		    m_node->ops_id == qbuf->ops_id)
+			ipp_put_mem_node(drm_dev, c_node, m_node);
 	}
+	mutex_unlock(&c_node->mem_lock);
 }
 
 int exynos_drm_ipp_queue_buf(struct drm_device *drm_dev, void *data,
@@ -998,7 +995,7 @@ int exynos_drm_ipp_queue_buf(struct drm_device *drm_dev, void *data,
 		}
 		break;
 	case IPP_BUF_DEQUEUE:
-		mutex_lock(&c_node->cmd_lock);
+		mutex_lock(&c_node->lock);
 
 		/* put event for destination buffer */
 		if (qbuf->ops_id == EXYNOS_DRM_OPS_DST)
@@ -1006,7 +1003,7 @@ int exynos_drm_ipp_queue_buf(struct drm_device *drm_dev, void *data,
 
 		ipp_clean_queue_buf(drm_dev, c_node, qbuf);
 
-		mutex_unlock(&c_node->cmd_lock);
+		mutex_unlock(&c_node->lock);
 		break;
 	default:
 		DRM_ERROR("invalid buffer control.\n");
@@ -1109,12 +1106,12 @@ int exynos_drm_ipp_cmd_ctrl(struct drm_device *drm_dev, void *data,
 	case IPP_CTRL_PLAY:
 		if (pm_runtime_suspended(ippdrv->dev))
 			pm_runtime_get_sync(ippdrv->dev);
+
 		c_node->state = IPP_STATE_START;
 
 		cmd_work = c_node->start_work;
 		cmd_work->ctrl = cmd_ctrl->ctrl;
 		ipp_handle_cmd_work(dev, ippdrv, cmd_work, c_node);
-		c_node->state = IPP_STATE_START;
 		break;
 	case IPP_CTRL_STOP:
 		cmd_work = c_node->stop_work;
@@ -1129,10 +1126,12 @@ int exynos_drm_ipp_cmd_ctrl(struct drm_device *drm_dev, void *data,
 
 		c_node->state = IPP_STATE_STOP;
 		ippdrv->dedicated = false;
-		ipp_clean_cmd_node(c_node);
+		mutex_lock(&ippdrv->cmd_lock);
+		ipp_clean_cmd_node(ctx, c_node);
 
 		if (list_empty(&ippdrv->cmd_list))
 			pm_runtime_put_sync(ippdrv->dev);
+		mutex_unlock(&ippdrv->cmd_lock);
 		break;
 	case IPP_CTRL_PAUSE:
 		cmd_work = c_node->stop_work;
@@ -1260,9 +1259,11 @@ static int ipp_start_property(struct exynos_drm_ippdrv *ippdrv,
 	/* store command info in ippdrv */
 	ippdrv->c_node = c_node;
 
+	mutex_lock(&c_node->mem_lock);
 	if (!ipp_check_mem_list(c_node)) {
 		DRM_DEBUG_KMS("empty memory.\n");
-		return -ENOMEM;
+		ret = -ENOMEM;
+		goto err_unlock;
 	}
 
 	/* set current property in ippdrv */
@@ -1270,7 +1271,7 @@ static int ipp_start_property(struct exynos_drm_ippdrv *ippdrv,
 	if (ret) {
 		DRM_ERROR("failed to set property.\n");
 		ippdrv->c_node = NULL;
-		return ret;
+		goto err_unlock;
 	}
 
 	/* check command */
@@ -1285,7 +1286,7 @@ static int ipp_start_property(struct exynos_drm_ippdrv *ippdrv,
 			if (!m_node) {
 				DRM_ERROR("failed to get node.\n");
 				ret = -EFAULT;
-				return ret;
+				goto err_unlock;
 			}
 
 			DRM_DEBUG_KMS("m_node[0x%x]\n", (int)m_node);
@@ -1293,7 +1294,7 @@ static int ipp_start_property(struct exynos_drm_ippdrv *ippdrv,
 			ret = ipp_set_mem_node(ippdrv, c_node, m_node);
 			if (ret) {
 				DRM_ERROR("failed to set m node.\n");
-				return ret;
+				goto err_unlock;
 			}
 		}
 		break;
@@ -1305,7 +1306,7 @@ static int ipp_start_property(struct exynos_drm_ippdrv *ippdrv,
 			ret = ipp_set_mem_node(ippdrv, c_node, m_node);
 			if (ret) {
 				DRM_ERROR("failed to set m node.\n");
-				return ret;
+				goto err_unlock;
 			}
 		}
 		break;
@@ -1317,14 +1318,16 @@ static int ipp_start_property(struct exynos_drm_ippdrv *ippdrv,
 			ret = ipp_set_mem_node(ippdrv, c_node, m_node);
 			if (ret) {
 				DRM_ERROR("failed to set m node.\n");
-				return ret;
+				goto err_unlock;
 			}
 		}
 		break;
 	default:
 		DRM_ERROR("invalid operations.\n");
-		return -EINVAL;
+		ret = -EINVAL;
+		goto err_unlock;
 	}
+	mutex_unlock(&c_node->mem_lock);
 
 	DRM_DEBUG_KMS("cmd[%d]\n", property->cmd);
 
@@ -1333,11 +1336,17 @@ static int ipp_start_property(struct exynos_drm_ippdrv *ippdrv,
 		ret = ippdrv->start(ippdrv->dev, property->cmd);
 		if (ret) {
 			DRM_ERROR("failed to start ops.\n");
+			ippdrv->c_node = NULL;
 			return ret;
 		}
 	}
 
 	return 0;
+
+err_unlock:
+	mutex_unlock(&c_node->mem_lock);
+	ippdrv->c_node = NULL;
+	return ret;
 }
 
 static int ipp_stop_property(struct drm_device *drm_dev,
@@ -1354,6 +1363,8 @@ static int ipp_stop_property(struct drm_device *drm_dev,
 	/* put event */
 	ipp_put_event(c_node, NULL);
 
+	mutex_lock(&c_node->mem_lock);
+
 	/* check command */
 	switch (property->cmd) {
 	case IPP_CMD_M2M:
@@ -1361,11 +1372,6 @@ static int ipp_stop_property(struct drm_device *drm_dev,
 			/* source/destination memory list */
 			head = &c_node->mem_list[i];
 
-			if (list_empty(head)) {
-				DRM_DEBUG_KMS("mem_list is empty.\n");
-				break;
-			}
-
 			list_for_each_entry_safe(m_node, tm_node,
 				head, list) {
 				ret = ipp_put_mem_node(drm_dev, c_node,
@@ -1381,11 +1387,6 @@ static int ipp_stop_property(struct drm_device *drm_dev,
 		/* destination memory list */
 		head = &c_node->mem_list[EXYNOS_DRM_OPS_DST];
 
-		if (list_empty(head)) {
-			DRM_DEBUG_KMS("mem_list is empty.\n");
-			break;
-		}
-
 		list_for_each_entry_safe(m_node, tm_node, head, list) {
 			ret = ipp_put_mem_node(drm_dev, c_node, m_node);
 			if (ret) {
@@ -1398,11 +1399,6 @@ static int ipp_stop_property(struct drm_device *drm_dev,
 		/* source memory list */
 		head = &c_node->mem_list[EXYNOS_DRM_OPS_SRC];
 
-		if (list_empty(head)) {
-			DRM_DEBUG_KMS("mem_list is empty.\n");
-			break;
-		}
-
 		list_for_each_entry_safe(m_node, tm_node, head, list) {
 			ret = ipp_put_mem_node(drm_dev, c_node, m_node);
 			if (ret) {
@@ -1418,6 +1414,8 @@ static int ipp_stop_property(struct drm_device *drm_dev,
 	}
 
 err_clear:
+	mutex_unlock(&c_node->mem_lock);
+
 	/* stop operations */
 	if (ippdrv->stop)
 		ippdrv->stop(ippdrv->dev, property->cmd);
@@ -1446,7 +1444,7 @@ void ipp_sched_cmd(struct work_struct *work)
 		return;
 	}
 
-	mutex_lock(&c_node->cmd_lock);
+	mutex_lock(&c_node->lock);
 
 	property = &c_node->property;
 
@@ -1494,7 +1492,7 @@ void ipp_sched_cmd(struct work_struct *work)
 	DRM_DEBUG_KMS("ctrl[%d] done.\n", cmd_work->ctrl);
 
 err_unlock:
-	mutex_unlock(&c_node->cmd_lock);
+	mutex_unlock(&c_node->lock);
 }
 
 static int ipp_send_event(struct exynos_drm_ippdrv *ippdrv,
@@ -1524,14 +1522,18 @@ static int ipp_send_event(struct exynos_drm_ippdrv *ippdrv,
 		return -EINVAL;
 	}
 
+	mutex_lock(&c_node->event_lock);
 	if (list_empty(&c_node->event_list)) {
 		DRM_DEBUG_KMS("event list is empty.\n");
-		return 0;
+		ret = 0;
+		goto err_event_unlock;
 	}
 
+	mutex_lock(&c_node->mem_lock);
 	if (!ipp_check_mem_list(c_node)) {
 		DRM_DEBUG_KMS("empty memory.\n");
-		return 0;
+		ret = 0;
+		goto err_mem_unlock;
 	}
 
 	/* check command */
@@ -1545,7 +1547,8 @@ static int ipp_send_event(struct exynos_drm_ippdrv *ippdrv,
 				struct drm_exynos_ipp_mem_node, list);
 			if (!m_node) {
 				DRM_ERROR("empty memory node.\n");
-				return -ENOMEM;
+				ret = -ENOMEM;
+				goto err_mem_unlock;
 			}
 
 			tbuf_id[i] = m_node->buf_id;
@@ -1567,7 +1570,8 @@ static int ipp_send_event(struct exynos_drm_ippdrv *ippdrv,
 		m_node = ipp_find_mem_node(c_node, &qbuf);
 		if (!m_node) {
 			DRM_ERROR("empty memory node.\n");
-			return -ENOMEM;
+			ret = -ENOMEM;
+			goto err_mem_unlock;
 		}
 
 		tbuf_id[EXYNOS_DRM_OPS_DST] = m_node->buf_id;
@@ -1584,7 +1588,8 @@ static int ipp_send_event(struct exynos_drm_ippdrv *ippdrv,
 			struct drm_exynos_ipp_mem_node, list);
 		if (!m_node) {
 			DRM_ERROR("empty memory node.\n");
-			return -ENOMEM;
+			ret = -ENOMEM;
+			goto err_mem_unlock;
 		}
 
 		tbuf_id[EXYNOS_DRM_OPS_SRC] = m_node->buf_id;
@@ -1595,8 +1600,10 @@ static int ipp_send_event(struct exynos_drm_ippdrv *ippdrv,
 		break;
 	default:
 		DRM_ERROR("invalid operations.\n");
-		return -EINVAL;
+		ret = -EINVAL;
+		goto err_mem_unlock;
 	}
+	mutex_unlock(&c_node->mem_lock);
 
 	if (tbuf_id[EXYNOS_DRM_OPS_DST] != buf_id[EXYNOS_DRM_OPS_DST])
 		DRM_ERROR("failed to match buf_id[%d %d]prop_id[%d]\n",
@@ -1611,11 +1618,6 @@ static int ipp_send_event(struct exynos_drm_ippdrv *ippdrv,
 	e = list_first_entry(&c_node->event_list,
 		struct drm_exynos_ipp_send_event, base.link);
 
-	if (!e) {
-		DRM_ERROR("empty event.\n");
-		return -EINVAL;
-	}
-
 	do_gettimeofday(&now);
 	DRM_DEBUG_KMS("tv_sec[%ld]tv_usec[%ld]\n", now.tv_sec, now.tv_usec);
 	e->event.tv_sec = now.tv_sec;
@@ -1630,11 +1632,18 @@ static int ipp_send_event(struct exynos_drm_ippdrv *ippdrv,
 	list_move_tail(&e->base.link, &e->base.file_priv->event_list);
 	wake_up_interruptible(&e->base.file_priv->event_wait);
 	spin_unlock_irqrestore(&drm_dev->event_lock, flags);
+	mutex_unlock(&c_node->event_lock);
 
 	DRM_DEBUG_KMS("done cmd[%d]prop_id[%d]buf_id[%d]\n",
 		property->cmd, property->prop_id, tbuf_id[EXYNOS_DRM_OPS_DST]);
 
 	return 0;
+
+err_mem_unlock:
+	mutex_unlock(&c_node->mem_lock);
+err_event_unlock:
+	mutex_unlock(&c_node->event_lock);
+	return ret;
 }
 
 void ipp_sched_event(struct work_struct *work)
@@ -1676,8 +1685,6 @@ void ipp_sched_event(struct work_struct *work)
 		goto err_completion;
 	}
 
-	mutex_lock(&c_node->event_lock);
-
 	ret = ipp_send_event(ippdrv, c_node, event_work->buf_id);
 	if (ret) {
 		DRM_ERROR("failed to send event.\n");
@@ -1687,8 +1694,6 @@ void ipp_sched_event(struct work_struct *work)
 err_completion:
 	if (ipp_is_m2m_cmd(c_node->property.cmd))
 		complete(&c_node->start_complete);
-
-	mutex_unlock(&c_node->event_lock);
 }
 
 static int ipp_subdrv_probe(struct drm_device *drm_dev, struct device *dev)
@@ -1699,23 +1704,21 @@ static int ipp_subdrv_probe(struct drm_device *drm_dev, struct device *dev)
 
 	/* get ipp driver entry */
 	list_for_each_entry(ippdrv, &exynos_drm_ippdrv_list, drv_list) {
+		u32 ipp_id;
+
 		ippdrv->drm_dev = drm_dev;
 
 		ret = ipp_create_id(&ctx->ipp_idr, &ctx->ipp_lock, ippdrv,
-			&ippdrv->ipp_id);
-		if (ret) {
+				    &ipp_id);
+		if (ret || ipp_id == 0) {
 			DRM_ERROR("failed to create id.\n");
-			goto err_idr;
+			goto err;
 		}
 
 		DRM_DEBUG_KMS("count[%d]ippdrv[0x%x]ipp_id[%d]\n",
-			count++, (int)ippdrv, ippdrv->ipp_id);
+			count++, (int)ippdrv, ipp_id);
 
-		if (ippdrv->ipp_id == 0) {
-			DRM_ERROR("failed to get ipp_id[%d]\n",
-				ippdrv->ipp_id);
-			goto err_idr;
-		}
+		ippdrv->prop_list.ipp_id = ipp_id;
 
 		/* store parent device for node */
 		ippdrv->parent_dev = dev;
@@ -1724,39 +1727,46 @@ static int ipp_subdrv_probe(struct drm_device *drm_dev, struct device *dev)
 		ippdrv->event_workq = ctx->event_workq;
 		ippdrv->sched_event = ipp_sched_event;
 		INIT_LIST_HEAD(&ippdrv->cmd_list);
+		mutex_init(&ippdrv->cmd_lock);
 
 		if (is_drm_iommu_supported(drm_dev)) {
 			ret = drm_iommu_attach_device(drm_dev, ippdrv->dev);
 			if (ret) {
 				DRM_ERROR("failed to activate iommu\n");
-				goto err_iommu;
+				goto err;
 			}
 		}
 	}
 
 	return 0;
 
-err_iommu:
+err:
 	/* get ipp driver entry */
-	list_for_each_entry_reverse(ippdrv, &exynos_drm_ippdrv_list, drv_list)
+	list_for_each_entry_continue_reverse(ippdrv, &exynos_drm_ippdrv_list,
+						drv_list) {
 		if (is_drm_iommu_supported(drm_dev))
 			drm_iommu_detach_device(drm_dev, ippdrv->dev);
 
-err_idr:
-	idr_destroy(&ctx->ipp_idr);
-	idr_destroy(&ctx->prop_idr);
+		ipp_remove_id(&ctx->ipp_idr, &ctx->ipp_lock,
+				ippdrv->prop_list.ipp_id);
+	}
+
 	return ret;
 }
 
 static void ipp_subdrv_remove(struct drm_device *drm_dev, struct device *dev)
 {
 	struct exynos_drm_ippdrv *ippdrv;
+	struct ipp_context *ctx = get_ipp_context(dev);
 
 	/* get ipp driver entry */
 	list_for_each_entry(ippdrv, &exynos_drm_ippdrv_list, drv_list) {
 		if (is_drm_iommu_supported(drm_dev))
 			drm_iommu_detach_device(drm_dev, ippdrv->dev);
 
+		ipp_remove_id(&ctx->ipp_idr, &ctx->ipp_lock,
+				ippdrv->prop_list.ipp_id);
+
 		ippdrv->drm_dev = NULL;
 		exynos_drm_ippdrv_unregister(ippdrv);
 	}
@@ -1787,20 +1797,14 @@ static void ipp_subdrv_close(struct drm_device *drm_dev, struct device *dev,
 	struct drm_exynos_file_private *file_priv = file->driver_priv;
 	struct exynos_drm_ipp_private *priv = file_priv->ipp_priv;
 	struct exynos_drm_ippdrv *ippdrv = NULL;
+	struct ipp_context *ctx = get_ipp_context(dev);
 	struct drm_exynos_ipp_cmd_node *c_node, *tc_node;
 	int count = 0;
 
 	DRM_DEBUG_KMS("for priv[0x%x]\n", (int)priv);
 
-	if (list_empty(&exynos_drm_ippdrv_list)) {
-		DRM_DEBUG_KMS("ippdrv_list is empty.\n");
-		goto err_clear;
-	}
-
 	list_for_each_entry(ippdrv, &exynos_drm_ippdrv_list, drv_list) {
-		if (list_empty(&ippdrv->cmd_list))
-			continue;
-
+		mutex_lock(&ippdrv->cmd_lock);
 		list_for_each_entry_safe(c_node, tc_node,
 			&ippdrv->cmd_list, list) {
 			DRM_DEBUG_KMS("count[%d]ippdrv[0x%x]\n",
@@ -1820,14 +1824,14 @@ static void ipp_subdrv_close(struct drm_device *drm_dev, struct device *dev,
 				}
 
 				ippdrv->dedicated = false;
-				ipp_clean_cmd_node(c_node);
+				ipp_clean_cmd_node(ctx, c_node);
 				if (list_empty(&ippdrv->cmd_list))
 					pm_runtime_put_sync(ippdrv->dev);
 			}
 		}
+		mutex_unlock(&ippdrv->cmd_lock);
 	}
 
-err_clear:
 	kfree(priv);
 	return;
 }
diff --git a/drivers/gpu/drm/exynos/exynos_drm_ipp.h b/drivers/gpu/drm/exynos/exynos_drm_ipp.h
index ab1634befc05..7aaeaae757c2 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_ipp.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_ipp.h
@@ -52,7 +52,7 @@ struct drm_exynos_ipp_cmd_work {
  * @list: list head to command queue information.
  * @event_list: list head of event.
  * @mem_list: list head to source,destination memory queue information.
- * @cmd_lock: lock for synchronization of access to ioctl.
+ * @lock: lock for synchronization of access to ioctl.
  * @mem_lock: lock for synchronization of access to memory nodes.
  * @event_lock: lock for synchronization of access to scheduled event.
  * @start_complete: completion of start of command.
@@ -68,7 +68,7 @@ struct drm_exynos_ipp_cmd_node {
 	struct list_head	list;
 	struct list_head	event_list;
 	struct list_head	mem_list[EXYNOS_DRM_OPS_MAX];
-	struct mutex	cmd_lock;
+	struct mutex	lock;
 	struct mutex	mem_lock;
 	struct mutex	event_lock;
 	struct completion	start_complete;
@@ -83,7 +83,7 @@ struct drm_exynos_ipp_cmd_node {
 /*
  * A structure of buffer information.
  *
- * @gem_objs: Y, Cb, Cr each gem object.
+ * @handles: Y, Cb, Cr each gem object handle.
  * @base: Y, Cb, Cr each planar address.
  */
 struct drm_exynos_ipp_buf_info {
@@ -142,12 +142,12 @@ struct exynos_drm_ipp_ops {
  * @parent_dev: parent device information.
  * @dev: platform device.
  * @drm_dev: drm device.
- * @ipp_id: id of ipp driver.
  * @dedicated: dedicated ipp device.
  * @ops: source, destination operations.
  * @event_workq: event work queue.
  * @c_node: current command information.
  * @cmd_list: list head for command information.
+ * @cmd_lock: lock for synchronization of access to cmd_list.
  * @prop_list: property informations of current ipp driver.
  * @check_property: check property about format, size, buffer.
  * @reset: reset ipp block.
@@ -160,13 +160,13 @@ struct exynos_drm_ippdrv {
 	struct device	*parent_dev;
 	struct device	*dev;
 	struct drm_device	*drm_dev;
-	u32	ipp_id;
 	bool	dedicated;
 	struct exynos_drm_ipp_ops	*ops[EXYNOS_DRM_OPS_MAX];
 	struct workqueue_struct	*event_workq;
 	struct drm_exynos_ipp_cmd_node *c_node;
 	struct list_head	cmd_list;
-	struct drm_exynos_ipp_prop_list *prop_list;
+	struct mutex	cmd_lock;
+	struct drm_exynos_ipp_prop_list prop_list;
 
 	int (*check_property)(struct device *dev,
 		struct drm_exynos_ipp_property *property);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_rotator.c b/drivers/gpu/drm/exynos/exynos_drm_rotator.c
index 7b901688defa..f01fbb6dc1f0 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_rotator.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_rotator.c
@@ -158,8 +158,9 @@ static irqreturn_t rotator_irq_handler(int irq, void *arg)
 			rot->cur_buf_id[EXYNOS_DRM_OPS_DST];
 		queue_work(ippdrv->event_workq,
 			(struct work_struct *)event_work);
-	} else
+	} else {
 		DRM_ERROR("the SFR is set illegally\n");
+	}
 
 	return IRQ_HANDLED;
 }
@@ -469,11 +470,7 @@ static struct exynos_drm_ipp_ops rot_dst_ops = {
 
 static int rotator_init_prop_list(struct exynos_drm_ippdrv *ippdrv)
 {
-	struct drm_exynos_ipp_prop_list *prop_list;
-
-	prop_list = devm_kzalloc(ippdrv->dev, sizeof(*prop_list), GFP_KERNEL);
-	if (!prop_list)
-		return -ENOMEM;
+	struct drm_exynos_ipp_prop_list *prop_list = &ippdrv->prop_list;
 
 	prop_list->version = 1;
 	prop_list->flip = (1 << EXYNOS_DRM_FLIP_VERTICAL) |
@@ -486,8 +483,6 @@ static int rotator_init_prop_list(struct exynos_drm_ippdrv *ippdrv)
 	prop_list->crop = 0;
 	prop_list->scale = 0;
 
-	ippdrv->prop_list = prop_list;
-
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_vidi.c b/drivers/gpu/drm/exynos/exynos_drm_vidi.c
index 852f2dadaebd..2fb8705d6461 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_vidi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_vidi.c
@@ -51,6 +51,7 @@ struct vidi_context {
 	struct drm_crtc			*crtc;
 	struct drm_encoder		*encoder;
 	struct drm_connector		connector;
+	struct exynos_drm_subdrv	subdrv;
 	struct vidi_win_data		win_data[WINDOWS_NR];
 	struct edid			*raw_edid;
 	unsigned int			clkdiv;
@@ -294,14 +295,13 @@ static void vidi_dpms(struct exynos_drm_manager *mgr, int mode)
 }
 
 static int vidi_mgr_initialize(struct exynos_drm_manager *mgr,
-			struct drm_device *drm_dev, int pipe)
+			struct drm_device *drm_dev)
 {
 	struct vidi_context *ctx = mgr->ctx;
+	struct exynos_drm_private *priv = drm_dev->dev_private;
 
-	DRM_ERROR("vidi initialize ct=%p dev=%p pipe=%d\n", ctx, drm_dev, pipe);
-
-	ctx->drm_dev = drm_dev;
-	ctx->pipe = pipe;
+	mgr->drm_dev = ctx->drm_dev = drm_dev;
+	mgr->pipe = ctx->pipe = priv->pipe++;
 
 	/*
 	 * enable drm irq mode.
@@ -324,7 +324,6 @@ static int vidi_mgr_initialize(struct exynos_drm_manager *mgr,
 }
 
 static struct exynos_drm_manager_ops vidi_manager_ops = {
-	.initialize = vidi_mgr_initialize,
 	.dpms = vidi_dpms,
 	.commit = vidi_commit,
 	.enable_vblank = vidi_enable_vblank,
@@ -533,12 +532,6 @@ static int vidi_get_modes(struct drm_connector *connector)
 	return drm_add_edid_modes(connector, edid);
 }
 
-static int vidi_mode_valid(struct drm_connector *connector,
-			struct drm_display_mode *mode)
-{
-	return MODE_OK;
-}
-
 static struct drm_encoder *vidi_best_encoder(struct drm_connector *connector)
 {
 	struct vidi_context *ctx = ctx_from_connector(connector);
@@ -548,7 +541,6 @@ static struct drm_encoder *vidi_best_encoder(struct drm_connector *connector)
 
 static struct drm_connector_helper_funcs vidi_connector_helper_funcs = {
 	.get_modes = vidi_get_modes,
-	.mode_valid = vidi_mode_valid,
 	.best_encoder = vidi_best_encoder,
 };
 
@@ -586,13 +578,38 @@ static struct exynos_drm_display vidi_display = {
 	.ops = &vidi_display_ops,
 };
 
+static int vidi_subdrv_probe(struct drm_device *drm_dev, struct device *dev)
+{
+	struct exynos_drm_manager *mgr = get_vidi_mgr(dev);
+	struct vidi_context *ctx = mgr->ctx;
+	struct drm_crtc *crtc = ctx->crtc;
+	int ret;
+
+	vidi_mgr_initialize(mgr, drm_dev);
+
+	ret = exynos_drm_crtc_create(&vidi_manager);
+	if (ret) {
+		DRM_ERROR("failed to create crtc.\n");
+		return ret;
+	}
+
+	ret = exynos_drm_create_enc_conn(drm_dev, &vidi_display);
+	if (ret) {
+		crtc->funcs->destroy(crtc);
+		DRM_ERROR("failed to create encoder and connector.\n");
+		return ret;
+	}
+
+	return 0;
+}
+
 static int vidi_probe(struct platform_device *pdev)
 {
-	struct device *dev = &pdev->dev;
+	struct exynos_drm_subdrv *subdrv;
 	struct vidi_context *ctx;
 	int ret;
 
-	ctx = devm_kzalloc(dev, sizeof(*ctx), GFP_KERNEL);
+	ctx = devm_kzalloc(&pdev->dev, sizeof(*ctx), GFP_KERNEL);
 	if (!ctx)
 		return -ENOMEM;
 
@@ -607,28 +624,43 @@ static int vidi_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, &vidi_manager);
 
-	ret = device_create_file(dev, &dev_attr_connection);
-	if (ret < 0)
-		DRM_INFO("failed to create connection sysfs.\n");
+	subdrv = &ctx->subdrv;
+	subdrv->dev = &pdev->dev;
+	subdrv->probe = vidi_subdrv_probe;
+
+	ret = exynos_drm_subdrv_register(subdrv);
+	if (ret < 0) {
+		dev_err(&pdev->dev, "failed to register drm vidi device\n");
+		return ret;
+	}
 
-	exynos_drm_manager_register(&vidi_manager);
-	exynos_drm_display_register(&vidi_display);
+	ret = device_create_file(&pdev->dev, &dev_attr_connection);
+	if (ret < 0) {
+		exynos_drm_subdrv_unregister(subdrv);
+		DRM_INFO("failed to create connection sysfs.\n");
+	}
 
 	return 0;
 }
 
 static int vidi_remove(struct platform_device *pdev)
 {
-	struct vidi_context *ctx = platform_get_drvdata(pdev);
-
-	exynos_drm_display_unregister(&vidi_display);
-	exynos_drm_manager_unregister(&vidi_manager);
+	struct exynos_drm_manager *mgr = platform_get_drvdata(pdev);
+	struct vidi_context *ctx = mgr->ctx;
+	struct drm_encoder *encoder = ctx->encoder;
+	struct drm_crtc *crtc = mgr->crtc;
 
 	if (ctx->raw_edid != (struct edid *)fake_edid_info) {
 		kfree(ctx->raw_edid);
 		ctx->raw_edid = NULL;
+
+		return -EINVAL;
 	}
 
+	crtc->funcs->destroy(crtc);
+	encoder->funcs->destroy(encoder);
+	drm_connector_cleanup(&ctx->connector);
+
 	return 0;
 }
 
@@ -640,3 +672,31 @@ struct platform_driver vidi_driver = {
 		.owner	= THIS_MODULE,
 	},
 };
+
+int exynos_drm_probe_vidi(void)
+{
+	struct platform_device *pdev;
+	int ret;
+
+	pdev = platform_device_register_simple("exynos-drm-vidi", -1, NULL, 0);
+	if (IS_ERR(pdev))
+		return PTR_ERR(pdev);
+
+	ret = platform_driver_register(&vidi_driver);
+	if (ret) {
+		platform_device_unregister(pdev);
+		return ret;
+	}
+
+	return ret;
+}
+
+void exynos_drm_remove_vidi(void)
+{
+	struct vidi_context *ctx = vidi_manager.ctx;
+	struct exynos_drm_subdrv *subdrv = &ctx->subdrv;
+	struct platform_device *pdev = to_platform_device(subdrv->dev);
+
+	platform_driver_unregister(&vidi_driver);
+	platform_device_unregister(pdev);
+}
diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index 9a6d652a3ef2..c104d0c9b385 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -33,13 +33,17 @@
 #include <linux/regulator/consumer.h>
 #include <linux/io.h>
 #include <linux/of.h>
-#include <linux/i2c.h>
+#include <linux/of_address.h>
 #include <linux/of_gpio.h>
 #include <linux/hdmi.h>
+#include <linux/component.h>
+#include <linux/mfd/syscon.h>
+#include <linux/regmap.h>
 
 #include <drm/exynos_drm.h>
 
 #include "exynos_drm_drv.h"
+#include "exynos_drm_crtc.h"
 #include "exynos_mixer.h"
 
 #include <linux/gpio.h>
@@ -48,6 +52,8 @@
 #define get_hdmi_display(dev)	platform_get_drvdata(to_platform_device(dev))
 #define ctx_from_connector(c)	container_of(c, struct hdmi_context, connector)
 
+#define HOTPLUG_DEBOUNCE_MS		1100
+
 /* AVI header and aspect ratio */
 #define HDMI_AVI_VERSION		0x02
 #define HDMI_AVI_LENGTH		0x0D
@@ -66,6 +72,8 @@ enum hdmi_type {
 
 struct hdmi_driver_data {
 	unsigned int type;
+	const struct hdmiphy_config *phy_confs;
+	unsigned int phy_conf_count;
 	unsigned int is_apb_phy:1;
 };
 
@@ -74,7 +82,6 @@ struct hdmi_resources {
 	struct clk			*sclk_hdmi;
 	struct clk			*sclk_pixel;
 	struct clk			*sclk_hdmiphy;
-	struct clk			*hdmiphy;
 	struct clk			*mout_hdmi;
 	struct regulator_bulk_data	*regul_bulk;
 	int				regul_count;
@@ -185,17 +192,23 @@ struct hdmi_context {
 
 	void __iomem			*regs;
 	int				irq;
+	struct delayed_work		hotplug_work;
 
 	struct i2c_adapter		*ddc_adpt;
 	struct i2c_client		*hdmiphy_port;
 
 	/* current hdmiphy conf regs */
+	struct drm_display_mode		current_mode;
 	struct hdmi_conf_regs		mode_conf;
 
 	struct hdmi_resources		res;
 
 	int				hpd_gpio;
+	void __iomem			*regs_hdmiphy;
+	const struct hdmiphy_config		*phy_confs;
+	unsigned int			phy_conf_count;
 
+	struct regmap			*pmureg;
 	enum hdmi_type			type;
 };
 
@@ -204,14 +217,6 @@ struct hdmiphy_config {
 	u8 conf[32];
 };
 
-struct hdmi_driver_data exynos4212_hdmi_driver_data = {
-	.type	= HDMI_TYPE14,
-};
-
-struct hdmi_driver_data exynos5_hdmi_driver_data = {
-	.type	= HDMI_TYPE14,
-};
-
 /* list of phy config settings */
 static const struct hdmiphy_config hdmiphy_v13_configs[] = {
 	{
@@ -319,18 +324,18 @@ static const struct hdmiphy_config hdmiphy_v14_configs[] = {
 	{
 		.pixel_clock = 71000000,
 		.conf = {
-			0x01, 0x91, 0x1e, 0x15, 0x40, 0x3c, 0xce, 0x08,
-			0x04, 0x20, 0xb2, 0xd8, 0x45, 0xa0, 0xac, 0x80,
-			0x06, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
+			0x01, 0xd1, 0x3b, 0x35, 0x40, 0x0c, 0x04, 0x08,
+			0x85, 0xa0, 0x63, 0xd9, 0x45, 0xa0, 0xac, 0x80,
+			0x08, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
 			0x54, 0xad, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
 		},
 	},
 	{
 		.pixel_clock = 73250000,
 		.conf = {
-			0x01, 0xd1, 0x1f, 0x15, 0x40, 0x18, 0xe9, 0x08,
-			0x02, 0xa0, 0xb7, 0xd8, 0x45, 0xa0, 0xac, 0x80,
-			0x06, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
+			0x01, 0xd1, 0x3d, 0x35, 0x40, 0x18, 0x02, 0x08,
+			0x83, 0xa0, 0x6e, 0xd9, 0x45, 0xa0, 0xac, 0x80,
+			0x08, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
 			0x54, 0xa8, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
 		},
 	},
@@ -362,15 +367,6 @@ static const struct hdmiphy_config hdmiphy_v14_configs[] = {
 		},
 	},
 	{
-		.pixel_clock = 88750000,
-		.conf = {
-			0x01, 0x91, 0x25, 0x17, 0x40, 0x30, 0xfe, 0x08,
-			0x06, 0x20, 0xde, 0xd8, 0x45, 0xa0, 0xac, 0x80,
-			0x06, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
-			0x54, 0x8a, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
-		},
-	},
-	{
 		.pixel_clock = 106500000,
 		.conf = {
 			0x01, 0xd1, 0x2c, 0x12, 0x40, 0x0c, 0x09, 0x08,
@@ -391,18 +387,18 @@ static const struct hdmiphy_config hdmiphy_v14_configs[] = {
 	{
 		.pixel_clock = 115500000,
 		.conf = {
-			0x01, 0xd1, 0x30, 0x1a, 0x40, 0x40, 0x10, 0x04,
-			0x04, 0xa0, 0x21, 0xd9, 0x45, 0xa0, 0xac, 0x80,
-			0x06, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
+			0x01, 0xd1, 0x30, 0x12, 0x40, 0x40, 0x10, 0x08,
+			0x80, 0x80, 0x21, 0xd9, 0x45, 0xa0, 0xac, 0x80,
+			0x08, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
 			0x54, 0xaa, 0x25, 0x03, 0x00, 0x00, 0x01, 0x80,
 		},
 	},
 	{
 		.pixel_clock = 119000000,
 		.conf = {
-			0x01, 0x91, 0x32, 0x14, 0x40, 0x60, 0xd8, 0x08,
-			0x06, 0x20, 0x2a, 0xd9, 0x45, 0xa0, 0xac, 0x80,
-			0x06, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
+			0x01, 0xd1, 0x32, 0x1a, 0x40, 0x30, 0xd8, 0x08,
+			0x04, 0xa0, 0x2a, 0xd9, 0x45, 0xa0, 0xac, 0x80,
+			0x08, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
 			0x54, 0x9d, 0x25, 0x03, 0x00, 0x00, 0x01, 0x80,
 		},
 	},
@@ -426,6 +422,183 @@ static const struct hdmiphy_config hdmiphy_v14_configs[] = {
 	},
 };
 
+static const struct hdmiphy_config hdmiphy_5420_configs[] = {
+	{
+		.pixel_clock = 25200000,
+		.conf = {
+			0x01, 0x52, 0x3F, 0x55, 0x40, 0x01, 0x00, 0xC8,
+			0x82, 0xC8, 0xBD, 0xD8, 0x45, 0xA0, 0xAC, 0x80,
+			0x06, 0x80, 0x01, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0xF4, 0x24, 0x00, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 27000000,
+		.conf = {
+			0x01, 0xD1, 0x22, 0x51, 0x40, 0x08, 0xFC, 0xE0,
+			0x98, 0xE8, 0xCB, 0xD8, 0x45, 0xA0, 0xAC, 0x80,
+			0x06, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0xE4, 0x24, 0x00, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 27027000,
+		.conf = {
+			0x01, 0xD1, 0x2D, 0x72, 0x40, 0x64, 0x12, 0xC8,
+			0x43, 0xE8, 0x0E, 0xD9, 0x45, 0xA0, 0xAC, 0x80,
+			0x26, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0xE3, 0x24, 0x00, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 36000000,
+		.conf = {
+			0x01, 0x51, 0x2D, 0x55, 0x40, 0x40, 0x00, 0xC8,
+			0x02, 0xC8, 0x0E, 0xD9, 0x45, 0xA0, 0xAC, 0x80,
+			0x08, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0xAB, 0x24, 0x00, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 40000000,
+		.conf = {
+			0x01, 0xD1, 0x21, 0x31, 0x40, 0x3C, 0x28, 0xC8,
+			0x87, 0xE8, 0xC8, 0xD8, 0x45, 0xA0, 0xAC, 0x80,
+			0x08, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0x9A, 0x24, 0x00, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 65000000,
+		.conf = {
+			0x01, 0xD1, 0x36, 0x34, 0x40, 0x0C, 0x04, 0xC8,
+			0x82, 0xE8, 0x45, 0xD9, 0x45, 0xA0, 0xAC, 0x80,
+			0x08, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0xBD, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 71000000,
+		.conf = {
+			0x01, 0xD1, 0x3B, 0x35, 0x40, 0x0C, 0x04, 0xC8,
+			0x85, 0xE8, 0x63, 0xD9, 0x45, 0xA0, 0xAC, 0x80,
+			0x08, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0x57, 0x24, 0x00, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 73250000,
+		.conf = {
+			0x01, 0xD1, 0x1F, 0x10, 0x40, 0x78, 0x8D, 0xC8,
+			0x81, 0xE8, 0xB7, 0xD8, 0x45, 0xA0, 0xAC, 0x80,
+			0x56, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0xA8, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 74176000,
+		.conf = {
+			0x01, 0xD1, 0x1F, 0x10, 0x40, 0x5B, 0xEF, 0xC8,
+			0x81, 0xE8, 0xB9, 0xD8, 0x45, 0xA0, 0xAC, 0x80,
+			0x56, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0xA6, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 74250000,
+		.conf = {
+			0x01, 0xD1, 0x1F, 0x10, 0x40, 0x40, 0xF8, 0x08,
+			0x81, 0xE8, 0xBA, 0xD8, 0x45, 0xA0, 0xAC, 0x80,
+			0x26, 0x80, 0x09, 0x84, 0x05, 0x22, 0x24, 0x66,
+			0x54, 0xA5, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 83500000,
+		.conf = {
+			0x01, 0xD1, 0x23, 0x11, 0x40, 0x0C, 0xFB, 0xC8,
+			0x85, 0xE8, 0xD1, 0xD8, 0x45, 0xA0, 0xAC, 0x80,
+			0x08, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0x4A, 0x24, 0x00, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 88750000,
+		.conf = {
+			0x01, 0xD1, 0x25, 0x11, 0x40, 0x18, 0xFF, 0xC8,
+			0x83, 0xE8, 0xDE, 0xD8, 0x45, 0xA0, 0xAC, 0x80,
+			0x08, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0x45, 0x24, 0x00, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 106500000,
+		.conf = {
+			0x01, 0xD1, 0x2C, 0x12, 0x40, 0x0C, 0x09, 0xC8,
+			0x84, 0xE8, 0x0A, 0xD9, 0x45, 0xA0, 0xAC, 0x80,
+			0x08, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0x73, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 108000000,
+		.conf = {
+			0x01, 0x51, 0x2D, 0x15, 0x40, 0x01, 0x00, 0xC8,
+			0x82, 0xC8, 0x0E, 0xD9, 0x45, 0xA0, 0xAC, 0x80,
+			0x08, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0xC7, 0x25, 0x03, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 115500000,
+		.conf = {
+			0x01, 0xD1, 0x30, 0x14, 0x40, 0x0C, 0x03, 0xC8,
+			0x88, 0xE8, 0x21, 0xD9, 0x45, 0xA0, 0xAC, 0x80,
+			0x08, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0x6A, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 146250000,
+		.conf = {
+			0x01, 0xD1, 0x3D, 0x15, 0x40, 0x18, 0xFD, 0xC8,
+			0x83, 0xE8, 0x6E, 0xD9, 0x45, 0xA0, 0xAC, 0x80,
+			0x08, 0x80, 0x09, 0x84, 0x05, 0x02, 0x24, 0x66,
+			0x54, 0x54, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
+		},
+	},
+	{
+		.pixel_clock = 148500000,
+		.conf = {
+			0x01, 0xD1, 0x1F, 0x00, 0x40, 0x40, 0xF8, 0x08,
+			0x81, 0xE8, 0xBA, 0xD8, 0x45, 0xA0, 0xAC, 0x80,
+			0x26, 0x80, 0x09, 0x84, 0x05, 0x22, 0x24, 0x66,
+			0x54, 0x4B, 0x25, 0x03, 0x00, 0x80, 0x01, 0x80,
+		},
+	},
+};
+
+static struct hdmi_driver_data exynos5420_hdmi_driver_data = {
+	.type		= HDMI_TYPE14,
+	.phy_confs	= hdmiphy_5420_configs,
+	.phy_conf_count	= ARRAY_SIZE(hdmiphy_5420_configs),
+	.is_apb_phy	= 1,
+};
+
+static struct hdmi_driver_data exynos4212_hdmi_driver_data = {
+	.type		= HDMI_TYPE14,
+	.phy_confs	= hdmiphy_v14_configs,
+	.phy_conf_count	= ARRAY_SIZE(hdmiphy_v14_configs),
+	.is_apb_phy	= 0,
+};
+
+static struct hdmi_driver_data exynos5_hdmi_driver_data = {
+	.type		= HDMI_TYPE14,
+	.phy_confs	= hdmiphy_v13_configs,
+	.phy_conf_count	= ARRAY_SIZE(hdmiphy_v13_configs),
+	.is_apb_phy	= 0,
+};
+
 static inline u32 hdmi_reg_read(struct hdmi_context *hdata, u32 reg_id)
 {
 	return readl(hdata->regs + reg_id);
@@ -445,6 +618,48 @@ static inline void hdmi_reg_writemask(struct hdmi_context *hdata,
 	writel(value, hdata->regs + reg_id);
 }
 
+static int hdmiphy_reg_writeb(struct hdmi_context *hdata,
+			u32 reg_offset, u8 value)
+{
+	if (hdata->hdmiphy_port) {
+		u8 buffer[2];
+		int ret;
+
+		buffer[0] = reg_offset;
+		buffer[1] = value;
+
+		ret = i2c_master_send(hdata->hdmiphy_port, buffer, 2);
+		if (ret == 2)
+			return 0;
+		return ret;
+	} else {
+		writeb(value, hdata->regs_hdmiphy + (reg_offset<<2));
+		return 0;
+	}
+}
+
+static int hdmiphy_reg_write_buf(struct hdmi_context *hdata,
+			u32 reg_offset, const u8 *buf, u32 len)
+{
+	if ((reg_offset + len) > 32)
+		return -EINVAL;
+
+	if (hdata->hdmiphy_port) {
+		int ret;
+
+		ret = i2c_master_send(hdata->hdmiphy_port, buf, len);
+		if (ret == len)
+			return 0;
+		return ret;
+	} else {
+		int i;
+		for (i = 0; i < len; i++)
+			writeb(buf[i], hdata->regs_hdmiphy +
+				((reg_offset + i)<<2));
+		return 0;
+	}
+}
+
 static void hdmi_v13_regs_dump(struct hdmi_context *hdata, char *prefix)
 {
 #define DUMPREG(reg_id) \
@@ -809,6 +1024,8 @@ static enum drm_connector_status hdmi_detect(struct drm_connector *connector,
 {
 	struct hdmi_context *hdata = ctx_from_connector(connector);
 
+	hdata->hpd = gpio_get_value(hdata->hpd_gpio);
+
 	return hdata->hpd ? connector_status_connected :
 			connector_status_disconnected;
 }
@@ -848,20 +1065,10 @@ static int hdmi_get_modes(struct drm_connector *connector)
 
 static int hdmi_find_phy_conf(struct hdmi_context *hdata, u32 pixel_clock)
 {
-	const struct hdmiphy_config *confs;
-	int count, i;
-
-	if (hdata->type == HDMI_TYPE13) {
-		confs = hdmiphy_v13_configs;
-		count = ARRAY_SIZE(hdmiphy_v13_configs);
-	} else if (hdata->type == HDMI_TYPE14) {
-		confs = hdmiphy_v14_configs;
-		count = ARRAY_SIZE(hdmiphy_v14_configs);
-	} else
-		return -EINVAL;
+	int i;
 
-	for (i = 0; i < count; i++)
-		if (confs[i].pixel_clock == pixel_clock)
+	for (i = 0; i < hdata->phy_conf_count; i++)
+		if (hdata->phy_confs[i].pixel_clock == pixel_clock)
 			return i;
 
 	DRM_DEBUG_KMS("Could not find phy config for %d\n", pixel_clock);
@@ -928,16 +1135,6 @@ static int hdmi_create_connector(struct exynos_drm_display *display,
 	return 0;
 }
 
-static int hdmi_initialize(struct exynos_drm_display *display,
-			struct drm_device *drm_dev)
-{
-	struct hdmi_context *hdata = display->ctx;
-
-	hdata->drm_dev = drm_dev;
-
-	return 0;
-}
-
 static void hdmi_mode_fixup(struct exynos_drm_display *display,
 				struct drm_connector *connector,
 				const struct drm_display_mode *mode,
@@ -1136,20 +1333,15 @@ static void hdmi_audio_control(struct hdmi_context *hdata, bool onoff)
 			HDMI_ASP_EN : HDMI_ASP_DIS, HDMI_ASP_MASK);
 }
 
-static void hdmi_conf_reset(struct hdmi_context *hdata)
+static void hdmi_start(struct hdmi_context *hdata, bool start)
 {
-	u32 reg;
+	u32 val = start ? HDMI_TG_EN : 0;
 
-	if (hdata->type == HDMI_TYPE13)
-		reg = HDMI_V13_CORE_RSTOUT;
-	else
-		reg = HDMI_CORE_RSTOUT;
+	if (hdata->current_mode.flags & DRM_MODE_FLAG_INTERLACE)
+		val |= HDMI_FIELD_EN;
 
-	/* resetting HDMI core */
-	hdmi_reg_writemask(hdata, reg,  0, HDMI_CORE_SW_RSTOUT);
-	usleep_range(10000, 12000);
-	hdmi_reg_writemask(hdata, reg, ~0, HDMI_CORE_SW_RSTOUT);
-	usleep_range(10000, 12000);
+	hdmi_reg_writemask(hdata, HDMI_CON_0, val, HDMI_EN);
+	hdmi_reg_writemask(hdata, HDMI_TG_CMD, val, HDMI_TG_EN | HDMI_FIELD_EN);
 }
 
 static void hdmi_conf_init(struct hdmi_context *hdata)
@@ -1163,6 +1355,8 @@ static void hdmi_conf_init(struct hdmi_context *hdata)
 	/* choose HDMI mode */
 	hdmi_reg_writemask(hdata, HDMI_MODE_SEL,
 		HDMI_MODE_HDMI_EN, HDMI_MODE_MASK);
+	/* Apply Video preable and Guard band in HDMI mode only */
+	hdmi_reg_writeb(hdata, HDMI_CON_2, 0);
 	/* disable bluescreen */
 	hdmi_reg_writemask(hdata, HDMI_CON_0, 0, HDMI_BLUE_SCR_EN);
 
@@ -1286,12 +1480,7 @@ static void hdmi_v13_mode_apply(struct hdmi_context *hdata)
 	clk_prepare_enable(hdata->res.sclk_hdmi);
 
 	/* enable HDMI and timing generator */
-	hdmi_reg_writemask(hdata, HDMI_CON_0, ~0, HDMI_EN);
-	if (core->int_pro_mode[0])
-		hdmi_reg_writemask(hdata, HDMI_TG_CMD, ~0, HDMI_TG_EN |
-				HDMI_FIELD_EN);
-	else
-		hdmi_reg_writemask(hdata, HDMI_TG_CMD, ~0, HDMI_TG_EN);
+	hdmi_start(hdata, true);
 }
 
 static void hdmi_v14_mode_apply(struct hdmi_context *hdata)
@@ -1453,12 +1642,7 @@ static void hdmi_v14_mode_apply(struct hdmi_context *hdata)
 	clk_prepare_enable(hdata->res.sclk_hdmi);
 
 	/* enable HDMI and timing generator */
-	hdmi_reg_writemask(hdata, HDMI_CON_0, ~0, HDMI_EN);
-	if (core->int_pro_mode[0])
-		hdmi_reg_writemask(hdata, HDMI_TG_CMD, ~0, HDMI_TG_EN |
-				HDMI_FIELD_EN);
-	else
-		hdmi_reg_writemask(hdata, HDMI_TG_CMD, ~0, HDMI_TG_EN);
+	hdmi_start(hdata, true);
 }
 
 static void hdmi_mode_apply(struct hdmi_context *hdata)
@@ -1499,32 +1683,51 @@ static void hdmiphy_conf_reset(struct hdmi_context *hdata)
 
 static void hdmiphy_poweron(struct hdmi_context *hdata)
 {
-	if (hdata->type == HDMI_TYPE14)
-		hdmi_reg_writemask(hdata, HDMI_PHY_CON_0, 0,
-			HDMI_PHY_POWER_OFF_EN);
+	if (hdata->type != HDMI_TYPE14)
+		return;
+
+	DRM_DEBUG_KMS("\n");
+
+	/* For PHY Mode Setting */
+	hdmiphy_reg_writeb(hdata, HDMIPHY_MODE_SET_DONE,
+				HDMI_PHY_ENABLE_MODE_SET);
+	/* Phy Power On */
+	hdmiphy_reg_writeb(hdata, HDMIPHY_POWER,
+				HDMI_PHY_POWER_ON);
+	/* For PHY Mode Setting */
+	hdmiphy_reg_writeb(hdata, HDMIPHY_MODE_SET_DONE,
+				HDMI_PHY_DISABLE_MODE_SET);
+	/* PHY SW Reset */
+	hdmiphy_conf_reset(hdata);
 }
 
 static void hdmiphy_poweroff(struct hdmi_context *hdata)
 {
-	if (hdata->type == HDMI_TYPE14)
-		hdmi_reg_writemask(hdata, HDMI_PHY_CON_0, ~0,
-			HDMI_PHY_POWER_OFF_EN);
+	if (hdata->type != HDMI_TYPE14)
+		return;
+
+	DRM_DEBUG_KMS("\n");
+
+	/* PHY SW Reset */
+	hdmiphy_conf_reset(hdata);
+	/* For PHY Mode Setting */
+	hdmiphy_reg_writeb(hdata, HDMIPHY_MODE_SET_DONE,
+				HDMI_PHY_ENABLE_MODE_SET);
+
+	/* PHY Power Off */
+	hdmiphy_reg_writeb(hdata, HDMIPHY_POWER,
+				HDMI_PHY_POWER_OFF);
+
+	/* For PHY Mode Setting */
+	hdmiphy_reg_writeb(hdata, HDMIPHY_MODE_SET_DONE,
+				HDMI_PHY_DISABLE_MODE_SET);
 }
 
 static void hdmiphy_conf_apply(struct hdmi_context *hdata)
 {
-	const u8 *hdmiphy_data;
-	u8 buffer[32];
-	u8 operation[2];
-	u8 read_buffer[32] = {0, };
 	int ret;
 	int i;
 
-	if (!hdata->hdmiphy_port) {
-		DRM_ERROR("hdmiphy is not attached\n");
-		return;
-	}
-
 	/* pixel clock */
 	i = hdmi_find_phy_conf(hdata, hdata->mode_conf.pixel_clock);
 	if (i < 0) {
@@ -1532,39 +1735,21 @@ static void hdmiphy_conf_apply(struct hdmi_context *hdata)
 		return;
 	}
 
-	if (hdata->type == HDMI_TYPE13)
-		hdmiphy_data = hdmiphy_v13_configs[i].conf;
-	else
-		hdmiphy_data = hdmiphy_v14_configs[i].conf;
-
-	memcpy(buffer, hdmiphy_data, 32);
-	ret = i2c_master_send(hdata->hdmiphy_port, buffer, 32);
-	if (ret != 32) {
-		DRM_ERROR("failed to configure HDMIPHY via I2C\n");
+	ret = hdmiphy_reg_write_buf(hdata, 0, hdata->phy_confs[i].conf, 32);
+	if (ret) {
+		DRM_ERROR("failed to configure hdmiphy\n");
 		return;
 	}
 
 	usleep_range(10000, 12000);
 
-	/* operation mode */
-	operation[0] = 0x1f;
-	operation[1] = 0x80;
-
-	ret = i2c_master_send(hdata->hdmiphy_port, operation, 2);
-	if (ret != 2) {
+	ret = hdmiphy_reg_writeb(hdata, HDMIPHY_MODE_SET_DONE,
+				HDMI_PHY_DISABLE_MODE_SET);
+	if (ret) {
 		DRM_ERROR("failed to enable hdmiphy\n");
 		return;
 	}
 
-	ret = i2c_master_recv(hdata->hdmiphy_port, read_buffer, 32);
-	if (ret < 0) {
-		DRM_ERROR("failed to read hdmiphy config\n");
-		return;
-	}
-
-	for (i = 0; i < ret; i++)
-		DRM_DEBUG_KMS("hdmiphy[0x%02x] write[0x%02x] - "
-			"recv [0x%02x]\n", i, buffer[i], read_buffer[i]);
 }
 
 static void hdmi_conf_apply(struct hdmi_context *hdata)
@@ -1573,7 +1758,7 @@ static void hdmi_conf_apply(struct hdmi_context *hdata)
 	hdmiphy_conf_apply(hdata);
 
 	mutex_lock(&hdata->hdmi_mutex);
-	hdmi_conf_reset(hdata);
+	hdmi_start(hdata, false);
 	hdmi_conf_init(hdata);
 	mutex_unlock(&hdata->hdmi_mutex);
 
@@ -1814,6 +1999,9 @@ static void hdmi_mode_set(struct exynos_drm_display *display,
 		m->vrefresh, (m->flags & DRM_MODE_FLAG_INTERLACE) ?
 		"INTERLACED" : "PROGERESSIVE");
 
+	/* preserve mode information for later use. */
+	drm_mode_copy(&hdata->current_mode, mode);
+
 	if (hdata->type == HDMI_TYPE13)
 		hdmi_v13_mode_set(hdata, mode);
 	else
@@ -1854,7 +2042,10 @@ static void hdmi_poweron(struct exynos_drm_display *display)
 	if (regulator_bulk_enable(res->regul_count, res->regul_bulk))
 		DRM_DEBUG_KMS("failed to enable regulator bulk\n");
 
-	clk_prepare_enable(res->hdmiphy);
+	/* set pmu hdmiphy control bit to enable hdmiphy */
+	regmap_update_bits(hdata->pmureg, PMU_HDMI_PHY_CONTROL,
+			PMU_HDMI_PHY_ENABLE_BIT, 1);
+
 	clk_prepare_enable(res->hdmi);
 	clk_prepare_enable(res->sclk_hdmi);
 
@@ -1872,16 +2063,20 @@ static void hdmi_poweroff(struct exynos_drm_display *display)
 		goto out;
 	mutex_unlock(&hdata->hdmi_mutex);
 
-	/*
-	 * The TV power domain needs any condition of hdmiphy to turn off and
-	 * its reset state seems to meet the condition.
-	 */
-	hdmiphy_conf_reset(hdata);
+	/* HDMI System Disable */
+	hdmi_reg_writemask(hdata, HDMI_CON_0, 0, HDMI_EN);
+
 	hdmiphy_poweroff(hdata);
 
+	cancel_delayed_work(&hdata->hotplug_work);
+
 	clk_disable_unprepare(res->sclk_hdmi);
 	clk_disable_unprepare(res->hdmi);
-	clk_disable_unprepare(res->hdmiphy);
+
+	/* reset pmu hdmiphy control bit to disable hdmiphy */
+	regmap_update_bits(hdata->pmureg, PMU_HDMI_PHY_CONTROL,
+			PMU_HDMI_PHY_ENABLE_BIT, 0);
+
 	regulator_bulk_disable(res->regul_count, res->regul_bulk);
 
 	pm_runtime_put_sync(hdata->dev);
@@ -1913,7 +2108,6 @@ static void hdmi_dpms(struct exynos_drm_display *display, int mode)
 }
 
 static struct exynos_drm_display_ops hdmi_display_ops = {
-	.initialize	= hdmi_initialize,
 	.create_connector = hdmi_create_connector,
 	.mode_fixup	= hdmi_mode_fixup,
 	.mode_set	= hdmi_mode_set,
@@ -1926,9 +2120,11 @@ static struct exynos_drm_display hdmi_display = {
 	.ops = &hdmi_display_ops,
 };
 
-static irqreturn_t hdmi_irq_thread(int irq, void *arg)
+static void hdmi_hotplug_work_func(struct work_struct *work)
 {
-	struct hdmi_context *hdata = arg;
+	struct hdmi_context *hdata;
+
+	hdata = container_of(work, struct hdmi_context, hotplug_work.work);
 
 	mutex_lock(&hdata->hdmi_mutex);
 	hdata->hpd = gpio_get_value(hdata->hpd_gpio);
@@ -1936,6 +2132,14 @@ static irqreturn_t hdmi_irq_thread(int irq, void *arg)
 
 	if (hdata->drm_dev)
 		drm_helper_hpd_irq_event(hdata->drm_dev);
+}
+
+static irqreturn_t hdmi_irq_thread(int irq, void *arg)
+{
+	struct hdmi_context *hdata = arg;
+
+	mod_delayed_work(system_wq, &hdata->hotplug_work,
+			msecs_to_jiffies(HOTPLUG_DEBOUNCE_MS));
 
 	return IRQ_HANDLED;
 }
@@ -1954,37 +2158,35 @@ static int hdmi_resources_init(struct hdmi_context *hdata)
 
 	DRM_DEBUG_KMS("HDMI resource init\n");
 
-	memset(res, 0, sizeof(*res));
-
 	/* get clocks, power */
 	res->hdmi = devm_clk_get(dev, "hdmi");
 	if (IS_ERR(res->hdmi)) {
 		DRM_ERROR("failed to get clock 'hdmi'\n");
+		ret = PTR_ERR(res->hdmi);
 		goto fail;
 	}
 	res->sclk_hdmi = devm_clk_get(dev, "sclk_hdmi");
 	if (IS_ERR(res->sclk_hdmi)) {
 		DRM_ERROR("failed to get clock 'sclk_hdmi'\n");
+		ret = PTR_ERR(res->sclk_hdmi);
 		goto fail;
 	}
 	res->sclk_pixel = devm_clk_get(dev, "sclk_pixel");
 	if (IS_ERR(res->sclk_pixel)) {
 		DRM_ERROR("failed to get clock 'sclk_pixel'\n");
+		ret = PTR_ERR(res->sclk_pixel);
 		goto fail;
 	}
 	res->sclk_hdmiphy = devm_clk_get(dev, "sclk_hdmiphy");
 	if (IS_ERR(res->sclk_hdmiphy)) {
 		DRM_ERROR("failed to get clock 'sclk_hdmiphy'\n");
-		goto fail;
-	}
-	res->hdmiphy = devm_clk_get(dev, "hdmiphy");
-	if (IS_ERR(res->hdmiphy)) {
-		DRM_ERROR("failed to get clock 'hdmiphy'\n");
+		ret = PTR_ERR(res->sclk_hdmiphy);
 		goto fail;
 	}
 	res->mout_hdmi = devm_clk_get(dev, "mout_hdmi");
 	if (IS_ERR(res->mout_hdmi)) {
 		DRM_ERROR("failed to get clock 'mout_hdmi'\n");
+		ret = PTR_ERR(res->mout_hdmi);
 		goto fail;
 	}
 
@@ -1992,8 +2194,10 @@ static int hdmi_resources_init(struct hdmi_context *hdata)
 
 	res->regul_bulk = devm_kzalloc(dev, ARRAY_SIZE(supply) *
 		sizeof(res->regul_bulk[0]), GFP_KERNEL);
-	if (!res->regul_bulk)
+	if (!res->regul_bulk) {
+		ret = -ENOMEM;
 		goto fail;
+	}
 	for (i = 0; i < ARRAY_SIZE(supply); ++i) {
 		res->regul_bulk[i].supply = supply[i];
 		res->regul_bulk[i].consumer = NULL;
@@ -2001,14 +2205,14 @@ static int hdmi_resources_init(struct hdmi_context *hdata)
 	ret = devm_regulator_bulk_get(dev, ARRAY_SIZE(supply), res->regul_bulk);
 	if (ret) {
 		DRM_ERROR("failed to get regulators\n");
-		goto fail;
+		return ret;
 	}
 	res->regul_count = ARRAY_SIZE(supply);
 
-	return 0;
+	return ret;
 fail:
 	DRM_ERROR("HDMI resource init - failed\n");
-	return -ENODEV;
+	return ret;
 }
 
 static struct s5p_hdmi_platform_data *drm_hdmi_dt_parse_pdata
@@ -2043,42 +2247,105 @@ static struct of_device_id hdmi_match_types[] = {
 		.compatible = "samsung,exynos4212-hdmi",
 		.data = &exynos4212_hdmi_driver_data,
 	}, {
+		.compatible = "samsung,exynos5420-hdmi",
+		.data = &exynos5420_hdmi_driver_data,
+	}, {
 		/* end node */
 	}
 };
 
+static int hdmi_bind(struct device *dev, struct device *master, void *data)
+{
+	struct drm_device *drm_dev = data;
+	struct hdmi_context *hdata;
+
+	hdata = hdmi_display.ctx;
+	hdata->drm_dev = drm_dev;
+
+	return exynos_drm_create_enc_conn(drm_dev, &hdmi_display);
+}
+
+static void hdmi_unbind(struct device *dev, struct device *master, void *data)
+{
+	struct exynos_drm_display *display = get_hdmi_display(dev);
+	struct drm_encoder *encoder = display->encoder;
+	struct hdmi_context *hdata = display->ctx;
+
+	encoder->funcs->destroy(encoder);
+	drm_connector_cleanup(&hdata->connector);
+}
+
+static const struct component_ops hdmi_component_ops = {
+	.bind	= hdmi_bind,
+	.unbind = hdmi_unbind,
+};
+
+static struct device_node *hdmi_legacy_ddc_dt_binding(struct device *dev)
+{
+	const char *compatible_str = "samsung,exynos4210-hdmiddc";
+	struct device_node *np;
+
+	np = of_find_compatible_node(NULL, NULL, compatible_str);
+	if (np)
+		return of_get_next_parent(np);
+
+	return NULL;
+}
+
+static struct device_node *hdmi_legacy_phy_dt_binding(struct device *dev)
+{
+	const char *compatible_str = "samsung,exynos4212-hdmiphy";
+
+	return of_find_compatible_node(NULL, NULL, compatible_str);
+}
+
 static int hdmi_probe(struct platform_device *pdev)
 {
+	struct device_node *ddc_node, *phy_node;
+	struct s5p_hdmi_platform_data *pdata;
+	struct hdmi_driver_data *drv_data;
+	const struct of_device_id *match;
 	struct device *dev = &pdev->dev;
 	struct hdmi_context *hdata;
-	struct s5p_hdmi_platform_data *pdata;
 	struct resource *res;
-	const struct of_device_id *match;
-	struct device_node *ddc_node, *phy_node;
-	struct hdmi_driver_data *drv_data;
 	int ret;
 
-	 if (!dev->of_node)
-		return -ENODEV;
+	ret = exynos_drm_component_add(&pdev->dev, EXYNOS_DEVICE_TYPE_CONNECTOR,
+					hdmi_display.type);
+	if (ret)
+		return ret;
+
+	if (!dev->of_node) {
+		ret = -ENODEV;
+		goto err_del_component;
+	}
 
 	pdata = drm_hdmi_dt_parse_pdata(dev);
-	if (!pdata)
-		return -EINVAL;
+	if (!pdata) {
+		ret = -EINVAL;
+		goto err_del_component;
+	}
 
 	hdata = devm_kzalloc(dev, sizeof(struct hdmi_context), GFP_KERNEL);
-	if (!hdata)
-		return -ENOMEM;
+	if (!hdata) {
+		ret = -ENOMEM;
+		goto err_del_component;
+	}
 
 	mutex_init(&hdata->hdmi_mutex);
 
 	platform_set_drvdata(pdev, &hdmi_display);
 
 	match = of_match_node(hdmi_match_types, dev->of_node);
-	if (!match)
-		return -ENODEV;
+	if (!match) {
+		ret = -ENODEV;
+		goto err_del_component;
+	}
 
 	drv_data = (struct hdmi_driver_data *)match->data;
 	hdata->type = drv_data->type;
+	hdata->phy_confs = drv_data->phy_confs;
+	hdata->phy_conf_count = drv_data->phy_conf_count;
 
 	hdata->hpd_gpio = pdata->hpd_gpio;
 	hdata->dev = dev;
@@ -2086,35 +2353,44 @@ static int hdmi_probe(struct platform_device *pdev)
 	ret = hdmi_resources_init(hdata);
 	if (ret) {
 		DRM_ERROR("hdmi_resources_init failed\n");
-		return -EINVAL;
+		return ret;
 	}
 
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	hdata->regs = devm_ioremap_resource(dev, res);
-	if (IS_ERR(hdata->regs))
-		return PTR_ERR(hdata->regs);
+	if (IS_ERR(hdata->regs)) {
+		ret = PTR_ERR(hdata->regs);
+		goto err_del_component;
+	}
 
 	ret = devm_gpio_request(dev, hdata->hpd_gpio, "HPD");
 	if (ret) {
 		DRM_ERROR("failed to request HPD gpio\n");
-		return ret;
+		goto err_del_component;
 	}
 
+	ddc_node = hdmi_legacy_ddc_dt_binding(dev);
+	if (ddc_node)
+		goto out_get_ddc_adpt;
+
 	/* DDC i2c driver */
 	ddc_node = of_parse_phandle(dev->of_node, "ddc", 0);
 	if (!ddc_node) {
 		DRM_ERROR("Failed to find ddc node in device tree\n");
-		return -ENODEV;
+		ret = -ENODEV;
+		goto err_del_component;
 	}
+
+out_get_ddc_adpt:
 	hdata->ddc_adpt = of_find_i2c_adapter_by_node(ddc_node);
 	if (!hdata->ddc_adpt) {
 		DRM_ERROR("Failed to get ddc i2c adapter by node\n");
-		return -ENODEV;
+		return -EPROBE_DEFER;
 	}
 
-	/* Not support APB PHY yet. */
-	if (drv_data->is_apb_phy)
-		return -EPERM;
+	phy_node = hdmi_legacy_phy_dt_binding(dev);
+	if (phy_node)
+		goto out_get_phy_port;
 
 	/* hdmiphy i2c driver */
 	phy_node = of_parse_phandle(dev->of_node, "phy", 0);
@@ -2123,11 +2399,22 @@ static int hdmi_probe(struct platform_device *pdev)
 		ret = -ENODEV;
 		goto err_ddc;
 	}
-	hdata->hdmiphy_port = of_find_i2c_device_by_node(phy_node);
-	if (!hdata->hdmiphy_port) {
-		DRM_ERROR("Failed to get hdmi phy i2c client from node\n");
-		ret = -ENODEV;
-		goto err_ddc;
+
+out_get_phy_port:
+	if (drv_data->is_apb_phy) {
+		hdata->regs_hdmiphy = of_iomap(phy_node, 0);
+		if (!hdata->regs_hdmiphy) {
+			DRM_ERROR("failed to ioremap hdmi phy\n");
+			ret = -ENOMEM;
+			goto err_ddc;
+		}
+	} else {
+		hdata->hdmiphy_port = of_find_i2c_device_by_node(phy_node);
+		if (!hdata->hdmiphy_port) {
+			DRM_ERROR("Failed to get hdmi phy i2c client\n");
+			ret = -EPROBE_DEFER;
+			goto err_ddc;
+		}
 	}
 
 	hdata->irq = gpio_to_irq(hdata->hpd_gpio);
@@ -2139,6 +2426,8 @@ static int hdmi_probe(struct platform_device *pdev)
 
 	hdata->hpd = gpio_get_value(hdata->hpd_gpio);
 
+	INIT_DELAYED_WORK(&hdata->hotplug_work, hdmi_hotplug_work_func);
+
 	ret = devm_request_threaded_irq(dev, hdata->irq, NULL,
 			hdmi_irq_thread, IRQF_TRIGGER_RISING |
 			IRQF_TRIGGER_FALLING | IRQF_ONESHOT,
@@ -2148,30 +2437,51 @@ static int hdmi_probe(struct platform_device *pdev)
 		goto err_hdmiphy;
 	}
 
-	pm_runtime_enable(dev);
+	hdata->pmureg = syscon_regmap_lookup_by_phandle(dev->of_node,
+			"samsung,syscon-phandle");
+	if (IS_ERR(hdata->pmureg)) {
+		DRM_ERROR("syscon regmap lookup failed.\n");
+		ret = -EPROBE_DEFER;
+		goto err_hdmiphy;
+	}
 
+	pm_runtime_enable(dev);
 	hdmi_display.ctx = hdata;
-	exynos_drm_display_register(&hdmi_display);
 
-	return 0;
+	ret = component_add(&pdev->dev, &hdmi_component_ops);
+	if (ret)
+		goto err_disable_pm_runtime;
+
+	return ret;
+
+err_disable_pm_runtime:
+	pm_runtime_disable(dev);
 
 err_hdmiphy:
-	put_device(&hdata->hdmiphy_port->dev);
+	if (hdata->hdmiphy_port)
+		put_device(&hdata->hdmiphy_port->dev);
 err_ddc:
 	put_device(&hdata->ddc_adpt->dev);
+
+err_del_component:
+	exynos_drm_component_del(&pdev->dev, EXYNOS_DEVICE_TYPE_CONNECTOR);
+
 	return ret;
 }
 
 static int hdmi_remove(struct platform_device *pdev)
 {
-	struct device *dev = &pdev->dev;
-	struct exynos_drm_display *display = get_hdmi_display(dev);
-	struct hdmi_context *hdata = display->ctx;
+	struct hdmi_context *hdata = hdmi_display.ctx;
+
+	cancel_delayed_work_sync(&hdata->hotplug_work);
 
 	put_device(&hdata->hdmiphy_port->dev);
 	put_device(&hdata->ddc_adpt->dev);
+
 	pm_runtime_disable(&pdev->dev);
+	component_del(&pdev->dev, &hdmi_component_ops);
 
+	exynos_drm_component_del(&pdev->dev, EXYNOS_DEVICE_TYPE_CONNECTOR);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.h b/drivers/gpu/drm/exynos/exynos_hdmi.h
deleted file mode 100644
index 0ddf3957de15..000000000000
--- a/drivers/gpu/drm/exynos/exynos_hdmi.h
+++ /dev/null
@@ -1,23 +0,0 @@
-/*
- *
- * Copyright (c) 2011 Samsung Electronics Co., Ltd.
- * Authors:
- *	Inki Dae <inki.dae@samsung.com>
- *	Seung-Woo Kim <sw0312.kim@samsung.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- */
-
-#ifndef _EXYNOS_HDMI_H_
-#define _EXYNOS_HDMI_H_
-
-void hdmi_attach_ddc_client(struct i2c_client *ddc);
-void hdmi_attach_hdmiphy_client(struct i2c_client *hdmiphy);
-
-extern struct i2c_driver hdmiphy_driver;
-extern struct i2c_driver ddc_driver;
-
-#endif
diff --git a/drivers/gpu/drm/exynos/exynos_hdmiphy.c b/drivers/gpu/drm/exynos/exynos_hdmiphy.c
deleted file mode 100644
index 59abb1494ceb..000000000000
--- a/drivers/gpu/drm/exynos/exynos_hdmiphy.c
+++ /dev/null
@@ -1,65 +0,0 @@
-/*
- * Copyright (C) 2011 Samsung Electronics Co.Ltd
- * Authors:
- *	Seung-Woo Kim <sw0312.kim@samsung.com>
- *	Inki Dae <inki.dae@samsung.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- *
- */
-
-#include <drm/drmP.h>
-
-#include <linux/kernel.h>
-#include <linux/i2c.h>
-#include <linux/of.h>
-
-#include "exynos_drm_drv.h"
-#include "exynos_hdmi.h"
-
-
-static int hdmiphy_probe(struct i2c_client *client,
-	const struct i2c_device_id *id)
-{
-	hdmi_attach_hdmiphy_client(client);
-
-	dev_info(&client->adapter->dev, "attached s5p_hdmiphy "
-		"into i2c adapter successfully\n");
-
-	return 0;
-}
-
-static int hdmiphy_remove(struct i2c_client *client)
-{
-	dev_info(&client->adapter->dev, "detached s5p_hdmiphy "
-		"from i2c adapter successfully\n");
-
-	return 0;
-}
-
-static struct of_device_id hdmiphy_match_types[] = {
-	{
-		.compatible = "samsung,exynos5-hdmiphy",
-	}, {
-		.compatible = "samsung,exynos4210-hdmiphy",
-	}, {
-		.compatible = "samsung,exynos4212-hdmiphy",
-	}, {
-		/* end node */
-	}
-};
-
-struct i2c_driver hdmiphy_driver = {
-	.driver = {
-		.name	= "exynos-hdmiphy",
-		.owner	= THIS_MODULE,
-		.of_match_table = hdmiphy_match_types,
-	},
-	.probe		= hdmiphy_probe,
-	.remove		= hdmiphy_remove,
-	.command		= NULL,
-};
-EXPORT_SYMBOL(hdmiphy_driver);
diff --git a/drivers/gpu/drm/exynos/exynos_mixer.c b/drivers/gpu/drm/exynos/exynos_mixer.c
index ce288818d2c0..4c5aed7e54c8 100644
--- a/drivers/gpu/drm/exynos/exynos_mixer.c
+++ b/drivers/gpu/drm/exynos/exynos_mixer.c
@@ -31,6 +31,7 @@
 #include <linux/clk.h>
 #include <linux/regulator/consumer.h>
 #include <linux/of.h>
+#include <linux/component.h>
 
 #include <drm/exynos_drm.h>
 
@@ -830,13 +831,15 @@ static int vp_resources_init(struct mixer_context *mixer_ctx)
 }
 
 static int mixer_initialize(struct exynos_drm_manager *mgr,
-			struct drm_device *drm_dev, int pipe)
+			struct drm_device *drm_dev)
 {
 	int ret;
 	struct mixer_context *mixer_ctx = mgr->ctx;
+	struct exynos_drm_private *priv;
+	priv = drm_dev->dev_private;
 
-	mixer_ctx->drm_dev = drm_dev;
-	mixer_ctx->pipe = pipe;
+	mgr->drm_dev = mixer_ctx->drm_dev = drm_dev;
+	mgr->pipe = mixer_ctx->pipe = priv->pipe++;
 
 	/* acquire resources: regs, irqs, clocks */
 	ret = mixer_resources_init(mixer_ctx);
@@ -1142,8 +1145,6 @@ int mixer_check_mode(struct drm_display_mode *mode)
 }
 
 static struct exynos_drm_manager_ops mixer_manager_ops = {
-	.initialize		= mixer_initialize,
-	.remove			= mixer_mgr_remove,
 	.dpms			= mixer_dpms,
 	.enable_vblank		= mixer_enable_vblank,
 	.disable_vblank		= mixer_disable_vblank,
@@ -1200,11 +1201,13 @@ static struct of_device_id mixer_match_types[] = {
 	}
 };
 
-static int mixer_probe(struct platform_device *pdev)
+static int mixer_bind(struct device *dev, struct device *manager, void *data)
 {
-	struct device *dev = &pdev->dev;
+	struct platform_device *pdev = to_platform_device(dev);
+	struct drm_device *drm_dev = data;
 	struct mixer_context *ctx;
 	struct mixer_drv_data *drv;
+	int ret;
 
 	dev_info(dev, "probe start\n");
 
@@ -1233,19 +1236,61 @@ static int mixer_probe(struct platform_device *pdev)
 	atomic_set(&ctx->wait_vsync_event, 0);
 
 	mixer_manager.ctx = ctx;
+	ret = mixer_initialize(&mixer_manager, drm_dev);
+	if (ret)
+		return ret;
+
 	platform_set_drvdata(pdev, &mixer_manager);
-	exynos_drm_manager_register(&mixer_manager);
+	ret = exynos_drm_crtc_create(&mixer_manager);
+	if (ret) {
+		mixer_mgr_remove(&mixer_manager);
+		return ret;
+	}
 
 	pm_runtime_enable(dev);
 
 	return 0;
 }
 
-static int mixer_remove(struct platform_device *pdev)
+static void mixer_unbind(struct device *dev, struct device *master, void *data)
+{
+	struct exynos_drm_manager *mgr = dev_get_drvdata(dev);
+	struct drm_crtc *crtc = mgr->crtc;
+
+	dev_info(dev, "remove successful\n");
+
+	mixer_mgr_remove(mgr);
+
+	pm_runtime_disable(dev);
+
+	crtc->funcs->destroy(crtc);
+}
+
+static const struct component_ops mixer_component_ops = {
+	.bind	= mixer_bind,
+	.unbind	= mixer_unbind,
+};
+
+static int mixer_probe(struct platform_device *pdev)
 {
-	dev_info(&pdev->dev, "remove successful\n");
+	int ret;
 
-	pm_runtime_disable(&pdev->dev);
+	ret = exynos_drm_component_add(&pdev->dev, EXYNOS_DEVICE_TYPE_CRTC,
+					mixer_manager.type);
+	if (ret)
+		return ret;
+
+	ret = component_add(&pdev->dev, &mixer_component_ops);
+	if (ret)
+		exynos_drm_component_del(&pdev->dev, EXYNOS_DEVICE_TYPE_CRTC);
+
+	return ret;
+}
+
+static int mixer_remove(struct platform_device *pdev)
+{
+	component_del(&pdev->dev, &mixer_component_ops);
+	exynos_drm_component_del(&pdev->dev, EXYNOS_DEVICE_TYPE_CRTC);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/exynos/regs-hdmi.h b/drivers/gpu/drm/exynos/regs-hdmi.h
index ef1b3eb3ba6e..3f35ac6d8a47 100644
--- a/drivers/gpu/drm/exynos/regs-hdmi.h
+++ b/drivers/gpu/drm/exynos/regs-hdmi.h
@@ -578,4 +578,20 @@
 #define HDMI_TG_VACT_ST4_H		HDMI_TG_BASE(0x0074)
 #define HDMI_TG_3D			HDMI_TG_BASE(0x00F0)
 
+/* HDMI PHY Registers Offsets*/
+#define HDMIPHY_POWER		(0x74 >> 2)
+#define HDMIPHY_MODE_SET_DONE		(0x7c >> 2)
+
+/* HDMI PHY Values */
+#define HDMI_PHY_POWER_ON              0x80
+#define HDMI_PHY_POWER_OFF             0xff
+
+/* HDMI PHY Values */
+#define HDMI_PHY_DISABLE_MODE_SET	0x80
+#define HDMI_PHY_ENABLE_MODE_SET	0x00
+
+/* PMU Registers for PHY */
+#define PMU_HDMI_PHY_CONTROL		0x700
+#define PMU_HDMI_PHY_ENABLE_BIT		BIT(0)
+
 #endif /* SAMSUNG_REGS_HDMI_H */
diff --git a/drivers/gpu/drm/gma500/mdfld_dsi_pkg_sender.c b/drivers/gpu/drm/gma500/mdfld_dsi_pkg_sender.c
index 489ffd2c66e5..87885d8c06e8 100644
--- a/drivers/gpu/drm/gma500/mdfld_dsi_pkg_sender.c
+++ b/drivers/gpu/drm/gma500/mdfld_dsi_pkg_sender.c
@@ -148,7 +148,7 @@ static int handle_dsi_error(struct mdfld_dsi_pkg_sender *sender, u32 mask)
 		break;
 	case BIT(14):
 		/*wait for all fifo empty*/
-		/*wait_for_all_fifos_empty(sender)*/;
+		/*wait_for_all_fifos_empty(sender)*/
 		break;
 	case BIT(15):
 		dev_dbg(sender->dev->dev, "No Action required\n");
diff --git a/drivers/gpu/drm/gma500/psb_drv.c b/drivers/gpu/drm/gma500/psb_drv.c
index b686e56646eb..6e8fe9ec02b5 100644
--- a/drivers/gpu/drm/gma500/psb_drv.c
+++ b/drivers/gpu/drm/gma500/psb_drv.c
@@ -112,11 +112,9 @@ static void psb_driver_lastclose(struct drm_device *dev)
 	struct drm_psb_private *dev_priv = dev->dev_private;
 	struct psb_fbdev *fbdev = dev_priv->fbdev;
 
-	drm_modeset_lock_all(dev);
-	ret = drm_fb_helper_restore_fbdev_mode(&fbdev->psb_fb_helper);
+	ret = drm_fb_helper_restore_fbdev_mode_unlocked(&fbdev->psb_fb_helper);
 	if (ret)
 		DRM_DEBUG("failed to restore crtc mode\n");
-	drm_modeset_unlock_all(dev);
 
 	return;
 }
@@ -354,7 +352,7 @@ static int psb_driver_load(struct drm_device *dev, unsigned long flags)
 	PSB_WVDC32(0xFFFFFFFF, PSB_INT_MASK_R);
 	spin_unlock_irqrestore(&dev_priv->irqmask_lock, irqflags);
 
-	drm_irq_install(dev);
+	drm_irq_install(dev, dev->pdev->irq);
 
 	dev->vblank_disable_allowed = true;
 	dev->max_vblank_count = 0xffffff; /* only 24 bits of frame count */
@@ -479,7 +477,7 @@ static struct drm_driver driver = {
 	.lastclose = psb_driver_lastclose,
 	.preclose = psb_driver_preclose,
 
-	.num_ioctls = DRM_ARRAY_SIZE(psb_ioctls),
+	.num_ioctls = ARRAY_SIZE(psb_ioctls),
 	.device_is_agp = psb_driver_device_is_agp,
 	.irq_preinstall = psb_irq_preinstall,
 	.irq_postinstall = psb_irq_postinstall,
diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c
index 48af5cac1902..240c331405b9 100644
--- a/drivers/gpu/drm/i2c/tda998x_drv.c
+++ b/drivers/gpu/drm/i2c/tda998x_drv.c
@@ -568,11 +568,11 @@ static irqreturn_t tda998x_irq_thread(int irq, void *data)
 
 static uint8_t tda998x_cksum(uint8_t *buf, size_t bytes)
 {
-	uint8_t sum = 0;
+	int sum = 0;
 
 	while (bytes--)
-		sum += *buf++;
-	return (255 - sum) + 1;
+		sum -= *buf++;
+	return sum;
 }
 
 #define HB(x) (x)
diff --git a/drivers/gpu/drm/i810/i810_dma.c b/drivers/gpu/drm/i810/i810_dma.c
index aeace37415aa..e88bac1d781f 100644
--- a/drivers/gpu/drm/i810/i810_dma.c
+++ b/drivers/gpu/drm/i810/i810_dma.c
@@ -1251,7 +1251,7 @@ const struct drm_ioctl_desc i810_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(I810_FLIP, i810_flip_bufs, DRM_AUTH|DRM_UNLOCKED),
 };
 
-int i810_max_ioctl = DRM_ARRAY_SIZE(i810_ioctls);
+int i810_max_ioctl = ARRAY_SIZE(i810_ioctls);
 
 /**
  * Determine if the device really is AGP or not.
diff --git a/drivers/gpu/drm/i915/Kconfig b/drivers/gpu/drm/i915/Kconfig
index bea2d67196fb..437e1824d0bf 100644
--- a/drivers/gpu/drm/i915/Kconfig
+++ b/drivers/gpu/drm/i915/Kconfig
@@ -5,6 +5,7 @@ config DRM_I915
 	depends on (AGP || AGP=n)
 	select INTEL_GTT
 	select AGP_INTEL if AGP
+	select INTERVAL_TREE
 	# we need shmfs for the swappable backing store, and in particular
 	# the shmem_readpage() which depends upon tmpfs
 	select SHMEM
@@ -71,7 +72,7 @@ config DRM_I915_PRELIMINARY_HW_SUPPORT
 
 config DRM_I915_UMS
 	bool "Enable userspace modesetting on Intel hardware (DEPRECATED)"
-	depends on DRM_I915
+	depends on DRM_I915 && BROKEN
 	default n
 	help
 	  Choose this option if you still need userspace modesetting.
diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index b1445b73465b..cad1683d8bb5 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -18,6 +18,7 @@ i915-$(CONFIG_DEBUG_FS) += i915_debugfs.o
 # GEM code
 i915-y += i915_cmd_parser.o \
 	  i915_gem_context.o \
+	  i915_gem_render_state.o \
 	  i915_gem_debug.o \
 	  i915_gem_dmabuf.o \
 	  i915_gem_evict.o \
@@ -26,12 +27,18 @@ i915-y += i915_cmd_parser.o \
 	  i915_gem.o \
 	  i915_gem_stolen.o \
 	  i915_gem_tiling.o \
+	  i915_gem_userptr.o \
 	  i915_gpu_error.o \
 	  i915_irq.o \
 	  i915_trace_points.o \
 	  intel_ringbuffer.o \
 	  intel_uncore.o
 
+# autogenerated null render state
+i915-y += intel_renderstate_gen6.o \
+	  intel_renderstate_gen7.o \
+	  intel_renderstate_gen8.o
+
 # modesetting core code
 i915-y += intel_bios.o \
 	  intel_display.o \
@@ -55,6 +62,7 @@ i915-y += dvo_ch7017.o \
 	  intel_dsi_cmd.o \
 	  intel_dsi.o \
 	  intel_dsi_pll.o \
+	  intel_dsi_panel_vbt.o \
 	  intel_dvo.o \
 	  intel_hdmi.o \
 	  intel_i2c.o \
diff --git a/drivers/gpu/drm/i915/dvo_ch7xxx.c b/drivers/gpu/drm/i915/dvo_ch7xxx.c
index a0f5bdd69491..80449f475960 100644
--- a/drivers/gpu/drm/i915/dvo_ch7xxx.c
+++ b/drivers/gpu/drm/i915/dvo_ch7xxx.c
@@ -160,7 +160,7 @@ static bool ch7xxx_readb(struct intel_dvo_device *dvo, int addr, uint8_t *ch)
 	if (i2c_transfer(adapter, msgs, 2) == 2) {
 		*ch = in_buf[0];
 		return true;
-	};
+	}
 
 	if (!ch7xxx->quiet) {
 		DRM_DEBUG_KMS("Unable to read register 0x%02x from %s:%02x.\n",
diff --git a/drivers/gpu/drm/i915/dvo_ivch.c b/drivers/gpu/drm/i915/dvo_ivch.c
index 0f1865d7d4d8..0f2587ff347c 100644
--- a/drivers/gpu/drm/i915/dvo_ivch.c
+++ b/drivers/gpu/drm/i915/dvo_ivch.c
@@ -195,7 +195,7 @@ static bool ivch_read(struct intel_dvo_device *dvo, int addr, uint16_t *data)
 	if (i2c_transfer(adapter, msgs, 3) == 3) {
 		*data = (in_buf[1] << 8) | in_buf[0];
 		return true;
-	};
+	}
 
 	if (!priv->quiet) {
 		DRM_DEBUG_KMS("Unable to read register 0x%02x from "
diff --git a/drivers/gpu/drm/i915/dvo_ns2501.c b/drivers/gpu/drm/i915/dvo_ns2501.c
index 8155ded79079..74f2af7c2d3e 100644
--- a/drivers/gpu/drm/i915/dvo_ns2501.c
+++ b/drivers/gpu/drm/i915/dvo_ns2501.c
@@ -121,7 +121,7 @@ static bool ns2501_readb(struct intel_dvo_device *dvo, int addr, uint8_t * ch)
 	if (i2c_transfer(adapter, msgs, 2) == 2) {
 		*ch = in_buf[0];
 		return true;
-	};
+	}
 
 	if (!ns->quiet) {
 		DRM_DEBUG_KMS
@@ -233,9 +233,8 @@ static enum drm_mode_status ns2501_mode_valid(struct intel_dvo_device *dvo,
 					      struct drm_display_mode *mode)
 {
 	DRM_DEBUG_KMS
-	    ("%s: is mode valid (hdisplay=%d,htotal=%d,vdisplay=%d,vtotal=%d)\n",
-	     __FUNCTION__, mode->hdisplay, mode->htotal, mode->vdisplay,
-	     mode->vtotal);
+	    ("is mode valid (hdisplay=%d,htotal=%d,vdisplay=%d,vtotal=%d)\n",
+	     mode->hdisplay, mode->htotal, mode->vdisplay, mode->vtotal);
 
 	/*
 	 * Currently, these are all the modes I have data from.
@@ -261,9 +260,8 @@ static void ns2501_mode_set(struct intel_dvo_device *dvo,
 	struct ns2501_priv *ns = (struct ns2501_priv *)(dvo->dev_priv);
 
 	DRM_DEBUG_KMS
-	    ("%s: set mode (hdisplay=%d,htotal=%d,vdisplay=%d,vtotal=%d).\n",
-	     __FUNCTION__, mode->hdisplay, mode->htotal, mode->vdisplay,
-	     mode->vtotal);
+	    ("set mode (hdisplay=%d,htotal=%d,vdisplay=%d,vtotal=%d).\n",
+	     mode->hdisplay, mode->htotal, mode->vdisplay, mode->vtotal);
 
 	/*
 	 * Where do I find the native resolution for which scaling is not required???
@@ -277,8 +275,7 @@ static void ns2501_mode_set(struct intel_dvo_device *dvo,
 		if (mode->hdisplay == 800 && mode->vdisplay == 600) {
 			/* mode 277 */
 			ns->reg_8_shadow &= ~NS2501_8_BPAS;
-			DRM_DEBUG_KMS("%s: switching to 800x600\n",
-				      __FUNCTION__);
+			DRM_DEBUG_KMS("switching to 800x600\n");
 
 			/*
 			 * No, I do not know where this data comes from.
@@ -341,8 +338,7 @@ static void ns2501_mode_set(struct intel_dvo_device *dvo,
 
 		} else if (mode->hdisplay == 640 && mode->vdisplay == 480) {
 			/* mode 274 */
-			DRM_DEBUG_KMS("%s: switching to 640x480\n",
-				      __FUNCTION__);
+			DRM_DEBUG_KMS("switching to 640x480\n");
 			/*
 			 * No, I do not know where this data comes from.
 			 * It is just what the video bios left in the DVO, so
@@ -406,8 +402,7 @@ static void ns2501_mode_set(struct intel_dvo_device *dvo,
 
 		} else if (mode->hdisplay == 1024 && mode->vdisplay == 768) {
 			/* mode 280 */
-			DRM_DEBUG_KMS("%s: switching to 1024x768\n",
-				      __FUNCTION__);
+			DRM_DEBUG_KMS("switching to 1024x768\n");
 			/*
 			 * This might or might not work, actually. I'm silently
 			 * assuming here that the native panel resolution is
@@ -458,8 +453,7 @@ static void ns2501_dpms(struct intel_dvo_device *dvo, bool enable)
 	struct ns2501_priv *ns = (struct ns2501_priv *)(dvo->dev_priv);
 	unsigned char ch;
 
-	DRM_DEBUG_KMS("%s: Trying set the dpms of the DVO to %i\n",
-		      __FUNCTION__, enable);
+	DRM_DEBUG_KMS("Trying set the dpms of the DVO to %i\n", enable);
 
 	ch = ns->reg_8_shadow;
 
diff --git a/drivers/gpu/drm/i915/dvo_sil164.c b/drivers/gpu/drm/i915/dvo_sil164.c
index 7b3e9e936200..fa0114967076 100644
--- a/drivers/gpu/drm/i915/dvo_sil164.c
+++ b/drivers/gpu/drm/i915/dvo_sil164.c
@@ -93,7 +93,7 @@ static bool sil164_readb(struct intel_dvo_device *dvo, int addr, uint8_t *ch)
 	if (i2c_transfer(adapter, msgs, 2) == 2) {
 		*ch = in_buf[0];
 		return true;
-	};
+	}
 
 	if (!sil->quiet) {
 		DRM_DEBUG_KMS("Unable to read register 0x%02x from %s:%02x.\n",
diff --git a/drivers/gpu/drm/i915/dvo_tfp410.c b/drivers/gpu/drm/i915/dvo_tfp410.c
index 12ea4b164692..7853719a0e81 100644
--- a/drivers/gpu/drm/i915/dvo_tfp410.c
+++ b/drivers/gpu/drm/i915/dvo_tfp410.c
@@ -118,7 +118,7 @@ static bool tfp410_readb(struct intel_dvo_device *dvo, int addr, uint8_t *ch)
 	if (i2c_transfer(adapter, msgs, 2) == 2) {
 		*ch = in_buf[0];
 		return true;
-	};
+	}
 
 	if (!tfp->quiet) {
 		DRM_DEBUG_KMS("Unable to read register 0x%02x from %s:%02x.\n",
diff --git a/drivers/gpu/drm/i915/i915_cmd_parser.c b/drivers/gpu/drm/i915/i915_cmd_parser.c
index 4cf6d020d513..9d7954366bd2 100644
--- a/drivers/gpu/drm/i915/i915_cmd_parser.c
+++ b/drivers/gpu/drm/i915/i915_cmd_parser.c
@@ -28,7 +28,7 @@
 #include "i915_drv.h"
 
 /**
- * DOC: i915 batch buffer command parser
+ * DOC: batch buffer command parser
  *
  * Motivation:
  * Certain OpenGL features (e.g. transform feedback, performance monitoring)
@@ -86,6 +86,367 @@
  * general bitmasking mechanism.
  */
 
+#define STD_MI_OPCODE_MASK  0xFF800000
+#define STD_3D_OPCODE_MASK  0xFFFF0000
+#define STD_2D_OPCODE_MASK  0xFFC00000
+#define STD_MFX_OPCODE_MASK 0xFFFF0000
+
+#define CMD(op, opm, f, lm, fl, ...)				\
+	{							\
+		.flags = (fl) | ((f) ? CMD_DESC_FIXED : 0),	\
+		.cmd = { (op), (opm) }, 			\
+		.length = { (lm) },				\
+		__VA_ARGS__					\
+	}
+
+/* Convenience macros to compress the tables */
+#define SMI STD_MI_OPCODE_MASK
+#define S3D STD_3D_OPCODE_MASK
+#define S2D STD_2D_OPCODE_MASK
+#define SMFX STD_MFX_OPCODE_MASK
+#define F true
+#define S CMD_DESC_SKIP
+#define R CMD_DESC_REJECT
+#define W CMD_DESC_REGISTER
+#define B CMD_DESC_BITMASK
+#define M CMD_DESC_MASTER
+
+/*            Command                          Mask   Fixed Len   Action
+	      ---------------------------------------------------------- */
+static const struct drm_i915_cmd_descriptor common_cmds[] = {
+	CMD(  MI_NOOP,                          SMI,    F,  1,      S  ),
+	CMD(  MI_USER_INTERRUPT,                SMI,    F,  1,      R  ),
+	CMD(  MI_WAIT_FOR_EVENT,                SMI,    F,  1,      M  ),
+	CMD(  MI_ARB_CHECK,                     SMI,    F,  1,      S  ),
+	CMD(  MI_REPORT_HEAD,                   SMI,    F,  1,      S  ),
+	CMD(  MI_SUSPEND_FLUSH,                 SMI,    F,  1,      S  ),
+	CMD(  MI_SEMAPHORE_MBOX,                SMI,   !F,  0xFF,   R  ),
+	CMD(  MI_STORE_DWORD_INDEX,             SMI,   !F,  0xFF,   R  ),
+	CMD(  MI_LOAD_REGISTER_IMM(1),          SMI,   !F,  0xFF,   W,
+	      .reg = { .offset = 1, .mask = 0x007FFFFC }               ),
+	CMD(  MI_STORE_REGISTER_MEM(1),         SMI,   !F,  0xFF,   W | B,
+	      .reg = { .offset = 1, .mask = 0x007FFFFC },
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_GLOBAL_GTT,
+			.expected = 0,
+	      }},						       ),
+	CMD(  MI_LOAD_REGISTER_MEM,             SMI,   !F,  0xFF,   W | B,
+	      .reg = { .offset = 1, .mask = 0x007FFFFC },
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_GLOBAL_GTT,
+			.expected = 0,
+	      }},						       ),
+	CMD(  MI_BATCH_BUFFER_START,            SMI,   !F,  0xFF,   S  ),
+};
+
+static const struct drm_i915_cmd_descriptor render_cmds[] = {
+	CMD(  MI_FLUSH,                         SMI,    F,  1,      S  ),
+	CMD(  MI_ARB_ON_OFF,                    SMI,    F,  1,      R  ),
+	CMD(  MI_PREDICATE,                     SMI,    F,  1,      S  ),
+	CMD(  MI_TOPOLOGY_FILTER,               SMI,    F,  1,      S  ),
+	CMD(  MI_DISPLAY_FLIP,                  SMI,   !F,  0xFF,   R  ),
+	CMD(  MI_SET_CONTEXT,                   SMI,   !F,  0xFF,   R  ),
+	CMD(  MI_URB_CLEAR,                     SMI,   !F,  0xFF,   S  ),
+	CMD(  MI_STORE_DWORD_IMM,               SMI,   !F,  0x3F,   B,
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_GLOBAL_GTT,
+			.expected = 0,
+	      }},						       ),
+	CMD(  MI_UPDATE_GTT,                    SMI,   !F,  0xFF,   R  ),
+	CMD(  MI_CLFLUSH,                       SMI,   !F,  0x3FF,  B,
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_GLOBAL_GTT,
+			.expected = 0,
+	      }},						       ),
+	CMD(  MI_REPORT_PERF_COUNT,             SMI,   !F,  0x3F,   B,
+	      .bits = {{
+			.offset = 1,
+			.mask = MI_REPORT_PERF_COUNT_GGTT,
+			.expected = 0,
+	      }},						       ),
+	CMD(  MI_CONDITIONAL_BATCH_BUFFER_END,  SMI,   !F,  0xFF,   B,
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_GLOBAL_GTT,
+			.expected = 0,
+	      }},						       ),
+	CMD(  GFX_OP_3DSTATE_VF_STATISTICS,     S3D,    F,  1,      S  ),
+	CMD(  PIPELINE_SELECT,                  S3D,    F,  1,      S  ),
+	CMD(  MEDIA_VFE_STATE,			S3D,   !F,  0xFFFF, B,
+	      .bits = {{
+			.offset = 2,
+			.mask = MEDIA_VFE_STATE_MMIO_ACCESS_MASK,
+			.expected = 0,
+	      }},						       ),
+	CMD(  GPGPU_OBJECT,                     S3D,   !F,  0xFF,   S  ),
+	CMD(  GPGPU_WALKER,                     S3D,   !F,  0xFF,   S  ),
+	CMD(  GFX_OP_3DSTATE_SO_DECL_LIST,      S3D,   !F,  0x1FF,  S  ),
+	CMD(  GFX_OP_PIPE_CONTROL(5),           S3D,   !F,  0xFF,   B,
+	      .bits = {{
+			.offset = 1,
+			.mask = (PIPE_CONTROL_MMIO_WRITE | PIPE_CONTROL_NOTIFY),
+			.expected = 0,
+	      },
+	      {
+			.offset = 1,
+		        .mask = (PIPE_CONTROL_GLOBAL_GTT_IVB |
+				 PIPE_CONTROL_STORE_DATA_INDEX),
+			.expected = 0,
+			.condition_offset = 1,
+			.condition_mask = PIPE_CONTROL_POST_SYNC_OP_MASK,
+	      }},						       ),
+};
+
+static const struct drm_i915_cmd_descriptor hsw_render_cmds[] = {
+	CMD(  MI_SET_PREDICATE,                 SMI,    F,  1,      S  ),
+	CMD(  MI_RS_CONTROL,                    SMI,    F,  1,      S  ),
+	CMD(  MI_URB_ATOMIC_ALLOC,              SMI,    F,  1,      S  ),
+	CMD(  MI_RS_CONTEXT,                    SMI,    F,  1,      S  ),
+	CMD(  MI_LOAD_SCAN_LINES_INCL,          SMI,   !F,  0x3F,   M  ),
+	CMD(  MI_LOAD_SCAN_LINES_EXCL,          SMI,   !F,  0x3F,   R  ),
+	CMD(  MI_LOAD_REGISTER_REG,             SMI,   !F,  0xFF,   R  ),
+	CMD(  MI_RS_STORE_DATA_IMM,             SMI,   !F,  0xFF,   S  ),
+	CMD(  MI_LOAD_URB_MEM,                  SMI,   !F,  0xFF,   S  ),
+	CMD(  MI_STORE_URB_MEM,                 SMI,   !F,  0xFF,   S  ),
+	CMD(  GFX_OP_3DSTATE_DX9_CONSTANTF_VS,  S3D,   !F,  0x7FF,  S  ),
+	CMD(  GFX_OP_3DSTATE_DX9_CONSTANTF_PS,  S3D,   !F,  0x7FF,  S  ),
+
+	CMD(  GFX_OP_3DSTATE_BINDING_TABLE_EDIT_VS,  S3D,   !F,  0x1FF,  S  ),
+	CMD(  GFX_OP_3DSTATE_BINDING_TABLE_EDIT_GS,  S3D,   !F,  0x1FF,  S  ),
+	CMD(  GFX_OP_3DSTATE_BINDING_TABLE_EDIT_HS,  S3D,   !F,  0x1FF,  S  ),
+	CMD(  GFX_OP_3DSTATE_BINDING_TABLE_EDIT_DS,  S3D,   !F,  0x1FF,  S  ),
+	CMD(  GFX_OP_3DSTATE_BINDING_TABLE_EDIT_PS,  S3D,   !F,  0x1FF,  S  ),
+};
+
+static const struct drm_i915_cmd_descriptor video_cmds[] = {
+	CMD(  MI_ARB_ON_OFF,                    SMI,    F,  1,      R  ),
+	CMD(  MI_STORE_DWORD_IMM,               SMI,   !F,  0xFF,   B,
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_GLOBAL_GTT,
+			.expected = 0,
+	      }},						       ),
+	CMD(  MI_UPDATE_GTT,                    SMI,   !F,  0x3F,   R  ),
+	CMD(  MI_FLUSH_DW,                      SMI,   !F,  0x3F,   B,
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_FLUSH_DW_NOTIFY,
+			.expected = 0,
+	      },
+	      {
+			.offset = 1,
+			.mask = MI_FLUSH_DW_USE_GTT,
+			.expected = 0,
+			.condition_offset = 0,
+			.condition_mask = MI_FLUSH_DW_OP_MASK,
+	      },
+	      {
+			.offset = 0,
+			.mask = MI_FLUSH_DW_STORE_INDEX,
+			.expected = 0,
+			.condition_offset = 0,
+			.condition_mask = MI_FLUSH_DW_OP_MASK,
+	      }},						       ),
+	CMD(  MI_CONDITIONAL_BATCH_BUFFER_END,  SMI,   !F,  0xFF,   B,
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_GLOBAL_GTT,
+			.expected = 0,
+	      }},						       ),
+	/*
+	 * MFX_WAIT doesn't fit the way we handle length for most commands.
+	 * It has a length field but it uses a non-standard length bias.
+	 * It is always 1 dword though, so just treat it as fixed length.
+	 */
+	CMD(  MFX_WAIT,                         SMFX,   F,  1,      S  ),
+};
+
+static const struct drm_i915_cmd_descriptor vecs_cmds[] = {
+	CMD(  MI_ARB_ON_OFF,                    SMI,    F,  1,      R  ),
+	CMD(  MI_STORE_DWORD_IMM,               SMI,   !F,  0xFF,   B,
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_GLOBAL_GTT,
+			.expected = 0,
+	      }},						       ),
+	CMD(  MI_UPDATE_GTT,                    SMI,   !F,  0x3F,   R  ),
+	CMD(  MI_FLUSH_DW,                      SMI,   !F,  0x3F,   B,
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_FLUSH_DW_NOTIFY,
+			.expected = 0,
+	      },
+	      {
+			.offset = 1,
+			.mask = MI_FLUSH_DW_USE_GTT,
+			.expected = 0,
+			.condition_offset = 0,
+			.condition_mask = MI_FLUSH_DW_OP_MASK,
+	      },
+	      {
+			.offset = 0,
+			.mask = MI_FLUSH_DW_STORE_INDEX,
+			.expected = 0,
+			.condition_offset = 0,
+			.condition_mask = MI_FLUSH_DW_OP_MASK,
+	      }},						       ),
+	CMD(  MI_CONDITIONAL_BATCH_BUFFER_END,  SMI,   !F,  0xFF,   B,
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_GLOBAL_GTT,
+			.expected = 0,
+	      }},						       ),
+};
+
+static const struct drm_i915_cmd_descriptor blt_cmds[] = {
+	CMD(  MI_DISPLAY_FLIP,                  SMI,   !F,  0xFF,   R  ),
+	CMD(  MI_STORE_DWORD_IMM,               SMI,   !F,  0x3FF,  B,
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_GLOBAL_GTT,
+			.expected = 0,
+	      }},						       ),
+	CMD(  MI_UPDATE_GTT,                    SMI,   !F,  0x3F,   R  ),
+	CMD(  MI_FLUSH_DW,                      SMI,   !F,  0x3F,   B,
+	      .bits = {{
+			.offset = 0,
+			.mask = MI_FLUSH_DW_NOTIFY,
+			.expected = 0,
+	      },
+	      {
+			.offset = 1,
+			.mask = MI_FLUSH_DW_USE_GTT,
+			.expected = 0,
+			.condition_offset = 0,
+			.condition_mask = MI_FLUSH_DW_OP_MASK,
+	      },
+	      {
+			.offset = 0,
+			.mask = MI_FLUSH_DW_STORE_INDEX,
+			.expected = 0,
+			.condition_offset = 0,
+			.condition_mask = MI_FLUSH_DW_OP_MASK,
+	      }},						       ),
+	CMD(  COLOR_BLT,                        S2D,   !F,  0x3F,   S  ),
+	CMD(  SRC_COPY_BLT,                     S2D,   !F,  0x3F,   S  ),
+};
+
+static const struct drm_i915_cmd_descriptor hsw_blt_cmds[] = {
+	CMD(  MI_LOAD_SCAN_LINES_INCL,          SMI,   !F,  0x3F,   M  ),
+	CMD(  MI_LOAD_SCAN_LINES_EXCL,          SMI,   !F,  0x3F,   R  ),
+};
+
+#undef CMD
+#undef SMI
+#undef S3D
+#undef S2D
+#undef SMFX
+#undef F
+#undef S
+#undef R
+#undef W
+#undef B
+#undef M
+
+static const struct drm_i915_cmd_table gen7_render_cmds[] = {
+	{ common_cmds, ARRAY_SIZE(common_cmds) },
+	{ render_cmds, ARRAY_SIZE(render_cmds) },
+};
+
+static const struct drm_i915_cmd_table hsw_render_ring_cmds[] = {
+	{ common_cmds, ARRAY_SIZE(common_cmds) },
+	{ render_cmds, ARRAY_SIZE(render_cmds) },
+	{ hsw_render_cmds, ARRAY_SIZE(hsw_render_cmds) },
+};
+
+static const struct drm_i915_cmd_table gen7_video_cmds[] = {
+	{ common_cmds, ARRAY_SIZE(common_cmds) },
+	{ video_cmds, ARRAY_SIZE(video_cmds) },
+};
+
+static const struct drm_i915_cmd_table hsw_vebox_cmds[] = {
+	{ common_cmds, ARRAY_SIZE(common_cmds) },
+	{ vecs_cmds, ARRAY_SIZE(vecs_cmds) },
+};
+
+static const struct drm_i915_cmd_table gen7_blt_cmds[] = {
+	{ common_cmds, ARRAY_SIZE(common_cmds) },
+	{ blt_cmds, ARRAY_SIZE(blt_cmds) },
+};
+
+static const struct drm_i915_cmd_table hsw_blt_ring_cmds[] = {
+	{ common_cmds, ARRAY_SIZE(common_cmds) },
+	{ blt_cmds, ARRAY_SIZE(blt_cmds) },
+	{ hsw_blt_cmds, ARRAY_SIZE(hsw_blt_cmds) },
+};
+
+/*
+ * Register whitelists, sorted by increasing register offset.
+ *
+ * Some registers that userspace accesses are 64 bits. The register
+ * access commands only allow 32-bit accesses. Hence, we have to include
+ * entries for both halves of the 64-bit registers.
+ */
+
+/* Convenience macro for adding 64-bit registers */
+#define REG64(addr) (addr), (addr + sizeof(u32))
+
+static const u32 gen7_render_regs[] = {
+	REG64(HS_INVOCATION_COUNT),
+	REG64(DS_INVOCATION_COUNT),
+	REG64(IA_VERTICES_COUNT),
+	REG64(IA_PRIMITIVES_COUNT),
+	REG64(VS_INVOCATION_COUNT),
+	REG64(GS_INVOCATION_COUNT),
+	REG64(GS_PRIMITIVES_COUNT),
+	REG64(CL_INVOCATION_COUNT),
+	REG64(CL_PRIMITIVES_COUNT),
+	REG64(PS_INVOCATION_COUNT),
+	REG64(PS_DEPTH_COUNT),
+	OACONTROL, /* Only allowed for LRI and SRM. See below. */
+	GEN7_3DPRIM_END_OFFSET,
+	GEN7_3DPRIM_START_VERTEX,
+	GEN7_3DPRIM_VERTEX_COUNT,
+	GEN7_3DPRIM_INSTANCE_COUNT,
+	GEN7_3DPRIM_START_INSTANCE,
+	GEN7_3DPRIM_BASE_VERTEX,
+	REG64(GEN7_SO_NUM_PRIMS_WRITTEN(0)),
+	REG64(GEN7_SO_NUM_PRIMS_WRITTEN(1)),
+	REG64(GEN7_SO_NUM_PRIMS_WRITTEN(2)),
+	REG64(GEN7_SO_NUM_PRIMS_WRITTEN(3)),
+	REG64(GEN7_SO_PRIM_STORAGE_NEEDED(0)),
+	REG64(GEN7_SO_PRIM_STORAGE_NEEDED(1)),
+	REG64(GEN7_SO_PRIM_STORAGE_NEEDED(2)),
+	REG64(GEN7_SO_PRIM_STORAGE_NEEDED(3)),
+	GEN7_SO_WRITE_OFFSET(0),
+	GEN7_SO_WRITE_OFFSET(1),
+	GEN7_SO_WRITE_OFFSET(2),
+	GEN7_SO_WRITE_OFFSET(3),
+};
+
+static const u32 gen7_blt_regs[] = {
+	BCS_SWCTRL,
+};
+
+static const u32 ivb_master_regs[] = {
+	FORCEWAKE_MT,
+	DERRMR,
+	GEN7_PIPE_DE_LOAD_SL(PIPE_A),
+	GEN7_PIPE_DE_LOAD_SL(PIPE_B),
+	GEN7_PIPE_DE_LOAD_SL(PIPE_C),
+};
+
+static const u32 hsw_master_regs[] = {
+	FORCEWAKE_MT,
+	DERRMR,
+};
+
+#undef REG64
+
 static u32 gen7_render_get_cmd_length_mask(u32 cmd_header)
 {
 	u32 client = (cmd_header & INSTR_CLIENT_MASK) >> INSTR_CLIENT_SHIFT;
@@ -137,15 +498,18 @@ static u32 gen7_blt_get_cmd_length_mask(u32 cmd_header)
 	return 0;
 }
 
-static void validate_cmds_sorted(struct intel_ring_buffer *ring)
+static bool validate_cmds_sorted(struct intel_engine_cs *ring,
+				 const struct drm_i915_cmd_table *cmd_tables,
+				 int cmd_table_count)
 {
 	int i;
+	bool ret = true;
 
-	if (!ring->cmd_tables || ring->cmd_table_count == 0)
-		return;
+	if (!cmd_tables || cmd_table_count == 0)
+		return true;
 
-	for (i = 0; i < ring->cmd_table_count; i++) {
-		const struct drm_i915_cmd_table *table = &ring->cmd_tables[i];
+	for (i = 0; i < cmd_table_count; i++) {
+		const struct drm_i915_cmd_table *table = &cmd_tables[i];
 		u32 previous = 0;
 		int j;
 
@@ -154,35 +518,107 @@ static void validate_cmds_sorted(struct intel_ring_buffer *ring)
 				&table->table[i];
 			u32 curr = desc->cmd.value & desc->cmd.mask;
 
-			if (curr < previous)
+			if (curr < previous) {
 				DRM_ERROR("CMD: table not sorted ring=%d table=%d entry=%d cmd=0x%08X prev=0x%08X\n",
 					  ring->id, i, j, curr, previous);
+				ret = false;
+			}
 
 			previous = curr;
 		}
 	}
+
+	return ret;
 }
 
-static void check_sorted(int ring_id, const u32 *reg_table, int reg_count)
+static bool check_sorted(int ring_id, const u32 *reg_table, int reg_count)
 {
 	int i;
 	u32 previous = 0;
+	bool ret = true;
 
 	for (i = 0; i < reg_count; i++) {
 		u32 curr = reg_table[i];
 
-		if (curr < previous)
+		if (curr < previous) {
 			DRM_ERROR("CMD: table not sorted ring=%d entry=%d reg=0x%08X prev=0x%08X\n",
 				  ring_id, i, curr, previous);
+			ret = false;
+		}
 
 		previous = curr;
 	}
+
+	return ret;
+}
+
+static bool validate_regs_sorted(struct intel_engine_cs *ring)
+{
+	return check_sorted(ring->id, ring->reg_table, ring->reg_count) &&
+		check_sorted(ring->id, ring->master_reg_table,
+			     ring->master_reg_count);
+}
+
+struct cmd_node {
+	const struct drm_i915_cmd_descriptor *desc;
+	struct hlist_node node;
+};
+
+/*
+ * Different command ranges have different numbers of bits for the opcode. For
+ * example, MI commands use bits 31:23 while 3D commands use bits 31:16. The
+ * problem is that, for example, MI commands use bits 22:16 for other fields
+ * such as GGTT vs PPGTT bits. If we include those bits in the mask then when
+ * we mask a command from a batch it could hash to the wrong bucket due to
+ * non-opcode bits being set. But if we don't include those bits, some 3D
+ * commands may hash to the same bucket due to not including opcode bits that
+ * make the command unique. For now, we will risk hashing to the same bucket.
+ *
+ * If we attempt to generate a perfect hash, we should be able to look at bits
+ * 31:29 of a command from a batch buffer and use the full mask for that
+ * client. The existing INSTR_CLIENT_MASK/SHIFT defines can be used for this.
+ */
+#define CMD_HASH_MASK STD_MI_OPCODE_MASK
+
+static int init_hash_table(struct intel_engine_cs *ring,
+			   const struct drm_i915_cmd_table *cmd_tables,
+			   int cmd_table_count)
+{
+	int i, j;
+
+	hash_init(ring->cmd_hash);
+
+	for (i = 0; i < cmd_table_count; i++) {
+		const struct drm_i915_cmd_table *table = &cmd_tables[i];
+
+		for (j = 0; j < table->count; j++) {
+			const struct drm_i915_cmd_descriptor *desc =
+				&table->table[j];
+			struct cmd_node *desc_node =
+				kmalloc(sizeof(*desc_node), GFP_KERNEL);
+
+			if (!desc_node)
+				return -ENOMEM;
+
+			desc_node->desc = desc;
+			hash_add(ring->cmd_hash, &desc_node->node,
+				 desc->cmd.value & CMD_HASH_MASK);
+		}
+	}
+
+	return 0;
 }
 
-static void validate_regs_sorted(struct intel_ring_buffer *ring)
+static void fini_hash_table(struct intel_engine_cs *ring)
 {
-	check_sorted(ring->id, ring->reg_table, ring->reg_count);
-	check_sorted(ring->id, ring->master_reg_table, ring->master_reg_count);
+	struct hlist_node *tmp;
+	struct cmd_node *desc_node;
+	int i;
+
+	hash_for_each_safe(ring->cmd_hash, i, tmp, desc_node, node) {
+		hash_del(&desc_node->node);
+		kfree(desc_node);
+	}
 }
 
 /**
@@ -190,25 +626,74 @@ static void validate_regs_sorted(struct intel_ring_buffer *ring)
  * @ring: the ringbuffer to initialize
  *
  * Optionally initializes fields related to batch buffer command parsing in the
- * struct intel_ring_buffer based on whether the platform requires software
+ * struct intel_engine_cs based on whether the platform requires software
  * command parsing.
+ *
+ * Return: non-zero if initialization fails
  */
-void i915_cmd_parser_init_ring(struct intel_ring_buffer *ring)
+int i915_cmd_parser_init_ring(struct intel_engine_cs *ring)
 {
+	const struct drm_i915_cmd_table *cmd_tables;
+	int cmd_table_count;
+	int ret;
+
 	if (!IS_GEN7(ring->dev))
-		return;
+		return 0;
 
 	switch (ring->id) {
 	case RCS:
+		if (IS_HASWELL(ring->dev)) {
+			cmd_tables = hsw_render_ring_cmds;
+			cmd_table_count =
+				ARRAY_SIZE(hsw_render_ring_cmds);
+		} else {
+			cmd_tables = gen7_render_cmds;
+			cmd_table_count = ARRAY_SIZE(gen7_render_cmds);
+		}
+
+		ring->reg_table = gen7_render_regs;
+		ring->reg_count = ARRAY_SIZE(gen7_render_regs);
+
+		if (IS_HASWELL(ring->dev)) {
+			ring->master_reg_table = hsw_master_regs;
+			ring->master_reg_count = ARRAY_SIZE(hsw_master_regs);
+		} else {
+			ring->master_reg_table = ivb_master_regs;
+			ring->master_reg_count = ARRAY_SIZE(ivb_master_regs);
+		}
+
 		ring->get_cmd_length_mask = gen7_render_get_cmd_length_mask;
 		break;
 	case VCS:
+		cmd_tables = gen7_video_cmds;
+		cmd_table_count = ARRAY_SIZE(gen7_video_cmds);
 		ring->get_cmd_length_mask = gen7_bsd_get_cmd_length_mask;
 		break;
 	case BCS:
+		if (IS_HASWELL(ring->dev)) {
+			cmd_tables = hsw_blt_ring_cmds;
+			cmd_table_count = ARRAY_SIZE(hsw_blt_ring_cmds);
+		} else {
+			cmd_tables = gen7_blt_cmds;
+			cmd_table_count = ARRAY_SIZE(gen7_blt_cmds);
+		}
+
+		ring->reg_table = gen7_blt_regs;
+		ring->reg_count = ARRAY_SIZE(gen7_blt_regs);
+
+		if (IS_HASWELL(ring->dev)) {
+			ring->master_reg_table = hsw_master_regs;
+			ring->master_reg_count = ARRAY_SIZE(hsw_master_regs);
+		} else {
+			ring->master_reg_table = ivb_master_regs;
+			ring->master_reg_count = ARRAY_SIZE(ivb_master_regs);
+		}
+
 		ring->get_cmd_length_mask = gen7_blt_get_cmd_length_mask;
 		break;
 	case VECS:
+		cmd_tables = hsw_vebox_cmds;
+		cmd_table_count = ARRAY_SIZE(hsw_vebox_cmds);
 		/* VECS can use the same length_mask function as VCS */
 		ring->get_cmd_length_mask = gen7_bsd_get_cmd_length_mask;
 		break;
@@ -218,18 +703,45 @@ void i915_cmd_parser_init_ring(struct intel_ring_buffer *ring)
 		BUG();
 	}
 
-	validate_cmds_sorted(ring);
-	validate_regs_sorted(ring);
+	BUG_ON(!validate_cmds_sorted(ring, cmd_tables, cmd_table_count));
+	BUG_ON(!validate_regs_sorted(ring));
+
+	ret = init_hash_table(ring, cmd_tables, cmd_table_count);
+	if (ret) {
+		DRM_ERROR("CMD: cmd_parser_init failed!\n");
+		fini_hash_table(ring);
+		return ret;
+	}
+
+	ring->needs_cmd_parser = true;
+
+	return 0;
+}
+
+/**
+ * i915_cmd_parser_fini_ring() - clean up cmd parser related fields
+ * @ring: the ringbuffer to clean up
+ *
+ * Releases any resources related to command parsing that may have been
+ * initialized for the specified ring.
+ */
+void i915_cmd_parser_fini_ring(struct intel_engine_cs *ring)
+{
+	if (!ring->needs_cmd_parser)
+		return;
+
+	fini_hash_table(ring);
 }
 
 static const struct drm_i915_cmd_descriptor*
-find_cmd_in_table(const struct drm_i915_cmd_table *table,
+find_cmd_in_table(struct intel_engine_cs *ring,
 		  u32 cmd_header)
 {
-	int i;
+	struct cmd_node *desc_node;
 
-	for (i = 0; i < table->count; i++) {
-		const struct drm_i915_cmd_descriptor *desc = &table->table[i];
+	hash_for_each_possible(ring->cmd_hash, desc_node, node,
+			       cmd_header & CMD_HASH_MASK) {
+		const struct drm_i915_cmd_descriptor *desc = desc_node->desc;
 		u32 masked_cmd = desc->cmd.mask & cmd_header;
 		u32 masked_value = desc->cmd.value & desc->cmd.mask;
 
@@ -249,20 +761,16 @@ find_cmd_in_table(const struct drm_i915_cmd_table *table,
  * ring's default length encoding and returns default_desc.
  */
 static const struct drm_i915_cmd_descriptor*
-find_cmd(struct intel_ring_buffer *ring,
+find_cmd(struct intel_engine_cs *ring,
 	 u32 cmd_header,
 	 struct drm_i915_cmd_descriptor *default_desc)
 {
+	const struct drm_i915_cmd_descriptor *desc;
 	u32 mask;
-	int i;
 
-	for (i = 0; i < ring->cmd_table_count; i++) {
-		const struct drm_i915_cmd_descriptor *desc;
-
-		desc = find_cmd_in_table(&ring->cmd_tables[i], cmd_header);
-		if (desc)
-			return desc;
-	}
+	desc = find_cmd_in_table(ring, cmd_header);
+	if (desc)
+		return desc;
 
 	mask = ring->get_cmd_length_mask(cmd_header);
 	if (!mask)
@@ -329,15 +837,112 @@ finish:
  *
  * Return: true if the ring requires software command parsing
  */
-bool i915_needs_cmd_parser(struct intel_ring_buffer *ring)
+bool i915_needs_cmd_parser(struct intel_engine_cs *ring)
 {
-	/* No command tables indicates a platform without parsing */
-	if (!ring->cmd_tables)
+	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+
+	if (!ring->needs_cmd_parser)
+		return false;
+
+	/*
+	 * XXX: VLV is Gen7 and therefore has cmd_tables, but has PPGTT
+	 * disabled. That will cause all of the parser's PPGTT checks to
+	 * fail. For now, disable parsing when PPGTT is off.
+	 */
+	if (!dev_priv->mm.aliasing_ppgtt)
 		return false;
 
 	return (i915.enable_cmd_parser == 1);
 }
 
+static bool check_cmd(const struct intel_engine_cs *ring,
+		      const struct drm_i915_cmd_descriptor *desc,
+		      const u32 *cmd,
+		      const bool is_master,
+		      bool *oacontrol_set)
+{
+	if (desc->flags & CMD_DESC_REJECT) {
+		DRM_DEBUG_DRIVER("CMD: Rejected command: 0x%08X\n", *cmd);
+		return false;
+	}
+
+	if ((desc->flags & CMD_DESC_MASTER) && !is_master) {
+		DRM_DEBUG_DRIVER("CMD: Rejected master-only command: 0x%08X\n",
+				 *cmd);
+		return false;
+	}
+
+	if (desc->flags & CMD_DESC_REGISTER) {
+		u32 reg_addr = cmd[desc->reg.offset] & desc->reg.mask;
+
+		/*
+		 * OACONTROL requires some special handling for writes. We
+		 * want to make sure that any batch which enables OA also
+		 * disables it before the end of the batch. The goal is to
+		 * prevent one process from snooping on the perf data from
+		 * another process. To do that, we need to check the value
+		 * that will be written to the register. Hence, limit
+		 * OACONTROL writes to only MI_LOAD_REGISTER_IMM commands.
+		 */
+		if (reg_addr == OACONTROL) {
+			if (desc->cmd.value == MI_LOAD_REGISTER_MEM)
+				return false;
+
+			if (desc->cmd.value == MI_LOAD_REGISTER_IMM(1))
+				*oacontrol_set = (cmd[2] != 0);
+		}
+
+		if (!valid_reg(ring->reg_table,
+			       ring->reg_count, reg_addr)) {
+			if (!is_master ||
+			    !valid_reg(ring->master_reg_table,
+				       ring->master_reg_count,
+				       reg_addr)) {
+				DRM_DEBUG_DRIVER("CMD: Rejected register 0x%08X in command: 0x%08X (ring=%d)\n",
+						 reg_addr,
+						 *cmd,
+						 ring->id);
+				return false;
+			}
+		}
+	}
+
+	if (desc->flags & CMD_DESC_BITMASK) {
+		int i;
+
+		for (i = 0; i < MAX_CMD_DESC_BITMASKS; i++) {
+			u32 dword;
+
+			if (desc->bits[i].mask == 0)
+				break;
+
+			if (desc->bits[i].condition_mask != 0) {
+				u32 offset =
+					desc->bits[i].condition_offset;
+				u32 condition = cmd[offset] &
+					desc->bits[i].condition_mask;
+
+				if (condition == 0)
+					continue;
+			}
+
+			dword = cmd[desc->bits[i].offset] &
+				desc->bits[i].mask;
+
+			if (dword != desc->bits[i].expected) {
+				DRM_DEBUG_DRIVER("CMD: Rejected command 0x%08X for bitmask 0x%08X (exp=0x%08X act=0x%08X) (ring=%d)\n",
+						 *cmd,
+						 desc->bits[i].mask,
+						 desc->bits[i].expected,
+						 dword, ring->id);
+				return false;
+			}
+		}
+	}
+
+	return true;
+}
+
 #define LENGTH_BIAS 2
 
 /**
@@ -352,7 +957,7 @@ bool i915_needs_cmd_parser(struct intel_ring_buffer *ring)
  *
  * Return: non-zero if the parser finds violations or otherwise fails
  */
-int i915_parse_cmds(struct intel_ring_buffer *ring,
+int i915_parse_cmds(struct intel_engine_cs *ring,
 		    struct drm_i915_gem_object *batch_obj,
 		    u32 batch_start_offset,
 		    bool is_master)
@@ -361,6 +966,7 @@ int i915_parse_cmds(struct intel_ring_buffer *ring,
 	u32 *cmd, *batch_base, *batch_end;
 	struct drm_i915_cmd_descriptor default_desc = { 0 };
 	int needs_clflush = 0;
+	bool oacontrol_set = false; /* OACONTROL tracking. See check_cmd() */
 
 	ret = i915_gem_obj_prepare_shmem_read(batch_obj, &needs_clflush);
 	if (ret) {
@@ -402,76 +1008,27 @@ int i915_parse_cmds(struct intel_ring_buffer *ring,
 			length = ((*cmd & desc->length.mask) + LENGTH_BIAS);
 
 		if ((batch_end - cmd) < length) {
-			DRM_DEBUG_DRIVER("CMD: Command length exceeds batch length: 0x%08X length=%d batchlen=%td\n",
+			DRM_DEBUG_DRIVER("CMD: Command length exceeds batch length: 0x%08X length=%u batchlen=%td\n",
 					 *cmd,
 					 length,
-					 (unsigned long)(batch_end - cmd));
+					 batch_end - cmd);
 			ret = -EINVAL;
 			break;
 		}
 
-		if (desc->flags & CMD_DESC_REJECT) {
-			DRM_DEBUG_DRIVER("CMD: Rejected command: 0x%08X\n", *cmd);
+		if (!check_cmd(ring, desc, cmd, is_master, &oacontrol_set)) {
 			ret = -EINVAL;
 			break;
 		}
 
-		if ((desc->flags & CMD_DESC_MASTER) && !is_master) {
-			DRM_DEBUG_DRIVER("CMD: Rejected master-only command: 0x%08X\n",
-					 *cmd);
-			ret = -EINVAL;
-			break;
-		}
-
-		if (desc->flags & CMD_DESC_REGISTER) {
-			u32 reg_addr = cmd[desc->reg.offset] & desc->reg.mask;
-
-			if (!valid_reg(ring->reg_table,
-				       ring->reg_count, reg_addr)) {
-				if (!is_master ||
-				    !valid_reg(ring->master_reg_table,
-					       ring->master_reg_count,
-					       reg_addr)) {
-					DRM_DEBUG_DRIVER("CMD: Rejected register 0x%08X in command: 0x%08X (ring=%d)\n",
-							 reg_addr,
-							 *cmd,
-							 ring->id);
-					ret = -EINVAL;
-					break;
-				}
-			}
-		}
-
-		if (desc->flags & CMD_DESC_BITMASK) {
-			int i;
-
-			for (i = 0; i < MAX_CMD_DESC_BITMASKS; i++) {
-				u32 dword;
-
-				if (desc->bits[i].mask == 0)
-					break;
-
-				dword = cmd[desc->bits[i].offset] &
-					desc->bits[i].mask;
-
-				if (dword != desc->bits[i].expected) {
-					DRM_DEBUG_DRIVER("CMD: Rejected command 0x%08X for bitmask 0x%08X (exp=0x%08X act=0x%08X) (ring=%d)\n",
-							 *cmd,
-							 desc->bits[i].mask,
-							 desc->bits[i].expected,
-							 dword, ring->id);
-					ret = -EINVAL;
-					break;
-				}
-			}
-
-			if (ret)
-				break;
-		}
-
 		cmd += length;
 	}
 
+	if (oacontrol_set) {
+		DRM_DEBUG_DRIVER("CMD: batch set OACONTROL but did not clear it\n");
+		ret = -EINVAL;
+	}
+
 	if (cmd >= batch_end) {
 		DRM_DEBUG_DRIVER("CMD: Got to the end of the buffer w/o a BBE cmd!\n");
 		ret = -EINVAL;
@@ -483,3 +1040,22 @@ int i915_parse_cmds(struct intel_ring_buffer *ring,
 
 	return ret;
 }
+
+/**
+ * i915_cmd_parser_get_version() - get the cmd parser version number
+ *
+ * The cmd parser maintains a simple increasing integer version number suitable
+ * for passing to userspace clients to determine what operations are permitted.
+ *
+ * Return: the current version number of the cmd parser
+ */
+int i915_cmd_parser_get_version(void)
+{
+	/*
+	 * Command parser version history
+	 *
+	 * 1. Initial version. Checks batches and reports violations, but leaves
+	 *    hardware parsing enabled (so does not allow new use cases).
+	 */
+	return 1;
+}
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 195fe5bc0aac..601caa88c092 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -79,7 +79,7 @@ drm_add_fake_info_node(struct drm_minor *minor,
 
 static int i915_capabilities(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	const struct intel_device_info *info = INTEL_INFO(dev);
 
@@ -172,7 +172,7 @@ describe_obj(struct seq_file *m, struct drm_i915_gem_object *obj)
 		seq_printf(m, " (%s)", obj->ring->name);
 }
 
-static void describe_ctx(struct seq_file *m, struct i915_hw_context *ctx)
+static void describe_ctx(struct seq_file *m, struct intel_context *ctx)
 {
 	seq_putc(m, ctx->is_initialized ? 'I' : 'i');
 	seq_putc(m, ctx->remap_slice ? 'R' : 'r');
@@ -181,7 +181,7 @@ static void describe_ctx(struct seq_file *m, struct i915_hw_context *ctx)
 
 static int i915_gem_object_list_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	uintptr_t list = (uintptr_t) node->info_ent->data;
 	struct list_head *head;
 	struct drm_device *dev = node->minor->dev;
@@ -239,7 +239,7 @@ static int obj_rank_by_stolen(void *priv,
 
 static int i915_gem_stolen_list_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_object *obj;
@@ -371,7 +371,7 @@ static int per_file_stats(int id, void *ptr, void *data)
 
 static int i915_gem_object_info(struct seq_file *m, void* data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 count, mappable_count, purgeable_count;
@@ -474,7 +474,7 @@ static int i915_gem_object_info(struct seq_file *m, void* data)
 
 static int i915_gem_gtt_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	uintptr_t list = (uintptr_t) node->info_ent->data;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -509,12 +509,12 @@ static int i915_gem_gtt_info(struct seq_file *m, void *data)
 
 static int i915_gem_pageflip_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	unsigned long flags;
 	struct intel_crtc *crtc;
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		const char pipe = pipe_name(crtc->pipe);
 		const char plane = plane_name(crtc->plane);
 		struct intel_unpin_work *work;
@@ -559,10 +559,10 @@ static int i915_gem_pageflip_info(struct seq_file *m, void *data)
 
 static int i915_gem_request_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	struct drm_i915_gem_request *gem_request;
 	int ret, count, i;
 
@@ -594,7 +594,7 @@ static int i915_gem_request_info(struct seq_file *m, void *data)
 }
 
 static void i915_ring_seqno_info(struct seq_file *m,
-				 struct intel_ring_buffer *ring)
+				 struct intel_engine_cs *ring)
 {
 	if (ring->get_seqno) {
 		seq_printf(m, "Current sequence (%s): %u\n",
@@ -604,10 +604,10 @@ static void i915_ring_seqno_info(struct seq_file *m,
 
 static int i915_gem_seqno_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int ret, i;
 
 	ret = mutex_lock_interruptible(&dev->struct_mutex);
@@ -627,10 +627,10 @@ static int i915_gem_seqno_info(struct seq_file *m, void *data)
 
 static int i915_interrupt_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int ret, i, pipe;
 
 	ret = mutex_lock_interruptible(&dev->struct_mutex);
@@ -638,7 +638,47 @@ static int i915_interrupt_info(struct seq_file *m, void *data)
 		return ret;
 	intel_runtime_pm_get(dev_priv);
 
-	if (INTEL_INFO(dev)->gen >= 8) {
+	if (IS_CHERRYVIEW(dev)) {
+		int i;
+		seq_printf(m, "Master Interrupt Control:\t%08x\n",
+			   I915_READ(GEN8_MASTER_IRQ));
+
+		seq_printf(m, "Display IER:\t%08x\n",
+			   I915_READ(VLV_IER));
+		seq_printf(m, "Display IIR:\t%08x\n",
+			   I915_READ(VLV_IIR));
+		seq_printf(m, "Display IIR_RW:\t%08x\n",
+			   I915_READ(VLV_IIR_RW));
+		seq_printf(m, "Display IMR:\t%08x\n",
+			   I915_READ(VLV_IMR));
+		for_each_pipe(pipe)
+			seq_printf(m, "Pipe %c stat:\t%08x\n",
+				   pipe_name(pipe),
+				   I915_READ(PIPESTAT(pipe)));
+
+		seq_printf(m, "Port hotplug:\t%08x\n",
+			   I915_READ(PORT_HOTPLUG_EN));
+		seq_printf(m, "DPFLIPSTAT:\t%08x\n",
+			   I915_READ(VLV_DPFLIPSTAT));
+		seq_printf(m, "DPINVGTT:\t%08x\n",
+			   I915_READ(DPINVGTT));
+
+		for (i = 0; i < 4; i++) {
+			seq_printf(m, "GT Interrupt IMR %d:\t%08x\n",
+				   i, I915_READ(GEN8_GT_IMR(i)));
+			seq_printf(m, "GT Interrupt IIR %d:\t%08x\n",
+				   i, I915_READ(GEN8_GT_IIR(i)));
+			seq_printf(m, "GT Interrupt IER %d:\t%08x\n",
+				   i, I915_READ(GEN8_GT_IER(i)));
+		}
+
+		seq_printf(m, "PCU interrupt mask:\t%08x\n",
+			   I915_READ(GEN8_PCU_IMR));
+		seq_printf(m, "PCU interrupt identity:\t%08x\n",
+			   I915_READ(GEN8_PCU_IIR));
+		seq_printf(m, "PCU interrupt enable:\t%08x\n",
+			   I915_READ(GEN8_PCU_IER));
+	} else if (INTEL_INFO(dev)->gen >= 8) {
 		seq_printf(m, "Master Interrupt Control:\t%08x\n",
 			   I915_READ(GEN8_MASTER_IRQ));
 
@@ -768,7 +808,7 @@ static int i915_interrupt_info(struct seq_file *m, void *data)
 
 static int i915_gem_fence_regs_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int i, ret;
@@ -797,10 +837,10 @@ static int i915_gem_fence_regs_info(struct seq_file *m, void *data)
 
 static int i915_hws_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	const u32 *hws;
 	int i;
 
@@ -945,7 +985,7 @@ DEFINE_SIMPLE_ATTRIBUTE(i915_next_seqno_fops,
 
 static int i915_rstdby_delays(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u16 crstanddelay;
@@ -966,9 +1006,9 @@ static int i915_rstdby_delays(struct seq_file *m, void *unused)
 	return 0;
 }
 
-static int i915_cur_delayinfo(struct seq_file *m, void *unused)
+static int i915_frequency_info(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret = 0;
@@ -991,6 +1031,7 @@ static int i915_cur_delayinfo(struct seq_file *m, void *unused)
 		u32 gt_perf_status = I915_READ(GEN6_GT_PERF_STATUS);
 		u32 rp_state_limits = I915_READ(GEN6_RP_STATE_LIMITS);
 		u32 rp_state_cap = I915_READ(GEN6_RP_STATE_CAP);
+		u32 rpmodectl, rpinclimit, rpdeclimit;
 		u32 rpstat, cagf, reqf;
 		u32 rpupei, rpcurup, rpprevup;
 		u32 rpdownei, rpcurdown, rpprevdown;
@@ -1011,6 +1052,10 @@ static int i915_cur_delayinfo(struct seq_file *m, void *unused)
 			reqf >>= 25;
 		reqf *= GT_FREQUENCY_MULTIPLIER;
 
+		rpmodectl = I915_READ(GEN6_RP_CONTROL);
+		rpinclimit = I915_READ(GEN6_RP_UP_THRESHOLD);
+		rpdeclimit = I915_READ(GEN6_RP_DOWN_THRESHOLD);
+
 		rpstat = I915_READ(GEN6_RPSTAT1);
 		rpupei = I915_READ(GEN6_RP_CUR_UP_EI);
 		rpcurup = I915_READ(GEN6_RP_CUR_UP);
@@ -1027,14 +1072,23 @@ static int i915_cur_delayinfo(struct seq_file *m, void *unused)
 		gen6_gt_force_wake_put(dev_priv, FORCEWAKE_ALL);
 		mutex_unlock(&dev->struct_mutex);
 
+		seq_printf(m, "PM IER=0x%08x IMR=0x%08x ISR=0x%08x IIR=0x%08x, MASK=0x%08x\n",
+			   I915_READ(GEN6_PMIER),
+			   I915_READ(GEN6_PMIMR),
+			   I915_READ(GEN6_PMISR),
+			   I915_READ(GEN6_PMIIR),
+			   I915_READ(GEN6_PMINTRMSK));
 		seq_printf(m, "GT_PERF_STATUS: 0x%08x\n", gt_perf_status);
-		seq_printf(m, "RPSTAT1: 0x%08x\n", rpstat);
 		seq_printf(m, "Render p-state ratio: %d\n",
 			   (gt_perf_status & 0xff00) >> 8);
 		seq_printf(m, "Render p-state VID: %d\n",
 			   gt_perf_status & 0xff);
 		seq_printf(m, "Render p-state limit: %d\n",
 			   rp_state_limits & 0xff);
+		seq_printf(m, "RPSTAT1: 0x%08x\n", rpstat);
+		seq_printf(m, "RPMODECTL: 0x%08x\n", rpmodectl);
+		seq_printf(m, "RPINCLIMIT: 0x%08x\n", rpinclimit);
+		seq_printf(m, "RPDECLIMIT: 0x%08x\n", rpdeclimit);
 		seq_printf(m, "RPNSWREQ: %dMHz\n", reqf);
 		seq_printf(m, "CAGF: %dMHz\n", cagf);
 		seq_printf(m, "RP CUR UP EI: %dus\n", rpupei &
@@ -1094,7 +1148,7 @@ out:
 
 static int i915_delayfreq_table(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 delayfreq;
@@ -1125,7 +1179,7 @@ static inline int MAP_TO_MV(int map)
 
 static int i915_inttoext_table(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 inttoext;
@@ -1149,7 +1203,7 @@ static int i915_inttoext_table(struct seq_file *m, void *unused)
 
 static int ironlake_drpc_info(struct seq_file *m)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 rgvmodectl, rstdbyctl;
@@ -1219,15 +1273,19 @@ static int ironlake_drpc_info(struct seq_file *m)
 static int vlv_drpc_info(struct seq_file *m)
 {
 
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 rpmodectl1, rcctl1;
 	unsigned fw_rendercount = 0, fw_mediacount = 0;
 
+	intel_runtime_pm_get(dev_priv);
+
 	rpmodectl1 = I915_READ(GEN6_RP_CONTROL);
 	rcctl1 = I915_READ(GEN6_RC_CONTROL);
 
+	intel_runtime_pm_put(dev_priv);
+
 	seq_printf(m, "Video Turbo Mode: %s\n",
 		   yesno(rpmodectl1 & GEN6_RP_MEDIA_TURBO));
 	seq_printf(m, "Turbo enabled: %s\n",
@@ -1247,6 +1305,11 @@ static int vlv_drpc_info(struct seq_file *m)
 			(I915_READ(VLV_GTLC_PW_STATUS) &
 				VLV_GTLC_PW_MEDIA_STATUS_MASK) ? "Up" : "Down");
 
+	seq_printf(m, "Render RC6 residency since boot: %u\n",
+		   I915_READ(VLV_GT_RENDER_RC6));
+	seq_printf(m, "Media RC6 residency since boot: %u\n",
+		   I915_READ(VLV_GT_MEDIA_RC6));
+
 	spin_lock_irq(&dev_priv->uncore.lock);
 	fw_rendercount = dev_priv->uncore.fw_rendercount;
 	fw_mediacount = dev_priv->uncore.fw_mediacount;
@@ -1263,7 +1326,7 @@ static int vlv_drpc_info(struct seq_file *m)
 static int gen6_drpc_info(struct seq_file *m)
 {
 
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 rpmodectl1, gt_core_status, rcctl1, rc6vids = 0;
@@ -1362,7 +1425,7 @@ static int gen6_drpc_info(struct seq_file *m)
 
 static int i915_drpc_info(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 
 	if (IS_VALLEYVIEW(dev))
@@ -1375,7 +1438,7 @@ static int i915_drpc_info(struct seq_file *m, void *unused)
 
 static int i915_fbc_status(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -1437,7 +1500,7 @@ static int i915_fbc_status(struct seq_file *m, void *unused)
 
 static int i915_ips_status(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -1460,7 +1523,7 @@ static int i915_ips_status(struct seq_file *m, void *unused)
 
 static int i915_sr_status(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	bool sr_enabled = false;
@@ -1486,7 +1549,7 @@ static int i915_sr_status(struct seq_file *m, void *unused)
 
 static int i915_emon_status(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long temp, chipset, gfx;
@@ -1514,7 +1577,7 @@ static int i915_emon_status(struct seq_file *m, void *unused)
 
 static int i915_ring_freq_table(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret = 0;
@@ -1557,7 +1620,7 @@ out:
 
 static int i915_gfxec(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret;
@@ -1577,7 +1640,7 @@ static int i915_gfxec(struct seq_file *m, void *unused)
 
 static int i915_opregion(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_opregion *opregion = &dev_priv->opregion;
@@ -1605,7 +1668,7 @@ out:
 
 static int i915_gem_framebuffer_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct intel_fbdev *ifbdev = NULL;
 	struct intel_framebuffer *fb;
@@ -1651,11 +1714,11 @@ static int i915_gem_framebuffer_info(struct seq_file *m, void *data)
 
 static int i915_context_status(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
-	struct i915_hw_context *ctx;
+	struct intel_engine_cs *ring;
+	struct intel_context *ctx;
 	int ret, i;
 
 	ret = mutex_lock_interruptible(&dev->mode_config.mutex);
@@ -1675,6 +1738,9 @@ static int i915_context_status(struct seq_file *m, void *unused)
 	}
 
 	list_for_each_entry(ctx, &dev_priv->context_list, link) {
+		if (ctx->obj == NULL)
+			continue;
+
 		seq_puts(m, "HW context ");
 		describe_ctx(m, ctx);
 		for_each_ring(ring, dev_priv, i)
@@ -1692,7 +1758,7 @@ static int i915_context_status(struct seq_file *m, void *unused)
 
 static int i915_gen6_forcewake_count_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned forcewake_count = 0, fw_rendercount = 0, fw_mediacount = 0;
@@ -1740,7 +1806,7 @@ static const char *swizzle_string(unsigned swizzle)
 
 static int i915_swizzle_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret;
@@ -1788,10 +1854,14 @@ static int i915_swizzle_info(struct seq_file *m, void *data)
 
 static int per_file_ctx(int id, void *ptr, void *data)
 {
-	struct i915_hw_context *ctx = ptr;
+	struct intel_context *ctx = ptr;
 	struct seq_file *m = data;
 	struct i915_hw_ppgtt *ppgtt = ctx_to_ppgtt(ctx);
 
+	if (i915_gem_context_is_default(ctx))
+		seq_puts(m, "  default context:\n");
+	else
+		seq_printf(m, "  context %d:\n", ctx->id);
 	ppgtt->debug_dump(ppgtt, m);
 
 	return 0;
@@ -1800,7 +1870,7 @@ static int per_file_ctx(int id, void *ptr, void *data)
 static void gen8_ppgtt_info(struct seq_file *m, struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	struct i915_hw_ppgtt *ppgtt = dev_priv->mm.aliasing_ppgtt;
 	int unused, i;
 
@@ -1816,8 +1886,7 @@ static void gen8_ppgtt_info(struct seq_file *m, struct drm_device *dev)
 			u64 pdp = I915_READ(ring->mmio_base + offset + 4);
 			pdp <<= 32;
 			pdp |= I915_READ(ring->mmio_base + offset);
-			for (i = 0; i < 4; i++)
-				seq_printf(m, "\tPDP%d 0x%016llx\n", i, pdp);
+			seq_printf(m, "\tPDP%d 0x%016llx\n", i, pdp);
 		}
 	}
 }
@@ -1825,7 +1894,7 @@ static void gen8_ppgtt_info(struct seq_file *m, struct drm_device *dev)
 static void gen6_ppgtt_info(struct seq_file *m, struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	struct drm_file *file;
 	int i;
 
@@ -1852,12 +1921,9 @@ static void gen6_ppgtt_info(struct seq_file *m, struct drm_device *dev)
 
 	list_for_each_entry_reverse(file, &dev->filelist, lhead) {
 		struct drm_i915_file_private *file_priv = file->driver_priv;
-		struct i915_hw_ppgtt *pvt_ppgtt;
 
-		pvt_ppgtt = ctx_to_ppgtt(file_priv->private_default_ctx);
 		seq_printf(m, "proc: %s\n",
 			   get_pid_task(file->pid, PIDTYPE_PID)->comm);
-		seq_puts(m, "  default context:\n");
 		idr_for_each(&file_priv->context_idr, per_file_ctx, m);
 	}
 	seq_printf(m, "ECOCHK: 0x%08x\n", I915_READ(GAM_ECOCHK));
@@ -1865,7 +1931,7 @@ static void gen6_ppgtt_info(struct seq_file *m, struct drm_device *dev)
 
 static int i915_ppgtt_info(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -1885,56 +1951,9 @@ static int i915_ppgtt_info(struct seq_file *m, void *data)
 	return 0;
 }
 
-static int i915_dpio_info(struct seq_file *m, void *data)
-{
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
-	struct drm_device *dev = node->minor->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	int ret;
-
-
-	if (!IS_VALLEYVIEW(dev)) {
-		seq_puts(m, "unsupported\n");
-		return 0;
-	}
-
-	ret = mutex_lock_interruptible(&dev_priv->dpio_lock);
-	if (ret)
-		return ret;
-
-	seq_printf(m, "DPIO_CTL: 0x%08x\n", I915_READ(DPIO_CTL));
-
-	seq_printf(m, "DPIO PLL DW3 CH0 : 0x%08x\n",
-		   vlv_dpio_read(dev_priv, PIPE_A, VLV_PLL_DW3(0)));
-	seq_printf(m, "DPIO PLL DW3 CH1: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, PIPE_A, VLV_PLL_DW3(1)));
-
-	seq_printf(m, "DPIO PLL DW5 CH0: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, PIPE_A, VLV_PLL_DW5(0)));
-	seq_printf(m, "DPIO PLL DW5 CH1: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, PIPE_A, VLV_PLL_DW5(1)));
-
-	seq_printf(m, "DPIO PLL DW7 CH0: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, PIPE_A, VLV_PLL_DW7(0)));
-	seq_printf(m, "DPIO PLL DW7 CH1: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, PIPE_A, VLV_PLL_DW7(1)));
-
-	seq_printf(m, "DPIO PLL DW10 CH0: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, PIPE_A, VLV_PLL_DW10(0)));
-	seq_printf(m, "DPIO PLL DW10 CH1: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, PIPE_A, VLV_PLL_DW10(1)));
-
-	seq_printf(m, "DPIO_FASTCLK_DISABLE: 0x%08x\n",
-		   vlv_dpio_read(dev_priv, PIPE_A, VLV_CMN_DW0));
-
-	mutex_unlock(&dev_priv->dpio_lock);
-
-	return 0;
-}
-
 static int i915_llc(struct seq_file *m, void *data)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -2040,11 +2059,11 @@ static int i915_energy_uJ(struct seq_file *m, void *data)
 
 static int i915_pc8_status(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	if (!IS_HASWELL(dev)) {
+	if (!IS_HASWELL(dev) && !IS_BROADWELL(dev)) {
 		seq_puts(m, "not supported\n");
 		return 0;
 	}
@@ -2115,7 +2134,7 @@ static const char *power_domain_str(enum intel_display_power_domain domain)
 
 static int i915_power_domain_info(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct i915_power_domains *power_domains = &dev_priv->power_domains;
@@ -2170,7 +2189,7 @@ static void intel_encoder_info(struct seq_file *m,
 			       struct intel_crtc *intel_crtc,
 			       struct intel_encoder *intel_encoder)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_crtc *crtc = &intel_crtc->base;
 	struct intel_connector *intel_connector;
@@ -2178,12 +2197,12 @@ static void intel_encoder_info(struct seq_file *m,
 
 	encoder = &intel_encoder->base;
 	seq_printf(m, "\tencoder %d: type: %s, connectors:\n",
-		   encoder->base.id, drm_get_encoder_name(encoder));
+		   encoder->base.id, encoder->name);
 	for_each_connector_on_encoder(dev, encoder, intel_connector) {
 		struct drm_connector *connector = &intel_connector->base;
 		seq_printf(m, "\t\tconnector %d: type: %s, status: %s",
 			   connector->base.id,
-			   drm_get_connector_name(connector),
+			   connector->name,
 			   drm_get_connector_status_name(connector->status));
 		if (connector->status == connector_status_connected) {
 			struct drm_display_mode *mode = &crtc->mode;
@@ -2197,7 +2216,7 @@ static void intel_encoder_info(struct seq_file *m,
 
 static void intel_crtc_info(struct seq_file *m, struct intel_crtc *intel_crtc)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_crtc *crtc = &intel_crtc->base;
 	struct intel_encoder *intel_encoder;
@@ -2254,7 +2273,7 @@ static void intel_connector_info(struct seq_file *m,
 	struct drm_display_mode *mode;
 
 	seq_printf(m, "connector %d: type %s, status: %s\n",
-		   connector->base.id, drm_get_connector_name(connector),
+		   connector->base.id, connector->name,
 		   drm_get_connector_status_name(connector->status));
 	if (connector->status == connector_status_connected) {
 		seq_printf(m, "\tname: %s\n", connector->display_info.name);
@@ -2286,10 +2305,8 @@ static bool cursor_active(struct drm_device *dev, int pipe)
 
 	if (IS_845G(dev) || IS_I865G(dev))
 		state = I915_READ(_CURACNTR) & CURSOR_ENABLE;
-	else if (INTEL_INFO(dev)->gen <= 6 || IS_VALLEYVIEW(dev))
-		state = I915_READ(CURCNTR(pipe)) & CURSOR_MODE;
 	else
-		state = I915_READ(CURCNTR_IVB(pipe)) & CURSOR_MODE;
+		state = I915_READ(CURCNTR(pipe)) & CURSOR_MODE;
 
 	return state;
 }
@@ -2299,10 +2316,7 @@ static bool cursor_position(struct drm_device *dev, int pipe, int *x, int *y)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 pos;
 
-	if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev) || IS_BROADWELL(dev))
-		pos = I915_READ(CURPOS_IVB(pipe));
-	else
-		pos = I915_READ(CURPOS(pipe));
+	pos = I915_READ(CURPOS(pipe));
 
 	*x = (pos >> CURSOR_X_SHIFT) & CURSOR_POS_MASK;
 	if (pos & (CURSOR_POS_SIGN << CURSOR_X_SHIFT))
@@ -2317,7 +2331,7 @@ static bool cursor_position(struct drm_device *dev, int pipe, int *x, int *y)
 
 static int i915_display_info(struct seq_file *m, void *unused)
 {
-	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *crtc;
@@ -2327,7 +2341,7 @@ static int i915_display_info(struct seq_file *m, void *unused)
 	drm_modeset_lock_all(dev);
 	seq_printf(m, "CRTC info\n");
 	seq_printf(m, "---------\n");
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		bool active;
 		int x, y;
 
@@ -2339,10 +2353,14 @@ static int i915_display_info(struct seq_file *m, void *unused)
 
 			active = cursor_position(dev, crtc->pipe, &x, &y);
 			seq_printf(m, "\tcursor visible? %s, position (%d, %d), addr 0x%08x, active? %s\n",
-				   yesno(crtc->cursor_visible),
+				   yesno(crtc->cursor_base),
 				   x, y, crtc->cursor_addr,
 				   yesno(active));
 		}
+
+		seq_printf(m, "\tunderrun reporting: cpu=%s pch=%s \n",
+			   yesno(!crtc->cpu_fifo_underrun_disabled),
+			   yesno(!crtc->pch_fifo_underrun_disabled));
 	}
 
 	seq_printf(m, "\n");
@@ -2595,7 +2613,7 @@ static int i9xx_pipe_crc_auto_source(struct drm_device *dev, enum pipe pipe,
 
 	*source = INTEL_PIPE_CRC_SOURCE_PIPE;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	list_for_each_entry(encoder, &dev->mode_config.encoder_list,
 			    base.head) {
 		if (!encoder->base.crtc)
@@ -2631,7 +2649,7 @@ static int i9xx_pipe_crc_auto_source(struct drm_device *dev, enum pipe pipe,
 			break;
 		}
 	}
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 
 	return ret;
 }
@@ -3106,7 +3124,7 @@ static const struct file_operations i915_display_crc_ctl_fops = {
 static void wm_latency_show(struct seq_file *m, const uint16_t wm[5])
 {
 	struct drm_device *dev = m->private;
-	int num_levels = IS_HASWELL(dev) || IS_BROADWELL(dev) ? 5 : 4;
+	int num_levels = ilk_wm_max_level(dev) + 1;
 	int level;
 
 	drm_modeset_lock_all(dev);
@@ -3189,7 +3207,7 @@ static ssize_t wm_latency_write(struct file *file, const char __user *ubuf,
 	struct seq_file *m = file->private_data;
 	struct drm_device *dev = m->private;
 	uint16_t new[5] = { 0 };
-	int num_levels = IS_HASWELL(dev) || IS_BROADWELL(dev) ? 5 : 4;
+	int num_levels = ilk_wm_max_level(dev) + 1;
 	int level;
 	int ret;
 	char tmp[32];
@@ -3286,9 +3304,15 @@ static int
 i915_wedged_set(void *data, u64 val)
 {
 	struct drm_device *dev = data;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	intel_runtime_pm_get(dev_priv);
 
 	i915_handle_error(dev, val,
 			  "Manually setting wedged to %llu", val);
+
+	intel_runtime_pm_put(dev_priv);
+
 	return 0;
 }
 
@@ -3774,7 +3798,7 @@ static const struct drm_info_list i915_debugfs_list[] = {
 	{"i915_gem_hws_bsd", i915_hws_info, 0, (void *)VCS},
 	{"i915_gem_hws_vebox", i915_hws_info, 0, (void *)VECS},
 	{"i915_rstdby_delays", i915_rstdby_delays, 0},
-	{"i915_cur_delayinfo", i915_cur_delayinfo, 0},
+	{"i915_frequency_info", i915_frequency_info, 0},
 	{"i915_delayfreq_table", i915_delayfreq_table, 0},
 	{"i915_inttoext_table", i915_inttoext_table, 0},
 	{"i915_drpc_info", i915_drpc_info, 0},
@@ -3790,7 +3814,6 @@ static const struct drm_info_list i915_debugfs_list[] = {
 	{"i915_gen6_forcewake_count", i915_gen6_forcewake_count_info, 0},
 	{"i915_swizzle_info", i915_swizzle_info, 0},
 	{"i915_ppgtt_info", i915_ppgtt_info, 0},
-	{"i915_dpio", i915_dpio_info, 0},
 	{"i915_llc", i915_llc, 0},
 	{"i915_edp_psr_status", i915_edp_psr_status, 0},
 	{"i915_sink_crc_eDP1", i915_sink_crc, 0},
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index eedb023af27d..4c22a5b7f4c5 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -44,6 +44,7 @@
 #include <acpi/video.h>
 #include <linux/pm.h>
 #include <linux/pm_runtime.h>
+#include <linux/oom.h>
 
 #define LP_RING(d) (&((struct drm_i915_private *)(d))->ring[RCS])
 
@@ -63,7 +64,7 @@
  * has access to the ring.
  */
 #define RING_LOCK_TEST_WITH_RETURN(dev, file) do {			\
-	if (LP_RING(dev->dev_private)->obj == NULL)			\
+	if (LP_RING(dev->dev_private)->buffer->obj == NULL)			\
 		LOCK_TEST_WITH_RETURN(dev, file);			\
 } while (0)
 
@@ -119,7 +120,7 @@ static void i915_write_hws_pga(struct drm_device *dev)
 static void i915_free_hws(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = LP_RING(dev_priv);
+	struct intel_engine_cs *ring = LP_RING(dev_priv);
 
 	if (dev_priv->status_page_dmah) {
 		drm_pci_free(dev, dev_priv->status_page_dmah);
@@ -139,7 +140,8 @@ void i915_kernel_lost_context(struct drm_device * dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_master_private *master_priv;
-	struct intel_ring_buffer *ring = LP_RING(dev_priv);
+	struct intel_engine_cs *ring = LP_RING(dev_priv);
+	struct intel_ringbuffer *ringbuf = ring->buffer;
 
 	/*
 	 * We should never lose context on the ring with modesetting
@@ -148,17 +150,17 @@ void i915_kernel_lost_context(struct drm_device * dev)
 	if (drm_core_check_feature(dev, DRIVER_MODESET))
 		return;
 
-	ring->head = I915_READ_HEAD(ring) & HEAD_ADDR;
-	ring->tail = I915_READ_TAIL(ring) & TAIL_ADDR;
-	ring->space = ring->head - (ring->tail + I915_RING_FREE_SPACE);
-	if (ring->space < 0)
-		ring->space += ring->size;
+	ringbuf->head = I915_READ_HEAD(ring) & HEAD_ADDR;
+	ringbuf->tail = I915_READ_TAIL(ring) & TAIL_ADDR;
+	ringbuf->space = ringbuf->head - (ringbuf->tail + I915_RING_FREE_SPACE);
+	if (ringbuf->space < 0)
+		ringbuf->space += ringbuf->size;
 
 	if (!dev->primary->master)
 		return;
 
 	master_priv = dev->primary->master->driver_priv;
-	if (ring->head == ring->tail && master_priv->sarea_priv)
+	if (ringbuf->head == ringbuf->tail && master_priv->sarea_priv)
 		master_priv->sarea_priv->perf_boxes |= I915_BOX_RING_EMPTY;
 }
 
@@ -201,7 +203,7 @@ static int i915_initialize(struct drm_device * dev, drm_i915_init_t * init)
 	}
 
 	if (init->ring_size != 0) {
-		if (LP_RING(dev_priv)->obj != NULL) {
+		if (LP_RING(dev_priv)->buffer->obj != NULL) {
 			i915_dma_cleanup(dev);
 			DRM_ERROR("Client tried to initialize ringbuffer in "
 				  "GEM mode\n");
@@ -234,11 +236,11 @@ static int i915_initialize(struct drm_device * dev, drm_i915_init_t * init)
 static int i915_dma_resume(struct drm_device * dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = LP_RING(dev_priv);
+	struct intel_engine_cs *ring = LP_RING(dev_priv);
 
 	DRM_DEBUG_DRIVER("%s\n", __func__);
 
-	if (ring->virtual_start == NULL) {
+	if (ring->buffer->virtual_start == NULL) {
 		DRM_ERROR("can not ioremap virtual address for"
 			  " ring buffer\n");
 		return -ENOMEM;
@@ -360,7 +362,7 @@ static int i915_emit_cmds(struct drm_device * dev, int *buffer, int dwords)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int i, ret;
 
-	if ((dwords+1) * sizeof(int) >= LP_RING(dev_priv)->size - 8)
+	if ((dwords+1) * sizeof(int) >= LP_RING(dev_priv)->buffer->size - 8)
 		return -EINVAL;
 
 	for (i = 0; i < dwords;) {
@@ -782,7 +784,7 @@ static int i915_wait_irq(struct drm_device * dev, int irq_nr)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_master_private *master_priv = dev->primary->master->driver_priv;
 	int ret = 0;
-	struct intel_ring_buffer *ring = LP_RING(dev_priv);
+	struct intel_engine_cs *ring = LP_RING(dev_priv);
 
 	DRM_DEBUG_DRIVER("irq_nr=%d breadcrumb=%d\n", irq_nr,
 		  READ_BREADCRUMB(dev_priv));
@@ -823,7 +825,7 @@ static int i915_irq_emit(struct drm_device *dev, void *data,
 	if (drm_core_check_feature(dev, DRIVER_MODESET))
 		return -ENODEV;
 
-	if (!dev_priv || !LP_RING(dev_priv)->virtual_start) {
+	if (!dev_priv || !LP_RING(dev_priv)->buffer->virtual_start) {
 		DRM_ERROR("called with no initialization\n");
 		return -EINVAL;
 	}
@@ -1017,6 +1019,9 @@ static int i915_getparam(struct drm_device *dev, void *data,
 	case I915_PARAM_HAS_EXEC_HANDLE_LUT:
 		value = 1;
 		break;
+	case I915_PARAM_CMD_PARSER_VERSION:
+		value = i915_cmd_parser_get_version();
+		break;
 	default:
 		DRM_DEBUG("Unknown parameter %d\n", param->param);
 		return -EINVAL;
@@ -1070,7 +1075,7 @@ static int i915_set_status_page(struct drm_device *dev, void *data,
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	drm_i915_hws_addr_t *hws = data;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 
 	if (drm_core_check_feature(dev, DRIVER_MODESET))
 		return -ENODEV;
@@ -1277,12 +1282,13 @@ static void i915_switcheroo_set_state(struct pci_dev *pdev, enum vga_switcheroo_
 static bool i915_switcheroo_can_switch(struct pci_dev *pdev)
 {
 	struct drm_device *dev = pci_get_drvdata(pdev);
-	bool can_switch;
 
-	spin_lock(&dev->count_lock);
-	can_switch = (dev->open_count == 0);
-	spin_unlock(&dev->count_lock);
-	return can_switch;
+	/*
+	 * FIXME: open_count is protected by drm_global_mutex but that would lead to
+	 * locking inversion with the driver load path. And the access here is
+	 * completely racy anyway. So don't bother with locking for now.
+	 */
+	return dev->open_count == 0;
 }
 
 static const struct vga_switcheroo_client_ops i915_switcheroo_ops = {
@@ -1326,7 +1332,7 @@ static int i915_load_modeset_init(struct drm_device *dev)
 
 	intel_power_domains_init_hw(dev_priv);
 
-	ret = drm_irq_install(dev);
+	ret = drm_irq_install(dev, dev->pdev->irq);
 	if (ret)
 		goto cleanup_gem_stolen;
 
@@ -1336,7 +1342,7 @@ static int i915_load_modeset_init(struct drm_device *dev)
 
 	ret = i915_gem_init(dev);
 	if (ret)
-		goto cleanup_power;
+		goto cleanup_irq;
 
 	INIT_WORK(&dev_priv->console_resume_work, intel_console_resume);
 
@@ -1345,10 +1351,8 @@ static int i915_load_modeset_init(struct drm_device *dev)
 	/* Always safe in the mode setting case. */
 	/* FIXME: do pre/post-mode set stuff in core KMS code */
 	dev->vblank_disable_allowed = true;
-	if (INTEL_INFO(dev)->num_pipes == 0) {
-		intel_display_power_put(dev_priv, POWER_DOMAIN_VGA);
+	if (INTEL_INFO(dev)->num_pipes == 0)
 		return 0;
-	}
 
 	ret = intel_fbdev_init(dev);
 	if (ret)
@@ -1383,8 +1387,7 @@ cleanup_gem:
 	mutex_unlock(&dev->struct_mutex);
 	WARN_ON(dev_priv->mm.aliasing_ppgtt);
 	drm_mm_takedown(&dev_priv->gtt.base.mm);
-cleanup_power:
-	intel_display_power_put(dev_priv, POWER_DOMAIN_VGA);
+cleanup_irq:
 	drm_irq_uninstall(dev);
 cleanup_gem_stolen:
 	i915_gem_cleanup_stolen(dev);
@@ -1739,8 +1742,8 @@ out_power_well:
 	intel_power_domains_remove(dev_priv);
 	drm_vblank_cleanup(dev);
 out_gem_unload:
-	if (dev_priv->mm.inactive_shrinker.scan_objects)
-		unregister_shrinker(&dev_priv->mm.inactive_shrinker);
+	WARN_ON(unregister_oom_notifier(&dev_priv->mm.oom_notifier));
+	unregister_shrinker(&dev_priv->mm.shrinker);
 
 	if (dev->pdev->msi_enabled)
 		pci_disable_msi(dev->pdev);
@@ -1791,8 +1794,8 @@ int i915_driver_unload(struct drm_device *dev)
 
 	i915_teardown_sysfs(dev);
 
-	if (dev_priv->mm.inactive_shrinker.scan_objects)
-		unregister_shrinker(&dev_priv->mm.inactive_shrinker);
+	WARN_ON(unregister_oom_notifier(&dev_priv->mm.oom_notifier));
+	unregister_shrinker(&dev_priv->mm.shrinker);
 
 	io_mapping_free(dev_priv->gtt.mappable);
 	arch_phys_wc_del(dev_priv->gtt.mtrr);
@@ -1864,7 +1867,7 @@ int i915_driver_unload(struct drm_device *dev)
 		kmem_cache_destroy(dev_priv->slab);
 
 	pci_dev_put(dev_priv->bridge_dev);
-	kfree(dev->dev_private);
+	kfree(dev_priv);
 
 	return 0;
 }
@@ -1925,6 +1928,8 @@ void i915_driver_postclose(struct drm_device *dev, struct drm_file *file)
 {
 	struct drm_i915_file_private *file_priv = file->driver_priv;
 
+	if (file_priv && file_priv->bsd_ring)
+		file_priv->bsd_ring = NULL;
 	kfree(file_priv);
 }
 
@@ -1978,9 +1983,10 @@ const struct drm_ioctl_desc i915_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(I915_GEM_CONTEXT_DESTROY, i915_gem_context_destroy_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(I915_REG_READ, i915_reg_read_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(I915_GET_RESET_STATS, i915_get_reset_stats_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_USERPTR, i915_gem_userptr_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
 };
 
-int i915_max_ioctl = DRM_ARRAY_SIZE(i915_ioctls);
+int i915_max_ioctl = ARRAY_SIZE(i915_ioctls);
 
 /*
  * This is really ugly: Because old userspace abused the linux agp interface to
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 82f4d1f47d3b..651e65e051c0 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -36,6 +36,7 @@
 
 #include <linux/console.h>
 #include <linux/module.h>
+#include <linux/pm_runtime.h>
 #include <drm/drm_crtc_helper.h>
 
 static struct drm_driver driver;
@@ -49,12 +50,30 @@ static struct drm_driver driver;
 	.dpll_md_offsets = { DPLL_A_MD_OFFSET, DPLL_B_MD_OFFSET }, \
 	.palette_offsets = { PALETTE_A_OFFSET, PALETTE_B_OFFSET }
 
+#define GEN_CHV_PIPEOFFSETS \
+	.pipe_offsets = { PIPE_A_OFFSET, PIPE_B_OFFSET, \
+			  CHV_PIPE_C_OFFSET }, \
+	.trans_offsets = { TRANSCODER_A_OFFSET, TRANSCODER_B_OFFSET, \
+			   CHV_TRANSCODER_C_OFFSET, }, \
+	.dpll_offsets = { DPLL_A_OFFSET, DPLL_B_OFFSET, \
+			  CHV_DPLL_C_OFFSET }, \
+	.dpll_md_offsets = { DPLL_A_MD_OFFSET, DPLL_B_MD_OFFSET, \
+			     CHV_DPLL_C_MD_OFFSET }, \
+	.palette_offsets = { PALETTE_A_OFFSET, PALETTE_B_OFFSET, \
+			     CHV_PALETTE_C_OFFSET }
+
+#define CURSOR_OFFSETS \
+	.cursor_offsets = { CURSOR_A_OFFSET, CURSOR_B_OFFSET, CHV_CURSOR_C_OFFSET }
+
+#define IVB_CURSOR_OFFSETS \
+	.cursor_offsets = { CURSOR_A_OFFSET, IVB_CURSOR_B_OFFSET, IVB_CURSOR_C_OFFSET }
 
 static const struct intel_device_info intel_i830_info = {
 	.gen = 2, .is_mobile = 1, .cursor_needs_physical = 1, .num_pipes = 2,
 	.has_overlay = 1, .overlay_needs_physical = 1,
 	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_845g_info = {
@@ -62,6 +81,7 @@ static const struct intel_device_info intel_845g_info = {
 	.has_overlay = 1, .overlay_needs_physical = 1,
 	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_i85x_info = {
@@ -71,6 +91,7 @@ static const struct intel_device_info intel_i85x_info = {
 	.has_fbc = 1,
 	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_i865g_info = {
@@ -78,6 +99,7 @@ static const struct intel_device_info intel_i865g_info = {
 	.has_overlay = 1, .overlay_needs_physical = 1,
 	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_i915g_info = {
@@ -85,6 +107,7 @@ static const struct intel_device_info intel_i915g_info = {
 	.has_overlay = 1, .overlay_needs_physical = 1,
 	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 static const struct intel_device_info intel_i915gm_info = {
 	.gen = 3, .is_mobile = 1, .num_pipes = 2,
@@ -94,12 +117,14 @@ static const struct intel_device_info intel_i915gm_info = {
 	.has_fbc = 1,
 	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 static const struct intel_device_info intel_i945g_info = {
 	.gen = 3, .has_hotplug = 1, .cursor_needs_physical = 1, .num_pipes = 2,
 	.has_overlay = 1, .overlay_needs_physical = 1,
 	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 static const struct intel_device_info intel_i945gm_info = {
 	.gen = 3, .is_i945gm = 1, .is_mobile = 1, .num_pipes = 2,
@@ -109,6 +134,7 @@ static const struct intel_device_info intel_i945gm_info = {
 	.has_fbc = 1,
 	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_i965g_info = {
@@ -117,6 +143,7 @@ static const struct intel_device_info intel_i965g_info = {
 	.has_overlay = 1,
 	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_i965gm_info = {
@@ -126,6 +153,7 @@ static const struct intel_device_info intel_i965gm_info = {
 	.supports_tv = 1,
 	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_g33_info = {
@@ -134,6 +162,7 @@ static const struct intel_device_info intel_g33_info = {
 	.has_overlay = 1,
 	.ring_mask = RENDER_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_g45_info = {
@@ -141,6 +170,7 @@ static const struct intel_device_info intel_g45_info = {
 	.has_pipe_cxsr = 1, .has_hotplug = 1,
 	.ring_mask = RENDER_RING | BSD_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_gm45_info = {
@@ -150,6 +180,7 @@ static const struct intel_device_info intel_gm45_info = {
 	.supports_tv = 1,
 	.ring_mask = RENDER_RING | BSD_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_pineview_info = {
@@ -157,6 +188,7 @@ static const struct intel_device_info intel_pineview_info = {
 	.need_gfx_hws = 1, .has_hotplug = 1,
 	.has_overlay = 1,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_ironlake_d_info = {
@@ -164,6 +196,7 @@ static const struct intel_device_info intel_ironlake_d_info = {
 	.need_gfx_hws = 1, .has_hotplug = 1,
 	.ring_mask = RENDER_RING | BSD_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_ironlake_m_info = {
@@ -172,6 +205,7 @@ static const struct intel_device_info intel_ironlake_m_info = {
 	.has_fbc = 1,
 	.ring_mask = RENDER_RING | BSD_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_sandybridge_d_info = {
@@ -181,6 +215,7 @@ static const struct intel_device_info intel_sandybridge_d_info = {
 	.ring_mask = RENDER_RING | BSD_RING | BLT_RING,
 	.has_llc = 1,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_sandybridge_m_info = {
@@ -190,6 +225,7 @@ static const struct intel_device_info intel_sandybridge_m_info = {
 	.ring_mask = RENDER_RING | BSD_RING | BLT_RING,
 	.has_llc = 1,
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 #define GEN7_FEATURES  \
@@ -203,6 +239,7 @@ static const struct intel_device_info intel_ivybridge_d_info = {
 	GEN7_FEATURES,
 	.is_ivybridge = 1,
 	GEN_DEFAULT_PIPEOFFSETS,
+	IVB_CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_ivybridge_m_info = {
@@ -210,6 +247,7 @@ static const struct intel_device_info intel_ivybridge_m_info = {
 	.is_ivybridge = 1,
 	.is_mobile = 1,
 	GEN_DEFAULT_PIPEOFFSETS,
+	IVB_CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_ivybridge_q_info = {
@@ -217,6 +255,7 @@ static const struct intel_device_info intel_ivybridge_q_info = {
 	.is_ivybridge = 1,
 	.num_pipes = 0, /* legal, last one wins */
 	GEN_DEFAULT_PIPEOFFSETS,
+	IVB_CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_valleyview_m_info = {
@@ -228,6 +267,7 @@ static const struct intel_device_info intel_valleyview_m_info = {
 	.has_fbc = 0, /* legal, last one wins */
 	.has_llc = 0, /* legal, last one wins */
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_valleyview_d_info = {
@@ -238,6 +278,7 @@ static const struct intel_device_info intel_valleyview_d_info = {
 	.has_fbc = 0, /* legal, last one wins */
 	.has_llc = 0, /* legal, last one wins */
 	GEN_DEFAULT_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_haswell_d_info = {
@@ -247,6 +288,7 @@ static const struct intel_device_info intel_haswell_d_info = {
 	.has_fpga_dbg = 1,
 	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	IVB_CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_haswell_m_info = {
@@ -257,6 +299,7 @@ static const struct intel_device_info intel_haswell_m_info = {
 	.has_fpga_dbg = 1,
 	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING,
 	GEN_DEFAULT_PIPEOFFSETS,
+	IVB_CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_broadwell_d_info = {
@@ -267,6 +310,7 @@ static const struct intel_device_info intel_broadwell_d_info = {
 	.has_ddi = 1,
 	.has_fbc = 1,
 	GEN_DEFAULT_PIPEOFFSETS,
+	IVB_CURSOR_OFFSETS,
 };
 
 static const struct intel_device_info intel_broadwell_m_info = {
@@ -277,6 +321,40 @@ static const struct intel_device_info intel_broadwell_m_info = {
 	.has_ddi = 1,
 	.has_fbc = 1,
 	GEN_DEFAULT_PIPEOFFSETS,
+	IVB_CURSOR_OFFSETS,
+};
+
+static const struct intel_device_info intel_broadwell_gt3d_info = {
+	.gen = 8, .num_pipes = 3,
+	.need_gfx_hws = 1, .has_hotplug = 1,
+	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING | BSD2_RING,
+	.has_llc = 1,
+	.has_ddi = 1,
+	.has_fbc = 1,
+	GEN_DEFAULT_PIPEOFFSETS,
+	IVB_CURSOR_OFFSETS,
+};
+
+static const struct intel_device_info intel_broadwell_gt3m_info = {
+	.gen = 8, .is_mobile = 1, .num_pipes = 3,
+	.need_gfx_hws = 1, .has_hotplug = 1,
+	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING | BSD2_RING,
+	.has_llc = 1,
+	.has_ddi = 1,
+	.has_fbc = 1,
+	GEN_DEFAULT_PIPEOFFSETS,
+	IVB_CURSOR_OFFSETS,
+};
+
+static const struct intel_device_info intel_cherryview_info = {
+	.is_preliminary = 1,
+	.gen = 8, .num_pipes = 3,
+	.need_gfx_hws = 1, .has_hotplug = 1,
+	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING,
+	.is_valleyview = 1,
+	.display_mmio_offset = VLV_DISPLAY_BASE,
+	GEN_CHV_PIPEOFFSETS,
+	CURSOR_OFFSETS,
 };
 
 /*
@@ -311,8 +389,11 @@ static const struct intel_device_info intel_broadwell_m_info = {
 	INTEL_HSW_M_IDS(&intel_haswell_m_info), \
 	INTEL_VLV_M_IDS(&intel_valleyview_m_info),	\
 	INTEL_VLV_D_IDS(&intel_valleyview_d_info),	\
-	INTEL_BDW_M_IDS(&intel_broadwell_m_info),	\
-	INTEL_BDW_D_IDS(&intel_broadwell_d_info)
+	INTEL_BDW_GT12M_IDS(&intel_broadwell_m_info),	\
+	INTEL_BDW_GT12D_IDS(&intel_broadwell_d_info),	\
+	INTEL_BDW_GT3M_IDS(&intel_broadwell_gt3m_info),	\
+	INTEL_BDW_GT3D_IDS(&intel_broadwell_gt3d_info), \
+	INTEL_CHV_IDS(&intel_cherryview_info)
 
 static const struct pci_device_id pciidlist[] = {		/* aka */
 	INTEL_PCI_IDS,
@@ -445,18 +526,20 @@ static int i915_drm_freeze(struct drm_device *dev)
 			return error;
 		}
 
-		cancel_delayed_work_sync(&dev_priv->rps.delayed_resume_work);
-
 		drm_irq_uninstall(dev);
 		dev_priv->enable_hotplug_processing = false;
+
+		intel_disable_gt_powersave(dev);
+
 		/*
 		 * Disable CRTCs directly since we want to preserve sw state
 		 * for _thaw.
 		 */
-		mutex_lock(&dev->mode_config.mutex);
-		list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
+		drm_modeset_lock_all(dev);
+		for_each_crtc(dev, crtc) {
 			dev_priv->display.crtc_disable(crtc);
-		mutex_unlock(&dev->mode_config.mutex);
+		}
+		drm_modeset_unlock_all(dev);
 
 		intel_modeset_suspend_hw(dev);
 	}
@@ -519,24 +602,6 @@ void intel_console_resume(struct work_struct *work)
 	console_unlock();
 }
 
-static void intel_resume_hotplug(struct drm_device *dev)
-{
-	struct drm_mode_config *mode_config = &dev->mode_config;
-	struct intel_encoder *encoder;
-
-	mutex_lock(&mode_config->mutex);
-	DRM_DEBUG_KMS("running encoder hotplug functions\n");
-
-	list_for_each_entry(encoder, &mode_config->encoder_list, base.head)
-		if (encoder->hot_plug)
-			encoder->hot_plug(encoder);
-
-	mutex_unlock(&mode_config->mutex);
-
-	/* Just fire off a uevent and let userspace tell us what to do */
-	drm_helper_hpd_irq_event(dev);
-}
-
 static int i915_drm_thaw_early(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -551,7 +616,6 @@ static int i915_drm_thaw_early(struct drm_device *dev)
 static int __i915_drm_thaw(struct drm_device *dev, bool restore_gtt_mappings)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	int error = 0;
 
 	if (drm_core_check_feature(dev, DRIVER_MODESET) &&
 	    restore_gtt_mappings) {
@@ -569,12 +633,14 @@ static int __i915_drm_thaw(struct drm_device *dev, bool restore_gtt_mappings)
 		drm_mode_config_reset(dev);
 
 		mutex_lock(&dev->struct_mutex);
-
-		error = i915_gem_init_hw(dev);
+		if (i915_gem_init_hw(dev)) {
+			DRM_ERROR("failed to re-initialize GPU, declaring wedged!\n");
+			atomic_set_mask(I915_WEDGED, &dev_priv->gpu_error.reset_counter);
+		}
 		mutex_unlock(&dev->struct_mutex);
 
 		/* We need working interrupts for modeset enabling ... */
-		drm_irq_install(dev);
+		drm_irq_install(dev, dev->pdev->irq);
 
 		intel_modeset_init_hw(dev);
 
@@ -591,7 +657,7 @@ static int __i915_drm_thaw(struct drm_device *dev, bool restore_gtt_mappings)
 		intel_hpd_init(dev);
 		dev_priv->enable_hotplug_processing = true;
 		/* Config may have changed between suspend and resume */
-		intel_resume_hotplug(dev);
+		drm_helper_hpd_irq_event(dev);
 	}
 
 	intel_opregion_init(dev);
@@ -613,7 +679,7 @@ static int __i915_drm_thaw(struct drm_device *dev, bool restore_gtt_mappings)
 	mutex_unlock(&dev_priv->modeset_restore_lock);
 
 	intel_runtime_pm_put(dev_priv);
-	return error;
+	return 0;
 }
 
 static int i915_drm_thaw(struct drm_device *dev)
@@ -746,18 +812,20 @@ int i915_reset(struct drm_device *dev)
 			return ret;
 		}
 
-		drm_irq_uninstall(dev);
-		drm_irq_install(dev);
+		/*
+		 * FIXME: This races pretty badly against concurrent holders of
+		 * ring interrupts. This is possible since we've started to drop
+		 * dev->struct_mutex in select places when waiting for the gpu.
+		 */
 
-		/* rps/rc6 re-init is necessary to restore state lost after the
-		 * reset and the re-install of drm irq. Skip for ironlake per
+		/*
+		 * rps/rc6 re-init is necessary to restore state lost after the
+		 * reset and the re-install of gt irqs. Skip for ironlake per
 		 * previous concerns that it doesn't respond well to some forms
-		 * of re-init after reset. */
-		if (INTEL_INFO(dev)->gen > 5) {
-			mutex_lock(&dev->struct_mutex);
-			intel_enable_gt_powersave(dev);
-			mutex_unlock(&dev->struct_mutex);
-		}
+		 * of re-init after reset.
+		 */
+		if (INTEL_INFO(dev)->gen > 5)
+			intel_reset_gt_powersave(dev);
 
 		intel_hpd_init(dev);
 	} else {
@@ -891,21 +959,453 @@ static int i915_pm_poweroff(struct device *dev)
 	return i915_drm_freeze(drm_dev);
 }
 
-static int i915_runtime_suspend(struct device *device)
+static int hsw_runtime_suspend(struct drm_i915_private *dev_priv)
+{
+	hsw_enable_pc8(dev_priv);
+
+	return 0;
+}
+
+static int snb_runtime_resume(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
+
+	intel_init_pch_refclk(dev);
+
+	return 0;
+}
+
+static int hsw_runtime_resume(struct drm_i915_private *dev_priv)
+{
+	hsw_disable_pc8(dev_priv);
+
+	return 0;
+}
+
+/*
+ * Save all Gunit registers that may be lost after a D3 and a subsequent
+ * S0i[R123] transition. The list of registers needing a save/restore is
+ * defined in the VLV2_S0IXRegs document. This documents marks all Gunit
+ * registers in the following way:
+ * - Driver: saved/restored by the driver
+ * - Punit : saved/restored by the Punit firmware
+ * - No, w/o marking: no need to save/restore, since the register is R/O or
+ *                    used internally by the HW in a way that doesn't depend
+ *                    keeping the content across a suspend/resume.
+ * - Debug : used for debugging
+ *
+ * We save/restore all registers marked with 'Driver', with the following
+ * exceptions:
+ * - Registers out of use, including also registers marked with 'Debug'.
+ *   These have no effect on the driver's operation, so we don't save/restore
+ *   them to reduce the overhead.
+ * - Registers that are fully setup by an initialization function called from
+ *   the resume path. For example many clock gating and RPS/RC6 registers.
+ * - Registers that provide the right functionality with their reset defaults.
+ *
+ * TODO: Except for registers that based on the above 3 criteria can be safely
+ * ignored, we save/restore all others, practically treating the HW context as
+ * a black-box for the driver. Further investigation is needed to reduce the
+ * saved/restored registers even further, by following the same 3 criteria.
+ */
+static void vlv_save_gunit_s0ix_state(struct drm_i915_private *dev_priv)
+{
+	struct vlv_s0ix_state *s = &dev_priv->vlv_s0ix_state;
+	int i;
+
+	/* GAM 0x4000-0x4770 */
+	s->wr_watermark		= I915_READ(GEN7_WR_WATERMARK);
+	s->gfx_prio_ctrl	= I915_READ(GEN7_GFX_PRIO_CTRL);
+	s->arb_mode		= I915_READ(ARB_MODE);
+	s->gfx_pend_tlb0	= I915_READ(GEN7_GFX_PEND_TLB0);
+	s->gfx_pend_tlb1	= I915_READ(GEN7_GFX_PEND_TLB1);
+
+	for (i = 0; i < ARRAY_SIZE(s->lra_limits); i++)
+		s->lra_limits[i] = I915_READ(GEN7_LRA_LIMITS_BASE + i * 4);
+
+	s->media_max_req_count	= I915_READ(GEN7_MEDIA_MAX_REQ_COUNT);
+	s->gfx_max_req_count	= I915_READ(GEN7_MEDIA_MAX_REQ_COUNT);
+
+	s->render_hwsp		= I915_READ(RENDER_HWS_PGA_GEN7);
+	s->ecochk		= I915_READ(GAM_ECOCHK);
+	s->bsd_hwsp		= I915_READ(BSD_HWS_PGA_GEN7);
+	s->blt_hwsp		= I915_READ(BLT_HWS_PGA_GEN7);
+
+	s->tlb_rd_addr		= I915_READ(GEN7_TLB_RD_ADDR);
+
+	/* MBC 0x9024-0x91D0, 0x8500 */
+	s->g3dctl		= I915_READ(VLV_G3DCTL);
+	s->gsckgctl		= I915_READ(VLV_GSCKGCTL);
+	s->mbctl		= I915_READ(GEN6_MBCTL);
+
+	/* GCP 0x9400-0x9424, 0x8100-0x810C */
+	s->ucgctl1		= I915_READ(GEN6_UCGCTL1);
+	s->ucgctl3		= I915_READ(GEN6_UCGCTL3);
+	s->rcgctl1		= I915_READ(GEN6_RCGCTL1);
+	s->rcgctl2		= I915_READ(GEN6_RCGCTL2);
+	s->rstctl		= I915_READ(GEN6_RSTCTL);
+	s->misccpctl		= I915_READ(GEN7_MISCCPCTL);
+
+	/* GPM 0xA000-0xAA84, 0x8000-0x80FC */
+	s->gfxpause		= I915_READ(GEN6_GFXPAUSE);
+	s->rpdeuhwtc		= I915_READ(GEN6_RPDEUHWTC);
+	s->rpdeuc		= I915_READ(GEN6_RPDEUC);
+	s->ecobus		= I915_READ(ECOBUS);
+	s->pwrdwnupctl		= I915_READ(VLV_PWRDWNUPCTL);
+	s->rp_down_timeout	= I915_READ(GEN6_RP_DOWN_TIMEOUT);
+	s->rp_deucsw		= I915_READ(GEN6_RPDEUCSW);
+	s->rcubmabdtmr		= I915_READ(GEN6_RCUBMABDTMR);
+	s->rcedata		= I915_READ(VLV_RCEDATA);
+	s->spare2gh		= I915_READ(VLV_SPAREG2H);
+
+	/* Display CZ domain, 0x4400C-0x4402C, 0x4F000-0x4F11F */
+	s->gt_imr		= I915_READ(GTIMR);
+	s->gt_ier		= I915_READ(GTIER);
+	s->pm_imr		= I915_READ(GEN6_PMIMR);
+	s->pm_ier		= I915_READ(GEN6_PMIER);
+
+	for (i = 0; i < ARRAY_SIZE(s->gt_scratch); i++)
+		s->gt_scratch[i] = I915_READ(GEN7_GT_SCRATCH_BASE + i * 4);
+
+	/* GT SA CZ domain, 0x100000-0x138124 */
+	s->tilectl		= I915_READ(TILECTL);
+	s->gt_fifoctl		= I915_READ(GTFIFOCTL);
+	s->gtlc_wake_ctrl	= I915_READ(VLV_GTLC_WAKE_CTRL);
+	s->gtlc_survive		= I915_READ(VLV_GTLC_SURVIVABILITY_REG);
+	s->pmwgicz		= I915_READ(VLV_PMWGICZ);
+
+	/* Gunit-Display CZ domain, 0x182028-0x1821CF */
+	s->gu_ctl0		= I915_READ(VLV_GU_CTL0);
+	s->gu_ctl1		= I915_READ(VLV_GU_CTL1);
+	s->clock_gate_dis2	= I915_READ(VLV_GUNIT_CLOCK_GATE2);
+
+	/*
+	 * Not saving any of:
+	 * DFT,		0x9800-0x9EC0
+	 * SARB,	0xB000-0xB1FC
+	 * GAC,		0x5208-0x524C, 0x14000-0x14C000
+	 * PCI CFG
+	 */
+}
+
+static void vlv_restore_gunit_s0ix_state(struct drm_i915_private *dev_priv)
+{
+	struct vlv_s0ix_state *s = &dev_priv->vlv_s0ix_state;
+	u32 val;
+	int i;
+
+	/* GAM 0x4000-0x4770 */
+	I915_WRITE(GEN7_WR_WATERMARK,	s->wr_watermark);
+	I915_WRITE(GEN7_GFX_PRIO_CTRL,	s->gfx_prio_ctrl);
+	I915_WRITE(ARB_MODE,		s->arb_mode | (0xffff << 16));
+	I915_WRITE(GEN7_GFX_PEND_TLB0,	s->gfx_pend_tlb0);
+	I915_WRITE(GEN7_GFX_PEND_TLB1,	s->gfx_pend_tlb1);
+
+	for (i = 0; i < ARRAY_SIZE(s->lra_limits); i++)
+		I915_WRITE(GEN7_LRA_LIMITS_BASE + i * 4, s->lra_limits[i]);
+
+	I915_WRITE(GEN7_MEDIA_MAX_REQ_COUNT, s->media_max_req_count);
+	I915_WRITE(GEN7_MEDIA_MAX_REQ_COUNT, s->gfx_max_req_count);
+
+	I915_WRITE(RENDER_HWS_PGA_GEN7,	s->render_hwsp);
+	I915_WRITE(GAM_ECOCHK,		s->ecochk);
+	I915_WRITE(BSD_HWS_PGA_GEN7,	s->bsd_hwsp);
+	I915_WRITE(BLT_HWS_PGA_GEN7,	s->blt_hwsp);
+
+	I915_WRITE(GEN7_TLB_RD_ADDR,	s->tlb_rd_addr);
+
+	/* MBC 0x9024-0x91D0, 0x8500 */
+	I915_WRITE(VLV_G3DCTL,		s->g3dctl);
+	I915_WRITE(VLV_GSCKGCTL,	s->gsckgctl);
+	I915_WRITE(GEN6_MBCTL,		s->mbctl);
+
+	/* GCP 0x9400-0x9424, 0x8100-0x810C */
+	I915_WRITE(GEN6_UCGCTL1,	s->ucgctl1);
+	I915_WRITE(GEN6_UCGCTL3,	s->ucgctl3);
+	I915_WRITE(GEN6_RCGCTL1,	s->rcgctl1);
+	I915_WRITE(GEN6_RCGCTL2,	s->rcgctl2);
+	I915_WRITE(GEN6_RSTCTL,		s->rstctl);
+	I915_WRITE(GEN7_MISCCPCTL,	s->misccpctl);
+
+	/* GPM 0xA000-0xAA84, 0x8000-0x80FC */
+	I915_WRITE(GEN6_GFXPAUSE,	s->gfxpause);
+	I915_WRITE(GEN6_RPDEUHWTC,	s->rpdeuhwtc);
+	I915_WRITE(GEN6_RPDEUC,		s->rpdeuc);
+	I915_WRITE(ECOBUS,		s->ecobus);
+	I915_WRITE(VLV_PWRDWNUPCTL,	s->pwrdwnupctl);
+	I915_WRITE(GEN6_RP_DOWN_TIMEOUT,s->rp_down_timeout);
+	I915_WRITE(GEN6_RPDEUCSW,	s->rp_deucsw);
+	I915_WRITE(GEN6_RCUBMABDTMR,	s->rcubmabdtmr);
+	I915_WRITE(VLV_RCEDATA,		s->rcedata);
+	I915_WRITE(VLV_SPAREG2H,	s->spare2gh);
+
+	/* Display CZ domain, 0x4400C-0x4402C, 0x4F000-0x4F11F */
+	I915_WRITE(GTIMR,		s->gt_imr);
+	I915_WRITE(GTIER,		s->gt_ier);
+	I915_WRITE(GEN6_PMIMR,		s->pm_imr);
+	I915_WRITE(GEN6_PMIER,		s->pm_ier);
+
+	for (i = 0; i < ARRAY_SIZE(s->gt_scratch); i++)
+		I915_WRITE(GEN7_GT_SCRATCH_BASE + i * 4, s->gt_scratch[i]);
+
+	/* GT SA CZ domain, 0x100000-0x138124 */
+	I915_WRITE(TILECTL,			s->tilectl);
+	I915_WRITE(GTFIFOCTL,			s->gt_fifoctl);
+	/*
+	 * Preserve the GT allow wake and GFX force clock bit, they are not
+	 * be restored, as they are used to control the s0ix suspend/resume
+	 * sequence by the caller.
+	 */
+	val = I915_READ(VLV_GTLC_WAKE_CTRL);
+	val &= VLV_GTLC_ALLOWWAKEREQ;
+	val |= s->gtlc_wake_ctrl & ~VLV_GTLC_ALLOWWAKEREQ;
+	I915_WRITE(VLV_GTLC_WAKE_CTRL, val);
+
+	val = I915_READ(VLV_GTLC_SURVIVABILITY_REG);
+	val &= VLV_GFX_CLK_FORCE_ON_BIT;
+	val |= s->gtlc_survive & ~VLV_GFX_CLK_FORCE_ON_BIT;
+	I915_WRITE(VLV_GTLC_SURVIVABILITY_REG, val);
+
+	I915_WRITE(VLV_PMWGICZ,			s->pmwgicz);
+
+	/* Gunit-Display CZ domain, 0x182028-0x1821CF */
+	I915_WRITE(VLV_GU_CTL0,			s->gu_ctl0);
+	I915_WRITE(VLV_GU_CTL1,			s->gu_ctl1);
+	I915_WRITE(VLV_GUNIT_CLOCK_GATE2,	s->clock_gate_dis2);
+}
+
+int vlv_force_gfx_clock(struct drm_i915_private *dev_priv, bool force_on)
+{
+	u32 val;
+	int err;
+
+	val = I915_READ(VLV_GTLC_SURVIVABILITY_REG);
+	WARN_ON(!!(val & VLV_GFX_CLK_FORCE_ON_BIT) == force_on);
+
+#define COND (I915_READ(VLV_GTLC_SURVIVABILITY_REG) & VLV_GFX_CLK_STATUS_BIT)
+	/* Wait for a previous force-off to settle */
+	if (force_on) {
+		err = wait_for(!COND, 20);
+		if (err) {
+			DRM_ERROR("timeout waiting for GFX clock force-off (%08x)\n",
+				  I915_READ(VLV_GTLC_SURVIVABILITY_REG));
+			return err;
+		}
+	}
+
+	val = I915_READ(VLV_GTLC_SURVIVABILITY_REG);
+	val &= ~VLV_GFX_CLK_FORCE_ON_BIT;
+	if (force_on)
+		val |= VLV_GFX_CLK_FORCE_ON_BIT;
+	I915_WRITE(VLV_GTLC_SURVIVABILITY_REG, val);
+
+	if (!force_on)
+		return 0;
+
+	err = wait_for(COND, 20);
+	if (err)
+		DRM_ERROR("timeout waiting for GFX clock force-on (%08x)\n",
+			  I915_READ(VLV_GTLC_SURVIVABILITY_REG));
+
+	return err;
+#undef COND
+}
+
+static int vlv_allow_gt_wake(struct drm_i915_private *dev_priv, bool allow)
+{
+	u32 val;
+	int err = 0;
+
+	val = I915_READ(VLV_GTLC_WAKE_CTRL);
+	val &= ~VLV_GTLC_ALLOWWAKEREQ;
+	if (allow)
+		val |= VLV_GTLC_ALLOWWAKEREQ;
+	I915_WRITE(VLV_GTLC_WAKE_CTRL, val);
+	POSTING_READ(VLV_GTLC_WAKE_CTRL);
+
+#define COND (!!(I915_READ(VLV_GTLC_PW_STATUS) & VLV_GTLC_ALLOWWAKEACK) == \
+	      allow)
+	err = wait_for(COND, 1);
+	if (err)
+		DRM_ERROR("timeout disabling GT waking\n");
+	return err;
+#undef COND
+}
+
+static int vlv_wait_for_gt_wells(struct drm_i915_private *dev_priv,
+				 bool wait_for_on)
+{
+	u32 mask;
+	u32 val;
+	int err;
+
+	mask = VLV_GTLC_PW_MEDIA_STATUS_MASK | VLV_GTLC_PW_RENDER_STATUS_MASK;
+	val = wait_for_on ? mask : 0;
+#define COND ((I915_READ(VLV_GTLC_PW_STATUS) & mask) == val)
+	if (COND)
+		return 0;
+
+	DRM_DEBUG_KMS("waiting for GT wells to go %s (%08x)\n",
+			wait_for_on ? "on" : "off",
+			I915_READ(VLV_GTLC_PW_STATUS));
+
+	/*
+	 * RC6 transitioning can be delayed up to 2 msec (see
+	 * valleyview_enable_rps), use 3 msec for safety.
+	 */
+	err = wait_for(COND, 3);
+	if (err)
+		DRM_ERROR("timeout waiting for GT wells to go %s\n",
+			  wait_for_on ? "on" : "off");
+
+	return err;
+#undef COND
+}
+
+static void vlv_check_no_gt_access(struct drm_i915_private *dev_priv)
+{
+	if (!(I915_READ(VLV_GTLC_PW_STATUS) & VLV_GTLC_ALLOWWAKEERR))
+		return;
+
+	DRM_ERROR("GT register access while GT waking disabled\n");
+	I915_WRITE(VLV_GTLC_PW_STATUS, VLV_GTLC_ALLOWWAKEERR);
+}
+
+static int vlv_runtime_suspend(struct drm_i915_private *dev_priv)
+{
+	u32 mask;
+	int err;
+
+	/*
+	 * Bspec defines the following GT well on flags as debug only, so
+	 * don't treat them as hard failures.
+	 */
+	(void)vlv_wait_for_gt_wells(dev_priv, false);
+
+	mask = VLV_GTLC_RENDER_CTX_EXISTS | VLV_GTLC_MEDIA_CTX_EXISTS;
+	WARN_ON((I915_READ(VLV_GTLC_WAKE_CTRL) & mask) != mask);
+
+	vlv_check_no_gt_access(dev_priv);
+
+	err = vlv_force_gfx_clock(dev_priv, true);
+	if (err)
+		goto err1;
+
+	err = vlv_allow_gt_wake(dev_priv, false);
+	if (err)
+		goto err2;
+	vlv_save_gunit_s0ix_state(dev_priv);
+
+	err = vlv_force_gfx_clock(dev_priv, false);
+	if (err)
+		goto err2;
+
+	return 0;
+
+err2:
+	/* For safety always re-enable waking and disable gfx clock forcing */
+	vlv_allow_gt_wake(dev_priv, true);
+err1:
+	vlv_force_gfx_clock(dev_priv, false);
+
+	return err;
+}
+
+static int vlv_runtime_resume(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
+	int err;
+	int ret;
+
+	/*
+	 * If any of the steps fail just try to continue, that's the best we
+	 * can do at this point. Return the first error code (which will also
+	 * leave RPM permanently disabled).
+	 */
+	ret = vlv_force_gfx_clock(dev_priv, true);
+
+	vlv_restore_gunit_s0ix_state(dev_priv);
+
+	err = vlv_allow_gt_wake(dev_priv, true);
+	if (!ret)
+		ret = err;
+
+	err = vlv_force_gfx_clock(dev_priv, false);
+	if (!ret)
+		ret = err;
+
+	vlv_check_no_gt_access(dev_priv);
+
+	intel_init_clock_gating(dev);
+	i915_gem_restore_fences(dev);
+
+	return ret;
+}
+
+static int intel_runtime_suspend(struct device *device)
 {
 	struct pci_dev *pdev = to_pci_dev(device);
 	struct drm_device *dev = pci_get_drvdata(pdev);
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	int ret;
+
+	if (WARN_ON_ONCE(!(dev_priv->rps.enabled && intel_enable_rc6(dev))))
+		return -ENODEV;
 
 	WARN_ON(!HAS_RUNTIME_PM(dev));
 	assert_force_wake_inactive(dev_priv);
 
 	DRM_DEBUG_KMS("Suspending device\n");
 
-	if (HAS_PC8(dev))
-		hsw_enable_pc8(dev_priv);
+	/*
+	 * We could deadlock here in case another thread holding struct_mutex
+	 * calls RPM suspend concurrently, since the RPM suspend will wait
+	 * first for this RPM suspend to finish. In this case the concurrent
+	 * RPM resume will be followed by its RPM suspend counterpart. Still
+	 * for consistency return -EAGAIN, which will reschedule this suspend.
+	 */
+	if (!mutex_trylock(&dev->struct_mutex)) {
+		DRM_DEBUG_KMS("device lock contention, deffering suspend\n");
+		/*
+		 * Bump the expiration timestamp, otherwise the suspend won't
+		 * be rescheduled.
+		 */
+		pm_runtime_mark_last_busy(device);
 
+		return -EAGAIN;
+	}
+	/*
+	 * We are safe here against re-faults, since the fault handler takes
+	 * an RPM reference.
+	 */
 	i915_gem_release_all_mmaps(dev_priv);
+	mutex_unlock(&dev->struct_mutex);
+
+	/*
+	 * rps.work can't be rearmed here, since we get here only after making
+	 * sure the GPU is idle and the RPS freq is set to the minimum. See
+	 * intel_mark_idle().
+	 */
+	cancel_work_sync(&dev_priv->rps.work);
+	intel_runtime_pm_disable_interrupts(dev);
+
+	if (IS_GEN6(dev)) {
+		ret = 0;
+	} else if (IS_HASWELL(dev) || IS_BROADWELL(dev)) {
+		ret = hsw_runtime_suspend(dev_priv);
+	} else if (IS_VALLEYVIEW(dev)) {
+		ret = vlv_runtime_suspend(dev_priv);
+	} else {
+		ret = -ENODEV;
+		WARN_ON(1);
+	}
+
+	if (ret) {
+		DRM_ERROR("Runtime suspend failed, disabling it (%d)\n", ret);
+		intel_runtime_pm_restore_interrupts(dev);
+
+		return ret;
+	}
 
 	del_timer_sync(&dev_priv->gpu_error.hangcheck_timer);
 	dev_priv->pm.suspended = true;
@@ -923,11 +1423,12 @@ static int i915_runtime_suspend(struct device *device)
 	return 0;
 }
 
-static int i915_runtime_resume(struct device *device)
+static int intel_runtime_resume(struct device *device)
 {
 	struct pci_dev *pdev = to_pci_dev(device);
 	struct drm_device *dev = pci_get_drvdata(pdev);
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	int ret;
 
 	WARN_ON(!HAS_RUNTIME_PM(dev));
 
@@ -936,11 +1437,33 @@ static int i915_runtime_resume(struct device *device)
 	intel_opregion_notify_adapter(dev, PCI_D0);
 	dev_priv->pm.suspended = false;
 
-	if (HAS_PC8(dev))
-		hsw_disable_pc8(dev_priv);
+	if (IS_GEN6(dev)) {
+		ret = snb_runtime_resume(dev_priv);
+	} else if (IS_HASWELL(dev) || IS_BROADWELL(dev)) {
+		ret = hsw_runtime_resume(dev_priv);
+	} else if (IS_VALLEYVIEW(dev)) {
+		ret = vlv_runtime_resume(dev_priv);
+	} else {
+		WARN_ON(1);
+		ret = -ENODEV;
+	}
 
-	DRM_DEBUG_KMS("Device resumed\n");
-	return 0;
+	/*
+	 * No point of rolling back things in case of an error, as the best
+	 * we can do is to hope that things will still work (and disable RPM).
+	 */
+	i915_gem_init_swizzling(dev);
+	gen6_update_ring_freq(dev);
+
+	intel_runtime_pm_restore_interrupts(dev);
+	intel_reset_gt_powersave(dev);
+
+	if (ret)
+		DRM_ERROR("Runtime resume failed, disabling it (%d)\n", ret);
+	else
+		DRM_DEBUG_KMS("Device resumed\n");
+
+	return ret;
 }
 
 static const struct dev_pm_ops i915_pm_ops = {
@@ -954,8 +1477,8 @@ static const struct dev_pm_ops i915_pm_ops = {
 	.poweroff = i915_pm_poweroff,
 	.restore_early = i915_pm_resume_early,
 	.restore = i915_pm_resume,
-	.runtime_suspend = i915_runtime_suspend,
-	.runtime_resume = i915_runtime_resume,
+	.runtime_suspend = intel_runtime_suspend,
+	.runtime_resume = intel_runtime_resume,
 };
 
 static const struct vm_operations_struct i915_gem_vm_ops = {
@@ -1062,6 +1585,7 @@ static int __init i915_init(void)
 		driver.get_vblank_timestamp = NULL;
 #ifndef CONFIG_DRM_I915_UMS
 		/* Silently fail loading to not upset userspace. */
+		DRM_DEBUG_DRIVER("KMS and UMS disabled.\n");
 		return 0;
 #endif
 	}
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 388c028e223c..49414d30e8d4 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -35,11 +35,13 @@
 #include "i915_reg.h"
 #include "intel_bios.h"
 #include "intel_ringbuffer.h"
+#include "i915_gem_gtt.h"
 #include <linux/io-mapping.h>
 #include <linux/i2c.h>
 #include <linux/i2c-algo-bit.h>
 #include <drm/intel-gtt.h>
 #include <linux/backlight.h>
+#include <linux/hashtable.h>
 #include <linux/intel-iommu.h>
 #include <linux/kref.h>
 #include <linux/pm_qos.h>
@@ -91,7 +93,7 @@ enum port {
 };
 #define port_name(p) ((p) + 'A')
 
-#define I915_NUM_PHYS_VLV 1
+#define I915_NUM_PHYS_VLV 2
 
 enum dpio_channel {
 	DPIO_CH0,
@@ -162,6 +164,12 @@ enum hpd_pin {
 #define for_each_pipe(p) for ((p) = 0; (p) < INTEL_INFO(dev)->num_pipes; (p)++)
 #define for_each_sprite(p, s) for ((s) = 0; (s) < INTEL_INFO(dev)->num_sprites[(p)]; (s)++)
 
+#define for_each_crtc(dev, crtc) \
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
+
+#define for_each_intel_crtc(dev, intel_crtc) \
+	list_for_each_entry(intel_crtc, &dev->mode_config.crtc_list, base.head)
+
 #define for_each_encoder_on_crtc(dev, __crtc, intel_encoder) \
 	list_for_each_entry((intel_encoder), &(dev)->mode_config.encoder_list, base.head) \
 		if ((intel_encoder)->base.crtc == (__crtc))
@@ -171,6 +179,7 @@ enum hpd_pin {
 		if ((intel_connector)->base.encoder == (__encoder))
 
 struct drm_i915_private;
+struct i915_mmu_object;
 
 enum intel_dpll_id {
 	DPLL_ID_PRIVATE = -1, /* non-shared dpll in use */
@@ -312,7 +321,6 @@ struct drm_i915_error_state {
 	u32 gab_ctl;
 	u32 gfx_mode;
 	u32 extra_instdone[I915_NUM_INSTDONE_REG];
-	u32 pipestat[I915_MAX_PIPES];
 	u64 fence[I915_MAX_NUM_FENCES];
 	struct intel_overlay_error_state *overlay;
 	struct intel_display_error_state *display;
@@ -346,7 +354,7 @@ struct drm_i915_error_state {
 		u64 bbaddr;
 		u64 acthd;
 		u32 fault_reg;
-		u32 faddr;
+		u64 faddr;
 		u32 rc_psmi; /* sleep state */
 		u32 semaphore_mboxes[I915_NUM_RINGS - 1];
 
@@ -385,6 +393,7 @@ struct drm_i915_error_state {
 		u32 tiling:2;
 		u32 dirty:1;
 		u32 purgeable:1;
+		u32 userptr:1;
 		s32 ring:4;
 		u32 cache_level:3;
 	} **active_bo, **pinned_bo;
@@ -449,10 +458,11 @@ struct drm_i915_display_funcs {
 	int (*queue_flip)(struct drm_device *dev, struct drm_crtc *crtc,
 			  struct drm_framebuffer *fb,
 			  struct drm_i915_gem_object *obj,
+			  struct intel_engine_cs *ring,
 			  uint32_t flags);
-	int (*update_primary_plane)(struct drm_crtc *crtc,
-				    struct drm_framebuffer *fb,
-				    int x, int y);
+	void (*update_primary_plane)(struct drm_crtc *crtc,
+				     struct drm_framebuffer *fb,
+				     int x, int y);
 	void (*hpd_irq_setup)(struct drm_device *dev);
 	/* clock updates for mode set */
 	/* cursor updates */
@@ -545,6 +555,7 @@ struct intel_device_info {
 	int dpll_offsets[I915_MAX_PIPES];
 	int dpll_md_offsets[I915_MAX_PIPES];
 	int palette_offsets[I915_MAX_PIPES];
+	int cursor_offsets[I915_MAX_PIPES];
 };
 
 #undef DEFINE_FLAG
@@ -560,168 +571,6 @@ enum i915_cache_level {
 	I915_CACHE_WT, /* hsw:gt3e WriteThrough for scanouts */
 };
 
-typedef uint32_t gen6_gtt_pte_t;
-
-/**
- * A VMA represents a GEM BO that is bound into an address space. Therefore, a
- * VMA's presence cannot be guaranteed before binding, or after unbinding the
- * object into/from the address space.
- *
- * To make things as simple as possible (ie. no refcounting), a VMA's lifetime
- * will always be <= an objects lifetime. So object refcounting should cover us.
- */
-struct i915_vma {
-	struct drm_mm_node node;
-	struct drm_i915_gem_object *obj;
-	struct i915_address_space *vm;
-
-	/** This object's place on the active/inactive lists */
-	struct list_head mm_list;
-
-	struct list_head vma_link; /* Link in the object's VMA list */
-
-	/** This vma's place in the batchbuffer or on the eviction list */
-	struct list_head exec_list;
-
-	/**
-	 * Used for performing relocations during execbuffer insertion.
-	 */
-	struct hlist_node exec_node;
-	unsigned long exec_handle;
-	struct drm_i915_gem_exec_object2 *exec_entry;
-
-	/**
-	 * How many users have pinned this object in GTT space. The following
-	 * users can each hold at most one reference: pwrite/pread, pin_ioctl
-	 * (via user_pin_count), execbuffer (objects are not allowed multiple
-	 * times for the same batchbuffer), and the framebuffer code. When
-	 * switching/pageflipping, the framebuffer code has at most two buffers
-	 * pinned per crtc.
-	 *
-	 * In the worst case this is 1 + 1 + 1 + 2*2 = 7. That would fit into 3
-	 * bits with absolutely no headroom. So use 4 bits. */
-	unsigned int pin_count:4;
-#define DRM_I915_GEM_OBJECT_MAX_PIN_COUNT 0xf
-
-	/** Unmap an object from an address space. This usually consists of
-	 * setting the valid PTE entries to a reserved scratch page. */
-	void (*unbind_vma)(struct i915_vma *vma);
-	/* Map an object into an address space with the given cache flags. */
-#define GLOBAL_BIND (1<<0)
-	void (*bind_vma)(struct i915_vma *vma,
-			 enum i915_cache_level cache_level,
-			 u32 flags);
-};
-
-struct i915_address_space {
-	struct drm_mm mm;
-	struct drm_device *dev;
-	struct list_head global_link;
-	unsigned long start;		/* Start offset always 0 for dri2 */
-	size_t total;		/* size addr space maps (ex. 2GB for ggtt) */
-
-	struct {
-		dma_addr_t addr;
-		struct page *page;
-	} scratch;
-
-	/**
-	 * List of objects currently involved in rendering.
-	 *
-	 * Includes buffers having the contents of their GPU caches
-	 * flushed, not necessarily primitives.  last_rendering_seqno
-	 * represents when the rendering involved will be completed.
-	 *
-	 * A reference is held on the buffer while on this list.
-	 */
-	struct list_head active_list;
-
-	/**
-	 * LRU list of objects which are not in the ringbuffer and
-	 * are ready to unbind, but are still in the GTT.
-	 *
-	 * last_rendering_seqno is 0 while an object is in this list.
-	 *
-	 * A reference is not held on the buffer while on this list,
-	 * as merely being GTT-bound shouldn't prevent its being
-	 * freed, and we'll pull it off the list in the free path.
-	 */
-	struct list_head inactive_list;
-
-	/* FIXME: Need a more generic return type */
-	gen6_gtt_pte_t (*pte_encode)(dma_addr_t addr,
-				     enum i915_cache_level level,
-				     bool valid); /* Create a valid PTE */
-	void (*clear_range)(struct i915_address_space *vm,
-			    uint64_t start,
-			    uint64_t length,
-			    bool use_scratch);
-	void (*insert_entries)(struct i915_address_space *vm,
-			       struct sg_table *st,
-			       uint64_t start,
-			       enum i915_cache_level cache_level);
-	void (*cleanup)(struct i915_address_space *vm);
-};
-
-/* The Graphics Translation Table is the way in which GEN hardware translates a
- * Graphics Virtual Address into a Physical Address. In addition to the normal
- * collateral associated with any va->pa translations GEN hardware also has a
- * portion of the GTT which can be mapped by the CPU and remain both coherent
- * and correct (in cases like swizzling). That region is referred to as GMADR in
- * the spec.
- */
-struct i915_gtt {
-	struct i915_address_space base;
-	size_t stolen_size;		/* Total size of stolen memory */
-
-	unsigned long mappable_end;	/* End offset that we can CPU map */
-	struct io_mapping *mappable;	/* Mapping to our CPU mappable region */
-	phys_addr_t mappable_base;	/* PA of our GMADR */
-
-	/** "Graphics Stolen Memory" holds the global PTEs */
-	void __iomem *gsm;
-
-	bool do_idle_maps;
-
-	int mtrr;
-
-	/* global gtt ops */
-	int (*gtt_probe)(struct drm_device *dev, size_t *gtt_total,
-			  size_t *stolen, phys_addr_t *mappable_base,
-			  unsigned long *mappable_end);
-};
-#define gtt_total_entries(gtt) ((gtt).base.total >> PAGE_SHIFT)
-
-#define GEN8_LEGACY_PDPS 4
-struct i915_hw_ppgtt {
-	struct i915_address_space base;
-	struct kref ref;
-	struct drm_mm_node node;
-	unsigned num_pd_entries;
-	unsigned num_pd_pages; /* gen8+ */
-	union {
-		struct page **pt_pages;
-		struct page **gen8_pt_pages[GEN8_LEGACY_PDPS];
-	};
-	struct page *pd_pages;
-	union {
-		uint32_t pd_offset;
-		dma_addr_t pd_dma_addr[GEN8_LEGACY_PDPS];
-	};
-	union {
-		dma_addr_t *pt_dma_addr;
-		dma_addr_t *gen8_pt_dma_addr[4];
-	};
-
-	struct i915_hw_context *ctx;
-
-	int (*enable)(struct i915_hw_ppgtt *ppgtt);
-	int (*switch_mm)(struct i915_hw_ppgtt *ppgtt,
-			 struct intel_ring_buffer *ring,
-			 bool synchronous);
-	void (*debug_dump)(struct i915_hw_ppgtt *ppgtt, struct seq_file *m);
-};
-
 struct i915_ctx_hang_stats {
 	/* This context had batch pending when hang was declared */
 	unsigned batch_pending;
@@ -738,13 +587,13 @@ struct i915_ctx_hang_stats {
 
 /* This must match up with the value previously used for execbuf2.rsvd1. */
 #define DEFAULT_CONTEXT_ID 0
-struct i915_hw_context {
+struct intel_context {
 	struct kref ref;
 	int id;
 	bool is_initialized;
 	uint8_t remap_slice;
 	struct drm_i915_file_private *file_priv;
-	struct intel_ring_buffer *last_ring;
+	struct intel_engine_cs *last_ring;
 	struct drm_i915_gem_object *obj;
 	struct i915_ctx_hang_stats hang_stats;
 	struct i915_address_space *vm;
@@ -782,6 +631,10 @@ struct i915_fbc {
 	} no_fbc_reason;
 };
 
+struct i915_drrs {
+	struct intel_connector *connector;
+};
+
 struct i915_psr {
 	bool sink_support;
 	bool source_ok;
@@ -965,6 +818,67 @@ struct i915_suspend_saved_registers {
 	u32 savePCH_PORT_HOTPLUG;
 };
 
+struct vlv_s0ix_state {
+	/* GAM */
+	u32 wr_watermark;
+	u32 gfx_prio_ctrl;
+	u32 arb_mode;
+	u32 gfx_pend_tlb0;
+	u32 gfx_pend_tlb1;
+	u32 lra_limits[GEN7_LRA_LIMITS_REG_NUM];
+	u32 media_max_req_count;
+	u32 gfx_max_req_count;
+	u32 render_hwsp;
+	u32 ecochk;
+	u32 bsd_hwsp;
+	u32 blt_hwsp;
+	u32 tlb_rd_addr;
+
+	/* MBC */
+	u32 g3dctl;
+	u32 gsckgctl;
+	u32 mbctl;
+
+	/* GCP */
+	u32 ucgctl1;
+	u32 ucgctl3;
+	u32 rcgctl1;
+	u32 rcgctl2;
+	u32 rstctl;
+	u32 misccpctl;
+
+	/* GPM */
+	u32 gfxpause;
+	u32 rpdeuhwtc;
+	u32 rpdeuc;
+	u32 ecobus;
+	u32 pwrdwnupctl;
+	u32 rp_down_timeout;
+	u32 rp_deucsw;
+	u32 rcubmabdtmr;
+	u32 rcedata;
+	u32 spare2gh;
+
+	/* Display 1 CZ domain */
+	u32 gt_imr;
+	u32 gt_ier;
+	u32 pm_imr;
+	u32 pm_ier;
+	u32 gt_scratch[GEN7_GT_SCRATCH_REG_NUM];
+
+	/* GT SA CZ domain */
+	u32 tilectl;
+	u32 gt_fifoctl;
+	u32 gtlc_wake_ctrl;
+	u32 gtlc_survive;
+	u32 pmwgicz;
+
+	/* Display 2 CZ domain */
+	u32 gu_ctl0;
+	u32 gu_ctl1;
+	u32 clock_gate_dis2;
+};
+
 struct intel_gen6_power_mgmt {
 	/* work and pm_iir are protected by dev_priv->irq_lock */
 	struct work_struct work;
@@ -1074,6 +988,7 @@ struct i915_power_domains {
 	 * time are on. They are kept on until after the first modeset.
 	 */
 	bool init_power_on;
+	bool initializing;
 	int power_well_count;
 
 	struct mutex lock;
@@ -1132,7 +1047,8 @@ struct i915_gem_mm {
 	/** PPGTT used for aliasing the PPGTT with the GTT */
 	struct i915_hw_ppgtt *aliasing_ppgtt;
 
-	struct shrinker inactive_shrinker;
+	struct notifier_block oom_notifier;
+	struct shrinker shrinker;
 	bool shrinker_no_lock_stealing;
 
 	/** LRU list of objects with fence regs on them. */
@@ -1170,6 +1086,9 @@ struct i915_gem_mm {
 	 */
 	bool busy;
 
+	/* the indicator for dispatch video commands on two BSD rings */
+	int bsd_ring_dispatch_index;
+
 	/** Bit 6 swizzling required for X tiling */
 	uint32_t bit_6_swizzle_x;
 	/** Bit 6 swizzling required for Y tiling */
@@ -1245,8 +1164,12 @@ struct i915_gpu_error {
 	 */
 	wait_queue_head_t reset_queue;
 
-	/* For gpu hang simulation. */
-	unsigned int stop_rings;
+	/* Userspace knobs for gpu hang simulation;
+	 * combines both a ring mask, and extra flags
+	 */
+	u32 stop_rings;
+#define I915_STOP_RING_ALLOW_BAN       (1 << 31)
+#define I915_STOP_RING_ALLOW_WARN      (1 << 30)
 
 	/* For missed irq/seqno simulation. */
 	unsigned int test_irq_rings;
@@ -1266,6 +1189,12 @@ struct ddi_vbt_port_info {
 	uint8_t supports_dp:1;
 };
 
+enum drrs_support_type {
+	DRRS_NOT_SUPPORTED = 0,
+	STATIC_DRRS_SUPPORT = 1,
+	SEAMLESS_DRRS_SUPPORT = 2
+};
+
 struct intel_vbt_data {
 	struct drm_display_mode *lfp_lvds_vbt_mode; /* if any */
 	struct drm_display_mode *sdvo_lvds_vbt_mode; /* if any */
@@ -1278,9 +1207,12 @@ struct intel_vbt_data {
 	unsigned int lvds_use_ssc:1;
 	unsigned int display_clock_mode:1;
 	unsigned int fdi_rx_polarity_inverted:1;
+	unsigned int has_mipi:1;
 	int lvds_ssc_freq;
 	unsigned int bios_lvds_val; /* initial [PCH_]LVDS reg val in VBIOS */
 
+	enum drrs_support_type drrs_type;
+
 	/* eDP */
 	int edp_rate;
 	int edp_lanes;
@@ -1299,7 +1231,14 @@ struct intel_vbt_data {
 
 	/* MIPI DSI */
 	struct {
+		u16 port;
 		u16 panel_id;
+		struct mipi_config *config;
+		struct mipi_pps_data *pps;
+		u8 seq_version;
+		u32 size;
+		u8 *data;
+		u8 *sequence[MIPI_SEQ_MAX];
 	} dsi;
 
 	int crt_ddc_pin;
@@ -1351,23 +1290,13 @@ struct ilk_wm_values {
  * goes back to false exactly before we reenable the IRQs. We use this variable
  * to check if someone is trying to enable/disable IRQs while they're supposed
  * to be disabled. This shouldn't happen and we'll print some error messages in
- * case it happens, but if it actually happens we'll also update the variables
- * inside struct regsave so when we restore the IRQs they will contain the
- * latest expected values.
+ * case it happens.
  *
  * For more, read the Documentation/power/runtime_pm.txt.
  */
 struct i915_runtime_pm {
 	bool suspended;
 	bool irqs_disabled;
-
-	struct {
-		uint32_t deimr;
-		uint32_t sdeimr;
-		uint32_t gtimr;
-		uint32_t gtier;
-		uint32_t gen6_pmimr;
-	} regsave;
 };
 
 enum intel_pipe_crc_source {
@@ -1400,7 +1329,7 @@ struct intel_pipe_crc {
 	wait_queue_head_t wq;
 };
 
-typedef struct drm_i915_private {
+struct drm_i915_private {
 	struct drm_device *dev;
 	struct kmem_cache *slab;
 
@@ -1424,10 +1353,13 @@ typedef struct drm_i915_private {
 	 */
 	uint32_t gpio_mmio_base;
 
+	/* MMIO base address for MIPI regs */
+	uint32_t mipi_mmio_base;
+
 	wait_queue_head_t gmbus_wait_queue;
 
 	struct pci_dev *bridge_dev;
-	struct intel_ring_buffer ring[I915_NUM_RINGS];
+	struct intel_engine_cs ring[I915_NUM_RINGS];
 	uint32_t last_seqno, next_seqno;
 
 	drm_dma_handle_t *status_page_dmah;
@@ -1469,6 +1401,7 @@ typedef struct drm_i915_private {
 	struct timer_list hotplug_reenable_timer;
 
 	struct i915_fbc fbc;
+	struct i915_drrs drrs;
 	struct intel_opregion opregion;
 	struct intel_vbt_data vbt;
 
@@ -1486,6 +1419,7 @@ typedef struct drm_i915_private {
 	int num_fence_regs; /* 8 on pre-965, 16 otherwise */
 
 	unsigned int fsb_freq, mem_freq, is_ddr3;
+	unsigned int vlv_cdclk_freq;
 
 	/**
 	 * wq - Driver workqueue for GEM.
@@ -1509,9 +1443,12 @@ typedef struct drm_i915_private {
 	struct mutex modeset_restore_lock;
 
 	struct list_head vm_list; /* Global list of all address spaces */
-	struct i915_gtt gtt; /* VMA representing the global address space */
+	struct i915_gtt gtt; /* VM representing the global address space */
 
 	struct i915_gem_mm mm;
+#if defined(CONFIG_MMU_NOTIFIER)
+	DECLARE_HASHTABLE(mmu_notifiers, 7);
+#endif
 
 	/* Kernel Modesetting */
 
@@ -1580,6 +1517,7 @@ typedef struct drm_i915_private {
 
 	u32 suspend_count;
 	struct i915_suspend_saved_registers regfile;
+	struct vlv_s0ix_state vlv_s0ix_state;
 
 	struct {
 		/*
@@ -1605,7 +1543,12 @@ typedef struct drm_i915_private {
 	struct i915_dri1_state dri1;
 	/* Old ums support infrastructure, same warning applies. */
 	struct i915_ums_state ums;
-} drm_i915_private_t;
+
+	/*
+	 * NOTE: This is the dri1/ums dungeon, don't add stuff here. Your patch
+	 * will be rejected. Instead look for a better place.
+	 */
+};
 
 static inline struct drm_i915_private *to_i915(const struct drm_device *dev)
 {
@@ -1642,6 +1585,8 @@ struct drm_i915_gem_object_ops {
 	 */
 	int (*get_pages)(struct drm_i915_gem_object *);
 	void (*put_pages)(struct drm_i915_gem_object *);
+	int (*dmabuf_export)(struct drm_i915_gem_object *);
+	void (*release)(struct drm_i915_gem_object *);
 };
 
 struct drm_i915_gem_object {
@@ -1732,7 +1677,7 @@ struct drm_i915_gem_object {
 	void *dma_buf_vmapping;
 	int vmapping_count;
 
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 
 	/** Breadcrumb of last rendering to the buffer. */
 	uint32_t last_read_seqno;
@@ -1755,8 +1700,20 @@ struct drm_i915_gem_object {
 
 	/** for phy allocated objects */
 	drm_dma_handle_t *phys_handle;
-};
 
+	union {
+		struct i915_gem_userptr {
+			uintptr_t ptr;
+			unsigned read_only :1;
+			unsigned workers :4;
+#define I915_GEM_USERPTR_MAX_WORKERS 15
+
+			struct mm_struct *mm;
+			struct i915_mmu_object *mn;
+			struct work_struct *work;
+		} userptr;
+	};
+};
 #define to_intel_bo(x) container_of(x, struct drm_i915_gem_object, base)
 
 /**
@@ -1771,7 +1728,7 @@ struct drm_i915_gem_object {
  */
 struct drm_i915_gem_request {
 	/** On Which ring this request was generated */
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 
 	/** GEM sequence number associated with this request. */
 	uint32_t seqno;
@@ -1783,7 +1740,7 @@ struct drm_i915_gem_request {
 	u32 tail;
 
 	/** Context related to this request */
-	struct i915_hw_context *ctx;
+	struct intel_context *ctx;
 
 	/** Batch buffer related to this request if any */
 	struct drm_i915_gem_object *batch_obj;
@@ -1810,8 +1767,8 @@ struct drm_i915_file_private {
 	} mm;
 	struct idr context_idr;
 
-	struct i915_hw_context *private_default_ctx;
 	atomic_t rps_wait_boost;
+	struct  intel_engine_cs *bsd_ring;
 };
 
 /*
@@ -1879,11 +1836,17 @@ struct drm_i915_cmd_descriptor {
 	 * the expected value, the parser rejects it. Only valid if flags has
 	 * the CMD_DESC_BITMASK bit set. Only entries where mask is non-zero
 	 * are valid.
+	 *
+	 * If the check specifies a non-zero condition_mask then the parser
+	 * only performs the check when the bits specified by condition_mask
+	 * are non-zero.
 	 */
 	struct {
 		u32 offset;
 		u32 mask;
 		u32 expected;
+		u32 condition_offset;
+		u32 condition_mask;
 	} bits[MAX_CMD_DESC_BITMASKS];
 };
 
@@ -1925,8 +1888,9 @@ struct drm_i915_cmd_table {
 				 (dev)->pdev->device == 0x0106 || \
 				 (dev)->pdev->device == 0x010A)
 #define IS_VALLEYVIEW(dev)	(INTEL_INFO(dev)->is_valleyview)
+#define IS_CHERRYVIEW(dev)	(INTEL_INFO(dev)->is_valleyview && IS_GEN8(dev))
 #define IS_HASWELL(dev)	(INTEL_INFO(dev)->is_haswell)
-#define IS_BROADWELL(dev)	(INTEL_INFO(dev)->gen == 8)
+#define IS_BROADWELL(dev)	(!INTEL_INFO(dev)->is_valleyview && IS_GEN8(dev))
 #define IS_MOBILE(dev)		(INTEL_INFO(dev)->is_mobile)
 #define IS_HSW_EARLY_SDV(dev)	(IS_HASWELL(dev) && \
 				 ((dev)->pdev->device & 0xFF00) == 0x0C00)
@@ -1962,17 +1926,21 @@ struct drm_i915_cmd_table {
 #define BSD_RING		(1<<VCS)
 #define BLT_RING		(1<<BCS)
 #define VEBOX_RING		(1<<VECS)
-#define HAS_BSD(dev)            (INTEL_INFO(dev)->ring_mask & BSD_RING)
-#define HAS_BLT(dev)            (INTEL_INFO(dev)->ring_mask & BLT_RING)
-#define HAS_VEBOX(dev)            (INTEL_INFO(dev)->ring_mask & VEBOX_RING)
-#define HAS_LLC(dev)            (INTEL_INFO(dev)->has_llc)
-#define HAS_WT(dev)            (IS_HASWELL(dev) && to_i915(dev)->ellc_size)
+#define BSD2_RING		(1<<VCS2)
+#define HAS_BSD(dev)		(INTEL_INFO(dev)->ring_mask & BSD_RING)
+#define HAS_BSD2(dev)		(INTEL_INFO(dev)->ring_mask & BSD2_RING)
+#define HAS_BLT(dev)		(INTEL_INFO(dev)->ring_mask & BLT_RING)
+#define HAS_VEBOX(dev)		(INTEL_INFO(dev)->ring_mask & VEBOX_RING)
+#define HAS_LLC(dev)		(INTEL_INFO(dev)->has_llc)
+#define HAS_WT(dev)		((IS_HASWELL(dev) || IS_BROADWELL(dev)) && \
+				 to_i915(dev)->ellc_size)
 #define I915_NEED_GFX_HWS(dev)	(INTEL_INFO(dev)->need_gfx_hws)
 
 #define HAS_HW_CONTEXTS(dev)	(INTEL_INFO(dev)->gen >= 6)
-#define HAS_ALIASING_PPGTT(dev)	(INTEL_INFO(dev)->gen >= 6 && !IS_VALLEYVIEW(dev))
-#define HAS_PPGTT(dev)		(INTEL_INFO(dev)->gen >= 7 && !IS_VALLEYVIEW(dev) \
-				 && !IS_BROADWELL(dev))
+#define HAS_ALIASING_PPGTT(dev)	(INTEL_INFO(dev)->gen >= 6 && \
+				 (!IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev)))
+#define HAS_PPGTT(dev)		(INTEL_INFO(dev)->gen >= 7 \
+				 && !IS_GEN8(dev))
 #define USES_PPGTT(dev)		intel_enable_ppgtt(dev, false)
 #define USES_FULL_PPGTT(dev)	intel_enable_ppgtt(dev, true)
 
@@ -2010,8 +1978,8 @@ struct drm_i915_cmd_table {
 #define HAS_DDI(dev)		(INTEL_INFO(dev)->has_ddi)
 #define HAS_FPGA_DBG_UNCLAIMED(dev)	(INTEL_INFO(dev)->has_fpga_dbg)
 #define HAS_PSR(dev)		(IS_HASWELL(dev) || IS_BROADWELL(dev))
-#define HAS_PC8(dev)		(IS_HASWELL(dev)) /* XXX HSW:ULX */
-#define HAS_RUNTIME_PM(dev)	(IS_HASWELL(dev))
+#define HAS_RUNTIME_PM(dev)	(IS_GEN6(dev) || IS_HASWELL(dev) || \
+				 IS_BROADWELL(dev) || IS_VALLEYVIEW(dev))
 
 #define INTEL_PCH_DEVICE_ID_MASK		0xff00
 #define INTEL_PCH_IBX_DEVICE_ID_TYPE		0x3b00
@@ -2068,6 +2036,7 @@ struct i915_params {
 	bool prefault_disable;
 	bool reset;
 	bool disable_display;
+	bool disable_vtd_wa;
 };
 extern struct i915_params i915 __read_mostly;
 
@@ -2096,6 +2065,7 @@ extern unsigned long i915_chipset_val(struct drm_i915_private *dev_priv);
 extern unsigned long i915_mch_val(struct drm_i915_private *dev_priv);
 extern unsigned long i915_gfx_val(struct drm_i915_private *dev_priv);
 extern void i915_update_gfx_val(struct drm_i915_private *dev_priv);
+int vlv_force_gfx_clock(struct drm_i915_private *dev_priv, bool on);
 
 extern void intel_console_resume(struct work_struct *work);
 
@@ -2170,6 +2140,9 @@ int i915_gem_set_tiling(struct drm_device *dev, void *data,
 			struct drm_file *file_priv);
 int i915_gem_get_tiling(struct drm_device *dev, void *data,
 			struct drm_file *file_priv);
+int i915_gem_init_userptr(struct drm_device *dev);
+int i915_gem_userptr_ioctl(struct drm_device *dev, void *data,
+			   struct drm_file *file);
 int i915_gem_get_aperture_ioctl(struct drm_device *dev, void *data,
 				struct drm_file *file_priv);
 int i915_gem_wait_ioctl(struct drm_device *dev, void *data,
@@ -2227,9 +2200,9 @@ static inline void i915_gem_object_unpin_pages(struct drm_i915_gem_object *obj)
 
 int __must_check i915_mutex_lock_interruptible(struct drm_device *dev);
 int i915_gem_object_sync(struct drm_i915_gem_object *obj,
-			 struct intel_ring_buffer *to);
+			 struct intel_engine_cs *to);
 void i915_vma_move_to_active(struct i915_vma *vma,
-			     struct intel_ring_buffer *ring);
+			     struct intel_engine_cs *ring);
 int i915_gem_dumb_create(struct drm_file *file_priv,
 			 struct drm_device *dev,
 			 struct drm_mode_create_dumb *args);
@@ -2249,31 +2222,14 @@ int __must_check i915_gem_set_seqno(struct drm_device *dev, u32 seqno);
 int __must_check i915_gem_object_get_fence(struct drm_i915_gem_object *obj);
 int __must_check i915_gem_object_put_fence(struct drm_i915_gem_object *obj);
 
-static inline bool
-i915_gem_object_pin_fence(struct drm_i915_gem_object *obj)
-{
-	if (obj->fence_reg != I915_FENCE_REG_NONE) {
-		struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
-		dev_priv->fence_regs[obj->fence_reg].pin_count++;
-		return true;
-	} else
-		return false;
-}
-
-static inline void
-i915_gem_object_unpin_fence(struct drm_i915_gem_object *obj)
-{
-	if (obj->fence_reg != I915_FENCE_REG_NONE) {
-		struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
-		WARN_ON(dev_priv->fence_regs[obj->fence_reg].pin_count <= 0);
-		dev_priv->fence_regs[obj->fence_reg].pin_count--;
-	}
-}
+bool i915_gem_object_pin_fence(struct drm_i915_gem_object *obj);
+void i915_gem_object_unpin_fence(struct drm_i915_gem_object *obj);
 
 struct drm_i915_gem_request *
-i915_gem_find_active_request(struct intel_ring_buffer *ring);
+i915_gem_find_active_request(struct intel_engine_cs *ring);
 
 bool i915_gem_retire_requests(struct drm_device *dev);
+void i915_gem_retire_requests_ring(struct intel_engine_cs *ring);
 int __must_check i915_gem_check_wedge(struct i915_gpu_error *error,
 				      bool interruptible);
 static inline bool i915_reset_in_progress(struct i915_gpu_error *error)
@@ -2292,23 +2248,35 @@ static inline u32 i915_reset_count(struct i915_gpu_error *error)
 	return ((atomic_read(&error->reset_counter) & ~I915_WEDGED) + 1) / 2;
 }
 
+static inline bool i915_stop_ring_allow_ban(struct drm_i915_private *dev_priv)
+{
+	return dev_priv->gpu_error.stop_rings == 0 ||
+		dev_priv->gpu_error.stop_rings & I915_STOP_RING_ALLOW_BAN;
+}
+
+static inline bool i915_stop_ring_allow_warn(struct drm_i915_private *dev_priv)
+{
+	return dev_priv->gpu_error.stop_rings == 0 ||
+		dev_priv->gpu_error.stop_rings & I915_STOP_RING_ALLOW_WARN;
+}
+
 void i915_gem_reset(struct drm_device *dev);
 bool i915_gem_clflush_object(struct drm_i915_gem_object *obj, bool force);
 int __must_check i915_gem_object_finish_gpu(struct drm_i915_gem_object *obj);
 int __must_check i915_gem_init(struct drm_device *dev);
 int __must_check i915_gem_init_hw(struct drm_device *dev);
-int i915_gem_l3_remap(struct intel_ring_buffer *ring, int slice);
+int i915_gem_l3_remap(struct intel_engine_cs *ring, int slice);
 void i915_gem_init_swizzling(struct drm_device *dev);
 void i915_gem_cleanup_ringbuffer(struct drm_device *dev);
 int __must_check i915_gpu_idle(struct drm_device *dev);
 int __must_check i915_gem_suspend(struct drm_device *dev);
-int __i915_add_request(struct intel_ring_buffer *ring,
+int __i915_add_request(struct intel_engine_cs *ring,
 		       struct drm_file *file,
 		       struct drm_i915_gem_object *batch_obj,
 		       u32 *seqno);
 #define i915_add_request(ring, seqno) \
 	__i915_add_request(ring, NULL, NULL, seqno)
-int __must_check i915_wait_seqno(struct intel_ring_buffer *ring,
+int __must_check i915_wait_seqno(struct intel_engine_cs *ring,
 				 uint32_t seqno);
 int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf);
 int __must_check
@@ -2319,7 +2287,7 @@ i915_gem_object_set_to_cpu_domain(struct drm_i915_gem_object *obj, bool write);
 int __must_check
 i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
 				     u32 alignment,
-				     struct intel_ring_buffer *pipelined);
+				     struct intel_engine_cs *pipelined);
 void i915_gem_object_unpin_from_display_plane(struct drm_i915_gem_object *obj);
 int i915_gem_object_attach_phys(struct drm_i915_gem_object *obj,
 				int align);
@@ -2416,22 +2384,22 @@ void i915_gem_context_reset(struct drm_device *dev);
 int i915_gem_context_open(struct drm_device *dev, struct drm_file *file);
 int i915_gem_context_enable(struct drm_i915_private *dev_priv);
 void i915_gem_context_close(struct drm_device *dev, struct drm_file *file);
-int i915_switch_context(struct intel_ring_buffer *ring,
-			struct i915_hw_context *to);
-struct i915_hw_context *
+int i915_switch_context(struct intel_engine_cs *ring,
+			struct intel_context *to);
+struct intel_context *
 i915_gem_context_get(struct drm_i915_file_private *file_priv, u32 id);
 void i915_gem_context_free(struct kref *ctx_ref);
-static inline void i915_gem_context_reference(struct i915_hw_context *ctx)
+static inline void i915_gem_context_reference(struct intel_context *ctx)
 {
 	kref_get(&ctx->ref);
 }
 
-static inline void i915_gem_context_unreference(struct i915_hw_context *ctx)
+static inline void i915_gem_context_unreference(struct intel_context *ctx)
 {
 	kref_put(&ctx->ref, i915_gem_context_free);
 }
 
-static inline bool i915_gem_context_is_default(const struct i915_hw_context *c)
+static inline bool i915_gem_context_is_default(const struct intel_context *c)
 {
 	return c->id == DEFAULT_CONTEXT_ID;
 }
@@ -2441,6 +2409,8 @@ int i915_gem_context_create_ioctl(struct drm_device *dev, void *data,
 int i915_gem_context_destroy_ioctl(struct drm_device *dev, void *data,
 				   struct drm_file *file);
 
+/* i915_gem_render_state.c */
+int i915_gem_render_state_init(struct intel_engine_cs *ring);
 /* i915_gem_evict.c */
 int __must_check i915_gem_evict_something(struct drm_device *dev,
 					  struct i915_address_space *vm,
@@ -2453,23 +2423,12 @@ int __must_check i915_gem_evict_something(struct drm_device *dev,
 int i915_gem_evict_vm(struct i915_address_space *vm, bool do_idle);
 int i915_gem_evict_everything(struct drm_device *dev);
 
-/* i915_gem_gtt.c */
-void i915_check_and_clear_faults(struct drm_device *dev);
-void i915_gem_suspend_gtt_mappings(struct drm_device *dev);
-void i915_gem_restore_gtt_mappings(struct drm_device *dev);
-int __must_check i915_gem_gtt_prepare_object(struct drm_i915_gem_object *obj);
-void i915_gem_gtt_finish_object(struct drm_i915_gem_object *obj);
-void i915_gem_init_global_gtt(struct drm_device *dev);
-void i915_gem_setup_global_gtt(struct drm_device *dev, unsigned long start,
-			       unsigned long mappable_end, unsigned long end);
-int i915_gem_gtt_init(struct drm_device *dev);
+/* belongs in i915_gem_gtt.h */
 static inline void i915_gem_chipset_flush(struct drm_device *dev)
 {
 	if (INTEL_INFO(dev)->gen < 6)
 		intel_gtt_chipset_flush();
 }
-int i915_gem_init_ppgtt(struct drm_device *dev, struct i915_hw_ppgtt *ppgtt);
-bool intel_enable_ppgtt(struct drm_device *dev, bool full);
 
 /* i915_gem_stolen.c */
 int i915_gem_init_stolen(struct drm_device *dev);
@@ -2537,9 +2496,11 @@ void i915_get_extra_instdone(struct drm_device *dev, uint32_t *instdone);
 const char *i915_cache_level_str(int type);
 
 /* i915_cmd_parser.c */
-void i915_cmd_parser_init_ring(struct intel_ring_buffer *ring);
-bool i915_needs_cmd_parser(struct intel_ring_buffer *ring);
-int i915_parse_cmds(struct intel_ring_buffer *ring,
+int i915_cmd_parser_get_version(void);
+int i915_cmd_parser_init_ring(struct intel_engine_cs *ring);
+void i915_cmd_parser_fini_ring(struct intel_engine_cs *ring);
+bool i915_needs_cmd_parser(struct intel_engine_cs *ring);
+int i915_parse_cmds(struct intel_engine_cs *ring,
 		    struct drm_i915_gem_object *batch_obj,
 		    u32 batch_start_offset,
 		    bool is_master);
@@ -2688,20 +2649,6 @@ void vlv_flisdsi_write(struct drm_i915_private *dev_priv, u32 reg, u32 val);
 int vlv_gpu_freq(struct drm_i915_private *dev_priv, int val);
 int vlv_freq_opcode(struct drm_i915_private *dev_priv, int val);
 
-void vlv_force_wake_get(struct drm_i915_private *dev_priv, int fw_engine);
-void vlv_force_wake_put(struct drm_i915_private *dev_priv, int fw_engine);
-
-#define FORCEWAKE_VLV_RENDER_RANGE_OFFSET(reg) \
-	(((reg) >= 0x2000 && (reg) < 0x4000) ||\
-	((reg) >= 0x5000 && (reg) < 0x8000) ||\
-	((reg) >= 0xB000 && (reg) < 0x12000) ||\
-	((reg) >= 0x2E000 && (reg) < 0x30000))
-
-#define FORCEWAKE_VLV_MEDIA_RANGE_OFFSET(reg)\
-	(((reg) >= 0x12000 && (reg) < 0x14000) ||\
-	((reg) >= 0x22000 && (reg) < 0x24000) ||\
-	((reg) >= 0x30000 && (reg) < 0x40000))
-
 #define FORCEWAKE_RENDER	(1 << 0)
 #define FORCEWAKE_MEDIA		(1 << 1)
 #define FORCEWAKE_ALL		(FORCEWAKE_RENDER | FORCEWAKE_MEDIA)
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 3326770c9ed2..f36126383d26 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -31,6 +31,7 @@
 #include "i915_drv.h"
 #include "i915_trace.h"
 #include "intel_drv.h"
+#include <linux/oom.h>
 #include <linux/shmem_fs.h>
 #include <linux/slab.h>
 #include <linux/swap.h>
@@ -43,6 +44,8 @@ static void i915_gem_object_flush_cpu_write_domain(struct drm_i915_gem_object *o
 static __must_check int
 i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj,
 			       bool readonly);
+static void
+i915_gem_object_retire(struct drm_i915_gem_object *obj);
 
 static void i915_gem_write_fence(struct drm_device *dev, int reg,
 				 struct drm_i915_gem_object *obj);
@@ -50,14 +53,15 @@ static void i915_gem_object_update_fence(struct drm_i915_gem_object *obj,
 					 struct drm_i915_fence_reg *fence,
 					 bool enable);
 
-static unsigned long i915_gem_inactive_count(struct shrinker *shrinker,
+static unsigned long i915_gem_shrinker_count(struct shrinker *shrinker,
 					     struct shrink_control *sc);
-static unsigned long i915_gem_inactive_scan(struct shrinker *shrinker,
+static unsigned long i915_gem_shrinker_scan(struct shrinker *shrinker,
 					    struct shrink_control *sc);
+static int i915_gem_shrinker_oom(struct notifier_block *nb,
+				 unsigned long event,
+				 void *ptr);
 static unsigned long i915_gem_purge(struct drm_i915_private *dev_priv, long target);
 static unsigned long i915_gem_shrink_all(struct drm_i915_private *dev_priv);
-static void i915_gem_object_truncate(struct drm_i915_gem_object *obj);
-static void i915_gem_retire_requests_ring(struct intel_ring_buffer *ring);
 
 static bool cpu_cache_is_coherent(struct drm_device *dev,
 				  enum i915_cache_level level)
@@ -470,6 +474,8 @@ int i915_gem_obj_prepare_shmem_read(struct drm_i915_gem_object *obj,
 		ret = i915_gem_object_wait_rendering(obj, true);
 		if (ret)
 			return ret;
+
+		i915_gem_object_retire(obj);
 	}
 
 	ret = i915_gem_object_get_pages(obj);
@@ -885,6 +891,8 @@ i915_gem_shmem_pwrite(struct drm_device *dev,
 		ret = i915_gem_object_wait_rendering(obj, false);
 		if (ret)
 			return ret;
+
+		i915_gem_object_retire(obj);
 	}
 	/* Same trick applies to invalidate partially written cachelines read
 	 * before writing. */
@@ -1088,7 +1096,7 @@ i915_gem_check_wedge(struct i915_gpu_error *error,
  * equal.
  */
 static int
-i915_gem_check_olr(struct intel_ring_buffer *ring, u32 seqno)
+i915_gem_check_olr(struct intel_engine_cs *ring, u32 seqno)
 {
 	int ret;
 
@@ -1107,7 +1115,7 @@ static void fake_irq(unsigned long data)
 }
 
 static bool missed_irq(struct drm_i915_private *dev_priv,
-		       struct intel_ring_buffer *ring)
+		       struct intel_engine_cs *ring)
 {
 	return test_bit(ring->id, &dev_priv->gpu_error.missed_irq_rings);
 }
@@ -1138,7 +1146,7 @@ static bool can_wait_boost(struct drm_i915_file_private *file_priv)
  * Returns 0 if the seqno was found within the alloted time. Else returns the
  * errno with remaining time filled in timeout argument.
  */
-static int __wait_seqno(struct intel_ring_buffer *ring, u32 seqno,
+static int __wait_seqno(struct intel_engine_cs *ring, u32 seqno,
 			unsigned reset_counter,
 			bool interruptible,
 			struct timespec *timeout,
@@ -1245,7 +1253,7 @@ static int __wait_seqno(struct intel_ring_buffer *ring, u32 seqno,
  * request and object lists appropriately for that event.
  */
 int
-i915_wait_seqno(struct intel_ring_buffer *ring, uint32_t seqno)
+i915_wait_seqno(struct intel_engine_cs *ring, uint32_t seqno)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1270,9 +1278,10 @@ i915_wait_seqno(struct intel_ring_buffer *ring, uint32_t seqno)
 
 static int
 i915_gem_object_wait_rendering__tail(struct drm_i915_gem_object *obj,
-				     struct intel_ring_buffer *ring)
+				     struct intel_engine_cs *ring)
 {
-	i915_gem_retire_requests_ring(ring);
+	if (!obj->active)
+		return 0;
 
 	/* Manually manage the write flush as we may have not yet
 	 * retired the buffer.
@@ -1282,7 +1291,6 @@ i915_gem_object_wait_rendering__tail(struct drm_i915_gem_object *obj,
 	 * we know we have passed the last write.
 	 */
 	obj->last_write_seqno = 0;
-	obj->base.write_domain &= ~I915_GEM_GPU_DOMAINS;
 
 	return 0;
 }
@@ -1295,7 +1303,7 @@ static __must_check int
 i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj,
 			       bool readonly)
 {
-	struct intel_ring_buffer *ring = obj->ring;
+	struct intel_engine_cs *ring = obj->ring;
 	u32 seqno;
 	int ret;
 
@@ -1320,7 +1328,7 @@ i915_gem_object_wait_rendering__nonblocking(struct drm_i915_gem_object *obj,
 {
 	struct drm_device *dev = obj->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = obj->ring;
+	struct intel_engine_cs *ring = obj->ring;
 	unsigned reset_counter;
 	u32 seqno;
 	int ret;
@@ -1536,7 +1544,7 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 
 	/* Access to snoopable pages through the GTT is incoherent. */
 	if (obj->cache_level != I915_CACHE_NONE && !HAS_LLC(dev)) {
-		ret = -EINVAL;
+		ret = -EFAULT;
 		goto unlock;
 	}
 
@@ -1803,12 +1811,16 @@ i915_gem_mmap_gtt_ioctl(struct drm_device *dev, void *data,
 	return i915_gem_mmap_gtt(file, dev, args->handle, &args->offset);
 }
 
+static inline int
+i915_gem_object_is_purgeable(struct drm_i915_gem_object *obj)
+{
+	return obj->madv == I915_MADV_DONTNEED;
+}
+
 /* Immediately discard the backing storage */
 static void
 i915_gem_object_truncate(struct drm_i915_gem_object *obj)
 {
-	struct inode *inode;
-
 	i915_gem_object_free_mmap_offset(obj);
 
 	if (obj->base.filp == NULL)
@@ -1819,16 +1831,28 @@ i915_gem_object_truncate(struct drm_i915_gem_object *obj)
 	 * To do this we must instruct the shmfs to drop all of its
 	 * backing pages, *now*.
 	 */
-	inode = file_inode(obj->base.filp);
-	shmem_truncate_range(inode, 0, (loff_t)-1);
-
+	shmem_truncate_range(file_inode(obj->base.filp), 0, (loff_t)-1);
 	obj->madv = __I915_MADV_PURGED;
 }
 
-static inline int
-i915_gem_object_is_purgeable(struct drm_i915_gem_object *obj)
+/* Try to discard unwanted pages */
+static void
+i915_gem_object_invalidate(struct drm_i915_gem_object *obj)
 {
-	return obj->madv == I915_MADV_DONTNEED;
+	struct address_space *mapping;
+
+	switch (obj->madv) {
+	case I915_MADV_DONTNEED:
+		i915_gem_object_truncate(obj);
+	case __I915_MADV_PURGED:
+		return;
+	}
+
+	if (obj->base.filp == NULL)
+		return;
+
+	mapping = file_inode(obj->base.filp)->i_mapping,
+	invalidate_mapping_pages(mapping, 0, (loff_t)-1);
 }
 
 static void
@@ -1893,8 +1917,7 @@ i915_gem_object_put_pages(struct drm_i915_gem_object *obj)
 	ops->put_pages(obj);
 	obj->pages = NULL;
 
-	if (i915_gem_object_is_purgeable(obj))
-		i915_gem_object_truncate(obj);
+	i915_gem_object_invalidate(obj);
 
 	return 0;
 }
@@ -1903,58 +1926,58 @@ static unsigned long
 __i915_gem_shrink(struct drm_i915_private *dev_priv, long target,
 		  bool purgeable_only)
 {
-	struct list_head still_bound_list;
-	struct drm_i915_gem_object *obj, *next;
+	struct list_head still_in_list;
+	struct drm_i915_gem_object *obj;
 	unsigned long count = 0;
 
-	list_for_each_entry_safe(obj, next,
-				 &dev_priv->mm.unbound_list,
-				 global_list) {
-		if ((i915_gem_object_is_purgeable(obj) || !purgeable_only) &&
-		    i915_gem_object_put_pages(obj) == 0) {
-			count += obj->base.size >> PAGE_SHIFT;
-			if (count >= target)
-				return count;
-		}
-	}
-
 	/*
-	 * As we may completely rewrite the bound list whilst unbinding
+	 * As we may completely rewrite the (un)bound list whilst unbinding
 	 * (due to retiring requests) we have to strictly process only
 	 * one element of the list at the time, and recheck the list
 	 * on every iteration.
+	 *
+	 * In particular, we must hold a reference whilst removing the
+	 * object as we may end up waiting for and/or retiring the objects.
+	 * This might release the final reference (held by the active list)
+	 * and result in the object being freed from under us. This is
+	 * similar to the precautions the eviction code must take whilst
+	 * removing objects.
+	 *
+	 * Also note that although these lists do not hold a reference to
+	 * the object we can safely grab one here: The final object
+	 * unreferencing and the bound_list are both protected by the
+	 * dev->struct_mutex and so we won't ever be able to observe an
+	 * object on the bound_list with a reference count equals 0.
 	 */
-	INIT_LIST_HEAD(&still_bound_list);
+	INIT_LIST_HEAD(&still_in_list);
+	while (count < target && !list_empty(&dev_priv->mm.unbound_list)) {
+		obj = list_first_entry(&dev_priv->mm.unbound_list,
+				       typeof(*obj), global_list);
+		list_move_tail(&obj->global_list, &still_in_list);
+
+		if (!i915_gem_object_is_purgeable(obj) && purgeable_only)
+			continue;
+
+		drm_gem_object_reference(&obj->base);
+
+		if (i915_gem_object_put_pages(obj) == 0)
+			count += obj->base.size >> PAGE_SHIFT;
+
+		drm_gem_object_unreference(&obj->base);
+	}
+	list_splice(&still_in_list, &dev_priv->mm.unbound_list);
+
+	INIT_LIST_HEAD(&still_in_list);
 	while (count < target && !list_empty(&dev_priv->mm.bound_list)) {
 		struct i915_vma *vma, *v;
 
 		obj = list_first_entry(&dev_priv->mm.bound_list,
 				       typeof(*obj), global_list);
-		list_move_tail(&obj->global_list, &still_bound_list);
+		list_move_tail(&obj->global_list, &still_in_list);
 
 		if (!i915_gem_object_is_purgeable(obj) && purgeable_only)
 			continue;
 
-		/*
-		 * Hold a reference whilst we unbind this object, as we may
-		 * end up waiting for and retiring requests. This might
-		 * release the final reference (held by the active list)
-		 * and result in the object being freed from under us.
-		 * in this object being freed.
-		 *
-		 * Note 1: Shrinking the bound list is special since only active
-		 * (and hence bound objects) can contain such limbo objects, so
-		 * we don't need special tricks for shrinking the unbound list.
-		 * The only other place where we have to be careful with active
-		 * objects suddenly disappearing due to retiring requests is the
-		 * eviction code.
-		 *
-		 * Note 2: Even though the bound list doesn't hold a reference
-		 * to the object we can safely grab one here: The final object
-		 * unreferencing and the bound_list are both protected by the
-		 * dev->struct_mutex and so we won't ever be able to observe an
-		 * object on the bound_list with a reference count equals 0.
-		 */
 		drm_gem_object_reference(&obj->base);
 
 		list_for_each_entry_safe(vma, v, &obj->vma_list, vma_link)
@@ -1966,7 +1989,7 @@ __i915_gem_shrink(struct drm_i915_private *dev_priv, long target,
 
 		drm_gem_object_unreference(&obj->base);
 	}
-	list_splice(&still_bound_list, &dev_priv->mm.bound_list);
+	list_splice(&still_in_list, &dev_priv->mm.bound_list);
 
 	return count;
 }
@@ -1980,17 +2003,8 @@ i915_gem_purge(struct drm_i915_private *dev_priv, long target)
 static unsigned long
 i915_gem_shrink_all(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_gem_object *obj, *next;
-	long freed = 0;
-
 	i915_gem_evict_everything(dev_priv->dev);
-
-	list_for_each_entry_safe(obj, next, &dev_priv->mm.unbound_list,
-				 global_list) {
-		if (i915_gem_object_put_pages(obj) == 0)
-			freed += obj->base.size >> PAGE_SHIFT;
-	}
-	return freed;
+	return __i915_gem_shrink(dev_priv, LONG_MAX, false);
 }
 
 static int
@@ -2094,7 +2108,19 @@ err_pages:
 		page_cache_release(sg_page_iter_page(&sg_iter));
 	sg_free_table(st);
 	kfree(st);
-	return PTR_ERR(page);
+
+	/* shmemfs first checks if there is enough memory to allocate the page
+	 * and reports ENOSPC should there be insufficient, along with the usual
+	 * ENOMEM for a genuine allocation failure.
+	 *
+	 * We use ENOSPC in our driver to mean that we have run out of aperture
+	 * space and so want to translate the error from shmemfs back to our
+	 * usual understanding of ENOMEM.
+	 */
+	if (PTR_ERR(page) == -ENOSPC)
+		return -ENOMEM;
+	else
+		return PTR_ERR(page);
 }
 
 /* Ensure that the associated pages are gathered from the backing storage
@@ -2131,7 +2157,7 @@ i915_gem_object_get_pages(struct drm_i915_gem_object *obj)
 
 static void
 i915_gem_object_move_to_active(struct drm_i915_gem_object *obj,
-			       struct intel_ring_buffer *ring)
+			       struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = obj->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -2169,7 +2195,7 @@ i915_gem_object_move_to_active(struct drm_i915_gem_object *obj,
 }
 
 void i915_vma_move_to_active(struct i915_vma *vma,
-			     struct intel_ring_buffer *ring)
+			     struct intel_engine_cs *ring)
 {
 	list_move_tail(&vma->mm_list, &vma->vm->active_list);
 	return i915_gem_object_move_to_active(vma->obj, ring);
@@ -2207,11 +2233,24 @@ i915_gem_object_move_to_inactive(struct drm_i915_gem_object *obj)
 	WARN_ON(i915_verify_lists(dev));
 }
 
+static void
+i915_gem_object_retire(struct drm_i915_gem_object *obj)
+{
+	struct intel_engine_cs *ring = obj->ring;
+
+	if (ring == NULL)
+		return;
+
+	if (i915_seqno_passed(ring->get_seqno(ring, true),
+			      obj->last_read_seqno))
+		i915_gem_object_move_to_inactive(obj);
+}
+
 static int
 i915_gem_init_seqno(struct drm_device *dev, u32 seqno)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int ret, i, j;
 
 	/* Carefully retire all requests without writing to the rings */
@@ -2226,8 +2265,8 @@ i915_gem_init_seqno(struct drm_device *dev, u32 seqno)
 	for_each_ring(ring, dev_priv, i) {
 		intel_ring_init_seqno(ring, seqno);
 
-		for (j = 0; j < ARRAY_SIZE(ring->sync_seqno); j++)
-			ring->sync_seqno[j] = 0;
+		for (j = 0; j < ARRAY_SIZE(ring->semaphore.sync_seqno); j++)
+			ring->semaphore.sync_seqno[j] = 0;
 	}
 
 	return 0;
@@ -2277,7 +2316,7 @@ i915_gem_get_seqno(struct drm_device *dev, u32 *seqno)
 	return 0;
 }
 
-int __i915_add_request(struct intel_ring_buffer *ring,
+int __i915_add_request(struct intel_engine_cs *ring,
 		       struct drm_file *file,
 		       struct drm_i915_gem_object *obj,
 		       u32 *out_seqno)
@@ -2382,7 +2421,7 @@ i915_gem_request_remove_from_client(struct drm_i915_gem_request *request)
 }
 
 static bool i915_context_is_banned(struct drm_i915_private *dev_priv,
-				   const struct i915_hw_context *ctx)
+				   const struct intel_context *ctx)
 {
 	unsigned long elapsed;
 
@@ -2395,8 +2434,9 @@ static bool i915_context_is_banned(struct drm_i915_private *dev_priv,
 		if (!i915_gem_context_is_default(ctx)) {
 			DRM_DEBUG("context hanging too fast, banning!\n");
 			return true;
-		} else if (dev_priv->gpu_error.stop_rings == 0) {
-			DRM_ERROR("gpu hanging too fast, banning!\n");
+		} else if (i915_stop_ring_allow_ban(dev_priv)) {
+			if (i915_stop_ring_allow_warn(dev_priv))
+				DRM_ERROR("gpu hanging too fast, banning!\n");
 			return true;
 		}
 	}
@@ -2405,7 +2445,7 @@ static bool i915_context_is_banned(struct drm_i915_private *dev_priv,
 }
 
 static void i915_set_reset_status(struct drm_i915_private *dev_priv,
-				  struct i915_hw_context *ctx,
+				  struct intel_context *ctx,
 				  const bool guilty)
 {
 	struct i915_ctx_hang_stats *hs;
@@ -2436,7 +2476,7 @@ static void i915_gem_free_request(struct drm_i915_gem_request *request)
 }
 
 struct drm_i915_gem_request *
-i915_gem_find_active_request(struct intel_ring_buffer *ring)
+i915_gem_find_active_request(struct intel_engine_cs *ring)
 {
 	struct drm_i915_gem_request *request;
 	u32 completed_seqno;
@@ -2454,7 +2494,7 @@ i915_gem_find_active_request(struct intel_ring_buffer *ring)
 }
 
 static void i915_gem_reset_ring_status(struct drm_i915_private *dev_priv,
-				       struct intel_ring_buffer *ring)
+				       struct intel_engine_cs *ring)
 {
 	struct drm_i915_gem_request *request;
 	bool ring_hung;
@@ -2473,7 +2513,7 @@ static void i915_gem_reset_ring_status(struct drm_i915_private *dev_priv,
 }
 
 static void i915_gem_reset_ring_cleanup(struct drm_i915_private *dev_priv,
-					struct intel_ring_buffer *ring)
+					struct intel_engine_cs *ring)
 {
 	while (!list_empty(&ring->active_list)) {
 		struct drm_i915_gem_object *obj;
@@ -2501,6 +2541,11 @@ static void i915_gem_reset_ring_cleanup(struct drm_i915_private *dev_priv,
 
 		i915_gem_free_request(request);
 	}
+
+	/* These may not have been flush before the reset, do so now */
+	kfree(ring->preallocated_lazy_request);
+	ring->preallocated_lazy_request = NULL;
+	ring->outstanding_lazy_seqno = 0;
 }
 
 void i915_gem_restore_fences(struct drm_device *dev)
@@ -2527,7 +2572,7 @@ void i915_gem_restore_fences(struct drm_device *dev)
 void i915_gem_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int i;
 
 	/*
@@ -2541,8 +2586,6 @@ void i915_gem_reset(struct drm_device *dev)
 	for_each_ring(ring, dev_priv, i)
 		i915_gem_reset_ring_cleanup(dev_priv, ring);
 
-	i915_gem_cleanup_ringbuffer(dev);
-
 	i915_gem_context_reset(dev);
 
 	i915_gem_restore_fences(dev);
@@ -2551,8 +2594,8 @@ void i915_gem_reset(struct drm_device *dev)
 /**
  * This function clears the request list as sequence numbers are passed.
  */
-static void
-i915_gem_retire_requests_ring(struct intel_ring_buffer *ring)
+void
+i915_gem_retire_requests_ring(struct intel_engine_cs *ring)
 {
 	uint32_t seqno;
 
@@ -2597,7 +2640,7 @@ i915_gem_retire_requests_ring(struct intel_ring_buffer *ring)
 		 * of tail of the request to update the last known position
 		 * of the GPU head.
 		 */
-		ring->last_retired_head = request->tail;
+		ring->buffer->last_retired_head = request->tail;
 
 		i915_gem_free_request(request);
 	}
@@ -2615,7 +2658,7 @@ bool
 i915_gem_retire_requests(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	bool idle = true;
 	int i;
 
@@ -2709,7 +2752,7 @@ i915_gem_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_wait *args = data;
 	struct drm_i915_gem_object *obj;
-	struct intel_ring_buffer *ring = NULL;
+	struct intel_engine_cs *ring = NULL;
 	struct timespec timeout_stack, *timeout = NULL;
 	unsigned reset_counter;
 	u32 seqno = 0;
@@ -2780,9 +2823,9 @@ out:
  */
 int
 i915_gem_object_sync(struct drm_i915_gem_object *obj,
-		     struct intel_ring_buffer *to)
+		     struct intel_engine_cs *to)
 {
-	struct intel_ring_buffer *from = obj->ring;
+	struct intel_engine_cs *from = obj->ring;
 	u32 seqno;
 	int ret, idx;
 
@@ -2795,7 +2838,7 @@ i915_gem_object_sync(struct drm_i915_gem_object *obj,
 	idx = intel_ring_sync_index(from, to);
 
 	seqno = obj->last_read_seqno;
-	if (seqno <= from->sync_seqno[idx])
+	if (seqno <= from->semaphore.sync_seqno[idx])
 		return 0;
 
 	ret = i915_gem_check_olr(obj->ring, seqno);
@@ -2803,13 +2846,13 @@ i915_gem_object_sync(struct drm_i915_gem_object *obj,
 		return ret;
 
 	trace_i915_gem_ring_sync_to(from, to, seqno);
-	ret = to->sync_to(to, from, seqno);
+	ret = to->semaphore.sync_to(to, from, seqno);
 	if (!ret)
 		/* We use last_read_seqno because sync_to()
 		 * might have just caused seqno wrap under
 		 * the radar.
 		 */
-		from->sync_seqno[idx] = obj->last_read_seqno;
+		from->semaphore.sync_seqno[idx] = obj->last_read_seqno;
 
 	return ret;
 }
@@ -2865,12 +2908,14 @@ int i915_vma_unbind(struct i915_vma *vma)
 	 * cause memory corruption through use-after-free.
 	 */
 
-	i915_gem_object_finish_gtt(obj);
+	if (i915_is_ggtt(vma->vm)) {
+		i915_gem_object_finish_gtt(obj);
 
-	/* release the fence reg _after_ flushing */
-	ret = i915_gem_object_put_fence(obj);
-	if (ret)
-		return ret;
+		/* release the fence reg _after_ flushing */
+		ret = i915_gem_object_put_fence(obj);
+		if (ret)
+			return ret;
+	}
 
 	trace_i915_vma_unbind(vma);
 
@@ -2903,7 +2948,7 @@ int i915_vma_unbind(struct i915_vma *vma)
 int i915_gpu_idle(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int ret, i;
 
 	/* Flush everything onto the inactive list. */
@@ -3144,6 +3189,9 @@ i915_gem_object_put_fence(struct drm_i915_gem_object *obj)
 
 	fence = &dev_priv->fence_regs[obj->fence_reg];
 
+	if (WARN_ON(fence->pin_count))
+		return -EBUSY;
+
 	i915_gem_object_fence_lost(obj);
 	i915_gem_object_update_fence(obj, fence, false);
 
@@ -3548,6 +3596,7 @@ i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj, bool write)
 	if (ret)
 		return ret;
 
+	i915_gem_object_retire(obj);
 	i915_gem_object_flush_cpu_write_domain(obj, false);
 
 	/* Serialise direct access to this object with the barriers for
@@ -3646,6 +3695,7 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
 		 * in obj->write_domain and have been skipping the clflushes.
 		 * Just set it to the CPU cache for now.
 		 */
+		i915_gem_object_retire(obj);
 		WARN_ON(obj->base.write_domain & ~I915_GEM_DOMAIN_CPU);
 
 		old_read_domains = obj->base.read_domains;
@@ -3743,6 +3793,15 @@ unlock:
 
 static bool is_pin_display(struct drm_i915_gem_object *obj)
 {
+	struct i915_vma *vma;
+
+	if (list_empty(&obj->vma_list))
+		return false;
+
+	vma = i915_gem_obj_to_ggtt(obj);
+	if (!vma)
+		return false;
+
 	/* There are 3 sources that pin objects:
 	 *   1. The display engine (scanouts, sprites, cursors);
 	 *   2. Reservations for execbuffer;
@@ -3754,7 +3813,7 @@ static bool is_pin_display(struct drm_i915_gem_object *obj)
 	 * subtracting the potential reference by the user, any pin_count
 	 * remains, it must be due to another use by the display engine.
 	 */
-	return i915_gem_obj_to_ggtt(obj)->pin_count - !!obj->user_pin_count;
+	return vma->pin_count - !!obj->user_pin_count;
 }
 
 /*
@@ -3765,9 +3824,10 @@ static bool is_pin_display(struct drm_i915_gem_object *obj)
 int
 i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
 				     u32 alignment,
-				     struct intel_ring_buffer *pipelined)
+				     struct intel_engine_cs *pipelined)
 {
 	u32 old_read_domains, old_write_domain;
+	bool was_pin_display;
 	int ret;
 
 	if (pipelined != obj->ring) {
@@ -3779,6 +3839,7 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
 	/* Mark the pin_display early so that we account for the
 	 * display coherency whilst setting up the cache domains.
 	 */
+	was_pin_display = obj->pin_display;
 	obj->pin_display = true;
 
 	/* The display engine is not coherent with the LLC cache on gen6.  As
@@ -3821,7 +3882,8 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
 	return 0;
 
 err_unpin_display:
-	obj->pin_display = is_pin_display(obj);
+	WARN_ON(was_pin_display != is_pin_display(obj));
+	obj->pin_display = was_pin_display;
 	return ret;
 }
 
@@ -3868,6 +3930,7 @@ i915_gem_object_set_to_cpu_domain(struct drm_i915_gem_object *obj, bool write)
 	if (ret)
 		return ret;
 
+	i915_gem_object_retire(obj);
 	i915_gem_object_flush_gtt_write_domain(obj);
 
 	old_write_domain = obj->base.write_domain;
@@ -3917,7 +3980,7 @@ i915_gem_ring_throttle(struct drm_device *dev, struct drm_file *file)
 	struct drm_i915_file_private *file_priv = file->driver_priv;
 	unsigned long recent_enough = jiffies - msecs_to_jiffies(20);
 	struct drm_i915_gem_request *request;
-	struct intel_ring_buffer *ring = NULL;
+	struct intel_engine_cs *ring = NULL;
 	unsigned reset_counter;
 	u32 seqno = 0;
 	int ret;
@@ -3976,9 +4039,13 @@ i915_gem_object_pin(struct drm_i915_gem_object *obj,
 		    uint32_t alignment,
 		    uint64_t flags)
 {
+	struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
 	struct i915_vma *vma;
 	int ret;
 
+	if (WARN_ON(vm == &dev_priv->mm.aliasing_ppgtt->base))
+		return -ENODEV;
+
 	if (WARN_ON(flags & (PIN_GLOBAL | PIN_MAPPABLE) && !i915_is_ggtt(vm)))
 		return -EINVAL;
 
@@ -4032,6 +4099,32 @@ i915_gem_object_ggtt_unpin(struct drm_i915_gem_object *obj)
 		obj->pin_mappable = false;
 }
 
+bool
+i915_gem_object_pin_fence(struct drm_i915_gem_object *obj)
+{
+	if (obj->fence_reg != I915_FENCE_REG_NONE) {
+		struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
+		struct i915_vma *ggtt_vma = i915_gem_obj_to_ggtt(obj);
+
+		WARN_ON(!ggtt_vma ||
+			dev_priv->fence_regs[obj->fence_reg].pin_count >
+			ggtt_vma->pin_count);
+		dev_priv->fence_regs[obj->fence_reg].pin_count++;
+		return true;
+	} else
+		return false;
+}
+
+void
+i915_gem_object_unpin_fence(struct drm_i915_gem_object *obj)
+{
+	if (obj->fence_reg != I915_FENCE_REG_NONE) {
+		struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
+		WARN_ON(dev_priv->fence_regs[obj->fence_reg].pin_count <= 0);
+		dev_priv->fence_regs[obj->fence_reg].pin_count--;
+	}
+}
+
 int
 i915_gem_pin_ioctl(struct drm_device *dev, void *data,
 		   struct drm_file *file)
@@ -4292,6 +4385,30 @@ struct drm_i915_gem_object *i915_gem_alloc_object(struct drm_device *dev,
 	return obj;
 }
 
+static bool discard_backing_storage(struct drm_i915_gem_object *obj)
+{
+	/* If we are the last user of the backing storage (be it shmemfs
+	 * pages or stolen etc), we know that the pages are going to be
+	 * immediately released. In this case, we can then skip copying
+	 * back the contents from the GPU.
+	 */
+
+	if (obj->madv != I915_MADV_WILLNEED)
+		return false;
+
+	if (obj->base.filp == NULL)
+		return true;
+
+	/* At first glance, this looks racy, but then again so would be
+	 * userspace racing mmap against close. However, the first external
+	 * reference to the filp can only be obtained through the
+	 * i915_gem_mmap_ioctl() which safeguards us against the user
+	 * acquiring such a reference whilst we are in the middle of
+	 * freeing the object.
+	 */
+	return atomic_long_read(&obj->base.filp->f_count) == 1;
+}
+
 void i915_gem_free_object(struct drm_gem_object *gem_obj)
 {
 	struct drm_i915_gem_object *obj = to_intel_bo(gem_obj);
@@ -4329,6 +4446,8 @@ void i915_gem_free_object(struct drm_gem_object *gem_obj)
 
 	if (WARN_ON(obj->pages_pin_count))
 		obj->pages_pin_count = 0;
+	if (discard_backing_storage(obj))
+		obj->madv = I915_MADV_DONTNEED;
 	i915_gem_object_put_pages(obj);
 	i915_gem_object_free_mmap_offset(obj);
 	i915_gem_object_release_stolen(obj);
@@ -4338,6 +4457,9 @@ void i915_gem_free_object(struct drm_gem_object *gem_obj)
 	if (obj->base.import_attach)
 		drm_prime_gem_destroy(&obj->base, NULL);
 
+	if (obj->ops->release)
+		obj->ops->release(obj);
+
 	drm_gem_object_release(&obj->base);
 	i915_gem_info_remove_obj(dev_priv, obj->base.size);
 
@@ -4371,6 +4493,17 @@ void i915_gem_vma_destroy(struct i915_vma *vma)
 	kfree(vma);
 }
 
+static void
+i915_gem_stop_ringbuffers(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_engine_cs *ring;
+	int i;
+
+	for_each_ring(ring, dev_priv, i)
+		intel_stop_ring_buffer(ring);
+}
+
 int
 i915_gem_suspend(struct drm_device *dev)
 {
@@ -4392,7 +4525,7 @@ i915_gem_suspend(struct drm_device *dev)
 		i915_gem_evict_everything(dev);
 
 	i915_kernel_lost_context(dev);
-	i915_gem_cleanup_ringbuffer(dev);
+	i915_gem_stop_ringbuffers(dev);
 
 	/* Hack!  Don't let anybody do execbuf while we don't control the chip.
 	 * We need to replace this with a semaphore, or something.
@@ -4413,7 +4546,7 @@ err:
 	return ret;
 }
 
-int i915_gem_l3_remap(struct intel_ring_buffer *ring, int slice)
+int i915_gem_l3_remap(struct intel_engine_cs *ring, int slice)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -4512,13 +4645,20 @@ static int i915_gem_init_rings(struct drm_device *dev)
 			goto cleanup_blt_ring;
 	}
 
+	if (HAS_BSD2(dev)) {
+		ret = intel_init_bsd2_ring_buffer(dev);
+		if (ret)
+			goto cleanup_vebox_ring;
+	}
 
 	ret = i915_gem_set_seqno(dev, ((u32)~0 - 0x1000));
 	if (ret)
-		goto cleanup_vebox_ring;
+		goto cleanup_bsd2_ring;
 
 	return 0;
 
+cleanup_bsd2_ring:
+	intel_cleanup_ring_buffer(&dev_priv->ring[VCS2]);
 cleanup_vebox_ring:
 	intel_cleanup_ring_buffer(&dev_priv->ring[VECS]);
 cleanup_blt_ring:
@@ -4576,15 +4716,11 @@ i915_gem_init_hw(struct drm_device *dev)
 	 * the do_switch), but before enabling PPGTT. So don't move this.
 	 */
 	ret = i915_gem_context_enable(dev_priv);
-	if (ret) {
+	if (ret && ret != -EIO) {
 		DRM_ERROR("Context enable failed %d\n", ret);
-		goto err_out;
+		i915_gem_cleanup_ringbuffer(dev);
 	}
 
-	return 0;
-
-err_out:
-	i915_gem_cleanup_ringbuffer(dev);
 	return ret;
 }
 
@@ -4597,11 +4733,13 @@ int i915_gem_init(struct drm_device *dev)
 
 	if (IS_VALLEYVIEW(dev)) {
 		/* VLVA0 (potential hack), BIOS isn't actually waking us */
-		I915_WRITE(VLV_GTLC_WAKE_CTRL, 1);
-		if (wait_for((I915_READ(VLV_GTLC_PW_STATUS) & 1) == 1, 10))
+		I915_WRITE(VLV_GTLC_WAKE_CTRL, VLV_GTLC_ALLOWWAKEREQ);
+		if (wait_for((I915_READ(VLV_GTLC_PW_STATUS) &
+			      VLV_GTLC_ALLOWWAKEACK), 10))
 			DRM_DEBUG_DRIVER("allow wake ack timed out\n");
 	}
 
+	i915_gem_init_userptr(dev);
 	i915_gem_init_global_gtt(dev);
 
 	ret = i915_gem_context_init(dev);
@@ -4611,25 +4749,28 @@ int i915_gem_init(struct drm_device *dev)
 	}
 
 	ret = i915_gem_init_hw(dev);
-	mutex_unlock(&dev->struct_mutex);
-	if (ret) {
-		WARN_ON(dev_priv->mm.aliasing_ppgtt);
-		i915_gem_context_fini(dev);
-		drm_mm_takedown(&dev_priv->gtt.base.mm);
-		return ret;
+	if (ret == -EIO) {
+		/* Allow ring initialisation to fail by marking the GPU as
+		 * wedged. But we only want to do this where the GPU is angry,
+		 * for all other failure, such as an allocation failure, bail.
+		 */
+		DRM_ERROR("Failed to initialize GPU, declaring it wedged\n");
+		atomic_set_mask(I915_WEDGED, &dev_priv->gpu_error.reset_counter);
+		ret = 0;
 	}
+	mutex_unlock(&dev->struct_mutex);
 
 	/* Allow hardware batchbuffers unless told otherwise, but not for KMS. */
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		dev_priv->dri1.allow_batchbuffer = 1;
-	return 0;
+	return ret;
 }
 
 void
 i915_gem_cleanup_ringbuffer(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int i;
 
 	for_each_ring(ring, dev_priv, i)
@@ -4661,16 +4802,15 @@ i915_gem_entervt_ioctl(struct drm_device *dev, void *data,
 	}
 
 	BUG_ON(!list_empty(&dev_priv->gtt.base.active_list));
-	mutex_unlock(&dev->struct_mutex);
 
-	ret = drm_irq_install(dev);
+	ret = drm_irq_install(dev, dev->pdev->irq);
 	if (ret)
 		goto cleanup_ringbuffer;
+	mutex_unlock(&dev->struct_mutex);
 
 	return 0;
 
 cleanup_ringbuffer:
-	mutex_lock(&dev->struct_mutex);
 	i915_gem_cleanup_ringbuffer(dev);
 	dev_priv->ums.mm_suspended = 1;
 	mutex_unlock(&dev->struct_mutex);
@@ -4685,7 +4825,9 @@ i915_gem_leavevt_ioctl(struct drm_device *dev, void *data,
 	if (drm_core_check_feature(dev, DRIVER_MODESET))
 		return 0;
 
+	mutex_lock(&dev->struct_mutex);
 	drm_irq_uninstall(dev);
+	mutex_unlock(&dev->struct_mutex);
 
 	return i915_gem_suspend(dev);
 }
@@ -4704,7 +4846,7 @@ i915_gem_lastclose(struct drm_device *dev)
 }
 
 static void
-init_ring_lists(struct intel_ring_buffer *ring)
+init_ring_lists(struct intel_engine_cs *ring)
 {
 	INIT_LIST_HEAD(&ring->active_list);
 	INIT_LIST_HEAD(&ring->request_list);
@@ -4752,7 +4894,7 @@ i915_gem_load(struct drm_device *dev)
 	init_waitqueue_head(&dev_priv->gpu_error.reset_queue);
 
 	/* On GEN3 we really need to make sure the ARB C3 LP bit is set */
-	if (IS_GEN3(dev)) {
+	if (!drm_core_check_feature(dev, DRIVER_MODESET) && IS_GEN3(dev)) {
 		I915_WRITE(MI_ARB_STATE,
 			   _MASKED_BIT_ENABLE(MI_ARB_C3_LP_WRITE_ENABLE));
 	}
@@ -4779,10 +4921,13 @@ i915_gem_load(struct drm_device *dev)
 
 	dev_priv->mm.interruptible = true;
 
-	dev_priv->mm.inactive_shrinker.scan_objects = i915_gem_inactive_scan;
-	dev_priv->mm.inactive_shrinker.count_objects = i915_gem_inactive_count;
-	dev_priv->mm.inactive_shrinker.seeks = DEFAULT_SEEKS;
-	register_shrinker(&dev_priv->mm.inactive_shrinker);
+	dev_priv->mm.shrinker.scan_objects = i915_gem_shrinker_scan;
+	dev_priv->mm.shrinker.count_objects = i915_gem_shrinker_count;
+	dev_priv->mm.shrinker.seeks = DEFAULT_SEEKS;
+	register_shrinker(&dev_priv->mm.shrinker);
+
+	dev_priv->mm.oom_notifier.notifier_call = i915_gem_shrinker_oom;
+	register_oom_notifier(&dev_priv->mm.oom_notifier);
 }
 
 void i915_gem_release(struct drm_device *dev, struct drm_file *file)
@@ -4857,27 +5002,46 @@ static bool mutex_is_locked_by(struct mutex *mutex, struct task_struct *task)
 #endif
 }
 
+static bool i915_gem_shrinker_lock(struct drm_device *dev, bool *unlock)
+{
+	if (!mutex_trylock(&dev->struct_mutex)) {
+		if (!mutex_is_locked_by(&dev->struct_mutex, current))
+			return false;
+
+		if (to_i915(dev)->mm.shrinker_no_lock_stealing)
+			return false;
+
+		*unlock = false;
+	} else
+		*unlock = true;
+
+	return true;
+}
+
+static int num_vma_bound(struct drm_i915_gem_object *obj)
+{
+	struct i915_vma *vma;
+	int count = 0;
+
+	list_for_each_entry(vma, &obj->vma_list, vma_link)
+		if (drm_mm_node_allocated(&vma->node))
+			count++;
+
+	return count;
+}
+
 static unsigned long
-i915_gem_inactive_count(struct shrinker *shrinker, struct shrink_control *sc)
+i915_gem_shrinker_count(struct shrinker *shrinker, struct shrink_control *sc)
 {
 	struct drm_i915_private *dev_priv =
-		container_of(shrinker,
-			     struct drm_i915_private,
-			     mm.inactive_shrinker);
+		container_of(shrinker, struct drm_i915_private, mm.shrinker);
 	struct drm_device *dev = dev_priv->dev;
 	struct drm_i915_gem_object *obj;
-	bool unlock = true;
 	unsigned long count;
+	bool unlock;
 
-	if (!mutex_trylock(&dev->struct_mutex)) {
-		if (!mutex_is_locked_by(&dev->struct_mutex, current))
-			return 0;
-
-		if (dev_priv->mm.shrinker_no_lock_stealing)
-			return 0;
-
-		unlock = false;
-	}
+	if (!i915_gem_shrinker_lock(dev, &unlock))
+		return 0;
 
 	count = 0;
 	list_for_each_entry(obj, &dev_priv->mm.unbound_list, global_list)
@@ -4885,10 +5049,8 @@ i915_gem_inactive_count(struct shrinker *shrinker, struct shrink_control *sc)
 			count += obj->base.size >> PAGE_SHIFT;
 
 	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
-		if (obj->active)
-			continue;
-
-		if (!i915_gem_obj_is_pinned(obj) && obj->pages_pin_count == 0)
+		if (!i915_gem_obj_is_pinned(obj) &&
+		    obj->pages_pin_count == num_vma_bound(obj))
 			count += obj->base.size >> PAGE_SHIFT;
 	}
 
@@ -4961,44 +5123,99 @@ unsigned long i915_gem_obj_size(struct drm_i915_gem_object *o,
 }
 
 static unsigned long
-i915_gem_inactive_scan(struct shrinker *shrinker, struct shrink_control *sc)
+i915_gem_shrinker_scan(struct shrinker *shrinker, struct shrink_control *sc)
 {
 	struct drm_i915_private *dev_priv =
-		container_of(shrinker,
-			     struct drm_i915_private,
-			     mm.inactive_shrinker);
+		container_of(shrinker, struct drm_i915_private, mm.shrinker);
 	struct drm_device *dev = dev_priv->dev;
 	unsigned long freed;
-	bool unlock = true;
+	bool unlock;
 
-	if (!mutex_trylock(&dev->struct_mutex)) {
-		if (!mutex_is_locked_by(&dev->struct_mutex, current))
-			return SHRINK_STOP;
-
-		if (dev_priv->mm.shrinker_no_lock_stealing)
-			return SHRINK_STOP;
-
-		unlock = false;
-	}
+	if (!i915_gem_shrinker_lock(dev, &unlock))
+		return SHRINK_STOP;
 
 	freed = i915_gem_purge(dev_priv, sc->nr_to_scan);
 	if (freed < sc->nr_to_scan)
 		freed += __i915_gem_shrink(dev_priv,
 					   sc->nr_to_scan - freed,
 					   false);
-	if (freed < sc->nr_to_scan)
-		freed += i915_gem_shrink_all(dev_priv);
-
 	if (unlock)
 		mutex_unlock(&dev->struct_mutex);
 
 	return freed;
 }
 
+static int
+i915_gem_shrinker_oom(struct notifier_block *nb, unsigned long event, void *ptr)
+{
+	struct drm_i915_private *dev_priv =
+		container_of(nb, struct drm_i915_private, mm.oom_notifier);
+	struct drm_device *dev = dev_priv->dev;
+	struct drm_i915_gem_object *obj;
+	unsigned long timeout = msecs_to_jiffies(5000) + 1;
+	unsigned long pinned, bound, unbound, freed;
+	bool was_interruptible;
+	bool unlock;
+
+	while (!i915_gem_shrinker_lock(dev, &unlock) && --timeout)
+		schedule_timeout_killable(1);
+	if (timeout == 0) {
+		pr_err("Unable to purge GPU memory due lock contention.\n");
+		return NOTIFY_DONE;
+	}
+
+	was_interruptible = dev_priv->mm.interruptible;
+	dev_priv->mm.interruptible = false;
+
+	freed = i915_gem_shrink_all(dev_priv);
+
+	dev_priv->mm.interruptible = was_interruptible;
+
+	/* Because we may be allocating inside our own driver, we cannot
+	 * assert that there are no objects with pinned pages that are not
+	 * being pointed to by hardware.
+	 */
+	unbound = bound = pinned = 0;
+	list_for_each_entry(obj, &dev_priv->mm.unbound_list, global_list) {
+		if (!obj->base.filp) /* not backed by a freeable object */
+			continue;
+
+		if (obj->pages_pin_count)
+			pinned += obj->base.size;
+		else
+			unbound += obj->base.size;
+	}
+	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
+		if (!obj->base.filp)
+			continue;
+
+		if (obj->pages_pin_count)
+			pinned += obj->base.size;
+		else
+			bound += obj->base.size;
+	}
+
+	if (unlock)
+		mutex_unlock(&dev->struct_mutex);
+
+	pr_info("Purging GPU memory, %lu bytes freed, %lu bytes still pinned.\n",
+		freed, pinned);
+	if (unbound || bound)
+		pr_err("%lu and %lu bytes still available in the "
+		       "bound and unbound GPU page lists.\n",
+		       bound, unbound);
+
+	*(unsigned long *)ptr += freed;
+	return NOTIFY_DONE;
+}
+
 struct i915_vma *i915_gem_obj_to_ggtt(struct drm_i915_gem_object *obj)
 {
 	struct i915_vma *vma;
 
+	/* This WARN has probably outlived its usefulness (callers already
+	 * WARN if they don't find the GGTT vma they expect). When removing,
+	 * remember to remove the pre-check in is_pin_display() as well */
 	if (WARN_ON(list_empty(&obj->vma_list)))
 		return NULL;
 
diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
index d72db15afa02..3ffe308d5893 100644
--- a/drivers/gpu/drm/i915/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/i915_gem_context.c
@@ -178,7 +178,7 @@ static int get_context_size(struct drm_device *dev)
 
 void i915_gem_context_free(struct kref *ctx_ref)
 {
-	struct i915_hw_context *ctx = container_of(ctx_ref,
+	struct intel_context *ctx = container_of(ctx_ref,
 						   typeof(*ctx), ref);
 	struct i915_hw_ppgtt *ppgtt = NULL;
 
@@ -199,7 +199,7 @@ void i915_gem_context_free(struct kref *ctx_ref)
 }
 
 static struct i915_hw_ppgtt *
-create_vm_for_ctx(struct drm_device *dev, struct i915_hw_context *ctx)
+create_vm_for_ctx(struct drm_device *dev, struct intel_context *ctx)
 {
 	struct i915_hw_ppgtt *ppgtt;
 	int ret;
@@ -218,12 +218,12 @@ create_vm_for_ctx(struct drm_device *dev, struct i915_hw_context *ctx)
 	return ppgtt;
 }
 
-static struct i915_hw_context *
+static struct intel_context *
 __create_hw_context(struct drm_device *dev,
 		  struct drm_i915_file_private *file_priv)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct i915_hw_context *ctx;
+	struct intel_context *ctx;
 	int ret;
 
 	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
@@ -240,7 +240,15 @@ __create_hw_context(struct drm_device *dev,
 			goto err_out;
 		}
 
-		if (INTEL_INFO(dev)->gen >= 7) {
+		/*
+		 * Try to make the context utilize L3 as well as LLC.
+		 *
+		 * On VLV we don't have L3 controls in the PTEs so we
+		 * shouldn't touch the cache level, especially as that
+		 * would make the object snooped which might have a
+		 * negative performance impact.
+		 */
+		if (INTEL_INFO(dev)->gen >= 7 && !IS_VALLEYVIEW(dev)) {
 			ret = i915_gem_object_set_cache_level(ctx->obj,
 							      I915_CACHE_L3_LLC);
 			/* Failure shouldn't ever happen this early */
@@ -277,14 +285,14 @@ err_out:
  * context state of the GPU for applications that don't utilize HW contexts, as
  * well as an idle case.
  */
-static struct i915_hw_context *
+static struct intel_context *
 i915_gem_create_context(struct drm_device *dev,
 			struct drm_i915_file_private *file_priv,
 			bool create_vm)
 {
 	const bool is_global_default_ctx = file_priv == NULL;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct i915_hw_context *ctx;
+	struct intel_context *ctx;
 	int ret = 0;
 
 	BUG_ON(!mutex_is_locked(&dev->struct_mutex));
@@ -356,8 +364,8 @@ void i915_gem_context_reset(struct drm_device *dev)
 	/* Prevent the hardware from restoring the last context (which hung) on
 	 * the next switch */
 	for (i = 0; i < I915_NUM_RINGS; i++) {
-		struct intel_ring_buffer *ring = &dev_priv->ring[i];
-		struct i915_hw_context *dctx = ring->default_context;
+		struct intel_engine_cs *ring = &dev_priv->ring[i];
+		struct intel_context *dctx = ring->default_context;
 
 		/* Do a fake switch to the default context */
 		if (ring->last_context == dctx)
@@ -383,7 +391,7 @@ void i915_gem_context_reset(struct drm_device *dev)
 int i915_gem_context_init(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct i915_hw_context *ctx;
+	struct intel_context *ctx;
 	int i;
 
 	/* Init should only be called once per module load. Eventually the
@@ -418,7 +426,7 @@ int i915_gem_context_init(struct drm_device *dev)
 void i915_gem_context_fini(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct i915_hw_context *dctx = dev_priv->ring[RCS].default_context;
+	struct intel_context *dctx = dev_priv->ring[RCS].default_context;
 	int i;
 
 	if (dctx->obj) {
@@ -441,10 +449,12 @@ void i915_gem_context_fini(struct drm_device *dev)
 			i915_gem_context_unreference(dctx);
 			dev_priv->ring[RCS].last_context = NULL;
 		}
+
+		i915_gem_object_ggtt_unpin(dctx->obj);
 	}
 
 	for (i = 0; i < I915_NUM_RINGS; i++) {
-		struct intel_ring_buffer *ring = &dev_priv->ring[i];
+		struct intel_engine_cs *ring = &dev_priv->ring[i];
 
 		if (ring->last_context)
 			i915_gem_context_unreference(ring->last_context);
@@ -453,13 +463,12 @@ void i915_gem_context_fini(struct drm_device *dev)
 		ring->last_context = NULL;
 	}
 
-	i915_gem_object_ggtt_unpin(dctx->obj);
 	i915_gem_context_unreference(dctx);
 }
 
 int i915_gem_context_enable(struct drm_i915_private *dev_priv)
 {
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int ret, i;
 
 	/* This is the only place the aliasing PPGTT gets enabled, which means
@@ -486,11 +495,7 @@ int i915_gem_context_enable(struct drm_i915_private *dev_priv)
 
 static int context_idr_cleanup(int id, void *p, void *data)
 {
-	struct i915_hw_context *ctx = p;
-
-	/* Ignore the default context because close will handle it */
-	if (i915_gem_context_is_default(ctx))
-		return 0;
+	struct intel_context *ctx = p;
 
 	i915_gem_context_unreference(ctx);
 	return 0;
@@ -499,17 +504,17 @@ static int context_idr_cleanup(int id, void *p, void *data)
 int i915_gem_context_open(struct drm_device *dev, struct drm_file *file)
 {
 	struct drm_i915_file_private *file_priv = file->driver_priv;
+	struct intel_context *ctx;
 
 	idr_init(&file_priv->context_idr);
 
 	mutex_lock(&dev->struct_mutex);
-	file_priv->private_default_ctx =
-		i915_gem_create_context(dev, file_priv, USES_FULL_PPGTT(dev));
+	ctx = i915_gem_create_context(dev, file_priv, USES_FULL_PPGTT(dev));
 	mutex_unlock(&dev->struct_mutex);
 
-	if (IS_ERR(file_priv->private_default_ctx)) {
+	if (IS_ERR(ctx)) {
 		idr_destroy(&file_priv->context_idr);
-		return PTR_ERR(file_priv->private_default_ctx);
+		return PTR_ERR(ctx);
 	}
 
 	return 0;
@@ -521,16 +526,14 @@ void i915_gem_context_close(struct drm_device *dev, struct drm_file *file)
 
 	idr_for_each(&file_priv->context_idr, context_idr_cleanup, NULL);
 	idr_destroy(&file_priv->context_idr);
-
-	i915_gem_context_unreference(file_priv->private_default_ctx);
 }
 
-struct i915_hw_context *
+struct intel_context *
 i915_gem_context_get(struct drm_i915_file_private *file_priv, u32 id)
 {
-	struct i915_hw_context *ctx;
+	struct intel_context *ctx;
 
-	ctx = (struct i915_hw_context *)idr_find(&file_priv->context_idr, id);
+	ctx = (struct intel_context *)idr_find(&file_priv->context_idr, id);
 	if (!ctx)
 		return ERR_PTR(-ENOENT);
 
@@ -538,8 +541,8 @@ i915_gem_context_get(struct drm_i915_file_private *file_priv, u32 id)
 }
 
 static inline int
-mi_set_context(struct intel_ring_buffer *ring,
-	       struct i915_hw_context *new_context,
+mi_set_context(struct intel_engine_cs *ring,
+	       struct intel_context *new_context,
 	       u32 hw_flags)
 {
 	int ret;
@@ -549,7 +552,7 @@ mi_set_context(struct intel_ring_buffer *ring,
 	 * explicitly, so we rely on the value at ring init, stored in
 	 * itlb_before_ctx_switch.
 	 */
-	if (IS_GEN6(ring->dev) && ring->itlb_before_ctx_switch) {
+	if (IS_GEN6(ring->dev)) {
 		ret = ring->flush(ring, I915_GEM_GPU_DOMAINS, 0);
 		if (ret)
 			return ret;
@@ -559,8 +562,8 @@ mi_set_context(struct intel_ring_buffer *ring,
 	if (ret)
 		return ret;
 
-	/* WaProgramMiArbOnOffAroundMiSetContext:ivb,vlv,hsw */
-	if (IS_GEN7(ring->dev))
+	/* WaProgramMiArbOnOffAroundMiSetContext:ivb,vlv,hsw,bdw,chv */
+	if (INTEL_INFO(ring->dev)->gen >= 7)
 		intel_ring_emit(ring, MI_ARB_ON_OFF | MI_ARB_DISABLE);
 	else
 		intel_ring_emit(ring, MI_NOOP);
@@ -578,7 +581,7 @@ mi_set_context(struct intel_ring_buffer *ring,
 	 */
 	intel_ring_emit(ring, MI_NOOP);
 
-	if (IS_GEN7(ring->dev))
+	if (INTEL_INFO(ring->dev)->gen >= 7)
 		intel_ring_emit(ring, MI_ARB_ON_OFF | MI_ARB_ENABLE);
 	else
 		intel_ring_emit(ring, MI_NOOP);
@@ -588,11 +591,11 @@ mi_set_context(struct intel_ring_buffer *ring,
 	return ret;
 }
 
-static int do_switch(struct intel_ring_buffer *ring,
-		     struct i915_hw_context *to)
+static int do_switch(struct intel_engine_cs *ring,
+		     struct intel_context *to)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
-	struct i915_hw_context *from = ring->last_context;
+	struct intel_context *from = ring->last_context;
 	struct i915_hw_ppgtt *ppgtt = ctx_to_ppgtt(to);
 	u32 hw_flags = 0;
 	int ret, i;
@@ -693,13 +696,19 @@ static int do_switch(struct intel_ring_buffer *ring,
 		i915_gem_context_unreference(from);
 	}
 
-	to->is_initialized = true;
-
 done:
 	i915_gem_context_reference(to);
 	ring->last_context = to;
 	to->last_ring = ring;
 
+	if (ring->id == RCS && !to->is_initialized && from == NULL) {
+		ret = i915_gem_render_state_init(ring);
+		if (ret)
+			DRM_ERROR("init render state: %d\n", ret);
+	}
+
+	to->is_initialized = true;
+
 	return 0;
 
 unpin_out:
@@ -718,8 +727,8 @@ unpin_out:
  * it will have a refoucnt > 1. This allows us to destroy the context abstract
  * object while letting the normal object tracking destroy the backing BO.
  */
-int i915_switch_context(struct intel_ring_buffer *ring,
-			struct i915_hw_context *to)
+int i915_switch_context(struct intel_engine_cs *ring,
+			struct intel_context *to)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 
@@ -748,7 +757,7 @@ int i915_gem_context_create_ioctl(struct drm_device *dev, void *data,
 {
 	struct drm_i915_gem_context_create *args = data;
 	struct drm_i915_file_private *file_priv = file->driver_priv;
-	struct i915_hw_context *ctx;
+	struct intel_context *ctx;
 	int ret;
 
 	if (!hw_context_enabled(dev))
@@ -774,7 +783,7 @@ int i915_gem_context_destroy_ioctl(struct drm_device *dev, void *data,
 {
 	struct drm_i915_gem_context_destroy *args = data;
 	struct drm_i915_file_private *file_priv = file->driver_priv;
-	struct i915_hw_context *ctx;
+	struct intel_context *ctx;
 	int ret;
 
 	if (args->ctx_id == DEFAULT_CONTEXT_ID)
diff --git a/drivers/gpu/drm/i915/i915_gem_dmabuf.c b/drivers/gpu/drm/i915/i915_gem_dmabuf.c
index 9bb533e0d762..580aa42443ed 100644
--- a/drivers/gpu/drm/i915/i915_gem_dmabuf.c
+++ b/drivers/gpu/drm/i915/i915_gem_dmabuf.c
@@ -161,12 +161,8 @@ static void i915_gem_dmabuf_vunmap(struct dma_buf *dma_buf, void *vaddr)
 {
 	struct drm_i915_gem_object *obj = dma_buf_to_obj(dma_buf);
 	struct drm_device *dev = obj->base.dev;
-	int ret;
-
-	ret = i915_mutex_lock_interruptible(dev);
-	if (ret)
-		return;
 
+	mutex_lock(&dev->struct_mutex);
 	if (--obj->vmapping_count == 0) {
 		vunmap(obj->dma_buf_vmapping);
 		obj->dma_buf_vmapping = NULL;
@@ -233,6 +229,14 @@ static const struct dma_buf_ops i915_dmabuf_ops =  {
 struct dma_buf *i915_gem_prime_export(struct drm_device *dev,
 				      struct drm_gem_object *gem_obj, int flags)
 {
+	struct drm_i915_gem_object *obj = to_intel_bo(gem_obj);
+
+	if (obj->ops->dmabuf_export) {
+		int ret = obj->ops->dmabuf_export(obj);
+		if (ret)
+			return ERR_PTR(ret);
+	}
+
 	return dma_buf_export(gem_obj, &i915_dmabuf_ops, gem_obj->size, flags);
 }
 
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 20fef6c50267..3a30133f93e8 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -265,10 +265,12 @@ static inline int use_cpu_reloc(struct drm_i915_gem_object *obj)
 
 static int
 relocate_entry_cpu(struct drm_i915_gem_object *obj,
-		   struct drm_i915_gem_relocation_entry *reloc)
+		   struct drm_i915_gem_relocation_entry *reloc,
+		   uint64_t target_offset)
 {
 	struct drm_device *dev = obj->base.dev;
 	uint32_t page_offset = offset_in_page(reloc->offset);
+	uint64_t delta = reloc->delta + target_offset;
 	char *vaddr;
 	int ret;
 
@@ -278,7 +280,7 @@ relocate_entry_cpu(struct drm_i915_gem_object *obj,
 
 	vaddr = kmap_atomic(i915_gem_object_get_page(obj,
 				reloc->offset >> PAGE_SHIFT));
-	*(uint32_t *)(vaddr + page_offset) = reloc->delta;
+	*(uint32_t *)(vaddr + page_offset) = lower_32_bits(delta);
 
 	if (INTEL_INFO(dev)->gen >= 8) {
 		page_offset = offset_in_page(page_offset + sizeof(uint32_t));
@@ -289,7 +291,7 @@ relocate_entry_cpu(struct drm_i915_gem_object *obj,
 			    (reloc->offset + sizeof(uint32_t)) >> PAGE_SHIFT));
 		}
 
-		*(uint32_t *)(vaddr + page_offset) = 0;
+		*(uint32_t *)(vaddr + page_offset) = upper_32_bits(delta);
 	}
 
 	kunmap_atomic(vaddr);
@@ -299,10 +301,12 @@ relocate_entry_cpu(struct drm_i915_gem_object *obj,
 
 static int
 relocate_entry_gtt(struct drm_i915_gem_object *obj,
-		   struct drm_i915_gem_relocation_entry *reloc)
+		   struct drm_i915_gem_relocation_entry *reloc,
+		   uint64_t target_offset)
 {
 	struct drm_device *dev = obj->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	uint64_t delta = reloc->delta + target_offset;
 	uint32_t __iomem *reloc_entry;
 	void __iomem *reloc_page;
 	int ret;
@@ -321,7 +325,7 @@ relocate_entry_gtt(struct drm_i915_gem_object *obj,
 			reloc->offset & PAGE_MASK);
 	reloc_entry = (uint32_t __iomem *)
 		(reloc_page + offset_in_page(reloc->offset));
-	iowrite32(reloc->delta, reloc_entry);
+	iowrite32(lower_32_bits(delta), reloc_entry);
 
 	if (INTEL_INFO(dev)->gen >= 8) {
 		reloc_entry += 1;
@@ -334,7 +338,7 @@ relocate_entry_gtt(struct drm_i915_gem_object *obj,
 			reloc_entry = reloc_page;
 		}
 
-		iowrite32(0, reloc_entry);
+		iowrite32(upper_32_bits(delta), reloc_entry);
 	}
 
 	io_mapping_unmap_atomic(reloc_page);
@@ -351,7 +355,7 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj,
 	struct drm_gem_object *target_obj;
 	struct drm_i915_gem_object *target_i915_obj;
 	struct i915_vma *target_vma;
-	uint32_t target_offset;
+	uint64_t target_offset;
 	int ret;
 
 	/* we've already hold a reference to all valid objects */
@@ -429,11 +433,10 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj,
 	if (obj->active && in_atomic())
 		return -EFAULT;
 
-	reloc->delta += target_offset;
 	if (use_cpu_reloc(obj))
-		ret = relocate_entry_cpu(obj, reloc);
+		ret = relocate_entry_cpu(obj, reloc, target_offset);
 	else
-		ret = relocate_entry_gtt(obj, reloc);
+		ret = relocate_entry_gtt(obj, reloc, target_offset);
 
 	if (ret)
 		return ret;
@@ -541,7 +544,7 @@ need_reloc_mappable(struct i915_vma *vma)
 
 static int
 i915_gem_execbuffer_reserve_vma(struct i915_vma *vma,
-				struct intel_ring_buffer *ring,
+				struct intel_engine_cs *ring,
 				bool *need_reloc)
 {
 	struct drm_i915_gem_object *obj = vma->obj;
@@ -628,7 +631,7 @@ eb_vma_misplaced(struct i915_vma *vma, bool has_fenced_gpu_access)
 }
 
 static int
-i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring,
+i915_gem_execbuffer_reserve(struct intel_engine_cs *ring,
 			    struct list_head *vmas,
 			    bool *need_relocs)
 {
@@ -642,6 +645,8 @@ i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring,
 	if (list_empty(vmas))
 		return 0;
 
+	i915_gem_retire_requests_ring(ring);
+
 	vm = list_first_entry(vmas, struct i915_vma, exec_list)->vm;
 
 	INIT_LIST_HEAD(&ordered_vmas);
@@ -727,7 +732,7 @@ static int
 i915_gem_execbuffer_relocate_slow(struct drm_device *dev,
 				  struct drm_i915_gem_execbuffer2 *args,
 				  struct drm_file *file,
-				  struct intel_ring_buffer *ring,
+				  struct intel_engine_cs *ring,
 				  struct eb_vmas *eb,
 				  struct drm_i915_gem_exec_object2 *exec)
 {
@@ -843,7 +848,7 @@ err:
 }
 
 static int
-i915_gem_execbuffer_move_to_gpu(struct intel_ring_buffer *ring,
+i915_gem_execbuffer_move_to_gpu(struct intel_engine_cs *ring,
 				struct list_head *vmas)
 {
 	struct i915_vma *vma;
@@ -926,11 +931,11 @@ validate_exec_list(struct drm_i915_gem_exec_object2 *exec,
 	return 0;
 }
 
-static struct i915_hw_context *
+static struct intel_context *
 i915_gem_validate_context(struct drm_device *dev, struct drm_file *file,
-			  struct intel_ring_buffer *ring, const u32 ctx_id)
+			  struct intel_engine_cs *ring, const u32 ctx_id)
 {
-	struct i915_hw_context *ctx = NULL;
+	struct intel_context *ctx = NULL;
 	struct i915_ctx_hang_stats *hs;
 
 	if (ring->id != RCS && ctx_id != DEFAULT_CONTEXT_ID)
@@ -951,7 +956,7 @@ i915_gem_validate_context(struct drm_device *dev, struct drm_file *file,
 
 static void
 i915_gem_execbuffer_move_to_active(struct list_head *vmas,
-				   struct intel_ring_buffer *ring)
+				   struct intel_engine_cs *ring)
 {
 	struct i915_vma *vma;
 
@@ -974,6 +979,9 @@ i915_gem_execbuffer_move_to_active(struct list_head *vmas,
 			if (i915_gem_obj_ggtt_bound(obj) &&
 			    i915_gem_obj_to_ggtt(obj)->pin_count)
 				intel_mark_fb_busy(obj, ring);
+
+			/* update for the implicit flush after a batch */
+			obj->base.write_domain &= ~I915_GEM_GPU_DOMAINS;
 		}
 
 		trace_i915_gem_object_change_domain(obj, old_read, old_write);
@@ -983,7 +991,7 @@ i915_gem_execbuffer_move_to_active(struct list_head *vmas,
 static void
 i915_gem_execbuffer_retire_commands(struct drm_device *dev,
 				    struct drm_file *file,
-				    struct intel_ring_buffer *ring,
+				    struct intel_engine_cs *ring,
 				    struct drm_i915_gem_object *obj)
 {
 	/* Unconditionally force add_request to emit a full flush. */
@@ -995,13 +1003,15 @@ i915_gem_execbuffer_retire_commands(struct drm_device *dev,
 
 static int
 i915_reset_gen7_sol_offsets(struct drm_device *dev,
-			    struct intel_ring_buffer *ring)
+			    struct intel_engine_cs *ring)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret, i;
 
-	if (!IS_GEN7(dev) || ring != &dev_priv->ring[RCS])
-		return 0;
+	if (!IS_GEN7(dev) || ring != &dev_priv->ring[RCS]) {
+		DRM_DEBUG("sol reset is gen7/rcs only\n");
+		return -EINVAL;
+	}
 
 	ret = intel_ring_begin(ring, 4 * 3);
 	if (ret)
@@ -1018,6 +1028,37 @@ i915_reset_gen7_sol_offsets(struct drm_device *dev,
 	return 0;
 }
 
+/**
+ * Find one BSD ring to dispatch the corresponding BSD command.
+ * The Ring ID is returned.
+ */
+static int gen8_dispatch_bsd_ring(struct drm_device *dev,
+				  struct drm_file *file)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_file_private *file_priv = file->driver_priv;
+
+	/* Check whether the file_priv is using one ring */
+	if (file_priv->bsd_ring)
+		return file_priv->bsd_ring->id;
+	else {
+		/* If no, use the ping-pong mechanism to select one ring */
+		int ring_id;
+
+		mutex_lock(&dev->struct_mutex);
+		if (dev_priv->mm.bsd_ring_dispatch_index == 0) {
+			ring_id = VCS;
+			dev_priv->mm.bsd_ring_dispatch_index = 1;
+		} else {
+			ring_id = VCS2;
+			dev_priv->mm.bsd_ring_dispatch_index = 0;
+		}
+		file_priv->bsd_ring = &dev_priv->ring[ring_id];
+		mutex_unlock(&dev->struct_mutex);
+		return ring_id;
+	}
+}
+
 static struct drm_i915_gem_object *
 eb_get_batch(struct eb_vmas *eb)
 {
@@ -1047,11 +1088,11 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	struct eb_vmas *eb;
 	struct drm_i915_gem_object *batch_obj;
 	struct drm_clip_rect *cliprects = NULL;
-	struct intel_ring_buffer *ring;
-	struct i915_hw_context *ctx;
+	struct intel_engine_cs *ring;
+	struct intel_context *ctx;
 	struct i915_address_space *vm;
 	const u32 ctx_id = i915_execbuffer2_get_context_id(*args);
-	u32 exec_start = args->batch_start_offset, exec_len;
+	u64 exec_start = args->batch_start_offset, exec_len;
 	u32 mask, flags;
 	int ret, mode, i;
 	bool need_relocs;
@@ -1073,7 +1114,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	if (args->flags & I915_EXEC_IS_PINNED)
 		flags |= I915_DISPATCH_PINNED;
 
-	if ((args->flags & I915_EXEC_RING_MASK) > I915_NUM_RINGS) {
+	if ((args->flags & I915_EXEC_RING_MASK) > LAST_USER_RING) {
 		DRM_DEBUG("execbuf with unknown ring: %d\n",
 			  (int)(args->flags & I915_EXEC_RING_MASK));
 		return -EINVAL;
@@ -1081,7 +1122,14 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 
 	if ((args->flags & I915_EXEC_RING_MASK) == I915_EXEC_DEFAULT)
 		ring = &dev_priv->ring[RCS];
-	else
+	else if ((args->flags & I915_EXEC_RING_MASK) == I915_EXEC_BSD) {
+		if (HAS_BSD2(dev)) {
+			int ring_id;
+			ring_id = gen8_dispatch_bsd_ring(dev, file);
+			ring = &dev_priv->ring[ring_id];
+		} else
+			ring = &dev_priv->ring[VCS];
+	} else
 		ring = &dev_priv->ring[(args->flags & I915_EXEC_RING_MASK) - 1];
 
 	if (!intel_ring_initialized(ring)) {
@@ -1096,14 +1144,22 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	case I915_EXEC_CONSTANTS_REL_GENERAL:
 	case I915_EXEC_CONSTANTS_ABSOLUTE:
 	case I915_EXEC_CONSTANTS_REL_SURFACE:
-		if (ring == &dev_priv->ring[RCS] &&
-		    mode != dev_priv->relative_constants_mode) {
-			if (INTEL_INFO(dev)->gen < 4)
+		if (mode != 0 && ring != &dev_priv->ring[RCS]) {
+			DRM_DEBUG("non-0 rel constants mode on non-RCS\n");
+			return -EINVAL;
+		}
+
+		if (mode != dev_priv->relative_constants_mode) {
+			if (INTEL_INFO(dev)->gen < 4) {
+				DRM_DEBUG("no rel constants on pre-gen4\n");
 				return -EINVAL;
+			}
 
 			if (INTEL_INFO(dev)->gen > 5 &&
-			    mode == I915_EXEC_CONSTANTS_REL_SURFACE)
+			    mode == I915_EXEC_CONSTANTS_REL_SURFACE) {
+				DRM_DEBUG("rel surface constants mode invalid on gen5+\n");
 				return -EINVAL;
+			}
 
 			/* The HW changed the meaning on this bit on gen6 */
 			if (INTEL_INFO(dev)->gen >= 6)
@@ -1151,6 +1207,16 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 			ret = -EFAULT;
 			goto pre_mutex_err;
 		}
+	} else {
+		if (args->DR4 == 0xffffffff) {
+			DRM_DEBUG("UXA submitting garbage DR4, fixing up\n");
+			args->DR4 = 0;
+		}
+
+		if (args->DR1 || args->DR4 || args->cliprects_ptr) {
+			DRM_DEBUG("0 cliprects but dirt in cliprects fields\n");
+			return -EINVAL;
+		}
 	}
 
 	intel_runtime_pm_get(dev_priv);
@@ -1170,7 +1236,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 		mutex_unlock(&dev->struct_mutex);
 		ret = PTR_ERR(ctx);
 		goto pre_mutex_err;
-	} 
+	}
 
 	i915_gem_context_reference(ctx);
 
@@ -1180,6 +1246,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 
 	eb = eb_create(args);
 	if (eb == NULL) {
+		i915_gem_context_unreference(ctx);
 		mutex_unlock(&dev->struct_mutex);
 		ret = -ENOMEM;
 		goto pre_mutex_err;
@@ -1430,6 +1497,11 @@ i915_gem_execbuffer2(struct drm_device *dev, void *data,
 		return -EINVAL;
 	}
 
+	if (args->rsvd2 != 0) {
+		DRM_DEBUG("dirty rvsd2 field\n");
+		return -EINVAL;
+	}
+
 	exec2_list = kmalloc(sizeof(*exec2_list)*args->buffer_count,
 			     GFP_TEMPORARY | __GFP_NOWARN | __GFP_NORETRY);
 	if (exec2_list == NULL)
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 5deb22864c52..eec820aec022 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -30,7 +30,8 @@
 #include "i915_trace.h"
 #include "intel_drv.h"
 
-static void gen8_setup_private_ppat(struct drm_i915_private *dev_priv);
+static void bdw_setup_private_ppat(struct drm_i915_private *dev_priv);
+static void chv_setup_private_ppat(struct drm_i915_private *dev_priv);
 
 bool intel_enable_ppgtt(struct drm_device *dev, bool full)
 {
@@ -65,59 +66,6 @@ static int sanitize_enable_ppgtt(struct drm_device *dev, int enable_ppgtt)
 	return HAS_ALIASING_PPGTT(dev) ? 1 : 0;
 }
 
-#define GEN6_PPGTT_PD_ENTRIES 512
-#define I915_PPGTT_PT_ENTRIES (PAGE_SIZE / sizeof(gen6_gtt_pte_t))
-typedef uint64_t gen8_gtt_pte_t;
-typedef gen8_gtt_pte_t gen8_ppgtt_pde_t;
-
-/* PPGTT stuff */
-#define GEN6_GTT_ADDR_ENCODE(addr)	((addr) | (((addr) >> 28) & 0xff0))
-#define HSW_GTT_ADDR_ENCODE(addr)	((addr) | (((addr) >> 28) & 0x7f0))
-
-#define GEN6_PDE_VALID			(1 << 0)
-/* gen6+ has bit 11-4 for physical addr bit 39-32 */
-#define GEN6_PDE_ADDR_ENCODE(addr)	GEN6_GTT_ADDR_ENCODE(addr)
-
-#define GEN6_PTE_VALID			(1 << 0)
-#define GEN6_PTE_UNCACHED		(1 << 1)
-#define HSW_PTE_UNCACHED		(0)
-#define GEN6_PTE_CACHE_LLC		(2 << 1)
-#define GEN7_PTE_CACHE_L3_LLC		(3 << 1)
-#define GEN6_PTE_ADDR_ENCODE(addr)	GEN6_GTT_ADDR_ENCODE(addr)
-#define HSW_PTE_ADDR_ENCODE(addr)	HSW_GTT_ADDR_ENCODE(addr)
-
-/* Cacheability Control is a 4-bit value. The low three bits are stored in *
- * bits 3:1 of the PTE, while the fourth bit is stored in bit 11 of the PTE.
- */
-#define HSW_CACHEABILITY_CONTROL(bits)	((((bits) & 0x7) << 1) | \
-					 (((bits) & 0x8) << (11 - 3)))
-#define HSW_WB_LLC_AGE3			HSW_CACHEABILITY_CONTROL(0x2)
-#define HSW_WB_LLC_AGE0			HSW_CACHEABILITY_CONTROL(0x3)
-#define HSW_WB_ELLC_LLC_AGE0		HSW_CACHEABILITY_CONTROL(0xb)
-#define HSW_WB_ELLC_LLC_AGE3		HSW_CACHEABILITY_CONTROL(0x8)
-#define HSW_WT_ELLC_LLC_AGE0		HSW_CACHEABILITY_CONTROL(0x6)
-#define HSW_WT_ELLC_LLC_AGE3		HSW_CACHEABILITY_CONTROL(0x7)
-
-#define GEN8_PTES_PER_PAGE		(PAGE_SIZE / sizeof(gen8_gtt_pte_t))
-#define GEN8_PDES_PER_PAGE		(PAGE_SIZE / sizeof(gen8_ppgtt_pde_t))
-
-/* GEN8 legacy style addressis defined as a 3 level page table:
- * 31:30 | 29:21 | 20:12 |  11:0
- * PDPE  |  PDE  |  PTE  | offset
- * The difference as compared to normal x86 3 level page table is the PDPEs are
- * programmed via register.
- */
-#define GEN8_PDPE_SHIFT			30
-#define GEN8_PDPE_MASK			0x3
-#define GEN8_PDE_SHIFT			21
-#define GEN8_PDE_MASK			0x1ff
-#define GEN8_PTE_SHIFT			12
-#define GEN8_PTE_MASK			0x1ff
-
-#define PPAT_UNCACHED_INDEX		(_PAGE_PWT | _PAGE_PCD)
-#define PPAT_CACHED_PDE_INDEX		0 /* WB LLC */
-#define PPAT_CACHED_INDEX		_PAGE_PAT /* WB LLCeLLC */
-#define PPAT_DISPLAY_ELLC_INDEX		_PAGE_PCD /* WT eLLC */
 
 static void ppgtt_bind_vma(struct i915_vma *vma,
 			   enum i915_cache_level cache_level,
@@ -131,10 +79,19 @@ static inline gen8_gtt_pte_t gen8_pte_encode(dma_addr_t addr,
 {
 	gen8_gtt_pte_t pte = valid ? _PAGE_PRESENT | _PAGE_RW : 0;
 	pte |= addr;
-	if (level != I915_CACHE_NONE)
-		pte |= PPAT_CACHED_INDEX;
-	else
+
+	switch (level) {
+	case I915_CACHE_NONE:
 		pte |= PPAT_UNCACHED_INDEX;
+		break;
+	case I915_CACHE_WT:
+		pte |= PPAT_DISPLAY_ELLC_INDEX;
+		break;
+	default:
+		pte |= PPAT_CACHED_INDEX;
+		break;
+	}
+
 	return pte;
 }
 
@@ -197,9 +154,6 @@ static gen6_gtt_pte_t ivb_pte_encode(dma_addr_t addr,
 	return pte;
 }
 
-#define BYT_PTE_WRITEABLE		(1 << 1)
-#define BYT_PTE_SNOOPED_BY_CPU_CACHES	(1 << 2)
-
 static gen6_gtt_pte_t byt_pte_encode(dma_addr_t addr,
 				     enum i915_cache_level level,
 				     bool valid)
@@ -253,7 +207,7 @@ static gen6_gtt_pte_t iris_pte_encode(dma_addr_t addr,
 }
 
 /* Broadwell Page Directory Pointer Descriptors */
-static int gen8_write_pdp(struct intel_ring_buffer *ring, unsigned entry,
+static int gen8_write_pdp(struct intel_engine_cs *ring, unsigned entry,
 			   uint64_t val, bool synchronous)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
@@ -283,7 +237,7 @@ static int gen8_write_pdp(struct intel_ring_buffer *ring, unsigned entry,
 }
 
 static int gen8_mm_switch(struct i915_hw_ppgtt *ppgtt,
-			  struct intel_ring_buffer *ring,
+			  struct intel_engine_cs *ring,
 			  bool synchronous)
 {
 	int i, ret;
@@ -332,6 +286,8 @@ static void gen8_ppgtt_clear_range(struct i915_address_space *vm,
 			num_entries--;
 		}
 
+		if (!HAS_LLC(ppgtt->base.dev))
+			drm_clflush_virt_range(pt_vaddr, PAGE_SIZE);
 		kunmap_atomic(pt_vaddr);
 
 		pte = 0;
@@ -368,6 +324,8 @@ static void gen8_ppgtt_insert_entries(struct i915_address_space *vm,
 			gen8_pte_encode(sg_page_iter_dma_address(&sg_iter),
 					cache_level, true);
 		if (++pte == GEN8_PTES_PER_PAGE) {
+			if (!HAS_LLC(ppgtt->base.dev))
+				drm_clflush_virt_range(pt_vaddr, PAGE_SIZE);
 			kunmap_atomic(pt_vaddr);
 			pt_vaddr = NULL;
 			if (++pde == GEN8_PDES_PER_PAGE) {
@@ -377,8 +335,11 @@ static void gen8_ppgtt_insert_entries(struct i915_address_space *vm,
 			pte = 0;
 		}
 	}
-	if (pt_vaddr)
+	if (pt_vaddr) {
+		if (!HAS_LLC(ppgtt->base.dev))
+			drm_clflush_virt_range(pt_vaddr, PAGE_SIZE);
 		kunmap_atomic(pt_vaddr);
+	}
 }
 
 static void gen8_free_page_tables(struct page **pt_pages)
@@ -641,6 +602,8 @@ static int gen8_ppgtt_init(struct i915_hw_ppgtt *ppgtt, uint64_t size)
 			pd_vaddr[j] = gen8_pde_encode(ppgtt->base.dev, addr,
 						      I915_CACHE_LLC);
 		}
+		if (!HAS_LLC(ppgtt->base.dev))
+			drm_clflush_virt_range(pd_vaddr, PAGE_SIZE);
 		kunmap_atomic(pd_vaddr);
 	}
 
@@ -753,7 +716,7 @@ static uint32_t get_pd_offset(struct i915_hw_ppgtt *ppgtt)
 }
 
 static int hsw_mm_switch(struct i915_hw_ppgtt *ppgtt,
-			 struct intel_ring_buffer *ring,
+			 struct intel_engine_cs *ring,
 			 bool synchronous)
 {
 	struct drm_device *dev = ppgtt->base.dev;
@@ -797,7 +760,7 @@ static int hsw_mm_switch(struct i915_hw_ppgtt *ppgtt,
 }
 
 static int gen7_mm_switch(struct i915_hw_ppgtt *ppgtt,
-			  struct intel_ring_buffer *ring,
+			  struct intel_engine_cs *ring,
 			  bool synchronous)
 {
 	struct drm_device *dev = ppgtt->base.dev;
@@ -848,7 +811,7 @@ static int gen7_mm_switch(struct i915_hw_ppgtt *ppgtt,
 }
 
 static int gen6_mm_switch(struct i915_hw_ppgtt *ppgtt,
-			  struct intel_ring_buffer *ring,
+			  struct intel_engine_cs *ring,
 			  bool synchronous)
 {
 	struct drm_device *dev = ppgtt->base.dev;
@@ -869,7 +832,7 @@ static int gen8_ppgtt_enable(struct i915_hw_ppgtt *ppgtt)
 {
 	struct drm_device *dev = ppgtt->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int j, ret;
 
 	for_each_ring(ring, dev_priv, j) {
@@ -899,7 +862,7 @@ static int gen7_ppgtt_enable(struct i915_hw_ppgtt *ppgtt)
 {
 	struct drm_device *dev = ppgtt->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	uint32_t ecochk, ecobits;
 	int i;
 
@@ -938,7 +901,7 @@ static int gen6_ppgtt_enable(struct i915_hw_ppgtt *ppgtt)
 {
 	struct drm_device *dev = ppgtt->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	uint32_t ecochk, gab_ctl, ecobits;
 	int i;
 
@@ -1067,8 +1030,6 @@ static void gen6_ppgtt_cleanup(struct i915_address_space *vm)
 
 static int gen6_ppgtt_allocate_page_directories(struct i915_hw_ppgtt *ppgtt)
 {
-#define GEN6_PD_ALIGN (PAGE_SIZE * 16)
-#define GEN6_PD_SIZE (GEN6_PPGTT_PD_ENTRIES * PAGE_SIZE)
 	struct drm_device *dev = ppgtt->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	bool retried = false;
@@ -1084,8 +1045,7 @@ alloc:
 						  &ppgtt->node, GEN6_PD_SIZE,
 						  GEN6_PD_ALIGN, 0,
 						  0, dev_priv->gtt.base.total,
-						  DRM_MM_SEARCH_DEFAULT,
-						  DRM_MM_CREATE_DEFAULT);
+						  DRM_MM_TOPDOWN);
 	if (ret == -ENOSPC && !retried) {
 		ret = i915_gem_evict_something(dev, &dev_priv->gtt.base,
 					       GEN6_PD_SIZE, GEN6_PD_ALIGN,
@@ -1311,7 +1271,7 @@ static void undo_idling(struct drm_i915_private *dev_priv, bool interruptible)
 void i915_check_and_clear_faults(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int i;
 
 	if (INTEL_INFO(dev)->gen < 6)
@@ -1386,7 +1346,11 @@ void i915_gem_restore_gtt_mappings(struct drm_device *dev)
 
 
 	if (INTEL_INFO(dev)->gen >= 8) {
-		gen8_setup_private_ppat(dev_priv);
+		if (IS_CHERRYVIEW(dev))
+			chv_setup_private_ppat(dev_priv);
+		else
+			bdw_setup_private_ppat(dev_priv);
+
 		return;
 	}
 
@@ -1438,7 +1402,7 @@ static void gen8_ggtt_insert_entries(struct i915_address_space *vm,
 		(gen8_gtt_pte_t __iomem *)dev_priv->gtt.gsm + first_entry;
 	int i = 0;
 	struct sg_page_iter sg_iter;
-	dma_addr_t addr;
+	dma_addr_t addr = 0;
 
 	for_each_sg_page(st->sgl, &sg_iter, st->nents, 0) {
 		addr = sg_dma_address(sg_iter.sg) +
@@ -1811,9 +1775,27 @@ static inline unsigned int gen8_get_total_gtt_size(u16 bdw_gmch_ctl)
 	bdw_gmch_ctl &= BDW_GMCH_GGMS_MASK;
 	if (bdw_gmch_ctl)
 		bdw_gmch_ctl = 1 << bdw_gmch_ctl;
+
+#ifdef CONFIG_X86_32
+	/* Limit 32b platforms to a 2GB GGTT: 4 << 20 / pte size * PAGE_SIZE */
+	if (bdw_gmch_ctl > 4)
+		bdw_gmch_ctl = 4;
+#endif
+
 	return bdw_gmch_ctl << 20;
 }
 
+static inline unsigned int chv_get_total_gtt_size(u16 gmch_ctrl)
+{
+	gmch_ctrl >>= SNB_GMCH_GGMS_SHIFT;
+	gmch_ctrl &= SNB_GMCH_GGMS_MASK;
+
+	if (gmch_ctrl)
+		return 1 << (20 + gmch_ctrl);
+
+	return 0;
+}
+
 static inline size_t gen6_get_stolen_size(u16 snb_gmch_ctl)
 {
 	snb_gmch_ctl >>= SNB_GMCH_GMS_SHIFT;
@@ -1828,6 +1810,24 @@ static inline size_t gen8_get_stolen_size(u16 bdw_gmch_ctl)
 	return bdw_gmch_ctl << 25; /* 32 MB units */
 }
 
+static size_t chv_get_stolen_size(u16 gmch_ctrl)
+{
+	gmch_ctrl >>= SNB_GMCH_GMS_SHIFT;
+	gmch_ctrl &= SNB_GMCH_GMS_MASK;
+
+	/*
+	 * 0x0  to 0x10: 32MB increments starting at 0MB
+	 * 0x11 to 0x16: 4MB increments starting at 8MB
+	 * 0x17 to 0x1d: 4MB increments start at 36MB
+	 */
+	if (gmch_ctrl < 0x11)
+		return gmch_ctrl << 25;
+	else if (gmch_ctrl < 0x17)
+		return (gmch_ctrl - 0x11 + 2) << 22;
+	else
+		return (gmch_ctrl - 0x17 + 9) << 22;
+}
+
 static int ggtt_probe_common(struct drm_device *dev,
 			     size_t gtt_size)
 {
@@ -1858,19 +1858,8 @@ static int ggtt_probe_common(struct drm_device *dev,
 /* The GGTT and PPGTT need a private PPAT setup in order to handle cacheability
  * bits. When using advanced contexts each context stores its own PAT, but
  * writing this data shouldn't be harmful even in those cases. */
-static void gen8_setup_private_ppat(struct drm_i915_private *dev_priv)
-{
-#define GEN8_PPAT_UC		(0<<0)
-#define GEN8_PPAT_WC		(1<<0)
-#define GEN8_PPAT_WT		(2<<0)
-#define GEN8_PPAT_WB		(3<<0)
-#define GEN8_PPAT_ELLC_OVERRIDE	(0<<2)
-/* FIXME(BDW): Bspec is completely confused about cache control bits. */
-#define GEN8_PPAT_LLC		(1<<2)
-#define GEN8_PPAT_LLCELLC	(2<<2)
-#define GEN8_PPAT_LLCeLLC	(3<<2)
-#define GEN8_PPAT_AGE(x)	(x<<4)
-#define GEN8_PPAT(i, x) ((uint64_t) (x) << ((i) * 8))
+static void bdw_setup_private_ppat(struct drm_i915_private *dev_priv)
+{
 	uint64_t pat;
 
 	pat = GEN8_PPAT(0, GEN8_PPAT_WB | GEN8_PPAT_LLC)     | /* for normal objects, no eLLC */
@@ -1888,6 +1877,33 @@ static void gen8_setup_private_ppat(struct drm_i915_private *dev_priv)
 	I915_WRITE(GEN8_PRIVATE_PAT + 4, pat >> 32);
 }
 
+static void chv_setup_private_ppat(struct drm_i915_private *dev_priv)
+{
+	uint64_t pat;
+
+	/*
+	 * Map WB on BDW to snooped on CHV.
+	 *
+	 * Only the snoop bit has meaning for CHV, the rest is
+	 * ignored.
+	 *
+	 * Note that the harware enforces snooping for all page
+	 * table accesses. The snoop bit is actually ignored for
+	 * PDEs.
+	 */
+	pat = GEN8_PPAT(0, CHV_PPAT_SNOOP) |
+	      GEN8_PPAT(1, 0) |
+	      GEN8_PPAT(2, 0) |
+	      GEN8_PPAT(3, 0) |
+	      GEN8_PPAT(4, CHV_PPAT_SNOOP) |
+	      GEN8_PPAT(5, CHV_PPAT_SNOOP) |
+	      GEN8_PPAT(6, CHV_PPAT_SNOOP) |
+	      GEN8_PPAT(7, CHV_PPAT_SNOOP);
+
+	I915_WRITE(GEN8_PRIVATE_PAT, pat);
+	I915_WRITE(GEN8_PRIVATE_PAT + 4, pat >> 32);
+}
+
 static int gen8_gmch_probe(struct drm_device *dev,
 			   size_t *gtt_total,
 			   size_t *stolen,
@@ -1908,12 +1924,20 @@ static int gen8_gmch_probe(struct drm_device *dev,
 
 	pci_read_config_word(dev->pdev, SNB_GMCH_CTRL, &snb_gmch_ctl);
 
-	*stolen = gen8_get_stolen_size(snb_gmch_ctl);
+	if (IS_CHERRYVIEW(dev)) {
+		*stolen = chv_get_stolen_size(snb_gmch_ctl);
+		gtt_size = chv_get_total_gtt_size(snb_gmch_ctl);
+	} else {
+		*stolen = gen8_get_stolen_size(snb_gmch_ctl);
+		gtt_size = gen8_get_total_gtt_size(snb_gmch_ctl);
+	}
 
-	gtt_size = gen8_get_total_gtt_size(snb_gmch_ctl);
 	*gtt_total = (gtt_size / sizeof(gen8_gtt_pte_t)) << PAGE_SHIFT;
 
-	gen8_setup_private_ppat(dev_priv);
+	if (IS_CHERRYVIEW(dev))
+		chv_setup_private_ppat(dev_priv);
+	else
+		bdw_setup_private_ppat(dev_priv);
 
 	ret = ggtt_probe_common(dev, gtt_size);
 
@@ -2043,6 +2067,10 @@ int i915_gem_gtt_init(struct drm_device *dev)
 		 gtt->base.total >> 20);
 	DRM_DEBUG_DRIVER("GMADR size = %ldM\n", gtt->mappable_end >> 20);
 	DRM_DEBUG_DRIVER("GTT stolen size = %zdM\n", gtt->stolen_size >> 20);
+#ifdef CONFIG_INTEL_IOMMU
+	if (intel_iommu_gfx_mapped)
+		DRM_INFO("VT-d active for gfx access\n");
+#endif
 	/*
 	 * i915.enable_ppgtt is read-only, so do an early pass to validate the
 	 * user's requested state against the hardware/driver capabilities.  We
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h
new file mode 100644
index 000000000000..1b96a06be3cb
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.h
@@ -0,0 +1,284 @@
+/*
+ * Copyright © 2014 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ * Please try to maintain the following order within this file unless it makes
+ * sense to do otherwise. From top to bottom:
+ * 1. typedefs
+ * 2. #defines, and macros
+ * 3. structure definitions
+ * 4. function prototypes
+ *
+ * Within each section, please try to order by generation in ascending order,
+ * from top to bottom (ie. gen6 on the top, gen8 on the bottom).
+ */
+
+#ifndef __I915_GEM_GTT_H__
+#define __I915_GEM_GTT_H__
+
+typedef uint32_t gen6_gtt_pte_t;
+typedef uint64_t gen8_gtt_pte_t;
+typedef gen8_gtt_pte_t gen8_ppgtt_pde_t;
+
+#define gtt_total_entries(gtt) ((gtt).base.total >> PAGE_SHIFT)
+
+#define I915_PPGTT_PT_ENTRIES		(PAGE_SIZE / sizeof(gen6_gtt_pte_t))
+/* gen6-hsw has bit 11-4 for physical addr bit 39-32 */
+#define GEN6_GTT_ADDR_ENCODE(addr)	((addr) | (((addr) >> 28) & 0xff0))
+#define GEN6_PTE_ADDR_ENCODE(addr)	GEN6_GTT_ADDR_ENCODE(addr)
+#define GEN6_PDE_ADDR_ENCODE(addr)	GEN6_GTT_ADDR_ENCODE(addr)
+#define GEN6_PTE_CACHE_LLC		(2 << 1)
+#define GEN6_PTE_UNCACHED		(1 << 1)
+#define GEN6_PTE_VALID			(1 << 0)
+
+#define GEN6_PPGTT_PD_ENTRIES		512
+#define GEN6_PD_SIZE			(GEN6_PPGTT_PD_ENTRIES * PAGE_SIZE)
+#define GEN6_PD_ALIGN			(PAGE_SIZE * 16)
+#define GEN6_PDE_VALID			(1 << 0)
+
+#define GEN7_PTE_CACHE_L3_LLC		(3 << 1)
+
+#define BYT_PTE_SNOOPED_BY_CPU_CACHES	(1 << 2)
+#define BYT_PTE_WRITEABLE		(1 << 1)
+
+/* Cacheability Control is a 4-bit value. The low three bits are stored in bits
+ * 3:1 of the PTE, while the fourth bit is stored in bit 11 of the PTE.
+ */
+#define HSW_CACHEABILITY_CONTROL(bits)	((((bits) & 0x7) << 1) | \
+					 (((bits) & 0x8) << (11 - 3)))
+#define HSW_WB_LLC_AGE3			HSW_CACHEABILITY_CONTROL(0x2)
+#define HSW_WB_LLC_AGE0			HSW_CACHEABILITY_CONTROL(0x3)
+#define HSW_WB_ELLC_LLC_AGE3		HSW_CACHEABILITY_CONTROL(0x8)
+#define HSW_WB_ELLC_LLC_AGE0		HSW_CACHEABILITY_CONTROL(0xb)
+#define HSW_WT_ELLC_LLC_AGE3		HSW_CACHEABILITY_CONTROL(0x7)
+#define HSW_WT_ELLC_LLC_AGE0		HSW_CACHEABILITY_CONTROL(0x6)
+#define HSW_PTE_UNCACHED		(0)
+#define HSW_GTT_ADDR_ENCODE(addr)	((addr) | (((addr) >> 28) & 0x7f0))
+#define HSW_PTE_ADDR_ENCODE(addr)	HSW_GTT_ADDR_ENCODE(addr)
+
+/* GEN8 legacy style address is defined as a 3 level page table:
+ * 31:30 | 29:21 | 20:12 |  11:0
+ * PDPE  |  PDE  |  PTE  | offset
+ * The difference as compared to normal x86 3 level page table is the PDPEs are
+ * programmed via register.
+ */
+#define GEN8_PDPE_SHIFT			30
+#define GEN8_PDPE_MASK			0x3
+#define GEN8_PDE_SHIFT			21
+#define GEN8_PDE_MASK			0x1ff
+#define GEN8_PTE_SHIFT			12
+#define GEN8_PTE_MASK			0x1ff
+#define GEN8_LEGACY_PDPS		4
+#define GEN8_PTES_PER_PAGE		(PAGE_SIZE / sizeof(gen8_gtt_pte_t))
+#define GEN8_PDES_PER_PAGE		(PAGE_SIZE / sizeof(gen8_ppgtt_pde_t))
+
+#define PPAT_UNCACHED_INDEX		(_PAGE_PWT | _PAGE_PCD)
+#define PPAT_CACHED_PDE_INDEX		0 /* WB LLC */
+#define PPAT_CACHED_INDEX		_PAGE_PAT /* WB LLCeLLC */
+#define PPAT_DISPLAY_ELLC_INDEX		_PAGE_PCD /* WT eLLC */
+
+#define CHV_PPAT_SNOOP			(1<<6)
+#define GEN8_PPAT_AGE(x)		(x<<4)
+#define GEN8_PPAT_LLCeLLC		(3<<2)
+#define GEN8_PPAT_LLCELLC		(2<<2)
+#define GEN8_PPAT_LLC			(1<<2)
+#define GEN8_PPAT_WB			(3<<0)
+#define GEN8_PPAT_WT			(2<<0)
+#define GEN8_PPAT_WC			(1<<0)
+#define GEN8_PPAT_UC			(0<<0)
+#define GEN8_PPAT_ELLC_OVERRIDE		(0<<2)
+#define GEN8_PPAT(i, x)			((uint64_t) (x) << ((i) * 8))
+
+enum i915_cache_level;
+/**
+ * A VMA represents a GEM BO that is bound into an address space. Therefore, a
+ * VMA's presence cannot be guaranteed before binding, or after unbinding the
+ * object into/from the address space.
+ *
+ * To make things as simple as possible (ie. no refcounting), a VMA's lifetime
+ * will always be <= an objects lifetime. So object refcounting should cover us.
+ */
+struct i915_vma {
+	struct drm_mm_node node;
+	struct drm_i915_gem_object *obj;
+	struct i915_address_space *vm;
+
+	/** This object's place on the active/inactive lists */
+	struct list_head mm_list;
+
+	struct list_head vma_link; /* Link in the object's VMA list */
+
+	/** This vma's place in the batchbuffer or on the eviction list */
+	struct list_head exec_list;
+
+	/**
+	 * Used for performing relocations during execbuffer insertion.
+	 */
+	struct hlist_node exec_node;
+	unsigned long exec_handle;
+	struct drm_i915_gem_exec_object2 *exec_entry;
+
+	/**
+	 * How many users have pinned this object in GTT space. The following
+	 * users can each hold at most one reference: pwrite/pread, pin_ioctl
+	 * (via user_pin_count), execbuffer (objects are not allowed multiple
+	 * times for the same batchbuffer), and the framebuffer code. When
+	 * switching/pageflipping, the framebuffer code has at most two buffers
+	 * pinned per crtc.
+	 *
+	 * In the worst case this is 1 + 1 + 1 + 2*2 = 7. That would fit into 3
+	 * bits with absolutely no headroom. So use 4 bits. */
+	unsigned int pin_count:4;
+#define DRM_I915_GEM_OBJECT_MAX_PIN_COUNT 0xf
+
+	/** Unmap an object from an address space. This usually consists of
+	 * setting the valid PTE entries to a reserved scratch page. */
+	void (*unbind_vma)(struct i915_vma *vma);
+	/* Map an object into an address space with the given cache flags. */
+#define GLOBAL_BIND (1<<0)
+	void (*bind_vma)(struct i915_vma *vma,
+			 enum i915_cache_level cache_level,
+			 u32 flags);
+};
+
+struct i915_address_space {
+	struct drm_mm mm;
+	struct drm_device *dev;
+	struct list_head global_link;
+	unsigned long start;		/* Start offset always 0 for dri2 */
+	size_t total;		/* size addr space maps (ex. 2GB for ggtt) */
+
+	struct {
+		dma_addr_t addr;
+		struct page *page;
+	} scratch;
+
+	/**
+	 * List of objects currently involved in rendering.
+	 *
+	 * Includes buffers having the contents of their GPU caches
+	 * flushed, not necessarily primitives.  last_rendering_seqno
+	 * represents when the rendering involved will be completed.
+	 *
+	 * A reference is held on the buffer while on this list.
+	 */
+	struct list_head active_list;
+
+	/**
+	 * LRU list of objects which are not in the ringbuffer and
+	 * are ready to unbind, but are still in the GTT.
+	 *
+	 * last_rendering_seqno is 0 while an object is in this list.
+	 *
+	 * A reference is not held on the buffer while on this list,
+	 * as merely being GTT-bound shouldn't prevent its being
+	 * freed, and we'll pull it off the list in the free path.
+	 */
+	struct list_head inactive_list;
+
+	/* FIXME: Need a more generic return type */
+	gen6_gtt_pte_t (*pte_encode)(dma_addr_t addr,
+				     enum i915_cache_level level,
+				     bool valid); /* Create a valid PTE */
+	void (*clear_range)(struct i915_address_space *vm,
+			    uint64_t start,
+			    uint64_t length,
+			    bool use_scratch);
+	void (*insert_entries)(struct i915_address_space *vm,
+			       struct sg_table *st,
+			       uint64_t start,
+			       enum i915_cache_level cache_level);
+	void (*cleanup)(struct i915_address_space *vm);
+};
+
+/* The Graphics Translation Table is the way in which GEN hardware translates a
+ * Graphics Virtual Address into a Physical Address. In addition to the normal
+ * collateral associated with any va->pa translations GEN hardware also has a
+ * portion of the GTT which can be mapped by the CPU and remain both coherent
+ * and correct (in cases like swizzling). That region is referred to as GMADR in
+ * the spec.
+ */
+struct i915_gtt {
+	struct i915_address_space base;
+	size_t stolen_size;		/* Total size of stolen memory */
+
+	unsigned long mappable_end;	/* End offset that we can CPU map */
+	struct io_mapping *mappable;	/* Mapping to our CPU mappable region */
+	phys_addr_t mappable_base;	/* PA of our GMADR */
+
+	/** "Graphics Stolen Memory" holds the global PTEs */
+	void __iomem *gsm;
+
+	bool do_idle_maps;
+
+	int mtrr;
+
+	/* global gtt ops */
+	int (*gtt_probe)(struct drm_device *dev, size_t *gtt_total,
+			  size_t *stolen, phys_addr_t *mappable_base,
+			  unsigned long *mappable_end);
+};
+
+struct i915_hw_ppgtt {
+	struct i915_address_space base;
+	struct kref ref;
+	struct drm_mm_node node;
+	unsigned num_pd_entries;
+	unsigned num_pd_pages; /* gen8+ */
+	union {
+		struct page **pt_pages;
+		struct page **gen8_pt_pages[GEN8_LEGACY_PDPS];
+	};
+	struct page *pd_pages;
+	union {
+		uint32_t pd_offset;
+		dma_addr_t pd_dma_addr[GEN8_LEGACY_PDPS];
+	};
+	union {
+		dma_addr_t *pt_dma_addr;
+		dma_addr_t *gen8_pt_dma_addr[4];
+	};
+
+	struct intel_context *ctx;
+
+	int (*enable)(struct i915_hw_ppgtt *ppgtt);
+	int (*switch_mm)(struct i915_hw_ppgtt *ppgtt,
+			 struct intel_engine_cs *ring,
+			 bool synchronous);
+	void (*debug_dump)(struct i915_hw_ppgtt *ppgtt, struct seq_file *m);
+};
+
+int i915_gem_gtt_init(struct drm_device *dev);
+void i915_gem_init_global_gtt(struct drm_device *dev);
+void i915_gem_setup_global_gtt(struct drm_device *dev, unsigned long start,
+			       unsigned long mappable_end, unsigned long end);
+
+bool intel_enable_ppgtt(struct drm_device *dev, bool full);
+int i915_gem_init_ppgtt(struct drm_device *dev, struct i915_hw_ppgtt *ppgtt);
+
+void i915_check_and_clear_faults(struct drm_device *dev);
+void i915_gem_suspend_gtt_mappings(struct drm_device *dev);
+void i915_gem_restore_gtt_mappings(struct drm_device *dev);
+
+int __must_check i915_gem_gtt_prepare_object(struct drm_i915_gem_object *obj);
+void i915_gem_gtt_finish_object(struct drm_i915_gem_object *obj);
+
+#endif
diff --git a/drivers/gpu/drm/i915/i915_gem_render_state.c b/drivers/gpu/drm/i915/i915_gem_render_state.c
new file mode 100644
index 000000000000..3521f998a178
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_gem_render_state.c
@@ -0,0 +1,198 @@
+/*
+ * Copyright © 2014 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ * Authors:
+ *    Mika Kuoppala <mika.kuoppala@intel.com>
+ *
+ */
+
+#include "i915_drv.h"
+#include "intel_renderstate.h"
+
+struct i915_render_state {
+	struct drm_i915_gem_object *obj;
+	unsigned long ggtt_offset;
+	void *batch;
+	u32 size;
+	u32 len;
+};
+
+static struct i915_render_state *render_state_alloc(struct drm_device *dev)
+{
+	struct i915_render_state *so;
+	struct page *page;
+	int ret;
+
+	so = kzalloc(sizeof(*so), GFP_KERNEL);
+	if (!so)
+		return ERR_PTR(-ENOMEM);
+
+	so->obj = i915_gem_alloc_object(dev, 4096);
+	if (so->obj == NULL) {
+		ret = -ENOMEM;
+		goto free;
+	}
+	so->size = 4096;
+
+	ret = i915_gem_obj_ggtt_pin(so->obj, 4096, 0);
+	if (ret)
+		goto free_gem;
+
+	BUG_ON(so->obj->pages->nents != 1);
+	page = sg_page(so->obj->pages->sgl);
+
+	so->batch = kmap(page);
+	if (!so->batch) {
+		ret = -ENOMEM;
+		goto unpin;
+	}
+
+	so->ggtt_offset = i915_gem_obj_ggtt_offset(so->obj);
+
+	return so;
+unpin:
+	i915_gem_object_ggtt_unpin(so->obj);
+free_gem:
+	drm_gem_object_unreference(&so->obj->base);
+free:
+	kfree(so);
+	return ERR_PTR(ret);
+}
+
+static void render_state_free(struct i915_render_state *so)
+{
+	kunmap(so->batch);
+	i915_gem_object_ggtt_unpin(so->obj);
+	drm_gem_object_unreference(&so->obj->base);
+	kfree(so);
+}
+
+static const struct intel_renderstate_rodata *
+render_state_get_rodata(struct drm_device *dev, const int gen)
+{
+	switch (gen) {
+	case 6:
+		return &gen6_null_state;
+	case 7:
+		return &gen7_null_state;
+	case 8:
+		return &gen8_null_state;
+	}
+
+	return NULL;
+}
+
+static int render_state_setup(const int gen,
+			      const struct intel_renderstate_rodata *rodata,
+			      struct i915_render_state *so)
+{
+	const u64 goffset = i915_gem_obj_ggtt_offset(so->obj);
+	u32 reloc_index = 0;
+	u32 * const d = so->batch;
+	unsigned int i = 0;
+	int ret;
+
+	if (!rodata || rodata->batch_items * 4 > so->size)
+		return -EINVAL;
+
+	ret = i915_gem_object_set_to_cpu_domain(so->obj, true);
+	if (ret)
+		return ret;
+
+	while (i < rodata->batch_items) {
+		u32 s = rodata->batch[i];
+
+		if (reloc_index < rodata->reloc_items &&
+		    i * 4  == rodata->reloc[reloc_index]) {
+
+			s += goffset & 0xffffffff;
+
+			/* We keep batch offsets max 32bit */
+			if (gen >= 8) {
+				if (i + 1 >= rodata->batch_items ||
+				    rodata->batch[i + 1] != 0)
+					return -EINVAL;
+
+				d[i] = s;
+				i++;
+				s = (goffset & 0xffffffff00000000ull) >> 32;
+			}
+
+			reloc_index++;
+		}
+
+		d[i] = s;
+		i++;
+	}
+
+	ret = i915_gem_object_set_to_gtt_domain(so->obj, false);
+	if (ret)
+		return ret;
+
+	if (rodata->reloc_items != reloc_index) {
+		DRM_ERROR("not all relocs resolved, %d out of %d\n",
+			  reloc_index, rodata->reloc_items);
+		return -EINVAL;
+	}
+
+	so->len = rodata->batch_items * 4;
+
+	return 0;
+}
+
+int i915_gem_render_state_init(struct intel_engine_cs *ring)
+{
+	const int gen = INTEL_INFO(ring->dev)->gen;
+	struct i915_render_state *so;
+	const struct intel_renderstate_rodata *rodata;
+	int ret;
+
+	if (WARN_ON(ring->id != RCS))
+		return -ENOENT;
+
+	rodata = render_state_get_rodata(ring->dev, gen);
+	if (rodata == NULL)
+		return 0;
+
+	so = render_state_alloc(ring->dev);
+	if (IS_ERR(so))
+		return PTR_ERR(so);
+
+	ret = render_state_setup(gen, rodata, so);
+	if (ret)
+		goto out;
+
+	ret = ring->dispatch_execbuffer(ring,
+					i915_gem_obj_ggtt_offset(so->obj),
+					so->len,
+					I915_DISPATCH_SECURE);
+	if (ret)
+		goto out;
+
+	i915_vma_move_to_active(i915_gem_obj_to_ggtt(so->obj), ring);
+
+	ret = __i915_add_request(ring, NULL, so->obj, NULL);
+	/* __i915_add_request moves object to inactive if it fails */
+out:
+	render_state_free(so);
+	return ret;
+}
diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c
new file mode 100644
index 000000000000..21ea92886a56
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_gem_userptr.c
@@ -0,0 +1,711 @@
+/*
+ * Copyright © 2012-2014 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ */
+
+#include "drmP.h"
+#include "i915_drm.h"
+#include "i915_drv.h"
+#include "i915_trace.h"
+#include "intel_drv.h"
+#include <linux/mmu_context.h>
+#include <linux/mmu_notifier.h>
+#include <linux/mempolicy.h>
+#include <linux/swap.h>
+
+#if defined(CONFIG_MMU_NOTIFIER)
+#include <linux/interval_tree.h>
+
+struct i915_mmu_notifier {
+	spinlock_t lock;
+	struct hlist_node node;
+	struct mmu_notifier mn;
+	struct rb_root objects;
+	struct drm_device *dev;
+	struct mm_struct *mm;
+	struct work_struct work;
+	unsigned long count;
+	unsigned long serial;
+};
+
+struct i915_mmu_object {
+	struct i915_mmu_notifier *mmu;
+	struct interval_tree_node it;
+	struct drm_i915_gem_object *obj;
+};
+
+static void i915_gem_userptr_mn_invalidate_range_start(struct mmu_notifier *_mn,
+						       struct mm_struct *mm,
+						       unsigned long start,
+						       unsigned long end)
+{
+	struct i915_mmu_notifier *mn = container_of(_mn, struct i915_mmu_notifier, mn);
+	struct interval_tree_node *it = NULL;
+	unsigned long serial = 0;
+
+	end--; /* interval ranges are inclusive, but invalidate range is exclusive */
+	while (start < end) {
+		struct drm_i915_gem_object *obj;
+
+		obj = NULL;
+		spin_lock(&mn->lock);
+		if (serial == mn->serial)
+			it = interval_tree_iter_next(it, start, end);
+		else
+			it = interval_tree_iter_first(&mn->objects, start, end);
+		if (it != NULL) {
+			obj = container_of(it, struct i915_mmu_object, it)->obj;
+			drm_gem_object_reference(&obj->base);
+			serial = mn->serial;
+		}
+		spin_unlock(&mn->lock);
+		if (obj == NULL)
+			return;
+
+		mutex_lock(&mn->dev->struct_mutex);
+		/* Cancel any active worker and force us to re-evaluate gup */
+		obj->userptr.work = NULL;
+
+		if (obj->pages != NULL) {
+			struct drm_i915_private *dev_priv = to_i915(mn->dev);
+			struct i915_vma *vma, *tmp;
+			bool was_interruptible;
+
+			was_interruptible = dev_priv->mm.interruptible;
+			dev_priv->mm.interruptible = false;
+
+			list_for_each_entry_safe(vma, tmp, &obj->vma_list, vma_link) {
+				int ret = i915_vma_unbind(vma);
+				WARN_ON(ret && ret != -EIO);
+			}
+			WARN_ON(i915_gem_object_put_pages(obj));
+
+			dev_priv->mm.interruptible = was_interruptible;
+		}
+
+		start = obj->userptr.ptr + obj->base.size;
+
+		drm_gem_object_unreference(&obj->base);
+		mutex_unlock(&mn->dev->struct_mutex);
+	}
+}
+
+static const struct mmu_notifier_ops i915_gem_userptr_notifier = {
+	.invalidate_range_start = i915_gem_userptr_mn_invalidate_range_start,
+};
+
+static struct i915_mmu_notifier *
+__i915_mmu_notifier_lookup(struct drm_device *dev, struct mm_struct *mm)
+{
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_mmu_notifier *mmu;
+
+	/* Protected by dev->struct_mutex */
+	hash_for_each_possible(dev_priv->mmu_notifiers, mmu, node, (unsigned long)mm)
+		if (mmu->mm == mm)
+			return mmu;
+
+	return NULL;
+}
+
+static struct i915_mmu_notifier *
+i915_mmu_notifier_get(struct drm_device *dev, struct mm_struct *mm)
+{
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct i915_mmu_notifier *mmu;
+	int ret;
+
+	lockdep_assert_held(&dev->struct_mutex);
+
+	mmu = __i915_mmu_notifier_lookup(dev, mm);
+	if (mmu)
+		return mmu;
+
+	mmu = kmalloc(sizeof(*mmu), GFP_KERNEL);
+	if (mmu == NULL)
+		return ERR_PTR(-ENOMEM);
+
+	spin_lock_init(&mmu->lock);
+	mmu->dev = dev;
+	mmu->mn.ops = &i915_gem_userptr_notifier;
+	mmu->mm = mm;
+	mmu->objects = RB_ROOT;
+	mmu->count = 0;
+	mmu->serial = 0;
+
+	/* Protected by mmap_sem (write-lock) */
+	ret = __mmu_notifier_register(&mmu->mn, mm);
+	if (ret) {
+		kfree(mmu);
+		return ERR_PTR(ret);
+	}
+
+	/* Protected by dev->struct_mutex */
+	hash_add(dev_priv->mmu_notifiers, &mmu->node, (unsigned long)mm);
+	return mmu;
+}
+
+static void
+__i915_mmu_notifier_destroy_worker(struct work_struct *work)
+{
+	struct i915_mmu_notifier *mmu = container_of(work, typeof(*mmu), work);
+	mmu_notifier_unregister(&mmu->mn, mmu->mm);
+	kfree(mmu);
+}
+
+static void
+__i915_mmu_notifier_destroy(struct i915_mmu_notifier *mmu)
+{
+	lockdep_assert_held(&mmu->dev->struct_mutex);
+
+	/* Protected by dev->struct_mutex */
+	hash_del(&mmu->node);
+
+	/* Our lock ordering is: mmap_sem, mmu_notifier_scru, struct_mutex.
+	 * We enter the function holding struct_mutex, therefore we need
+	 * to drop our mutex prior to calling mmu_notifier_unregister in
+	 * order to prevent lock inversion (and system-wide deadlock)
+	 * between the mmap_sem and struct-mutex. Hence we defer the
+	 * unregistration to a workqueue where we hold no locks.
+	 */
+	INIT_WORK(&mmu->work, __i915_mmu_notifier_destroy_worker);
+	schedule_work(&mmu->work);
+}
+
+static void __i915_mmu_notifier_update_serial(struct i915_mmu_notifier *mmu)
+{
+	if (++mmu->serial == 0)
+		mmu->serial = 1;
+}
+
+static void
+i915_mmu_notifier_del(struct i915_mmu_notifier *mmu,
+		      struct i915_mmu_object *mn)
+{
+	lockdep_assert_held(&mmu->dev->struct_mutex);
+
+	spin_lock(&mmu->lock);
+	interval_tree_remove(&mn->it, &mmu->objects);
+	__i915_mmu_notifier_update_serial(mmu);
+	spin_unlock(&mmu->lock);
+
+	/* Protected against _add() by dev->struct_mutex */
+	if (--mmu->count == 0)
+		__i915_mmu_notifier_destroy(mmu);
+}
+
+static int
+i915_mmu_notifier_add(struct i915_mmu_notifier *mmu,
+		      struct i915_mmu_object *mn)
+{
+	struct interval_tree_node *it;
+	int ret;
+
+	ret = i915_mutex_lock_interruptible(mmu->dev);
+	if (ret)
+		return ret;
+
+	/* Make sure we drop the final active reference (and thereby
+	 * remove the objects from the interval tree) before we do
+	 * the check for overlapping objects.
+	 */
+	i915_gem_retire_requests(mmu->dev);
+
+	/* Disallow overlapping userptr objects */
+	spin_lock(&mmu->lock);
+	it = interval_tree_iter_first(&mmu->objects,
+				      mn->it.start, mn->it.last);
+	if (it) {
+		struct drm_i915_gem_object *obj;
+
+		/* We only need to check the first object in the range as it
+		 * either has cancelled gup work queued and we need to
+		 * return back to the user to give time for the gup-workers
+		 * to flush their object references upon which the object will
+		 * be removed from the interval-tree, or the the range is
+		 * still in use by another client and the overlap is invalid.
+		 */
+
+		obj = container_of(it, struct i915_mmu_object, it)->obj;
+		ret = obj->userptr.workers ? -EAGAIN : -EINVAL;
+	} else {
+		interval_tree_insert(&mn->it, &mmu->objects);
+		__i915_mmu_notifier_update_serial(mmu);
+		ret = 0;
+	}
+	spin_unlock(&mmu->lock);
+	mutex_unlock(&mmu->dev->struct_mutex);
+
+	return ret;
+}
+
+static void
+i915_gem_userptr_release__mmu_notifier(struct drm_i915_gem_object *obj)
+{
+	struct i915_mmu_object *mn;
+
+	mn = obj->userptr.mn;
+	if (mn == NULL)
+		return;
+
+	i915_mmu_notifier_del(mn->mmu, mn);
+	obj->userptr.mn = NULL;
+}
+
+static int
+i915_gem_userptr_init__mmu_notifier(struct drm_i915_gem_object *obj,
+				    unsigned flags)
+{
+	struct i915_mmu_notifier *mmu;
+	struct i915_mmu_object *mn;
+	int ret;
+
+	if (flags & I915_USERPTR_UNSYNCHRONIZED)
+		return capable(CAP_SYS_ADMIN) ? 0 : -EPERM;
+
+	down_write(&obj->userptr.mm->mmap_sem);
+	ret = i915_mutex_lock_interruptible(obj->base.dev);
+	if (ret == 0) {
+		mmu = i915_mmu_notifier_get(obj->base.dev, obj->userptr.mm);
+		if (!IS_ERR(mmu))
+			mmu->count++; /* preemptive add to act as a refcount */
+		else
+			ret = PTR_ERR(mmu);
+		mutex_unlock(&obj->base.dev->struct_mutex);
+	}
+	up_write(&obj->userptr.mm->mmap_sem);
+	if (ret)
+		return ret;
+
+	mn = kzalloc(sizeof(*mn), GFP_KERNEL);
+	if (mn == NULL) {
+		ret = -ENOMEM;
+		goto destroy_mmu;
+	}
+
+	mn->mmu = mmu;
+	mn->it.start = obj->userptr.ptr;
+	mn->it.last = mn->it.start + obj->base.size - 1;
+	mn->obj = obj;
+
+	ret = i915_mmu_notifier_add(mmu, mn);
+	if (ret)
+		goto free_mn;
+
+	obj->userptr.mn = mn;
+	return 0;
+
+free_mn:
+	kfree(mn);
+destroy_mmu:
+	mutex_lock(&obj->base.dev->struct_mutex);
+	if (--mmu->count == 0)
+		__i915_mmu_notifier_destroy(mmu);
+	mutex_unlock(&obj->base.dev->struct_mutex);
+	return ret;
+}
+
+#else
+
+static void
+i915_gem_userptr_release__mmu_notifier(struct drm_i915_gem_object *obj)
+{
+}
+
+static int
+i915_gem_userptr_init__mmu_notifier(struct drm_i915_gem_object *obj,
+				    unsigned flags)
+{
+	if ((flags & I915_USERPTR_UNSYNCHRONIZED) == 0)
+		return -ENODEV;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	return 0;
+}
+#endif
+
+struct get_pages_work {
+	struct work_struct work;
+	struct drm_i915_gem_object *obj;
+	struct task_struct *task;
+};
+
+
+#if IS_ENABLED(CONFIG_SWIOTLB)
+#define swiotlb_active() swiotlb_nr_tbl()
+#else
+#define swiotlb_active() 0
+#endif
+
+static int
+st_set_pages(struct sg_table **st, struct page **pvec, int num_pages)
+{
+	struct scatterlist *sg;
+	int ret, n;
+
+	*st = kmalloc(sizeof(**st), GFP_KERNEL);
+	if (*st == NULL)
+		return -ENOMEM;
+
+	if (swiotlb_active()) {
+		ret = sg_alloc_table(*st, num_pages, GFP_KERNEL);
+		if (ret)
+			goto err;
+
+		for_each_sg((*st)->sgl, sg, num_pages, n)
+			sg_set_page(sg, pvec[n], PAGE_SIZE, 0);
+	} else {
+		ret = sg_alloc_table_from_pages(*st, pvec, num_pages,
+						0, num_pages << PAGE_SHIFT,
+						GFP_KERNEL);
+		if (ret)
+			goto err;
+	}
+
+	return 0;
+
+err:
+	kfree(*st);
+	*st = NULL;
+	return ret;
+}
+
+static void
+__i915_gem_userptr_get_pages_worker(struct work_struct *_work)
+{
+	struct get_pages_work *work = container_of(_work, typeof(*work), work);
+	struct drm_i915_gem_object *obj = work->obj;
+	struct drm_device *dev = obj->base.dev;
+	const int num_pages = obj->base.size >> PAGE_SHIFT;
+	struct page **pvec;
+	int pinned, ret;
+
+	ret = -ENOMEM;
+	pinned = 0;
+
+	pvec = kmalloc(num_pages*sizeof(struct page *),
+		       GFP_TEMPORARY | __GFP_NOWARN | __GFP_NORETRY);
+	if (pvec == NULL)
+		pvec = drm_malloc_ab(num_pages, sizeof(struct page *));
+	if (pvec != NULL) {
+		struct mm_struct *mm = obj->userptr.mm;
+
+		down_read(&mm->mmap_sem);
+		while (pinned < num_pages) {
+			ret = get_user_pages(work->task, mm,
+					     obj->userptr.ptr + pinned * PAGE_SIZE,
+					     num_pages - pinned,
+					     !obj->userptr.read_only, 0,
+					     pvec + pinned, NULL);
+			if (ret < 0)
+				break;
+
+			pinned += ret;
+		}
+		up_read(&mm->mmap_sem);
+	}
+
+	mutex_lock(&dev->struct_mutex);
+	if (obj->userptr.work != &work->work) {
+		ret = 0;
+	} else if (pinned == num_pages) {
+		ret = st_set_pages(&obj->pages, pvec, num_pages);
+		if (ret == 0) {
+			list_add_tail(&obj->global_list, &to_i915(dev)->mm.unbound_list);
+			pinned = 0;
+		}
+	}
+
+	obj->userptr.work = ERR_PTR(ret);
+	obj->userptr.workers--;
+	drm_gem_object_unreference(&obj->base);
+	mutex_unlock(&dev->struct_mutex);
+
+	release_pages(pvec, pinned, 0);
+	drm_free_large(pvec);
+
+	put_task_struct(work->task);
+	kfree(work);
+}
+
+static int
+i915_gem_userptr_get_pages(struct drm_i915_gem_object *obj)
+{
+	const int num_pages = obj->base.size >> PAGE_SHIFT;
+	struct page **pvec;
+	int pinned, ret;
+
+	/* If userspace should engineer that these pages are replaced in
+	 * the vma between us binding this page into the GTT and completion
+	 * of rendering... Their loss. If they change the mapping of their
+	 * pages they need to create a new bo to point to the new vma.
+	 *
+	 * However, that still leaves open the possibility of the vma
+	 * being copied upon fork. Which falls under the same userspace
+	 * synchronisation issue as a regular bo, except that this time
+	 * the process may not be expecting that a particular piece of
+	 * memory is tied to the GPU.
+	 *
+	 * Fortunately, we can hook into the mmu_notifier in order to
+	 * discard the page references prior to anything nasty happening
+	 * to the vma (discard or cloning) which should prevent the more
+	 * egregious cases from causing harm.
+	 */
+
+	pvec = NULL;
+	pinned = 0;
+	if (obj->userptr.mm == current->mm) {
+		pvec = kmalloc(num_pages*sizeof(struct page *),
+			       GFP_TEMPORARY | __GFP_NOWARN | __GFP_NORETRY);
+		if (pvec == NULL) {
+			pvec = drm_malloc_ab(num_pages, sizeof(struct page *));
+			if (pvec == NULL)
+				return -ENOMEM;
+		}
+
+		pinned = __get_user_pages_fast(obj->userptr.ptr, num_pages,
+					       !obj->userptr.read_only, pvec);
+	}
+	if (pinned < num_pages) {
+		if (pinned < 0) {
+			ret = pinned;
+			pinned = 0;
+		} else {
+			/* Spawn a worker so that we can acquire the
+			 * user pages without holding our mutex. Access
+			 * to the user pages requires mmap_sem, and we have
+			 * a strict lock ordering of mmap_sem, struct_mutex -
+			 * we already hold struct_mutex here and so cannot
+			 * call gup without encountering a lock inversion.
+			 *
+			 * Userspace will keep on repeating the operation
+			 * (thanks to EAGAIN) until either we hit the fast
+			 * path or the worker completes. If the worker is
+			 * cancelled or superseded, the task is still run
+			 * but the results ignored. (This leads to
+			 * complications that we may have a stray object
+			 * refcount that we need to be wary of when
+			 * checking for existing objects during creation.)
+			 * If the worker encounters an error, it reports
+			 * that error back to this function through
+			 * obj->userptr.work = ERR_PTR.
+			 */
+			ret = -EAGAIN;
+			if (obj->userptr.work == NULL &&
+			    obj->userptr.workers < I915_GEM_USERPTR_MAX_WORKERS) {
+				struct get_pages_work *work;
+
+				work = kmalloc(sizeof(*work), GFP_KERNEL);
+				if (work != NULL) {
+					obj->userptr.work = &work->work;
+					obj->userptr.workers++;
+
+					work->obj = obj;
+					drm_gem_object_reference(&obj->base);
+
+					work->task = current;
+					get_task_struct(work->task);
+
+					INIT_WORK(&work->work, __i915_gem_userptr_get_pages_worker);
+					schedule_work(&work->work);
+				} else
+					ret = -ENOMEM;
+			} else {
+				if (IS_ERR(obj->userptr.work)) {
+					ret = PTR_ERR(obj->userptr.work);
+					obj->userptr.work = NULL;
+				}
+			}
+		}
+	} else {
+		ret = st_set_pages(&obj->pages, pvec, num_pages);
+		if (ret == 0) {
+			obj->userptr.work = NULL;
+			pinned = 0;
+		}
+	}
+
+	release_pages(pvec, pinned, 0);
+	drm_free_large(pvec);
+	return ret;
+}
+
+static void
+i915_gem_userptr_put_pages(struct drm_i915_gem_object *obj)
+{
+	struct scatterlist *sg;
+	int i;
+
+	BUG_ON(obj->userptr.work != NULL);
+
+	if (obj->madv != I915_MADV_WILLNEED)
+		obj->dirty = 0;
+
+	for_each_sg(obj->pages->sgl, sg, obj->pages->nents, i) {
+		struct page *page = sg_page(sg);
+
+		if (obj->dirty)
+			set_page_dirty(page);
+
+		mark_page_accessed(page);
+		page_cache_release(page);
+	}
+	obj->dirty = 0;
+
+	sg_free_table(obj->pages);
+	kfree(obj->pages);
+}
+
+static void
+i915_gem_userptr_release(struct drm_i915_gem_object *obj)
+{
+	i915_gem_userptr_release__mmu_notifier(obj);
+
+	if (obj->userptr.mm) {
+		mmput(obj->userptr.mm);
+		obj->userptr.mm = NULL;
+	}
+}
+
+static int
+i915_gem_userptr_dmabuf_export(struct drm_i915_gem_object *obj)
+{
+	if (obj->userptr.mn)
+		return 0;
+
+	return i915_gem_userptr_init__mmu_notifier(obj, 0);
+}
+
+static const struct drm_i915_gem_object_ops i915_gem_userptr_ops = {
+	.dmabuf_export = i915_gem_userptr_dmabuf_export,
+	.get_pages = i915_gem_userptr_get_pages,
+	.put_pages = i915_gem_userptr_put_pages,
+	.release = i915_gem_userptr_release,
+};
+
+/**
+ * Creates a new mm object that wraps some normal memory from the process
+ * context - user memory.
+ *
+ * We impose several restrictions upon the memory being mapped
+ * into the GPU.
+ * 1. It must be page aligned (both start/end addresses, i.e ptr and size).
+ * 2. It cannot overlap any other userptr object in the same address space.
+ * 3. It must be normal system memory, not a pointer into another map of IO
+ *    space (e.g. it must not be a GTT mmapping of another object).
+ * 4. We only allow a bo as large as we could in theory map into the GTT,
+ *    that is we limit the size to the total size of the GTT.
+ * 5. The bo is marked as being snoopable. The backing pages are left
+ *    accessible directly by the CPU, but reads and writes by the GPU may
+ *    incur the cost of a snoop (unless you have an LLC architecture).
+ *
+ * Synchronisation between multiple users and the GPU is left to userspace
+ * through the normal set-domain-ioctl. The kernel will enforce that the
+ * GPU relinquishes the VMA before it is returned back to the system
+ * i.e. upon free(), munmap() or process termination. However, the userspace
+ * malloc() library may not immediately relinquish the VMA after free() and
+ * instead reuse it whilst the GPU is still reading and writing to the VMA.
+ * Caveat emptor.
+ *
+ * Also note, that the object created here is not currently a "first class"
+ * object, in that several ioctls are banned. These are the CPU access
+ * ioctls: mmap(), pwrite and pread. In practice, you are expected to use
+ * direct access via your pointer rather than use those ioctls.
+ *
+ * If you think this is a good interface to use to pass GPU memory between
+ * drivers, please use dma-buf instead. In fact, wherever possible use
+ * dma-buf instead.
+ */
+int
+i915_gem_userptr_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_gem_userptr *args = data;
+	struct drm_i915_gem_object *obj;
+	int ret;
+	u32 handle;
+
+	if (args->flags & ~(I915_USERPTR_READ_ONLY |
+			    I915_USERPTR_UNSYNCHRONIZED))
+		return -EINVAL;
+
+	if (offset_in_page(args->user_ptr | args->user_size))
+		return -EINVAL;
+
+	if (args->user_size > dev_priv->gtt.base.total)
+		return -E2BIG;
+
+	if (!access_ok(args->flags & I915_USERPTR_READ_ONLY ? VERIFY_READ : VERIFY_WRITE,
+		       (char __user *)(unsigned long)args->user_ptr, args->user_size))
+		return -EFAULT;
+
+	if (args->flags & I915_USERPTR_READ_ONLY) {
+		/* On almost all of the current hw, we cannot tell the GPU that a
+		 * page is readonly, so this is just a placeholder in the uAPI.
+		 */
+		return -ENODEV;
+	}
+
+	/* Allocate the new object */
+	obj = i915_gem_object_alloc(dev);
+	if (obj == NULL)
+		return -ENOMEM;
+
+	drm_gem_private_object_init(dev, &obj->base, args->user_size);
+	i915_gem_object_init(obj, &i915_gem_userptr_ops);
+	obj->cache_level = I915_CACHE_LLC;
+	obj->base.write_domain = I915_GEM_DOMAIN_CPU;
+	obj->base.read_domains = I915_GEM_DOMAIN_CPU;
+
+	obj->userptr.ptr = args->user_ptr;
+	obj->userptr.read_only = !!(args->flags & I915_USERPTR_READ_ONLY);
+
+	/* And keep a pointer to the current->mm for resolving the user pages
+	 * at binding. This means that we need to hook into the mmu_notifier
+	 * in order to detect if the mmu is destroyed.
+	 */
+	ret = -ENOMEM;
+	if ((obj->userptr.mm = get_task_mm(current)))
+		ret = i915_gem_userptr_init__mmu_notifier(obj, args->flags);
+	if (ret == 0)
+		ret = drm_gem_handle_create(file, &obj->base, &handle);
+
+	/* drop reference from allocate - handle holds it now */
+	drm_gem_object_unreference_unlocked(&obj->base);
+	if (ret)
+		return ret;
+
+	args->handle = handle;
+	return 0;
+}
+
+int
+i915_gem_init_userptr(struct drm_device *dev)
+{
+#if defined(CONFIG_MMU_NOTIFIER)
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	hash_init(dev_priv->mmu_notifiers);
+#endif
+	return 0;
+}
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 12f1d43b2d68..87ec60e181a7 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -42,6 +42,7 @@ static const char *ring_str(int ring)
 	case VCS: return "bsd";
 	case BCS: return "blt";
 	case VECS: return "vebox";
+	case VCS2: return "bsd2";
 	default: return "";
 	}
 }
@@ -204,6 +205,7 @@ static void print_error_buffers(struct drm_i915_error_state_buf *m,
 		err_puts(m, tiling_flag(err->tiling));
 		err_puts(m, dirty_flag(err->dirty));
 		err_puts(m, purgeable_flag(err->purgeable));
+		err_puts(m, err->userptr ? " userptr" : "");
 		err_puts(m, err->ring != -1 ? " " : "");
 		err_puts(m, ring_str(err->ring));
 		err_puts(m, i915_cache_level_str(err->cache_level));
@@ -257,7 +259,8 @@ static void i915_ring_error_state(struct drm_i915_error_state_buf *m,
 		err_printf(m, "  INSTPS: 0x%08x\n", ring->instps);
 	}
 	err_printf(m, "  INSTPM: 0x%08x\n", ring->instpm);
-	err_printf(m, "  FADDR: 0x%08x\n", ring->faddr);
+	err_printf(m, "  FADDR: 0x%08x %08x\n", upper_32_bits(ring->faddr),
+		   lower_32_bits(ring->faddr));
 	if (INTEL_INFO(dev)->gen >= 6) {
 		err_printf(m, "  RC PSMI: 0x%08x\n", ring->rc_psmi);
 		err_printf(m, "  FAULT_REG: 0x%08x\n", ring->fault_reg);
@@ -452,16 +455,7 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 			err_printf(m, "%s --- HW Context = 0x%08x\n",
 				   dev_priv->ring[i].name,
 				   obj->gtt_offset);
-			offset = 0;
-			for (elt = 0; elt < PAGE_SIZE/16; elt += 4) {
-				err_printf(m, "[%04x] %08x %08x %08x %08x\n",
-					   offset,
-					   obj->pages[0][elt],
-					   obj->pages[0][elt+1],
-					   obj->pages[0][elt+2],
-					   obj->pages[0][elt+3]);
-					offset += 16;
-			}
+			print_error_obj(m, obj);
 		}
 	}
 
@@ -648,6 +642,7 @@ static void capture_bo(struct drm_i915_error_buffer *err,
 	err->tiling = obj->tiling_mode;
 	err->dirty = obj->dirty;
 	err->purgeable = obj->madv != I915_MADV_WILLNEED;
+	err->userptr = obj->userptr.mm != NULL;
 	err->ring = obj->ring ? obj->ring->id : -1;
 	err->cache_level = obj->cache_level;
 }
@@ -752,7 +747,7 @@ static void i915_gem_record_fences(struct drm_device *dev,
 }
 
 static void i915_record_ring_state(struct drm_device *dev,
-				   struct intel_ring_buffer *ring,
+				   struct intel_engine_cs *ring,
 				   struct drm_i915_error_ring *ering)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -764,14 +759,14 @@ static void i915_record_ring_state(struct drm_device *dev,
 			= I915_READ(RING_SYNC_0(ring->mmio_base));
 		ering->semaphore_mboxes[1]
 			= I915_READ(RING_SYNC_1(ring->mmio_base));
-		ering->semaphore_seqno[0] = ring->sync_seqno[0];
-		ering->semaphore_seqno[1] = ring->sync_seqno[1];
+		ering->semaphore_seqno[0] = ring->semaphore.sync_seqno[0];
+		ering->semaphore_seqno[1] = ring->semaphore.sync_seqno[1];
 	}
 
 	if (HAS_VEBOX(dev)) {
 		ering->semaphore_mboxes[2] =
 			I915_READ(RING_SYNC_2(ring->mmio_base));
-		ering->semaphore_seqno[2] = ring->sync_seqno[2];
+		ering->semaphore_seqno[2] = ring->semaphore.sync_seqno[2];
 	}
 
 	if (INTEL_INFO(dev)->gen >= 4) {
@@ -781,8 +776,10 @@ static void i915_record_ring_state(struct drm_device *dev,
 		ering->instdone = I915_READ(RING_INSTDONE(ring->mmio_base));
 		ering->instps = I915_READ(RING_INSTPS(ring->mmio_base));
 		ering->bbaddr = I915_READ(RING_BBADDR(ring->mmio_base));
-		if (INTEL_INFO(dev)->gen >= 8)
+		if (INTEL_INFO(dev)->gen >= 8) {
+			ering->faddr |= (u64) I915_READ(RING_DMA_FADD_UDW(ring->mmio_base)) << 32;
 			ering->bbaddr |= (u64) I915_READ(RING_BBADDR_UDW(ring->mmio_base)) << 32;
+		}
 		ering->bbstate = I915_READ(RING_BBSTATE(ring->mmio_base));
 	} else {
 		ering->faddr = I915_READ(DMA_FADD_I8XX);
@@ -828,8 +825,8 @@ static void i915_record_ring_state(struct drm_device *dev,
 		ering->hws = I915_READ(mmio);
 	}
 
-	ering->cpu_ring_head = ring->head;
-	ering->cpu_ring_tail = ring->tail;
+	ering->cpu_ring_head = ring->buffer->head;
+	ering->cpu_ring_tail = ring->buffer->tail;
 
 	ering->hangcheck_score = ring->hangcheck.score;
 	ering->hangcheck_action = ring->hangcheck.action;
@@ -862,7 +859,7 @@ static void i915_record_ring_state(struct drm_device *dev,
 }
 
 
-static void i915_gem_record_active_context(struct intel_ring_buffer *ring,
+static void i915_gem_record_active_context(struct intel_engine_cs *ring,
 					   struct drm_i915_error_state *error,
 					   struct drm_i915_error_ring *ering)
 {
@@ -875,10 +872,7 @@ static void i915_gem_record_active_context(struct intel_ring_buffer *ring,
 
 	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
 		if ((error->ccid & PAGE_MASK) == i915_gem_obj_ggtt_offset(obj)) {
-			ering->ctx = i915_error_object_create_sized(dev_priv,
-								    obj,
-								    &dev_priv->gtt.base,
-								    1);
+			ering->ctx = i915_error_ggtt_object_create(dev_priv, obj);
 			break;
 		}
 	}
@@ -892,7 +886,7 @@ static void i915_gem_record_rings(struct drm_device *dev,
 	int i, count;
 
 	for (i = 0; i < I915_NUM_RINGS; i++) {
-		struct intel_ring_buffer *ring = &dev_priv->ring[i];
+		struct intel_engine_cs *ring = &dev_priv->ring[i];
 
 		if (ring->dev == NULL)
 			continue;
@@ -936,7 +930,7 @@ static void i915_gem_record_rings(struct drm_device *dev,
 		}
 
 		error->ring[i].ringbuffer =
-			i915_error_ggtt_object_create(dev_priv, ring->obj);
+			i915_error_ggtt_object_create(dev_priv, ring->buffer->obj);
 
 		if (ring->status_page.obj)
 			error->ring[i].hws_page =
@@ -1037,7 +1031,6 @@ static void i915_capture_reg_state(struct drm_i915_private *dev_priv,
 				   struct drm_i915_error_state *error)
 {
 	struct drm_device *dev = dev_priv->dev;
-	int pipe;
 
 	/* General organization
 	 * 1. Registers specific to a single generation
@@ -1062,9 +1055,6 @@ static void i915_capture_reg_state(struct drm_i915_private *dev_priv,
 		error->gfx_mode = I915_READ(GFX_MODE);
 	}
 
-	if (IS_GEN2(dev))
-		error->ier = I915_READ16(IER);
-
 	/* 2: Registers which belong to multiple generations */
 	if (INTEL_INFO(dev)->gen >= 7)
 		error->forcewake = I915_READ(FORCEWAKE_MT);
@@ -1088,9 +1078,10 @@ static void i915_capture_reg_state(struct drm_i915_private *dev_priv,
 	if (HAS_PCH_SPLIT(dev))
 		error->ier = I915_READ(DEIER) | I915_READ(GTIER);
 	else {
-		error->ier = I915_READ(IER);
-		for_each_pipe(pipe)
-			error->pipestat[pipe] = I915_READ(PIPESTAT(pipe));
+		if (IS_GEN2(dev))
+			error->ier = I915_READ16(IER);
+		else
+			error->ier = I915_READ(IER);
 	}
 
 	/* 4: Everything else */
diff --git a/drivers/gpu/drm/i915/i915_ioc32.c b/drivers/gpu/drm/i915/i915_ioc32.c
index 3c59584161c2..2e0613e26251 100644
--- a/drivers/gpu/drm/i915/i915_ioc32.c
+++ b/drivers/gpu/drm/i915/i915_ioc32.c
@@ -208,7 +208,7 @@ long i915_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 	if (nr < DRM_COMMAND_BASE)
 		return drm_compat_ioctl(filp, cmd, arg);
 
-	if (nr < DRM_COMMAND_BASE + DRM_ARRAY_SIZE(i915_compat_ioctls))
+	if (nr < DRM_COMMAND_BASE + ARRAY_SIZE(i915_compat_ioctls))
 		fn = i915_compat_ioctls[nr - DRM_COMMAND_BASE];
 
 	if (fn != NULL)
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 0b99de95593b..6f8017a7e937 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -80,17 +80,64 @@ static const u32 hpd_status_i915[] = { /* i915 and valleyview are the same */
 	[HPD_PORT_D] = PORTD_HOTPLUG_INT_STATUS
 };
 
+/* IIR can theoretically queue up two events. Be paranoid. */
+#define GEN8_IRQ_RESET_NDX(type, which) do { \
+	I915_WRITE(GEN8_##type##_IMR(which), 0xffffffff); \
+	POSTING_READ(GEN8_##type##_IMR(which)); \
+	I915_WRITE(GEN8_##type##_IER(which), 0); \
+	I915_WRITE(GEN8_##type##_IIR(which), 0xffffffff); \
+	POSTING_READ(GEN8_##type##_IIR(which)); \
+	I915_WRITE(GEN8_##type##_IIR(which), 0xffffffff); \
+	POSTING_READ(GEN8_##type##_IIR(which)); \
+} while (0)
+
+#define GEN5_IRQ_RESET(type) do { \
+	I915_WRITE(type##IMR, 0xffffffff); \
+	POSTING_READ(type##IMR); \
+	I915_WRITE(type##IER, 0); \
+	I915_WRITE(type##IIR, 0xffffffff); \
+	POSTING_READ(type##IIR); \
+	I915_WRITE(type##IIR, 0xffffffff); \
+	POSTING_READ(type##IIR); \
+} while (0)
+
+/*
+ * We should clear IMR at preinstall/uninstall, and just check at postinstall.
+ */
+#define GEN5_ASSERT_IIR_IS_ZERO(reg) do { \
+	u32 val = I915_READ(reg); \
+	if (val) { \
+		WARN(1, "Interrupt register 0x%x is not zero: 0x%08x\n", \
+		     (reg), val); \
+		I915_WRITE((reg), 0xffffffff); \
+		POSTING_READ(reg); \
+		I915_WRITE((reg), 0xffffffff); \
+		POSTING_READ(reg); \
+	} \
+} while (0)
+
+#define GEN8_IRQ_INIT_NDX(type, which, imr_val, ier_val) do { \
+	GEN5_ASSERT_IIR_IS_ZERO(GEN8_##type##_IIR(which)); \
+	I915_WRITE(GEN8_##type##_IMR(which), (imr_val)); \
+	I915_WRITE(GEN8_##type##_IER(which), (ier_val)); \
+	POSTING_READ(GEN8_##type##_IER(which)); \
+} while (0)
+
+#define GEN5_IRQ_INIT(type, imr_val, ier_val) do { \
+	GEN5_ASSERT_IIR_IS_ZERO(type##IIR); \
+	I915_WRITE(type##IMR, (imr_val)); \
+	I915_WRITE(type##IER, (ier_val)); \
+	POSTING_READ(type##IER); \
+} while (0)
+
 /* For display hotplug interrupt */
 static void
 ironlake_enable_display_irq(struct drm_i915_private *dev_priv, u32 mask)
 {
 	assert_spin_locked(&dev_priv->irq_lock);
 
-	if (dev_priv->pm.irqs_disabled) {
-		WARN(1, "IRQs disabled\n");
-		dev_priv->pm.regsave.deimr &= ~mask;
+	if (WARN_ON(dev_priv->pm.irqs_disabled))
 		return;
-	}
 
 	if ((dev_priv->irq_mask & mask) != 0) {
 		dev_priv->irq_mask &= ~mask;
@@ -104,11 +151,8 @@ ironlake_disable_display_irq(struct drm_i915_private *dev_priv, u32 mask)
 {
 	assert_spin_locked(&dev_priv->irq_lock);
 
-	if (dev_priv->pm.irqs_disabled) {
-		WARN(1, "IRQs disabled\n");
-		dev_priv->pm.regsave.deimr |= mask;
+	if (WARN_ON(dev_priv->pm.irqs_disabled))
 		return;
-	}
 
 	if ((dev_priv->irq_mask & mask) != mask) {
 		dev_priv->irq_mask |= mask;
@@ -129,13 +173,8 @@ static void ilk_update_gt_irq(struct drm_i915_private *dev_priv,
 {
 	assert_spin_locked(&dev_priv->irq_lock);
 
-	if (dev_priv->pm.irqs_disabled) {
-		WARN(1, "IRQs disabled\n");
-		dev_priv->pm.regsave.gtimr &= ~interrupt_mask;
-		dev_priv->pm.regsave.gtimr |= (~enabled_irq_mask &
-						interrupt_mask);
+	if (WARN_ON(dev_priv->pm.irqs_disabled))
 		return;
-	}
 
 	dev_priv->gt_irq_mask &= ~interrupt_mask;
 	dev_priv->gt_irq_mask |= (~enabled_irq_mask & interrupt_mask);
@@ -167,13 +206,8 @@ static void snb_update_pm_irq(struct drm_i915_private *dev_priv,
 
 	assert_spin_locked(&dev_priv->irq_lock);
 
-	if (dev_priv->pm.irqs_disabled) {
-		WARN(1, "IRQs disabled\n");
-		dev_priv->pm.regsave.gen6_pmimr &= ~interrupt_mask;
-		dev_priv->pm.regsave.gen6_pmimr |= (~enabled_irq_mask &
-						     interrupt_mask);
+	if (WARN_ON(dev_priv->pm.irqs_disabled))
 		return;
-	}
 
 	new_val = dev_priv->pm_irq_mask;
 	new_val &= ~interrupt_mask;
@@ -214,6 +248,46 @@ static bool ivb_can_enable_err_int(struct drm_device *dev)
 	return true;
 }
 
+/**
+  * bdw_update_pm_irq - update GT interrupt 2
+  * @dev_priv: driver private
+  * @interrupt_mask: mask of interrupt bits to update
+  * @enabled_irq_mask: mask of interrupt bits to enable
+  *
+  * Copied from the snb function, updated with relevant register offsets
+  */
+static void bdw_update_pm_irq(struct drm_i915_private *dev_priv,
+			      uint32_t interrupt_mask,
+			      uint32_t enabled_irq_mask)
+{
+	uint32_t new_val;
+
+	assert_spin_locked(&dev_priv->irq_lock);
+
+	if (WARN_ON(dev_priv->pm.irqs_disabled))
+		return;
+
+	new_val = dev_priv->pm_irq_mask;
+	new_val &= ~interrupt_mask;
+	new_val |= (~enabled_irq_mask & interrupt_mask);
+
+	if (new_val != dev_priv->pm_irq_mask) {
+		dev_priv->pm_irq_mask = new_val;
+		I915_WRITE(GEN8_GT_IMR(2), dev_priv->pm_irq_mask);
+		POSTING_READ(GEN8_GT_IMR(2));
+	}
+}
+
+void bdw_enable_pm_irq(struct drm_i915_private *dev_priv, uint32_t mask)
+{
+	bdw_update_pm_irq(dev_priv, mask, mask);
+}
+
+void bdw_disable_pm_irq(struct drm_i915_private *dev_priv, uint32_t mask)
+{
+	bdw_update_pm_irq(dev_priv, mask, 0);
+}
+
 static bool cpt_can_enable_serr_int(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -232,16 +306,51 @@ static bool cpt_can_enable_serr_int(struct drm_device *dev)
 	return true;
 }
 
-static void i9xx_clear_fifo_underrun(struct drm_device *dev, enum pipe pipe)
+void i9xx_check_fifo_underruns(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *crtc;
+	unsigned long flags;
+
+	spin_lock_irqsave(&dev_priv->irq_lock, flags);
+
+	for_each_intel_crtc(dev, crtc) {
+		u32 reg = PIPESTAT(crtc->pipe);
+		u32 pipestat;
+
+		if (crtc->cpu_fifo_underrun_disabled)
+			continue;
+
+		pipestat = I915_READ(reg) & 0xffff0000;
+		if ((pipestat & PIPE_FIFO_UNDERRUN_STATUS) == 0)
+			continue;
+
+		I915_WRITE(reg, pipestat | PIPE_FIFO_UNDERRUN_STATUS);
+		POSTING_READ(reg);
+
+		DRM_ERROR("pipe %c underrun\n", pipe_name(crtc->pipe));
+	}
+
+	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
+}
+
+static void i9xx_set_fifo_underrun_reporting(struct drm_device *dev,
+					     enum pipe pipe,
+					     bool enable, bool old)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 reg = PIPESTAT(pipe);
-	u32 pipestat = I915_READ(reg) & 0x7fff0000;
+	u32 pipestat = I915_READ(reg) & 0xffff0000;
 
 	assert_spin_locked(&dev_priv->irq_lock);
 
-	I915_WRITE(reg, pipestat | PIPE_FIFO_UNDERRUN_STATUS);
-	POSTING_READ(reg);
+	if (enable) {
+		I915_WRITE(reg, pipestat | PIPE_FIFO_UNDERRUN_STATUS);
+		POSTING_READ(reg);
+	} else {
+		if (old && pipestat & PIPE_FIFO_UNDERRUN_STATUS)
+			DRM_ERROR("pipe %c underrun\n", pipe_name(pipe));
+	}
 }
 
 static void ironlake_set_fifo_underrun_reporting(struct drm_device *dev,
@@ -258,7 +367,8 @@ static void ironlake_set_fifo_underrun_reporting(struct drm_device *dev,
 }
 
 static void ivybridge_set_fifo_underrun_reporting(struct drm_device *dev,
-						  enum pipe pipe, bool enable)
+						  enum pipe pipe,
+						  bool enable, bool old)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	if (enable) {
@@ -269,15 +379,12 @@ static void ivybridge_set_fifo_underrun_reporting(struct drm_device *dev,
 
 		ironlake_enable_display_irq(dev_priv, DE_ERR_INT_IVB);
 	} else {
-		bool was_enabled = !(I915_READ(DEIMR) & DE_ERR_INT_IVB);
-
-		/* Change the state _after_ we've read out the current one. */
 		ironlake_disable_display_irq(dev_priv, DE_ERR_INT_IVB);
 
-		if (!was_enabled &&
-		    (I915_READ(GEN7_ERR_INT) & ERR_INT_FIFO_UNDERRUN(pipe))) {
-			DRM_DEBUG_KMS("uncleared fifo underrun on pipe %c\n",
-				      pipe_name(pipe));
+		if (old &&
+		    I915_READ(GEN7_ERR_INT) & ERR_INT_FIFO_UNDERRUN(pipe)) {
+			DRM_ERROR("uncleared fifo underrun on pipe %c\n",
+				  pipe_name(pipe));
 		}
 	}
 }
@@ -313,14 +420,8 @@ static void ibx_display_interrupt_update(struct drm_i915_private *dev_priv,
 
 	assert_spin_locked(&dev_priv->irq_lock);
 
-	if (dev_priv->pm.irqs_disabled &&
-	    (interrupt_mask & SDE_HOTPLUG_MASK_CPT)) {
-		WARN(1, "IRQs disabled\n");
-		dev_priv->pm.regsave.sdeimr &= ~interrupt_mask;
-		dev_priv->pm.regsave.sdeimr |= (~enabled_irq_mask &
-						 interrupt_mask);
+	if (WARN_ON(dev_priv->pm.irqs_disabled))
 		return;
-	}
 
 	I915_WRITE(SDEIMR, sdeimr);
 	POSTING_READ(SDEIMR);
@@ -346,7 +447,7 @@ static void ibx_set_fifo_underrun_reporting(struct drm_device *dev,
 
 static void cpt_set_fifo_underrun_reporting(struct drm_device *dev,
 					    enum transcoder pch_transcoder,
-					    bool enable)
+					    bool enable, bool old)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -359,16 +460,12 @@ static void cpt_set_fifo_underrun_reporting(struct drm_device *dev,
 
 		ibx_enable_display_interrupt(dev_priv, SDE_ERROR_CPT);
 	} else {
-		uint32_t tmp = I915_READ(SERR_INT);
-		bool was_enabled = !(I915_READ(SDEIMR) & SDE_ERROR_CPT);
-
-		/* Change the state _after_ we've read out the current one. */
 		ibx_disable_display_interrupt(dev_priv, SDE_ERROR_CPT);
 
-		if (!was_enabled &&
-		    (tmp & SERR_INT_TRANS_FIFO_UNDERRUN(pch_transcoder))) {
-			DRM_DEBUG_KMS("uncleared pch fifo underrun on pch transcoder %c\n",
-				      transcoder_name(pch_transcoder));
+		if (old && I915_READ(SERR_INT) &
+		    SERR_INT_TRANS_FIFO_UNDERRUN(pch_transcoder)) {
+			DRM_ERROR("uncleared pch fifo underrun on pch transcoder %c\n",
+				  transcoder_name(pch_transcoder));
 		}
 	}
 }
@@ -387,34 +484,29 @@ static void cpt_set_fifo_underrun_reporting(struct drm_device *dev,
  *
  * Returns the previous state of underrun reporting.
  */
-bool __intel_set_cpu_fifo_underrun_reporting(struct drm_device *dev,
-					     enum pipe pipe, bool enable)
+static bool __intel_set_cpu_fifo_underrun_reporting(struct drm_device *dev,
+						    enum pipe pipe, bool enable)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_crtc *crtc = dev_priv->pipe_to_crtc_mapping[pipe];
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	bool ret;
+	bool old;
 
 	assert_spin_locked(&dev_priv->irq_lock);
 
-	ret = !intel_crtc->cpu_fifo_underrun_disabled;
-
-	if (enable == ret)
-		goto done;
-
+	old = !intel_crtc->cpu_fifo_underrun_disabled;
 	intel_crtc->cpu_fifo_underrun_disabled = !enable;
 
-	if (enable && (INTEL_INFO(dev)->gen < 5 || IS_VALLEYVIEW(dev)))
-		i9xx_clear_fifo_underrun(dev, pipe);
+	if (INTEL_INFO(dev)->gen < 5 || IS_VALLEYVIEW(dev))
+		i9xx_set_fifo_underrun_reporting(dev, pipe, enable, old);
 	else if (IS_GEN5(dev) || IS_GEN6(dev))
 		ironlake_set_fifo_underrun_reporting(dev, pipe, enable);
 	else if (IS_GEN7(dev))
-		ivybridge_set_fifo_underrun_reporting(dev, pipe, enable);
+		ivybridge_set_fifo_underrun_reporting(dev, pipe, enable, old);
 	else if (IS_GEN8(dev))
 		broadwell_set_fifo_underrun_reporting(dev, pipe, enable);
 
-done:
-	return ret;
+	return old;
 }
 
 bool intel_set_cpu_fifo_underrun_reporting(struct drm_device *dev,
@@ -463,7 +555,7 @@ bool intel_set_pch_fifo_underrun_reporting(struct drm_device *dev,
 	struct drm_crtc *crtc = dev_priv->pipe_to_crtc_mapping[pch_transcoder];
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	unsigned long flags;
-	bool ret;
+	bool old;
 
 	/*
 	 * NOTE: Pre-LPT has a fixed cpu pipe -> pch transcoder mapping, but LPT
@@ -476,21 +568,16 @@ bool intel_set_pch_fifo_underrun_reporting(struct drm_device *dev,
 
 	spin_lock_irqsave(&dev_priv->irq_lock, flags);
 
-	ret = !intel_crtc->pch_fifo_underrun_disabled;
-
-	if (enable == ret)
-		goto done;
-
+	old = !intel_crtc->pch_fifo_underrun_disabled;
 	intel_crtc->pch_fifo_underrun_disabled = !enable;
 
 	if (HAS_PCH_IBX(dev))
 		ibx_set_fifo_underrun_reporting(dev, pch_transcoder, enable);
 	else
-		cpt_set_fifo_underrun_reporting(dev, pch_transcoder, enable);
+		cpt_set_fifo_underrun_reporting(dev, pch_transcoder, enable, old);
 
-done:
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
-	return ret;
+	return old;
 }
 
 
@@ -503,8 +590,10 @@ __i915_enable_pipestat(struct drm_i915_private *dev_priv, enum pipe pipe,
 
 	assert_spin_locked(&dev_priv->irq_lock);
 
-	if (WARN_ON_ONCE(enable_mask & ~PIPESTAT_INT_ENABLE_MASK ||
-	                 status_mask & ~PIPESTAT_INT_STATUS_MASK))
+	if (WARN_ONCE(enable_mask & ~PIPESTAT_INT_ENABLE_MASK ||
+		      status_mask & ~PIPESTAT_INT_STATUS_MASK,
+		      "pipe %c: enable_mask=0x%x, status_mask=0x%x\n",
+		      pipe_name(pipe), enable_mask, status_mask))
 		return;
 
 	if ((pipestat & enable_mask) == enable_mask)
@@ -527,8 +616,10 @@ __i915_disable_pipestat(struct drm_i915_private *dev_priv, enum pipe pipe,
 
 	assert_spin_locked(&dev_priv->irq_lock);
 
-	if (WARN_ON_ONCE(enable_mask & ~PIPESTAT_INT_ENABLE_MASK ||
-	                 status_mask & ~PIPESTAT_INT_STATUS_MASK))
+	if (WARN_ONCE(enable_mask & ~PIPESTAT_INT_ENABLE_MASK ||
+		      status_mask & ~PIPESTAT_INT_STATUS_MASK,
+		      "pipe %c: enable_mask=0x%x, status_mask=0x%x\n",
+		      pipe_name(pipe), enable_mask, status_mask))
 		return;
 
 	if ((pipestat & enable_mask) == 0)
@@ -546,11 +637,17 @@ static u32 vlv_get_pipestat_enable_mask(struct drm_device *dev, u32 status_mask)
 	u32 enable_mask = status_mask << 16;
 
 	/*
-	 * On pipe A we don't support the PSR interrupt yet, on pipe B the
-	 * same bit MBZ.
+	 * On pipe A we don't support the PSR interrupt yet,
+	 * on pipe B and C the same bit MBZ.
 	 */
 	if (WARN_ON_ONCE(status_mask & PIPE_A_PSR_STATUS_VLV))
 		return 0;
+	/*
+	 * On pipe B and C we don't support the PSR interrupt yet, on pipe
+	 * A the same bit is for perf counters which we don't use either.
+	 */
+	if (WARN_ON_ONCE(status_mask & PIPE_B_PSR_STATUS_VLV))
+		return 0;
 
 	enable_mask &= ~(PIPE_FIFO_UNDERRUN_STATUS |
 			 SPRITE0_FLIP_DONE_INT_EN_VLV |
@@ -637,6 +734,56 @@ i915_pipe_enabled(struct drm_device *dev, int pipe)
 	}
 }
 
+/*
+ * This timing diagram depicts the video signal in and
+ * around the vertical blanking period.
+ *
+ * Assumptions about the fictitious mode used in this example:
+ *  vblank_start >= 3
+ *  vsync_start = vblank_start + 1
+ *  vsync_end = vblank_start + 2
+ *  vtotal = vblank_start + 3
+ *
+ *           start of vblank:
+ *           latch double buffered registers
+ *           increment frame counter (ctg+)
+ *           generate start of vblank interrupt (gen4+)
+ *           |
+ *           |          frame start:
+ *           |          generate frame start interrupt (aka. vblank interrupt) (gmch)
+ *           |          may be shifted forward 1-3 extra lines via PIPECONF
+ *           |          |
+ *           |          |  start of vsync:
+ *           |          |  generate vsync interrupt
+ *           |          |  |
+ * ___xxxx___    ___xxxx___    ___xxxx___    ___xxxx___    ___xxxx___    ___xxxx
+ *       .   \hs/   .      \hs/          \hs/          \hs/   .      \hs/
+ * ----va---> <-----------------vb--------------------> <--------va-------------
+ *       |          |       <----vs----->                     |
+ * -vbs-----> <---vbs+1---> <---vbs+2---> <-----0-----> <-----1-----> <-----2--- (scanline counter gen2)
+ * -vbs-2---> <---vbs-1---> <---vbs-----> <---vbs+1---> <---vbs+2---> <-----0--- (scanline counter gen3+)
+ * -vbs-2---> <---vbs-2---> <---vbs-1---> <---vbs-----> <---vbs+1---> <---vbs+2- (scanline counter hsw+ hdmi)
+ *       |          |                                         |
+ *       last visible pixel                                   first visible pixel
+ *                  |                                         increment frame counter (gen3/4)
+ *                  pixel counter = vblank_start * htotal     pixel counter = 0 (gen3/4)
+ *
+ * x  = horizontal active
+ * _  = horizontal blanking
+ * hs = horizontal sync
+ * va = vertical active
+ * vb = vertical blanking
+ * vs = vertical sync
+ * vbs = vblank_start (number)
+ *
+ * Summary:
+ * - most events happen at the start of horizontal sync
+ * - frame start happens at the start of horizontal blank, 1-4 lines
+ *   (depending on PIPECONF settings) after the start of vblank
+ * - gen3/4 pixel and frame counter are synchronized with the start
+ *   of horizontal active on the first line of vertical active
+ */
+
 static u32 i8xx_get_vblank_counter(struct drm_device *dev, int pipe)
 {
 	/* Gen2 doesn't have a hardware frame counter */
@@ -651,7 +798,7 @@ static u32 i915_get_vblank_counter(struct drm_device *dev, int pipe)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long high_frame;
 	unsigned long low_frame;
-	u32 high1, high2, low, pixel, vbl_start;
+	u32 high1, high2, low, pixel, vbl_start, hsync_start, htotal;
 
 	if (!i915_pipe_enabled(dev, pipe)) {
 		DRM_DEBUG_DRIVER("trying to get vblank count for disabled "
@@ -665,17 +812,28 @@ static u32 i915_get_vblank_counter(struct drm_device *dev, int pipe)
 		const struct drm_display_mode *mode =
 			&intel_crtc->config.adjusted_mode;
 
-		vbl_start = mode->crtc_vblank_start * mode->crtc_htotal;
+		htotal = mode->crtc_htotal;
+		hsync_start = mode->crtc_hsync_start;
+		vbl_start = mode->crtc_vblank_start;
+		if (mode->flags & DRM_MODE_FLAG_INTERLACE)
+			vbl_start = DIV_ROUND_UP(vbl_start, 2);
 	} else {
 		enum transcoder cpu_transcoder = (enum transcoder) pipe;
-		u32 htotal;
 
 		htotal = ((I915_READ(HTOTAL(cpu_transcoder)) >> 16) & 0x1fff) + 1;
+		hsync_start = (I915_READ(HSYNC(cpu_transcoder))  & 0x1fff) + 1;
 		vbl_start = (I915_READ(VBLANK(cpu_transcoder)) & 0x1fff) + 1;
-
-		vbl_start *= htotal;
+		if ((I915_READ(PIPECONF(cpu_transcoder)) &
+		     PIPECONF_INTERLACE_MASK) != PIPECONF_PROGRESSIVE)
+			vbl_start = DIV_ROUND_UP(vbl_start, 2);
 	}
 
+	/* Convert to pixel count */
+	vbl_start *= htotal;
+
+	/* Start of vblank event occurs at start of hsync */
+	vbl_start -= htotal - hsync_start;
+
 	high_frame = PIPEFRAME(pipe);
 	low_frame = PIPEFRAMEPIXEL(pipe);
 
@@ -719,24 +877,28 @@ static u32 gm45_get_vblank_counter(struct drm_device *dev, int pipe)
 /* raw reads, only for fast reads of display block, no need for forcewake etc. */
 #define __raw_i915_read32(dev_priv__, reg__) readl((dev_priv__)->regs + (reg__))
 
-static bool ilk_pipe_in_vblank_locked(struct drm_device *dev, enum pipe pipe)
+static int __intel_get_crtc_scanline(struct intel_crtc *crtc)
 {
+	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	uint32_t status;
-	int reg;
+	const struct drm_display_mode *mode = &crtc->config.adjusted_mode;
+	enum pipe pipe = crtc->pipe;
+	int position, vtotal;
 
-	if (INTEL_INFO(dev)->gen >= 8) {
-		status = GEN8_PIPE_VBLANK;
-		reg = GEN8_DE_PIPE_ISR(pipe);
-	} else if (INTEL_INFO(dev)->gen >= 7) {
-		status = DE_PIPE_VBLANK_IVB(pipe);
-		reg = DEISR;
-	} else {
-		status = DE_PIPE_VBLANK(pipe);
-		reg = DEISR;
-	}
+	vtotal = mode->crtc_vtotal;
+	if (mode->flags & DRM_MODE_FLAG_INTERLACE)
+		vtotal /= 2;
+
+	if (IS_GEN2(dev))
+		position = __raw_i915_read32(dev_priv, PIPEDSL(pipe)) & DSL_LINEMASK_GEN2;
+	else
+		position = __raw_i915_read32(dev_priv, PIPEDSL(pipe)) & DSL_LINEMASK_GEN3;
 
-	return __raw_i915_read32(dev_priv, reg) & status;
+	/*
+	 * See update_scanline_offset() for the details on the
+	 * scanline_offset adjustment.
+	 */
+	return (position + crtc->scanline_offset) % vtotal;
 }
 
 static int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
@@ -748,7 +910,7 @@ static int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	const struct drm_display_mode *mode = &intel_crtc->config.adjusted_mode;
 	int position;
-	int vbl_start, vbl_end, htotal, vtotal;
+	int vbl_start, vbl_end, hsync_start, htotal, vtotal;
 	bool in_vbl = true;
 	int ret = 0;
 	unsigned long irqflags;
@@ -760,6 +922,7 @@ static int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
 	}
 
 	htotal = mode->crtc_htotal;
+	hsync_start = mode->crtc_hsync_start;
 	vtotal = mode->crtc_vtotal;
 	vbl_start = mode->crtc_vblank_start;
 	vbl_end = mode->crtc_vblank_end;
@@ -778,7 +941,7 @@ static int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
 	 * following code must not block on uncore.lock.
 	 */
 	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
-	
+
 	/* preempt_disable_rt() should go right here in PREEMPT_RT patchset. */
 
 	/* Get optional system timestamp before query. */
@@ -789,68 +952,7 @@ static int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
 		/* No obvious pixelcount register. Only query vertical
 		 * scanout position from Display scan line register.
 		 */
-		if (IS_GEN2(dev))
-			position = __raw_i915_read32(dev_priv, PIPEDSL(pipe)) & DSL_LINEMASK_GEN2;
-		else
-			position = __raw_i915_read32(dev_priv, PIPEDSL(pipe)) & DSL_LINEMASK_GEN3;
-
-		if (HAS_DDI(dev)) {
-			/*
-			 * On HSW HDMI outputs there seems to be a 2 line
-			 * difference, whereas eDP has the normal 1 line
-			 * difference that earlier platforms have. External
-			 * DP is unknown. For now just check for the 2 line
-			 * difference case on all output types on HSW+.
-			 *
-			 * This might misinterpret the scanline counter being
-			 * one line too far along on eDP, but that's less
-			 * dangerous than the alternative since that would lead
-			 * the vblank timestamp code astray when it sees a
-			 * scanline count before vblank_start during a vblank
-			 * interrupt.
-			 */
-			in_vbl = ilk_pipe_in_vblank_locked(dev, pipe);
-			if ((in_vbl && (position == vbl_start - 2 ||
-					position == vbl_start - 1)) ||
-			    (!in_vbl && (position == vbl_end - 2 ||
-					 position == vbl_end - 1)))
-				position = (position + 2) % vtotal;
-		} else if (HAS_PCH_SPLIT(dev)) {
-			/*
-			 * The scanline counter increments at the leading edge
-			 * of hsync, ie. it completely misses the active portion
-			 * of the line. Fix up the counter at both edges of vblank
-			 * to get a more accurate picture whether we're in vblank
-			 * or not.
-			 */
-			in_vbl = ilk_pipe_in_vblank_locked(dev, pipe);
-			if ((in_vbl && position == vbl_start - 1) ||
-			    (!in_vbl && position == vbl_end - 1))
-				position = (position + 1) % vtotal;
-		} else {
-			/*
-			 * ISR vblank status bits don't work the way we'd want
-			 * them to work on non-PCH platforms (for
-			 * ilk_pipe_in_vblank_locked()), and there doesn't
-			 * appear any other way to determine if we're currently
-			 * in vblank.
-			 *
-			 * Instead let's assume that we're already in vblank if
-			 * we got called from the vblank interrupt and the
-			 * scanline counter value indicates that we're on the
-			 * line just prior to vblank start. This should result
-			 * in the correct answer, unless the vblank interrupt
-			 * delivery really got delayed for almost exactly one
-			 * full frame/field.
-			 */
-			if (flags & DRM_CALLED_FROM_VBLIRQ &&
-			    position == vbl_start - 1) {
-				position = (position + 1) % vtotal;
-
-				/* Signal this correction as "applied". */
-				ret |= 0x8;
-			}
-		}
+		position = __intel_get_crtc_scanline(intel_crtc);
 	} else {
 		/* Have access to pixelcount since start of frame.
 		 * We can split this into vertical and horizontal
@@ -862,6 +964,29 @@ static int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
 		vbl_start *= htotal;
 		vbl_end *= htotal;
 		vtotal *= htotal;
+
+		/*
+		 * In interlaced modes, the pixel counter counts all pixels,
+		 * so one field will have htotal more pixels. In order to avoid
+		 * the reported position from jumping backwards when the pixel
+		 * counter is beyond the length of the shorter field, just
+		 * clamp the position the length of the shorter field. This
+		 * matches how the scanline counter based position works since
+		 * the scanline counter doesn't count the two half lines.
+		 */
+		if (position >= vtotal)
+			position = vtotal - 1;
+
+		/*
+		 * Start of vblank interrupt is triggered at start of hsync,
+		 * just prior to the first active line of vblank. However we
+		 * consider lines to start at the leading edge of horizontal
+		 * active. So, should we get here before we've crossed into
+		 * the horizontal active of the first line in vblank, we would
+		 * not set the DRM_SCANOUTPOS_INVBL flag. In order to fix that,
+		 * always add htotal-hsync_start to the current pixel position.
+		 */
+		position = (position + htotal - hsync_start) % vtotal;
 	}
 
 	/* Get optional system timestamp after query. */
@@ -900,6 +1025,19 @@ static int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
 	return ret;
 }
 
+int intel_get_crtc_scanline(struct intel_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	unsigned long irqflags;
+	int position;
+
+	spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
+	position = __intel_get_crtc_scanline(crtc);
+	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
+
+	return position;
+}
+
 static int i915_get_vblank_timestamp(struct drm_device *dev, int pipe,
 			      int *max_error,
 			      struct timeval *vblank_time,
@@ -945,7 +1083,7 @@ static bool intel_hpd_irq_event(struct drm_device *dev,
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s] status updated from %s to %s\n",
 		      connector->base.id,
-		      drm_get_connector_name(connector),
+		      connector->name,
 		      drm_get_connector_status_name(old_status),
 		      drm_get_connector_status_name(connector->status));
 
@@ -990,7 +1128,7 @@ static void i915_hotplug_work_func(struct work_struct *work)
 		    connector->polled == DRM_CONNECTOR_POLL_HPD) {
 			DRM_INFO("HPD interrupt storm detected on connector %s: "
 				 "switching from hotplug detection to polling\n",
-				drm_get_connector_name(connector));
+				connector->name);
 			dev_priv->hpd_stats[intel_encoder->hpd_pin].hpd_mark = HPD_DISABLED;
 			connector->polled = DRM_CONNECTOR_POLL_CONNECT
 				| DRM_CONNECTOR_POLL_DISCONNECT;
@@ -998,7 +1136,7 @@ static void i915_hotplug_work_func(struct work_struct *work)
 		}
 		if (hpd_event_bits & (1 << intel_encoder->hpd_pin)) {
 			DRM_DEBUG_KMS("Connector %s (pin %i) received hotplug event.\n",
-				      drm_get_connector_name(connector), intel_encoder->hpd_pin);
+				      connector->name, intel_encoder->hpd_pin);
 		}
 	}
 	 /* if there were no outputs to poll, poll was disabled,
@@ -1073,9 +1211,9 @@ static void ironlake_rps_change_irq_handler(struct drm_device *dev)
 }
 
 static void notify_ring(struct drm_device *dev,
-			struct intel_ring_buffer *ring)
+			struct intel_engine_cs *ring)
 {
-	if (ring->obj == NULL)
+	if (!intel_ring_initialized(ring))
 		return;
 
 	trace_i915_gem_request_complete(ring);
@@ -1094,8 +1232,12 @@ static void gen6_pm_rps_work(struct work_struct *work)
 	spin_lock_irq(&dev_priv->irq_lock);
 	pm_iir = dev_priv->rps.pm_iir;
 	dev_priv->rps.pm_iir = 0;
-	/* Make sure not to corrupt PMIMR state used by ringbuffer code */
-	snb_enable_pm_irq(dev_priv, dev_priv->pm_rps_events);
+	if (IS_BROADWELL(dev_priv->dev))
+		bdw_enable_pm_irq(dev_priv, dev_priv->pm_rps_events);
+	else {
+		/* Make sure not to corrupt PMIMR state used by ringbuffer */
+		snb_enable_pm_irq(dev_priv, dev_priv->pm_rps_events);
+	}
 	spin_unlock_irq(&dev_priv->irq_lock);
 
 	/* Make sure we didn't queue anything we're not going to process. */
@@ -1292,6 +1434,19 @@ static void snb_gt_irq_handler(struct drm_device *dev,
 		ivybridge_parity_error_irq_handler(dev, gt_iir);
 }
 
+static void gen8_rps_irq_handler(struct drm_i915_private *dev_priv, u32 pm_iir)
+{
+	if ((pm_iir & dev_priv->pm_rps_events) == 0)
+		return;
+
+	spin_lock(&dev_priv->irq_lock);
+	dev_priv->rps.pm_iir |= pm_iir & dev_priv->pm_rps_events;
+	bdw_disable_pm_irq(dev_priv, pm_iir & dev_priv->pm_rps_events);
+	spin_unlock(&dev_priv->irq_lock);
+
+	queue_work(dev_priv->wq, &dev_priv->rps.work);
+}
+
 static irqreturn_t gen8_gt_irq_handler(struct drm_device *dev,
 				       struct drm_i915_private *dev_priv,
 				       u32 master_ctl)
@@ -1315,18 +1470,32 @@ static irqreturn_t gen8_gt_irq_handler(struct drm_device *dev,
 			DRM_ERROR("The master control interrupt lied (GT0)!\n");
 	}
 
-	if (master_ctl & GEN8_GT_VCS1_IRQ) {
+	if (master_ctl & (GEN8_GT_VCS1_IRQ | GEN8_GT_VCS2_IRQ)) {
 		tmp = I915_READ(GEN8_GT_IIR(1));
 		if (tmp) {
 			ret = IRQ_HANDLED;
 			vcs = tmp >> GEN8_VCS1_IRQ_SHIFT;
 			if (vcs & GT_RENDER_USER_INTERRUPT)
 				notify_ring(dev, &dev_priv->ring[VCS]);
+			vcs = tmp >> GEN8_VCS2_IRQ_SHIFT;
+			if (vcs & GT_RENDER_USER_INTERRUPT)
+				notify_ring(dev, &dev_priv->ring[VCS2]);
 			I915_WRITE(GEN8_GT_IIR(1), tmp);
 		} else
 			DRM_ERROR("The master control interrupt lied (GT1)!\n");
 	}
 
+	if (master_ctl & GEN8_GT_PM_IRQ) {
+		tmp = I915_READ(GEN8_GT_IIR(2));
+		if (tmp & dev_priv->pm_rps_events) {
+			ret = IRQ_HANDLED;
+			gen8_rps_irq_handler(dev_priv, tmp);
+			I915_WRITE(GEN8_GT_IIR(2),
+				   tmp & dev_priv->pm_rps_events);
+		} else
+			DRM_ERROR("The master control interrupt lied (PM)!\n");
+	}
+
 	if (master_ctl & GEN8_GT_VECS_IRQ) {
 		tmp = I915_READ(GEN8_GT_IIR(3));
 		if (tmp) {
@@ -1549,6 +1718,19 @@ static void gen6_rps_irq_handler(struct drm_i915_private *dev_priv, u32 pm_iir)
 	}
 }
 
+static bool intel_pipe_handle_vblank(struct drm_device *dev, enum pipe pipe)
+{
+	struct intel_crtc *crtc;
+
+	if (!drm_handle_vblank(dev, pipe))
+		return false;
+
+	crtc = to_intel_crtc(intel_get_crtc_for_pipe(dev, pipe));
+	wake_up(&crtc->vbl_wait);
+
+	return true;
+}
+
 static void valleyview_pipestat_irq_handler(struct drm_device *dev, u32 iir)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1578,6 +1760,9 @@ static void valleyview_pipestat_irq_handler(struct drm_device *dev, u32 iir)
 		case PIPE_B:
 			iir_bit = I915_DISPLAY_PIPE_B_EVENT_INTERRUPT;
 			break;
+		case PIPE_C:
+			iir_bit = I915_DISPLAY_PIPE_C_EVENT_INTERRUPT;
+			break;
 		}
 		if (iir & iir_bit)
 			mask |= dev_priv->pipestat_irq_mask[pipe];
@@ -1600,7 +1785,7 @@ static void valleyview_pipestat_irq_handler(struct drm_device *dev, u32 iir)
 
 	for_each_pipe(pipe) {
 		if (pipe_stats[pipe] & PIPE_START_VBLANK_INTERRUPT_STATUS)
-			drm_handle_vblank(dev, pipe);
+			intel_pipe_handle_vblank(dev, pipe);
 
 		if (pipe_stats[pipe] & PLANE_FLIP_DONE_INT_STATUS_VLV) {
 			intel_prepare_page_flip(dev, pipe);
@@ -1619,9 +1804,36 @@ static void valleyview_pipestat_irq_handler(struct drm_device *dev, u32 iir)
 		gmbus_irq_handler(dev);
 }
 
+static void i9xx_hpd_irq_handler(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 hotplug_status = I915_READ(PORT_HOTPLUG_STAT);
+
+	if (IS_G4X(dev)) {
+		u32 hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_G4X;
+
+		intel_hpd_irq_handler(dev, hotplug_trigger, hpd_status_g4x);
+	} else {
+		u32 hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_I915;
+
+		intel_hpd_irq_handler(dev, hotplug_trigger, hpd_status_i915);
+	}
+
+	if ((IS_G4X(dev) || IS_VALLEYVIEW(dev)) &&
+	    hotplug_status & DP_AUX_CHANNEL_MASK_INT_STATUS_G4X)
+		dp_aux_irq_handler(dev);
+
+	I915_WRITE(PORT_HOTPLUG_STAT, hotplug_status);
+	/*
+	 * Make sure hotplug status is cleared before we clear IIR, or else we
+	 * may miss hotplug events.
+	 */
+	POSTING_READ(PORT_HOTPLUG_STAT);
+}
+
 static irqreturn_t valleyview_irq_handler(int irq, void *arg)
 {
-	struct drm_device *dev = (struct drm_device *) arg;
+	struct drm_device *dev = arg;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 iir, gt_iir, pm_iir;
 	irqreturn_t ret = IRQ_NONE;
@@ -1641,19 +1853,8 @@ static irqreturn_t valleyview_irq_handler(int irq, void *arg)
 		valleyview_pipestat_irq_handler(dev, iir);
 
 		/* Consume port.  Then clear IIR or we'll miss events */
-		if (iir & I915_DISPLAY_PORT_INTERRUPT) {
-			u32 hotplug_status = I915_READ(PORT_HOTPLUG_STAT);
-			u32 hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_I915;
-
-			intel_hpd_irq_handler(dev, hotplug_trigger, hpd_status_i915);
-
-			if (hotplug_status & DP_AUX_CHANNEL_MASK_INT_STATUS_G4X)
-				dp_aux_irq_handler(dev);
-
-			I915_WRITE(PORT_HOTPLUG_STAT, hotplug_status);
-			I915_READ(PORT_HOTPLUG_STAT);
-		}
-
+		if (iir & I915_DISPLAY_PORT_INTERRUPT)
+			i9xx_hpd_irq_handler(dev);
 
 		if (pm_iir)
 			gen6_rps_irq_handler(dev_priv, pm_iir);
@@ -1667,6 +1868,40 @@ out:
 	return ret;
 }
 
+static irqreturn_t cherryview_irq_handler(int irq, void *arg)
+{
+	struct drm_device *dev = arg;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 master_ctl, iir;
+	irqreturn_t ret = IRQ_NONE;
+
+	for (;;) {
+		master_ctl = I915_READ(GEN8_MASTER_IRQ) & ~GEN8_MASTER_IRQ_CONTROL;
+		iir = I915_READ(VLV_IIR);
+
+		if (master_ctl == 0 && iir == 0)
+			break;
+
+		I915_WRITE(GEN8_MASTER_IRQ, 0);
+
+		gen8_gt_irq_handler(dev, dev_priv, master_ctl);
+
+		valleyview_pipestat_irq_handler(dev, iir);
+
+		/* Consume port.  Then clear IIR or we'll miss events */
+		i9xx_hpd_irq_handler(dev);
+
+		I915_WRITE(VLV_IIR, iir);
+
+		I915_WRITE(GEN8_MASTER_IRQ, DE_MASTER_IRQ_CONTROL);
+		POSTING_READ(GEN8_MASTER_IRQ);
+
+		ret = IRQ_HANDLED;
+	}
+
+	return ret;
+}
+
 static void ibx_irq_handler(struct drm_device *dev, u32 pch_iir)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1827,7 +2062,7 @@ static void ilk_display_irq_handler(struct drm_device *dev, u32 de_iir)
 
 	for_each_pipe(pipe) {
 		if (de_iir & DE_PIPE_VBLANK(pipe))
-			drm_handle_vblank(dev, pipe);
+			intel_pipe_handle_vblank(dev, pipe);
 
 		if (de_iir & DE_PIPE_FIFO_UNDERRUN(pipe))
 			if (intel_set_cpu_fifo_underrun_reporting(dev, pipe, false))
@@ -1877,7 +2112,7 @@ static void ivb_display_irq_handler(struct drm_device *dev, u32 de_iir)
 
 	for_each_pipe(pipe) {
 		if (de_iir & (DE_PIPE_VBLANK_IVB(pipe)))
-			drm_handle_vblank(dev, pipe);
+			intel_pipe_handle_vblank(dev, pipe);
 
 		/* plane/pipes map 1:1 on ilk+ */
 		if (de_iir & DE_PLANE_FLIP_DONE_IVB(pipe)) {
@@ -1899,7 +2134,7 @@ static void ivb_display_irq_handler(struct drm_device *dev, u32 de_iir)
 
 static irqreturn_t ironlake_irq_handler(int irq, void *arg)
 {
-	struct drm_device *dev = (struct drm_device *) arg;
+	struct drm_device *dev = arg;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 de_iir, gt_iir, de_ier, sde_ier = 0;
 	irqreturn_t ret = IRQ_NONE;
@@ -2020,9 +2255,9 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg)
 
 		pipe_iir = I915_READ(GEN8_DE_PIPE_IIR(pipe));
 		if (pipe_iir & GEN8_PIPE_VBLANK)
-			drm_handle_vblank(dev, pipe);
+			intel_pipe_handle_vblank(dev, pipe);
 
-		if (pipe_iir & GEN8_PIPE_FLIP_DONE) {
+		if (pipe_iir & GEN8_PIPE_PRIMARY_FLIP_DONE) {
 			intel_prepare_page_flip(dev, pipe);
 			intel_finish_page_flip_plane(dev, pipe);
 		}
@@ -2075,7 +2310,7 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg)
 static void i915_error_wake_up(struct drm_i915_private *dev_priv,
 			       bool reset_completed)
 {
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int i;
 
 	/*
@@ -2137,6 +2372,14 @@ static void i915_error_work_func(struct work_struct *work)
 				   reset_event);
 
 		/*
+		 * In most cases it's guaranteed that we get here with an RPM
+		 * reference held, for example because there is a pending GPU
+		 * request that won't finish until the reset is done. This
+		 * isn't the case at least when we get here by doing a
+		 * simulated reset via debugs, so get an RPM reference.
+		 */
+		intel_runtime_pm_get(dev_priv);
+		/*
 		 * All state reset _must_ be completed before we update the
 		 * reset counter, for otherwise waiters might miss the reset
 		 * pending state and not properly drop locks, resulting in
@@ -2146,6 +2389,8 @@ static void i915_error_work_func(struct work_struct *work)
 
 		intel_display_handle_reset(dev);
 
+		intel_runtime_pm_put(dev_priv);
+
 		if (ret == 0) {
 			/*
 			 * After all the gem state is reset, increment the reset
@@ -2383,10 +2628,6 @@ static int i915_enable_vblank(struct drm_device *dev, int pipe)
 	else
 		i915_enable_pipestat(dev_priv, pipe,
 				     PIPE_VBLANK_INTERRUPT_STATUS);
-
-	/* maintain vblank delivery even in deep C-states */
-	if (INTEL_INFO(dev)->gen == 3)
-		I915_WRITE(INSTPM, _MASKED_BIT_DISABLE(INSTPM_AGPBUSY_DIS));
 	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 
 	return 0;
@@ -2450,9 +2691,6 @@ static void i915_disable_vblank(struct drm_device *dev, int pipe)
 	unsigned long irqflags;
 
 	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	if (INTEL_INFO(dev)->gen == 3)
-		I915_WRITE(INSTPM, _MASKED_BIT_ENABLE(INSTPM_AGPBUSY_DIS));
-
 	i915_disable_pipestat(dev_priv, pipe,
 			      PIPE_VBLANK_INTERRUPT_STATUS |
 			      PIPE_START_VBLANK_INTERRUPT_STATUS);
@@ -2498,29 +2736,77 @@ static void gen8_disable_vblank(struct drm_device *dev, int pipe)
 }
 
 static u32
-ring_last_seqno(struct intel_ring_buffer *ring)
+ring_last_seqno(struct intel_engine_cs *ring)
 {
 	return list_entry(ring->request_list.prev,
 			  struct drm_i915_gem_request, list)->seqno;
 }
 
 static bool
-ring_idle(struct intel_ring_buffer *ring, u32 seqno)
+ring_idle(struct intel_engine_cs *ring, u32 seqno)
 {
 	return (list_empty(&ring->request_list) ||
 		i915_seqno_passed(seqno, ring_last_seqno(ring)));
 }
 
-static struct intel_ring_buffer *
-semaphore_waits_for(struct intel_ring_buffer *ring, u32 *seqno)
+static bool
+ipehr_is_semaphore_wait(struct drm_device *dev, u32 ipehr)
+{
+	if (INTEL_INFO(dev)->gen >= 8) {
+		/*
+		 * FIXME: gen8 semaphore support - currently we don't emit
+		 * semaphores on bdw anyway, but this needs to be addressed when
+		 * we merge that code.
+		 */
+		return false;
+	} else {
+		ipehr &= ~MI_SEMAPHORE_SYNC_MASK;
+		return ipehr == (MI_SEMAPHORE_MBOX | MI_SEMAPHORE_COMPARE |
+				 MI_SEMAPHORE_REGISTER);
+	}
+}
+
+static struct intel_engine_cs *
+semaphore_wait_to_signaller_ring(struct intel_engine_cs *ring, u32 ipehr)
+{
+	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	struct intel_engine_cs *signaller;
+	int i;
+
+	if (INTEL_INFO(dev_priv->dev)->gen >= 8) {
+		/*
+		 * FIXME: gen8 semaphore support - currently we don't emit
+		 * semaphores on bdw anyway, but this needs to be addressed when
+		 * we merge that code.
+		 */
+		return NULL;
+	} else {
+		u32 sync_bits = ipehr & MI_SEMAPHORE_SYNC_MASK;
+
+		for_each_ring(signaller, dev_priv, i) {
+			if(ring == signaller)
+				continue;
+
+			if (sync_bits == signaller->semaphore.mbox.wait[ring->id])
+				return signaller;
+		}
+	}
+
+	DRM_ERROR("No signaller ring found for ring %i, ipehr 0x%08x\n",
+		  ring->id, ipehr);
+
+	return NULL;
+}
+
+static struct intel_engine_cs *
+semaphore_waits_for(struct intel_engine_cs *ring, u32 *seqno)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 	u32 cmd, ipehr, head;
 	int i;
 
 	ipehr = I915_READ(RING_IPEHR(ring->mmio_base));
-	if ((ipehr & ~(0x3 << 16)) !=
-	    (MI_SEMAPHORE_MBOX | MI_SEMAPHORE_COMPARE | MI_SEMAPHORE_REGISTER))
+	if (!ipehr_is_semaphore_wait(ring->dev, ipehr))
 		return NULL;
 
 	/*
@@ -2538,10 +2824,10 @@ semaphore_waits_for(struct intel_ring_buffer *ring, u32 *seqno)
 		 * our ring is smaller than what the hardware (and hence
 		 * HEAD_ADDR) allows. Also handles wrap-around.
 		 */
-		head &= ring->size - 1;
+		head &= ring->buffer->size - 1;
 
 		/* This here seems to blow up */
-		cmd = ioread32(ring->virtual_start + head);
+		cmd = ioread32(ring->buffer->virtual_start + head);
 		if (cmd == ipehr)
 			break;
 
@@ -2551,14 +2837,14 @@ semaphore_waits_for(struct intel_ring_buffer *ring, u32 *seqno)
 	if (!i)
 		return NULL;
 
-	*seqno = ioread32(ring->virtual_start + head + 4) + 1;
-	return &dev_priv->ring[(ring->id + (((ipehr >> 17) & 1) + 1)) % 3];
+	*seqno = ioread32(ring->buffer->virtual_start + head + 4) + 1;
+	return semaphore_wait_to_signaller_ring(ring, ipehr);
 }
 
-static int semaphore_passed(struct intel_ring_buffer *ring)
+static int semaphore_passed(struct intel_engine_cs *ring)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
-	struct intel_ring_buffer *signaller;
+	struct intel_engine_cs *signaller;
 	u32 seqno, ctl;
 
 	ring->hangcheck.deadlock = true;
@@ -2577,7 +2863,7 @@ static int semaphore_passed(struct intel_ring_buffer *ring)
 
 static void semaphore_clear_deadlocks(struct drm_i915_private *dev_priv)
 {
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int i;
 
 	for_each_ring(ring, dev_priv, i)
@@ -2585,7 +2871,7 @@ static void semaphore_clear_deadlocks(struct drm_i915_private *dev_priv)
 }
 
 static enum intel_ring_hangcheck_action
-ring_stuck(struct intel_ring_buffer *ring, u64 acthd)
+ring_stuck(struct intel_engine_cs *ring, u64 acthd)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -2641,7 +2927,7 @@ static void i915_hangcheck_elapsed(unsigned long data)
 {
 	struct drm_device *dev = (struct drm_device *)data;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	int i;
 	int busy_count = 0, rings_hung = 0;
 	bool stuck[I915_NUM_RINGS] = { 0 };
@@ -2759,57 +3045,63 @@ void i915_queue_hangcheck(struct drm_device *dev)
 		  round_jiffies_up(jiffies + DRM_I915_HANGCHECK_JIFFIES));
 }
 
-static void ibx_irq_preinstall(struct drm_device *dev)
+static void ibx_irq_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	if (HAS_PCH_NOP(dev))
 		return;
 
-	/* south display irq */
-	I915_WRITE(SDEIMR, 0xffffffff);
-	/*
-	 * SDEIER is also touched by the interrupt handler to work around missed
-	 * PCH interrupts. Hence we can't update it after the interrupt handler
-	 * is enabled - instead we unconditionally enable all PCH interrupt
-	 * sources here, but then only unmask them as needed with SDEIMR.
-	 */
+	GEN5_IRQ_RESET(SDE);
+
+	if (HAS_PCH_CPT(dev) || HAS_PCH_LPT(dev))
+		I915_WRITE(SERR_INT, 0xffffffff);
+}
+
+/*
+ * SDEIER is also touched by the interrupt handler to work around missed PCH
+ * interrupts. Hence we can't update it after the interrupt handler is enabled -
+ * instead we unconditionally enable all PCH interrupt sources here, but then
+ * only unmask them as needed with SDEIMR.
+ *
+ * This function needs to be called before interrupts are enabled.
+ */
+static void ibx_irq_pre_postinstall(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	if (HAS_PCH_NOP(dev))
+		return;
+
+	WARN_ON(I915_READ(SDEIER) != 0);
 	I915_WRITE(SDEIER, 0xffffffff);
 	POSTING_READ(SDEIER);
 }
 
-static void gen5_gt_irq_preinstall(struct drm_device *dev)
+static void gen5_gt_irq_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	/* and GT */
-	I915_WRITE(GTIMR, 0xffffffff);
-	I915_WRITE(GTIER, 0x0);
-	POSTING_READ(GTIER);
-
-	if (INTEL_INFO(dev)->gen >= 6) {
-		/* and PM */
-		I915_WRITE(GEN6_PMIMR, 0xffffffff);
-		I915_WRITE(GEN6_PMIER, 0x0);
-		POSTING_READ(GEN6_PMIER);
-	}
+	GEN5_IRQ_RESET(GT);
+	if (INTEL_INFO(dev)->gen >= 6)
+		GEN5_IRQ_RESET(GEN6_PM);
 }
 
 /* drm_dma.h hooks
 */
-static void ironlake_irq_preinstall(struct drm_device *dev)
+static void ironlake_irq_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	I915_WRITE(HWSTAM, 0xeffe);
+	I915_WRITE(HWSTAM, 0xffffffff);
 
-	I915_WRITE(DEIMR, 0xffffffff);
-	I915_WRITE(DEIER, 0x0);
-	POSTING_READ(DEIER);
+	GEN5_IRQ_RESET(DE);
+	if (IS_GEN7(dev))
+		I915_WRITE(GEN7_ERR_INT, 0xffffffff);
 
-	gen5_gt_irq_preinstall(dev);
+	gen5_gt_irq_reset(dev);
 
-	ibx_irq_preinstall(dev);
+	ibx_irq_reset(dev);
 }
 
 static void valleyview_irq_preinstall(struct drm_device *dev)
@@ -2827,7 +3119,7 @@ static void valleyview_irq_preinstall(struct drm_device *dev)
 	I915_WRITE(GTIIR, I915_READ(GTIIR));
 	I915_WRITE(GTIIR, I915_READ(GTIIR));
 
-	gen5_gt_irq_preinstall(dev);
+	gen5_gt_irq_reset(dev);
 
 	I915_WRITE(DPINVGTT, 0xff);
 
@@ -2841,7 +3133,15 @@ static void valleyview_irq_preinstall(struct drm_device *dev)
 	POSTING_READ(VLV_IER);
 }
 
-static void gen8_irq_preinstall(struct drm_device *dev)
+static void gen8_gt_irq_reset(struct drm_i915_private *dev_priv)
+{
+	GEN8_IRQ_RESET_NDX(GT, 0);
+	GEN8_IRQ_RESET_NDX(GT, 1);
+	GEN8_IRQ_RESET_NDX(GT, 2);
+	GEN8_IRQ_RESET_NDX(GT, 3);
+}
+
+static void gen8_irq_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int pipe;
@@ -2849,43 +3149,44 @@ static void gen8_irq_preinstall(struct drm_device *dev)
 	I915_WRITE(GEN8_MASTER_IRQ, 0);
 	POSTING_READ(GEN8_MASTER_IRQ);
 
-	/* IIR can theoretically queue up two events. Be paranoid */
-#define GEN8_IRQ_INIT_NDX(type, which) do { \
-		I915_WRITE(GEN8_##type##_IMR(which), 0xffffffff); \
-		POSTING_READ(GEN8_##type##_IMR(which)); \
-		I915_WRITE(GEN8_##type##_IER(which), 0); \
-		I915_WRITE(GEN8_##type##_IIR(which), 0xffffffff); \
-		POSTING_READ(GEN8_##type##_IIR(which)); \
-		I915_WRITE(GEN8_##type##_IIR(which), 0xffffffff); \
-	} while (0)
-
-#define GEN8_IRQ_INIT(type) do { \
-		I915_WRITE(GEN8_##type##_IMR, 0xffffffff); \
-		POSTING_READ(GEN8_##type##_IMR); \
-		I915_WRITE(GEN8_##type##_IER, 0); \
-		I915_WRITE(GEN8_##type##_IIR, 0xffffffff); \
-		POSTING_READ(GEN8_##type##_IIR); \
-		I915_WRITE(GEN8_##type##_IIR, 0xffffffff); \
-	} while (0)
-
-	GEN8_IRQ_INIT_NDX(GT, 0);
-	GEN8_IRQ_INIT_NDX(GT, 1);
-	GEN8_IRQ_INIT_NDX(GT, 2);
-	GEN8_IRQ_INIT_NDX(GT, 3);
+	gen8_gt_irq_reset(dev_priv);
 
-	for_each_pipe(pipe) {
-		GEN8_IRQ_INIT_NDX(DE_PIPE, pipe);
-	}
+	for_each_pipe(pipe)
+		GEN8_IRQ_RESET_NDX(DE_PIPE, pipe);
+
+	GEN5_IRQ_RESET(GEN8_DE_PORT_);
+	GEN5_IRQ_RESET(GEN8_DE_MISC_);
+	GEN5_IRQ_RESET(GEN8_PCU_);
+
+	ibx_irq_reset(dev);
+}
+
+static void cherryview_irq_preinstall(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int pipe;
+
+	I915_WRITE(GEN8_MASTER_IRQ, 0);
+	POSTING_READ(GEN8_MASTER_IRQ);
 
-	GEN8_IRQ_INIT(DE_PORT);
-	GEN8_IRQ_INIT(DE_MISC);
-	GEN8_IRQ_INIT(PCU);
-#undef GEN8_IRQ_INIT
-#undef GEN8_IRQ_INIT_NDX
+	gen8_gt_irq_reset(dev_priv);
+
+	GEN5_IRQ_RESET(GEN8_PCU_);
 
 	POSTING_READ(GEN8_PCU_IIR);
 
-	ibx_irq_preinstall(dev);
+	I915_WRITE(DPINVGTT, DPINVGTT_STATUS_MASK_CHV);
+
+	I915_WRITE(PORT_HOTPLUG_EN, 0);
+	I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT));
+
+	for_each_pipe(pipe)
+		I915_WRITE(PIPESTAT(pipe), 0xffff);
+
+	I915_WRITE(VLV_IMR, 0xffffffff);
+	I915_WRITE(VLV_IER, 0x0);
+	I915_WRITE(VLV_IIR, 0xffffffff);
+	POSTING_READ(VLV_IIR);
 }
 
 static void ibx_hpd_irq_setup(struct drm_device *dev)
@@ -2931,15 +3232,12 @@ static void ibx_irq_postinstall(struct drm_device *dev)
 	if (HAS_PCH_NOP(dev))
 		return;
 
-	if (HAS_PCH_IBX(dev)) {
+	if (HAS_PCH_IBX(dev))
 		mask = SDE_GMBUS | SDE_AUX_MASK | SDE_POISON;
-	} else {
+	else
 		mask = SDE_GMBUS_CPT | SDE_AUX_MASK_CPT;
 
-		I915_WRITE(SERR_INT, I915_READ(SERR_INT));
-	}
-
-	I915_WRITE(SDEIIR, I915_READ(SDEIIR));
+	GEN5_ASSERT_IIR_IS_ZERO(SDEIIR);
 	I915_WRITE(SDEIMR, ~mask);
 }
 
@@ -2965,10 +3263,7 @@ static void gen5_gt_irq_postinstall(struct drm_device *dev)
 		gt_irqs |= GT_BLT_USER_INTERRUPT | GT_BSD_USER_INTERRUPT;
 	}
 
-	I915_WRITE(GTIIR, I915_READ(GTIIR));
-	I915_WRITE(GTIMR, dev_priv->gt_irq_mask);
-	I915_WRITE(GTIER, gt_irqs);
-	POSTING_READ(GTIER);
+	GEN5_IRQ_INIT(GT, dev_priv->gt_irq_mask, gt_irqs);
 
 	if (INTEL_INFO(dev)->gen >= 6) {
 		pm_irqs |= dev_priv->pm_rps_events;
@@ -2977,10 +3272,7 @@ static void gen5_gt_irq_postinstall(struct drm_device *dev)
 			pm_irqs |= PM_VEBOX_USER_INTERRUPT;
 
 		dev_priv->pm_irq_mask = 0xffffffff;
-		I915_WRITE(GEN6_PMIIR, I915_READ(GEN6_PMIIR));
-		I915_WRITE(GEN6_PMIMR, dev_priv->pm_irq_mask);
-		I915_WRITE(GEN6_PMIER, pm_irqs);
-		POSTING_READ(GEN6_PMIER);
+		GEN5_IRQ_INIT(GEN6_PM, dev_priv->pm_irq_mask, pm_irqs);
 	}
 }
 
@@ -2997,8 +3289,6 @@ static int ironlake_irq_postinstall(struct drm_device *dev)
 				DE_PLANEA_FLIP_DONE_IVB | DE_AUX_CHANNEL_A_IVB);
 		extra_mask = (DE_PIPEC_VBLANK_IVB | DE_PIPEB_VBLANK_IVB |
 			      DE_PIPEA_VBLANK_IVB | DE_ERR_INT_IVB);
-
-		I915_WRITE(GEN7_ERR_INT, I915_READ(GEN7_ERR_INT));
 	} else {
 		display_mask = (DE_MASTER_IRQ_CONTROL | DE_GSE | DE_PCH_EVENT |
 				DE_PLANEA_FLIP_DONE | DE_PLANEB_FLIP_DONE |
@@ -3011,11 +3301,11 @@ static int ironlake_irq_postinstall(struct drm_device *dev)
 
 	dev_priv->irq_mask = ~display_mask;
 
-	/* should always can generate irq */
-	I915_WRITE(DEIIR, I915_READ(DEIIR));
-	I915_WRITE(DEIMR, dev_priv->irq_mask);
-	I915_WRITE(DEIER, display_mask | extra_mask);
-	POSTING_READ(DEIER);
+	I915_WRITE(HWSTAM, 0xeffe);
+
+	ibx_irq_pre_postinstall(dev);
+
+	GEN5_IRQ_INIT(DE, dev_priv->irq_mask, display_mask | extra_mask);
 
 	gen5_gt_irq_postinstall(dev);
 
@@ -3175,21 +3465,16 @@ static void gen8_gt_irq_postinstall(struct drm_i915_private *dev_priv)
 		GT_RENDER_USER_INTERRUPT << GEN8_VECS_IRQ_SHIFT
 		};
 
-	for (i = 0; i < ARRAY_SIZE(gt_interrupts); i++) {
-		u32 tmp = I915_READ(GEN8_GT_IIR(i));
-		if (tmp)
-			DRM_ERROR("Interrupt (%d) should have been masked in pre-install 0x%08x\n",
-				  i, tmp);
-		I915_WRITE(GEN8_GT_IMR(i), ~gt_interrupts[i]);
-		I915_WRITE(GEN8_GT_IER(i), gt_interrupts[i]);
-	}
-	POSTING_READ(GEN8_GT_IER(0));
+	for (i = 0; i < ARRAY_SIZE(gt_interrupts); i++)
+		GEN8_IRQ_INIT_NDX(GT, i, ~gt_interrupts[i], gt_interrupts[i]);
+
+	dev_priv->pm_irq_mask = 0xffffffff;
 }
 
 static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 {
 	struct drm_device *dev = dev_priv->dev;
-	uint32_t de_pipe_masked = GEN8_PIPE_FLIP_DONE |
+	uint32_t de_pipe_masked = GEN8_PIPE_PRIMARY_FLIP_DONE |
 		GEN8_PIPE_CDCLK_CRC_DONE |
 		GEN8_DE_PIPE_IRQ_FAULT_ERRORS;
 	uint32_t de_pipe_enables = de_pipe_masked | GEN8_PIPE_VBLANK |
@@ -3199,25 +3484,19 @@ static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 	dev_priv->de_irq_mask[PIPE_B] = ~de_pipe_masked;
 	dev_priv->de_irq_mask[PIPE_C] = ~de_pipe_masked;
 
-	for_each_pipe(pipe) {
-		u32 tmp = I915_READ(GEN8_DE_PIPE_IIR(pipe));
-		if (tmp)
-			DRM_ERROR("Interrupt (%d) should have been masked in pre-install 0x%08x\n",
-				  pipe, tmp);
-		I915_WRITE(GEN8_DE_PIPE_IMR(pipe), dev_priv->de_irq_mask[pipe]);
-		I915_WRITE(GEN8_DE_PIPE_IER(pipe), de_pipe_enables);
-	}
-	POSTING_READ(GEN8_DE_PIPE_ISR(0));
+	for_each_pipe(pipe)
+		GEN8_IRQ_INIT_NDX(DE_PIPE, pipe, dev_priv->de_irq_mask[pipe],
+				  de_pipe_enables);
 
-	I915_WRITE(GEN8_DE_PORT_IMR, ~GEN8_AUX_CHANNEL_A);
-	I915_WRITE(GEN8_DE_PORT_IER, GEN8_AUX_CHANNEL_A);
-	POSTING_READ(GEN8_DE_PORT_IER);
+	GEN5_IRQ_INIT(GEN8_DE_PORT_, ~GEN8_AUX_CHANNEL_A, GEN8_AUX_CHANNEL_A);
 }
 
 static int gen8_irq_postinstall(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
+	ibx_irq_pre_postinstall(dev);
+
 	gen8_gt_irq_postinstall(dev_priv);
 	gen8_de_irq_postinstall(dev_priv);
 
@@ -3229,44 +3508,55 @@ static int gen8_irq_postinstall(struct drm_device *dev)
 	return 0;
 }
 
-static void gen8_irq_uninstall(struct drm_device *dev)
+static int cherryview_irq_postinstall(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 enable_mask = I915_DISPLAY_PORT_INTERRUPT |
+		I915_DISPLAY_PIPE_A_EVENT_INTERRUPT |
+		I915_DISPLAY_PIPE_B_EVENT_INTERRUPT |
+		I915_DISPLAY_PIPE_C_EVENT_INTERRUPT;
+	u32 pipestat_enable = PLANE_FLIP_DONE_INT_STATUS_VLV |
+		PIPE_CRC_DONE_INTERRUPT_STATUS;
+	unsigned long irqflags;
 	int pipe;
 
-	if (!dev_priv)
-		return;
+	/*
+	 * Leave vblank interrupts masked initially.  enable/disable will
+	 * toggle them based on usage.
+	 */
+	dev_priv->irq_mask = ~enable_mask;
 
-	I915_WRITE(GEN8_MASTER_IRQ, 0);
+	for_each_pipe(pipe)
+		I915_WRITE(PIPESTAT(pipe), 0xffff);
 
-#define GEN8_IRQ_FINI_NDX(type, which) do { \
-		I915_WRITE(GEN8_##type##_IMR(which), 0xffffffff); \
-		I915_WRITE(GEN8_##type##_IER(which), 0); \
-		I915_WRITE(GEN8_##type##_IIR(which), 0xffffffff); \
-	} while (0)
+	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
+	i915_enable_pipestat(dev_priv, PIPE_A, PIPE_GMBUS_INTERRUPT_STATUS);
+	for_each_pipe(pipe)
+		i915_enable_pipestat(dev_priv, pipe, pipestat_enable);
+	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 
-#define GEN8_IRQ_FINI(type) do { \
-		I915_WRITE(GEN8_##type##_IMR, 0xffffffff); \
-		I915_WRITE(GEN8_##type##_IER, 0); \
-		I915_WRITE(GEN8_##type##_IIR, 0xffffffff); \
-	} while (0)
+	I915_WRITE(VLV_IIR, 0xffffffff);
+	I915_WRITE(VLV_IMR, dev_priv->irq_mask);
+	I915_WRITE(VLV_IER, enable_mask);
 
-	GEN8_IRQ_FINI_NDX(GT, 0);
-	GEN8_IRQ_FINI_NDX(GT, 1);
-	GEN8_IRQ_FINI_NDX(GT, 2);
-	GEN8_IRQ_FINI_NDX(GT, 3);
+	gen8_gt_irq_postinstall(dev_priv);
 
-	for_each_pipe(pipe) {
-		GEN8_IRQ_FINI_NDX(DE_PIPE, pipe);
-	}
+	I915_WRITE(GEN8_MASTER_IRQ, MASTER_INTERRUPT_ENABLE);
+	POSTING_READ(GEN8_MASTER_IRQ);
 
-	GEN8_IRQ_FINI(DE_PORT);
-	GEN8_IRQ_FINI(DE_MISC);
-	GEN8_IRQ_FINI(PCU);
-#undef GEN8_IRQ_FINI
-#undef GEN8_IRQ_FINI_NDX
+	return 0;
+}
 
-	POSTING_READ(GEN8_PCU_IIR);
+static void gen8_irq_uninstall(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	if (!dev_priv)
+		return;
+
+	intel_hpd_irq_uninstall(dev_priv);
+
+	gen8_irq_reset(dev);
 }
 
 static void valleyview_irq_uninstall(struct drm_device *dev)
@@ -3278,6 +3568,8 @@ static void valleyview_irq_uninstall(struct drm_device *dev)
 	if (!dev_priv)
 		return;
 
+	I915_WRITE(VLV_MASTER_IER, 0);
+
 	intel_hpd_irq_uninstall(dev_priv);
 
 	for_each_pipe(pipe)
@@ -3300,35 +3592,67 @@ static void valleyview_irq_uninstall(struct drm_device *dev)
 	POSTING_READ(VLV_IER);
 }
 
-static void ironlake_irq_uninstall(struct drm_device *dev)
+static void cherryview_irq_uninstall(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	int pipe;
 
 	if (!dev_priv)
 		return;
 
-	intel_hpd_irq_uninstall(dev_priv);
+	I915_WRITE(GEN8_MASTER_IRQ, 0);
+	POSTING_READ(GEN8_MASTER_IRQ);
 
-	I915_WRITE(HWSTAM, 0xffffffff);
+#define GEN8_IRQ_FINI_NDX(type, which)				\
+do {								\
+	I915_WRITE(GEN8_##type##_IMR(which), 0xffffffff);	\
+	I915_WRITE(GEN8_##type##_IER(which), 0);		\
+	I915_WRITE(GEN8_##type##_IIR(which), 0xffffffff);	\
+	POSTING_READ(GEN8_##type##_IIR(which));			\
+	I915_WRITE(GEN8_##type##_IIR(which), 0xffffffff);	\
+} while (0)
+
+#define GEN8_IRQ_FINI(type)				\
+do {							\
+	I915_WRITE(GEN8_##type##_IMR, 0xffffffff);	\
+	I915_WRITE(GEN8_##type##_IER, 0);		\
+	I915_WRITE(GEN8_##type##_IIR, 0xffffffff);	\
+	POSTING_READ(GEN8_##type##_IIR);		\
+	I915_WRITE(GEN8_##type##_IIR, 0xffffffff);	\
+} while (0)
 
-	I915_WRITE(DEIMR, 0xffffffff);
-	I915_WRITE(DEIER, 0x0);
-	I915_WRITE(DEIIR, I915_READ(DEIIR));
-	if (IS_GEN7(dev))
-		I915_WRITE(GEN7_ERR_INT, I915_READ(GEN7_ERR_INT));
+	GEN8_IRQ_FINI_NDX(GT, 0);
+	GEN8_IRQ_FINI_NDX(GT, 1);
+	GEN8_IRQ_FINI_NDX(GT, 2);
+	GEN8_IRQ_FINI_NDX(GT, 3);
 
-	I915_WRITE(GTIMR, 0xffffffff);
-	I915_WRITE(GTIER, 0x0);
-	I915_WRITE(GTIIR, I915_READ(GTIIR));
+	GEN8_IRQ_FINI(PCU);
 
-	if (HAS_PCH_NOP(dev))
+#undef GEN8_IRQ_FINI
+#undef GEN8_IRQ_FINI_NDX
+
+	I915_WRITE(PORT_HOTPLUG_EN, 0);
+	I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT));
+
+	for_each_pipe(pipe)
+		I915_WRITE(PIPESTAT(pipe), 0xffff);
+
+	I915_WRITE(VLV_IMR, 0xffffffff);
+	I915_WRITE(VLV_IER, 0x0);
+	I915_WRITE(VLV_IIR, 0xffffffff);
+	POSTING_READ(VLV_IIR);
+}
+
+static void ironlake_irq_uninstall(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	if (!dev_priv)
 		return;
 
-	I915_WRITE(SDEIMR, 0xffffffff);
-	I915_WRITE(SDEIER, 0x0);
-	I915_WRITE(SDEIIR, I915_READ(SDEIIR));
-	if (HAS_PCH_CPT(dev) || HAS_PCH_LPT(dev))
-		I915_WRITE(SERR_INT, I915_READ(SERR_INT));
+	intel_hpd_irq_uninstall(dev_priv);
+
+	ironlake_irq_reset(dev);
 }
 
 static void i8xx_irq_preinstall(struct drm_device * dev)
@@ -3386,7 +3710,7 @@ static bool i8xx_handle_vblank(struct drm_device *dev,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u16 flip_pending = DISPLAY_PLANE_FLIP_PENDING(plane);
 
-	if (!drm_handle_vblank(dev, pipe))
+	if (!intel_pipe_handle_vblank(dev, pipe))
 		return false;
 
 	if ((iir & flip_pending) == 0)
@@ -3410,7 +3734,7 @@ static bool i8xx_handle_vblank(struct drm_device *dev,
 
 static irqreturn_t i8xx_irq_handler(int irq, void *arg)
 {
-	struct drm_device *dev = (struct drm_device *) arg;
+	struct drm_device *dev = arg;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u16 iir, new_iir;
 	u32 pipe_stats[2];
@@ -3571,7 +3895,7 @@ static bool i915_handle_vblank(struct drm_device *dev,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 flip_pending = DISPLAY_PLANE_FLIP_PENDING(plane);
 
-	if (!drm_handle_vblank(dev, pipe))
+	if (!intel_pipe_handle_vblank(dev, pipe))
 		return false;
 
 	if ((iir & flip_pending) == 0)
@@ -3595,7 +3919,7 @@ static bool i915_handle_vblank(struct drm_device *dev,
 
 static irqreturn_t i915_irq_handler(int irq, void *arg)
 {
-	struct drm_device *dev = (struct drm_device *) arg;
+	struct drm_device *dev = arg;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 iir, new_iir, pipe_stats[I915_MAX_PIPES];
 	unsigned long irqflags;
@@ -3636,16 +3960,9 @@ static irqreturn_t i915_irq_handler(int irq, void *arg)
 			break;
 
 		/* Consume port.  Then clear IIR or we'll miss events */
-		if ((I915_HAS_HOTPLUG(dev)) &&
-		    (iir & I915_DISPLAY_PORT_INTERRUPT)) {
-			u32 hotplug_status = I915_READ(PORT_HOTPLUG_STAT);
-			u32 hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_I915;
-
-			intel_hpd_irq_handler(dev, hotplug_trigger, hpd_status_i915);
-
-			I915_WRITE(PORT_HOTPLUG_STAT, hotplug_status);
-			POSTING_READ(PORT_HOTPLUG_STAT);
-		}
+		if (I915_HAS_HOTPLUG(dev) &&
+		    iir & I915_DISPLAY_PORT_INTERRUPT)
+			i9xx_hpd_irq_handler(dev);
 
 		I915_WRITE(IIR, iir & ~flip_mask);
 		new_iir = I915_READ(IIR); /* Flush posted writes */
@@ -3832,7 +4149,7 @@ static void i915_hpd_irq_setup(struct drm_device *dev)
 
 static irqreturn_t i965_irq_handler(int irq, void *arg)
 {
-	struct drm_device *dev = (struct drm_device *) arg;
+	struct drm_device *dev = arg;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 iir, new_iir;
 	u32 pipe_stats[I915_MAX_PIPES];
@@ -3879,22 +4196,8 @@ static irqreturn_t i965_irq_handler(int irq, void *arg)
 		ret = IRQ_HANDLED;
 
 		/* Consume port.  Then clear IIR or we'll miss events */
-		if (iir & I915_DISPLAY_PORT_INTERRUPT) {
-			u32 hotplug_status = I915_READ(PORT_HOTPLUG_STAT);
-			u32 hotplug_trigger = hotplug_status & (IS_G4X(dev) ?
-								  HOTPLUG_INT_STATUS_G4X :
-								  HOTPLUG_INT_STATUS_I915);
-
-			intel_hpd_irq_handler(dev, hotplug_trigger,
-					      IS_G4X(dev) ? hpd_status_g4x : hpd_status_i915);
-
-			if (IS_G4X(dev) &&
-			    (hotplug_status & DP_AUX_CHANNEL_MASK_INT_STATUS_G4X))
-				dp_aux_irq_handler(dev);
-
-			I915_WRITE(PORT_HOTPLUG_STAT, hotplug_status);
-			I915_READ(PORT_HOTPLUG_STAT);
-		}
+		if (iir & I915_DISPLAY_PORT_INTERRUPT)
+			i9xx_hpd_irq_handler(dev);
 
 		I915_WRITE(IIR, iir & ~flip_mask);
 		new_iir = I915_READ(IIR); /* Flush posted writes */
@@ -3997,7 +4300,7 @@ static void intel_hpd_irq_reenable(unsigned long data)
 			if (intel_connector->encoder->hpd_pin == i) {
 				if (connector->polled != intel_connector->polled)
 					DRM_DEBUG_DRIVER("Reenabling HPD on connector %s\n",
-							 drm_get_connector_name(connector));
+							 connector->name);
 				connector->polled = intel_connector->polled;
 				if (!connector->polled)
 					connector->polled = DRM_CONNECTOR_POLL_HPD;
@@ -4045,7 +4348,15 @@ void intel_irq_init(struct drm_device *dev)
 		dev->driver->get_scanout_position = i915_get_crtc_scanoutpos;
 	}
 
-	if (IS_VALLEYVIEW(dev)) {
+	if (IS_CHERRYVIEW(dev)) {
+		dev->driver->irq_handler = cherryview_irq_handler;
+		dev->driver->irq_preinstall = cherryview_irq_preinstall;
+		dev->driver->irq_postinstall = cherryview_irq_postinstall;
+		dev->driver->irq_uninstall = cherryview_irq_uninstall;
+		dev->driver->enable_vblank = valleyview_enable_vblank;
+		dev->driver->disable_vblank = valleyview_disable_vblank;
+		dev_priv->display.hpd_irq_setup = i915_hpd_irq_setup;
+	} else if (IS_VALLEYVIEW(dev)) {
 		dev->driver->irq_handler = valleyview_irq_handler;
 		dev->driver->irq_preinstall = valleyview_irq_preinstall;
 		dev->driver->irq_postinstall = valleyview_irq_postinstall;
@@ -4055,7 +4366,7 @@ void intel_irq_init(struct drm_device *dev)
 		dev_priv->display.hpd_irq_setup = i915_hpd_irq_setup;
 	} else if (IS_GEN8(dev)) {
 		dev->driver->irq_handler = gen8_irq_handler;
-		dev->driver->irq_preinstall = gen8_irq_preinstall;
+		dev->driver->irq_preinstall = gen8_irq_reset;
 		dev->driver->irq_postinstall = gen8_irq_postinstall;
 		dev->driver->irq_uninstall = gen8_irq_uninstall;
 		dev->driver->enable_vblank = gen8_enable_vblank;
@@ -4063,7 +4374,7 @@ void intel_irq_init(struct drm_device *dev)
 		dev_priv->display.hpd_irq_setup = ibx_hpd_irq_setup;
 	} else if (HAS_PCH_SPLIT(dev)) {
 		dev->driver->irq_handler = ironlake_irq_handler;
-		dev->driver->irq_preinstall = ironlake_irq_preinstall;
+		dev->driver->irq_preinstall = ironlake_irq_reset;
 		dev->driver->irq_postinstall = ironlake_irq_postinstall;
 		dev->driver->irq_uninstall = ironlake_irq_uninstall;
 		dev->driver->enable_vblank = ironlake_enable_vblank;
@@ -4121,57 +4432,20 @@ void intel_hpd_init(struct drm_device *dev)
 }
 
 /* Disable interrupts so we can allow runtime PM. */
-void hsw_runtime_pm_disable_interrupts(struct drm_device *dev)
+void intel_runtime_pm_disable_interrupts(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	unsigned long irqflags;
-
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-
-	dev_priv->pm.regsave.deimr = I915_READ(DEIMR);
-	dev_priv->pm.regsave.sdeimr = I915_READ(SDEIMR);
-	dev_priv->pm.regsave.gtimr = I915_READ(GTIMR);
-	dev_priv->pm.regsave.gtier = I915_READ(GTIER);
-	dev_priv->pm.regsave.gen6_pmimr = I915_READ(GEN6_PMIMR);
-
-	ironlake_disable_display_irq(dev_priv, 0xffffffff);
-	ibx_disable_display_interrupt(dev_priv, 0xffffffff);
-	ilk_disable_gt_irq(dev_priv, 0xffffffff);
-	snb_disable_pm_irq(dev_priv, 0xffffffff);
 
+	dev->driver->irq_uninstall(dev);
 	dev_priv->pm.irqs_disabled = true;
-
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 }
 
 /* Restore interrupts so we can recover from runtime PM. */
-void hsw_runtime_pm_restore_interrupts(struct drm_device *dev)
+void intel_runtime_pm_restore_interrupts(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	unsigned long irqflags;
-	uint32_t val;
-
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-
-	val = I915_READ(DEIMR);
-	WARN(val != 0xffffffff, "DEIMR is 0x%08x\n", val);
-
-	val = I915_READ(SDEIMR);
-	WARN(val != 0xffffffff, "SDEIMR is 0x%08x\n", val);
-
-	val = I915_READ(GTIMR);
-	WARN(val != 0xffffffff, "GTIMR is 0x%08x\n", val);
-
-	val = I915_READ(GEN6_PMIMR);
-	WARN(val != 0xffffffff, "GEN6_PMIMR is 0x%08x\n", val);
 
 	dev_priv->pm.irqs_disabled = false;
-
-	ironlake_enable_display_irq(dev_priv, ~dev_priv->pm.regsave.deimr);
-	ibx_enable_display_interrupt(dev_priv, ~dev_priv->pm.regsave.sdeimr);
-	ilk_enable_gt_irq(dev_priv, ~dev_priv->pm.regsave.gtimr);
-	snb_enable_pm_irq(dev_priv, ~dev_priv->pm.regsave.gen6_pmimr);
-	I915_WRITE(GTIER, dev_priv->pm.regsave.gtier);
-
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+	dev->driver->irq_preinstall(dev);
+	dev->driver->irq_postinstall(dev);
 }
diff --git a/drivers/gpu/drm/i915/i915_params.c b/drivers/gpu/drm/i915/i915_params.c
index d1d7980f0e01..d05a2afa17dc 100644
--- a/drivers/gpu/drm/i915/i915_params.c
+++ b/drivers/gpu/drm/i915/i915_params.c
@@ -46,7 +46,8 @@ struct i915_params i915 __read_mostly = {
 	.reset = true,
 	.invert_brightness = 0,
 	.disable_display = 0,
-	.enable_cmd_parser = 0,
+	.enable_cmd_parser = 1,
+	.disable_vtd_wa = 0,
 };
 
 module_param_named(modeset, i915.modeset, int, 0400);
@@ -149,6 +150,9 @@ MODULE_PARM_DESC(invert_brightness,
 module_param_named(disable_display, i915.disable_display, bool, 0600);
 MODULE_PARM_DESC(disable_display, "Disable display (default: false)");
 
+module_param_named(disable_vtd_wa, i915.disable_vtd_wa, bool, 0600);
+MODULE_PARM_DESC(disable_vtd_wa, "Disable all VT-d workarounds (default: false)");
+
 module_param_named(enable_cmd_parser, i915.enable_cmd_parser, int, 0600);
 MODULE_PARM_DESC(enable_cmd_parser,
-		 "Enable command parsing (1=enabled, 0=disabled [default])");
+		 "Enable command parsing (1=enabled [default], 0=disabled)");
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index c77af69c2d8f..e691b30b2817 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -29,6 +29,8 @@
 #define _TRANSCODER(tran, a, b) ((a) + (tran)*((b)-(a)))
 
 #define _PORT(port, a, b) ((a) + (port)*((b)-(a)))
+#define _PIPE3(pipe, a, b, c) (pipe < 2 ? _PIPE(pipe, a, b) : c)
+#define _PORT3(port, a, b, c) (port < 2 ? _PORT(port, a, b) : c)
 
 #define _MASKED_BIT_ENABLE(a) (((a) << 16) | (a))
 #define _MASKED_BIT_DISABLE(a) ((a) << 16)
@@ -77,13 +79,19 @@
 
 /* Graphics reset regs */
 #define I965_GDRST 0xc0 /* PCI config register */
-#define ILK_GDSR 0x2ca4 /* MCHBAR offset */
 #define  GRDOM_FULL	(0<<2)
 #define  GRDOM_RENDER	(1<<2)
 #define  GRDOM_MEDIA	(3<<2)
 #define  GRDOM_MASK	(3<<2)
 #define  GRDOM_RESET_ENABLE (1<<0)
 
+#define ILK_GDSR 0x2ca4 /* MCHBAR offset */
+#define  ILK_GRDOM_FULL		(0<<1)
+#define  ILK_GRDOM_RENDER	(1<<1)
+#define  ILK_GRDOM_MEDIA	(3<<1)
+#define  ILK_GRDOM_MASK		(3<<1)
+#define  ILK_GRDOM_RESET_ENABLE (1<<0)
+
 #define GEN6_MBCUNIT_SNPCR	0x900c /* for LLC config */
 #define   GEN6_MBC_SNPCR_SHIFT	21
 #define   GEN6_MBC_SNPCR_MASK	(3<<21)
@@ -92,6 +100,9 @@
 #define   GEN6_MBC_SNPCR_LOW	(2<<21)
 #define   GEN6_MBC_SNPCR_MIN	(3<<21) /* only 1/16th of the cache is shared */
 
+#define VLV_G3DCTL		0x9024
+#define VLV_GSCKGCTL		0x9028
+
 #define GEN6_MBCTL		0x0907c
 #define   GEN6_MBCTL_ENABLE_BOOT_FETCH	(1 << 4)
 #define   GEN6_MBCTL_CTX_FETCH_NEEDED	(1 << 3)
@@ -190,6 +201,8 @@
  * Memory interface instructions used by the kernel
  */
 #define MI_INSTR(opcode, flags) (((opcode) << 23) | (flags))
+/* Many MI commands use bit 22 of the header dword for GGTT vs PPGTT */
+#define  MI_GLOBAL_GTT    (1<<22)
 
 #define MI_NOOP			MI_INSTR(0, 0)
 #define MI_USER_INTERRUPT	MI_INSTR(0x02, 0)
@@ -244,7 +257,8 @@
 #define   MI_SEMAPHORE_SYNC_BVE	    (0<<16) /* VECS wait for BCS  (VEBSYNC) */
 #define   MI_SEMAPHORE_SYNC_VVE	    (1<<16) /* VECS wait for VCS  (VEVSYNC) */
 #define   MI_SEMAPHORE_SYNC_RVE	    (2<<16) /* VECS wait for RCS  (VERSYNC) */
-#define   MI_SEMAPHORE_SYNC_INVALID  (3<<16)
+#define   MI_SEMAPHORE_SYNC_INVALID (3<<16)
+#define   MI_SEMAPHORE_SYNC_MASK    (3<<16)
 #define MI_SET_CONTEXT		MI_INSTR(0x18, 0)
 #define   MI_MM_SPACE_GTT		(1<<8)
 #define   MI_MM_SPACE_PHYSICAL		(0<<8)
@@ -262,13 +276,16 @@
  * - One can actually load arbitrary many arbitrary registers: Simply issue x
  *   address/value pairs. Don't overdue it, though, x <= 2^4 must hold!
  */
-#define MI_LOAD_REGISTER_IMM(x)	MI_INSTR(0x22, 2*x-1)
-#define MI_STORE_REGISTER_MEM(x) MI_INSTR(0x24, 2*x-1)
+#define MI_LOAD_REGISTER_IMM(x)	MI_INSTR(0x22, 2*(x)-1)
+#define MI_STORE_REGISTER_MEM(x) MI_INSTR(0x24, 2*(x)-1)
+#define MI_STORE_REGISTER_MEM_GEN8(x) MI_INSTR(0x24, 3*(x)-1)
 #define   MI_SRM_LRM_GLOBAL_GTT		(1<<22)
 #define MI_FLUSH_DW		MI_INSTR(0x26, 1) /* for GEN6 */
 #define   MI_FLUSH_DW_STORE_INDEX	(1<<21)
 #define   MI_INVALIDATE_TLB		(1<<18)
 #define   MI_FLUSH_DW_OP_STOREDW	(1<<14)
+#define   MI_FLUSH_DW_OP_MASK		(3<<14)
+#define   MI_FLUSH_DW_NOTIFY		(1<<8)
 #define   MI_INVALIDATE_BSD		(1<<7)
 #define   MI_FLUSH_DW_USE_GTT		(1<<2)
 #define   MI_FLUSH_DW_USE_PPGTT		(0<<2)
@@ -330,9 +347,12 @@
 #define   DISPLAY_PLANE_B           (1<<20)
 #define GFX_OP_PIPE_CONTROL(len)	((0x3<<29)|(0x3<<27)|(0x2<<24)|(len-2))
 #define   PIPE_CONTROL_GLOBAL_GTT_IVB			(1<<24) /* gen7+ */
+#define   PIPE_CONTROL_MMIO_WRITE			(1<<23)
+#define   PIPE_CONTROL_STORE_DATA_INDEX			(1<<21)
 #define   PIPE_CONTROL_CS_STALL				(1<<20)
 #define   PIPE_CONTROL_TLB_INVALIDATE			(1<<18)
 #define   PIPE_CONTROL_QW_WRITE				(1<<14)
+#define   PIPE_CONTROL_POST_SYNC_OP_MASK                (3<<14)
 #define   PIPE_CONTROL_DEPTH_STALL			(1<<13)
 #define   PIPE_CONTROL_WRITE_FLUSH			(1<<12)
 #define   PIPE_CONTROL_RENDER_TARGET_CACHE_FLUSH	(1<<12) /* gen6+ */
@@ -347,6 +367,94 @@
 #define   PIPE_CONTROL_DEPTH_CACHE_FLUSH		(1<<0)
 #define   PIPE_CONTROL_GLOBAL_GTT (1<<2) /* in addr dword */
 
+/*
+ * Commands used only by the command parser
+ */
+#define MI_SET_PREDICATE        MI_INSTR(0x01, 0)
+#define MI_ARB_CHECK            MI_INSTR(0x05, 0)
+#define MI_RS_CONTROL           MI_INSTR(0x06, 0)
+#define MI_URB_ATOMIC_ALLOC     MI_INSTR(0x09, 0)
+#define MI_PREDICATE            MI_INSTR(0x0C, 0)
+#define MI_RS_CONTEXT           MI_INSTR(0x0F, 0)
+#define MI_TOPOLOGY_FILTER      MI_INSTR(0x0D, 0)
+#define MI_LOAD_SCAN_LINES_EXCL MI_INSTR(0x13, 0)
+#define MI_URB_CLEAR            MI_INSTR(0x19, 0)
+#define MI_UPDATE_GTT           MI_INSTR(0x23, 0)
+#define MI_CLFLUSH              MI_INSTR(0x27, 0)
+#define MI_REPORT_PERF_COUNT    MI_INSTR(0x28, 0)
+#define   MI_REPORT_PERF_COUNT_GGTT (1<<0)
+#define MI_LOAD_REGISTER_MEM    MI_INSTR(0x29, 0)
+#define MI_LOAD_REGISTER_REG    MI_INSTR(0x2A, 0)
+#define MI_RS_STORE_DATA_IMM    MI_INSTR(0x2B, 0)
+#define MI_LOAD_URB_MEM         MI_INSTR(0x2C, 0)
+#define MI_STORE_URB_MEM        MI_INSTR(0x2D, 0)
+#define MI_CONDITIONAL_BATCH_BUFFER_END MI_INSTR(0x36, 0)
+
+#define PIPELINE_SELECT                ((0x3<<29)|(0x1<<27)|(0x1<<24)|(0x4<<16))
+#define GFX_OP_3DSTATE_VF_STATISTICS   ((0x3<<29)|(0x1<<27)|(0x0<<24)|(0xB<<16))
+#define MEDIA_VFE_STATE                ((0x3<<29)|(0x2<<27)|(0x0<<24)|(0x0<<16))
+#define  MEDIA_VFE_STATE_MMIO_ACCESS_MASK (0x18)
+#define GPGPU_OBJECT                   ((0x3<<29)|(0x2<<27)|(0x1<<24)|(0x4<<16))
+#define GPGPU_WALKER                   ((0x3<<29)|(0x2<<27)|(0x1<<24)|(0x5<<16))
+#define GFX_OP_3DSTATE_DX9_CONSTANTF_VS \
+	((0x3<<29)|(0x3<<27)|(0x0<<24)|(0x39<<16))
+#define GFX_OP_3DSTATE_DX9_CONSTANTF_PS \
+	((0x3<<29)|(0x3<<27)|(0x0<<24)|(0x3A<<16))
+#define GFX_OP_3DSTATE_SO_DECL_LIST \
+	((0x3<<29)|(0x3<<27)|(0x1<<24)|(0x17<<16))
+
+#define GFX_OP_3DSTATE_BINDING_TABLE_EDIT_VS \
+	((0x3<<29)|(0x3<<27)|(0x0<<24)|(0x43<<16))
+#define GFX_OP_3DSTATE_BINDING_TABLE_EDIT_GS \
+	((0x3<<29)|(0x3<<27)|(0x0<<24)|(0x44<<16))
+#define GFX_OP_3DSTATE_BINDING_TABLE_EDIT_HS \
+	((0x3<<29)|(0x3<<27)|(0x0<<24)|(0x45<<16))
+#define GFX_OP_3DSTATE_BINDING_TABLE_EDIT_DS \
+	((0x3<<29)|(0x3<<27)|(0x0<<24)|(0x46<<16))
+#define GFX_OP_3DSTATE_BINDING_TABLE_EDIT_PS \
+	((0x3<<29)|(0x3<<27)|(0x0<<24)|(0x47<<16))
+
+#define MFX_WAIT  ((0x3<<29)|(0x1<<27)|(0x0<<16))
+
+#define COLOR_BLT     ((0x2<<29)|(0x40<<22))
+#define SRC_COPY_BLT  ((0x2<<29)|(0x43<<22))
+
+/*
+ * Registers used only by the command parser
+ */
+#define BCS_SWCTRL 0x22200
+
+#define HS_INVOCATION_COUNT 0x2300
+#define DS_INVOCATION_COUNT 0x2308
+#define IA_VERTICES_COUNT   0x2310
+#define IA_PRIMITIVES_COUNT 0x2318
+#define VS_INVOCATION_COUNT 0x2320
+#define GS_INVOCATION_COUNT 0x2328
+#define GS_PRIMITIVES_COUNT 0x2330
+#define CL_INVOCATION_COUNT 0x2338
+#define CL_PRIMITIVES_COUNT 0x2340
+#define PS_INVOCATION_COUNT 0x2348
+#define PS_DEPTH_COUNT      0x2350
+
+/* There are the 4 64-bit counter registers, one for each stream output */
+#define GEN7_SO_NUM_PRIMS_WRITTEN(n) (0x5200 + (n) * 8)
+
+#define GEN7_SO_PRIM_STORAGE_NEEDED(n)  (0x5240 + (n) * 8)
+
+#define GEN7_3DPRIM_END_OFFSET          0x2420
+#define GEN7_3DPRIM_START_VERTEX        0x2430
+#define GEN7_3DPRIM_VERTEX_COUNT        0x2434
+#define GEN7_3DPRIM_INSTANCE_COUNT      0x2438
+#define GEN7_3DPRIM_START_INSTANCE      0x243C
+#define GEN7_3DPRIM_BASE_VERTEX         0x2440
+
+#define OACONTROL 0x2360
+
+#define _GEN7_PIPEA_DE_LOAD_SL	0x70068
+#define _GEN7_PIPEB_DE_LOAD_SL	0x71068
+#define GEN7_PIPE_DE_LOAD_SL(pipe) _PIPE(pipe, \
+					 _GEN7_PIPEA_DE_LOAD_SL, \
+					 _GEN7_PIPEB_DE_LOAD_SL)
 
 /*
  * Reset registers
@@ -370,6 +478,7 @@
 #define   IOSF_PORT_PUNIT			0x4
 #define   IOSF_PORT_NC				0x11
 #define   IOSF_PORT_DPIO			0x12
+#define   IOSF_PORT_DPIO_2			0x1a
 #define   IOSF_PORT_GPIO_NC			0x13
 #define   IOSF_PORT_CCK				0x14
 #define   IOSF_PORT_CCU				0xA9
@@ -381,9 +490,6 @@
 /* See configdb bunit SB addr map */
 #define BUNIT_REG_BISOC				0x11
 
-#define PUNIT_OPCODE_REG_READ			6
-#define PUNIT_OPCODE_REG_WRITE			7
-
 #define PUNIT_REG_DSPFREQ			0x36
 #define   DSPFREQSTAT_SHIFT			30
 #define   DSPFREQSTAT_MASK			(0x3 << DSPFREQSTAT_SHIFT)
@@ -469,16 +575,91 @@ enum punit_power_well {
 #define  DSI_PLL_M1_DIV_MASK			(0x1ff << 0)
 #define CCK_DISPLAY_CLOCK_CONTROL		0x6b
 
-/*
- * DPIO - a special bus for various display related registers to hide behind
+/**
+ * DOC: DPIO
+ *
+ * VLV and CHV have slightly peculiar display PHYs for driving DP/HDMI
+ * ports. DPIO is the name given to such a display PHY. These PHYs
+ * don't follow the standard programming model using direct MMIO
+ * registers, and instead their registers must be accessed trough IOSF
+ * sideband. VLV has one such PHY for driving ports B and C, and CHV
+ * adds another PHY for driving port D. Each PHY responds to specific
+ * IOSF-SB port.
+ *
+ * Each display PHY is made up of one or two channels. Each channel
+ * houses a common lane part which contains the PLL and other common
+ * logic. CH0 common lane also contains the IOSF-SB logic for the
+ * Common Register Interface (CRI) ie. the DPIO registers. CRI clock
+ * must be running when any DPIO registers are accessed.
+ *
+ * In addition to having their own registers, the PHYs are also
+ * controlled through some dedicated signals from the display
+ * controller. These include PLL reference clock enable, PLL enable,
+ * and CRI clock selection, for example.
  *
- * DPIO is VLV only.
+ * Eeach channel also has two splines (also called data lanes), and
+ * each spline is made up of one Physical Access Coding Sub-Layer
+ * (PCS) block and two TX lanes. So each channel has two PCS blocks
+ * and four TX lanes. The TX lanes are used as DP lanes or TMDS
+ * data/clock pairs depending on the output type.
+ *
+ * Additionally the PHY also contains an AUX lane with AUX blocks
+ * for each channel. This is used for DP AUX communication, but
+ * this fact isn't really relevant for the driver since AUX is
+ * controlled from the display controller side. No DPIO registers
+ * need to be accessed during AUX communication,
+ *
+ * Generally the common lane corresponds to the pipe and
+ * the spline (PCS/TX) correponds to the port.
+ *
+ * For dual channel PHY (VLV/CHV):
+ *
+ *  pipe A == CMN/PLL/REF CH0
+ *
+ *  pipe B == CMN/PLL/REF CH1
+ *
+ *  port B == PCS/TX CH0
+ *
+ *  port C == PCS/TX CH1
+ *
+ * This is especially important when we cross the streams
+ * ie. drive port B with pipe B, or port C with pipe A.
+ *
+ * For single channel PHY (CHV):
+ *
+ *  pipe C == CMN/PLL/REF CH0
+ *
+ *  port D == PCS/TX CH0
+ *
+ * Note: digital port B is DDI0, digital port C is DDI1,
+ * digital port D is DDI2
+ */
+/*
+ * Dual channel PHY (VLV/CHV)
+ * ---------------------------------
+ * |      CH0      |      CH1      |
+ * |  CMN/PLL/REF  |  CMN/PLL/REF  |
+ * |---------------|---------------| Display PHY
+ * | PCS01 | PCS23 | PCS01 | PCS23 |
+ * |-------|-------|-------|-------|
+ * |TX0|TX1|TX2|TX3|TX0|TX1|TX2|TX3|
+ * ---------------------------------
+ * |     DDI0      |     DDI1      | DP/HDMI ports
+ * ---------------------------------
  *
- * Note: digital port B is DDI0, digital pot C is DDI1
+ * Single channel PHY (CHV)
+ * -----------------
+ * |      CH0      |
+ * |  CMN/PLL/REF  |
+ * |---------------| Display PHY
+ * | PCS01 | PCS23 |
+ * |-------|-------|
+ * |TX0|TX1|TX2|TX3|
+ * -----------------
+ * |     DDI2      | DP/HDMI port
+ * -----------------
  */
 #define DPIO_DEVFN			0
-#define DPIO_OPCODE_REG_WRITE		1
-#define DPIO_OPCODE_REG_READ		0
 
 #define DPIO_CTL			(VLV_DISPLAY_BASE + 0x2110)
 #define  DPIO_MODSEL1			(1<<3) /* if ref clk b == 27 */
@@ -555,14 +736,29 @@ enum punit_power_well {
 #define   DPIO_PCS_TX_LANE1_RESET	(1<<7)
 #define VLV_PCS_DW0(ch) _PORT(ch, _VLV_PCS_DW0_CH0, _VLV_PCS_DW0_CH1)
 
+#define _VLV_PCS01_DW0_CH0		0x200
+#define _VLV_PCS23_DW0_CH0		0x400
+#define _VLV_PCS01_DW0_CH1		0x2600
+#define _VLV_PCS23_DW0_CH1		0x2800
+#define VLV_PCS01_DW0(ch) _PORT(ch, _VLV_PCS01_DW0_CH0, _VLV_PCS01_DW0_CH1)
+#define VLV_PCS23_DW0(ch) _PORT(ch, _VLV_PCS23_DW0_CH0, _VLV_PCS23_DW0_CH1)
+
 #define _VLV_PCS_DW1_CH0		0x8204
 #define _VLV_PCS_DW1_CH1		0x8404
+#define   CHV_PCS_REQ_SOFTRESET_EN	(1<<23)
 #define   DPIO_PCS_CLK_CRI_RXEB_EIOS_EN	(1<<22)
 #define   DPIO_PCS_CLK_CRI_RXDIGFILTSG_EN (1<<21)
 #define   DPIO_PCS_CLK_DATAWIDTH_SHIFT	(6)
 #define   DPIO_PCS_CLK_SOFT_RESET	(1<<5)
 #define VLV_PCS_DW1(ch) _PORT(ch, _VLV_PCS_DW1_CH0, _VLV_PCS_DW1_CH1)
 
+#define _VLV_PCS01_DW1_CH0		0x204
+#define _VLV_PCS23_DW1_CH0		0x404
+#define _VLV_PCS01_DW1_CH1		0x2604
+#define _VLV_PCS23_DW1_CH1		0x2804
+#define VLV_PCS01_DW1(ch) _PORT(ch, _VLV_PCS01_DW1_CH0, _VLV_PCS01_DW1_CH1)
+#define VLV_PCS23_DW1(ch) _PORT(ch, _VLV_PCS23_DW1_CH0, _VLV_PCS23_DW1_CH1)
+
 #define _VLV_PCS_DW8_CH0		0x8220
 #define _VLV_PCS_DW8_CH1		0x8420
 #define VLV_PCS_DW8(ch) _PORT(ch, _VLV_PCS_DW8_CH0, _VLV_PCS_DW8_CH1)
@@ -578,6 +774,19 @@ enum punit_power_well {
 #define _VLV_PCS_DW9_CH1		0x8424
 #define	VLV_PCS_DW9(ch) _PORT(ch, _VLV_PCS_DW9_CH0, _VLV_PCS_DW9_CH1)
 
+#define _CHV_PCS_DW10_CH0		0x8228
+#define _CHV_PCS_DW10_CH1		0x8428
+#define   DPIO_PCS_SWING_CALC_TX0_TX2	(1<<30)
+#define   DPIO_PCS_SWING_CALC_TX1_TX3	(1<<31)
+#define CHV_PCS_DW10(ch) _PORT(ch, _CHV_PCS_DW10_CH0, _CHV_PCS_DW10_CH1)
+
+#define _VLV_PCS01_DW10_CH0		0x0228
+#define _VLV_PCS23_DW10_CH0		0x0428
+#define _VLV_PCS01_DW10_CH1		0x2628
+#define _VLV_PCS23_DW10_CH1		0x2828
+#define VLV_PCS01_DW10(port) _PORT(port, _VLV_PCS01_DW10_CH0, _VLV_PCS01_DW10_CH1)
+#define VLV_PCS23_DW10(port) _PORT(port, _VLV_PCS23_DW10_CH0, _VLV_PCS23_DW10_CH1)
+
 #define _VLV_PCS_DW11_CH0		0x822c
 #define _VLV_PCS_DW11_CH1		0x842c
 #define VLV_PCS_DW11(ch) _PORT(ch, _VLV_PCS_DW11_CH0, _VLV_PCS_DW11_CH1)
@@ -596,14 +805,21 @@ enum punit_power_well {
 
 #define _VLV_TX_DW2_CH0			0x8288
 #define _VLV_TX_DW2_CH1			0x8488
+#define   DPIO_SWING_MARGIN_SHIFT	16
+#define   DPIO_SWING_MARGIN_MASK	(0xff << DPIO_SWING_MARGIN_SHIFT)
+#define   DPIO_UNIQ_TRANS_SCALE_SHIFT	8
 #define VLV_TX_DW2(ch) _PORT(ch, _VLV_TX_DW2_CH0, _VLV_TX_DW2_CH1)
 
 #define _VLV_TX_DW3_CH0			0x828c
 #define _VLV_TX_DW3_CH1			0x848c
+/* The following bit for CHV phy */
+#define   DPIO_TX_UNIQ_TRANS_SCALE_EN	(1<<27)
 #define VLV_TX_DW3(ch) _PORT(ch, _VLV_TX_DW3_CH0, _VLV_TX_DW3_CH1)
 
 #define _VLV_TX_DW4_CH0			0x8290
 #define _VLV_TX_DW4_CH1			0x8490
+#define   DPIO_SWING_DEEMPH9P5_SHIFT	24
+#define   DPIO_SWING_DEEMPH9P5_MASK	(0xff << DPIO_SWING_DEEMPH9P5_SHIFT)
 #define VLV_TX_DW4(ch) _PORT(ch, _VLV_TX_DW4_CH0, _VLV_TX_DW4_CH1)
 
 #define _VLV_TX3_DW4_CH0		0x690
@@ -623,6 +839,73 @@ enum punit_power_well {
 #define _VLV_TX_DW14_CH1		0x84b8
 #define VLV_TX_DW14(ch) _PORT(ch, _VLV_TX_DW14_CH0, _VLV_TX_DW14_CH1)
 
+/* CHV dpPhy registers */
+#define _CHV_PLL_DW0_CH0		0x8000
+#define _CHV_PLL_DW0_CH1		0x8180
+#define CHV_PLL_DW0(ch) _PIPE(ch, _CHV_PLL_DW0_CH0, _CHV_PLL_DW0_CH1)
+
+#define _CHV_PLL_DW1_CH0		0x8004
+#define _CHV_PLL_DW1_CH1		0x8184
+#define   DPIO_CHV_N_DIV_SHIFT		8
+#define   DPIO_CHV_M1_DIV_BY_2		(0 << 0)
+#define CHV_PLL_DW1(ch) _PIPE(ch, _CHV_PLL_DW1_CH0, _CHV_PLL_DW1_CH1)
+
+#define _CHV_PLL_DW2_CH0		0x8008
+#define _CHV_PLL_DW2_CH1		0x8188
+#define CHV_PLL_DW2(ch) _PIPE(ch, _CHV_PLL_DW2_CH0, _CHV_PLL_DW2_CH1)
+
+#define _CHV_PLL_DW3_CH0		0x800c
+#define _CHV_PLL_DW3_CH1		0x818c
+#define  DPIO_CHV_FRAC_DIV_EN		(1 << 16)
+#define  DPIO_CHV_FIRST_MOD		(0 << 8)
+#define  DPIO_CHV_SECOND_MOD		(1 << 8)
+#define  DPIO_CHV_FEEDFWD_GAIN_SHIFT	0
+#define CHV_PLL_DW3(ch) _PIPE(ch, _CHV_PLL_DW3_CH0, _CHV_PLL_DW3_CH1)
+
+#define _CHV_PLL_DW6_CH0		0x8018
+#define _CHV_PLL_DW6_CH1		0x8198
+#define   DPIO_CHV_GAIN_CTRL_SHIFT	16
+#define	  DPIO_CHV_INT_COEFF_SHIFT	8
+#define   DPIO_CHV_PROP_COEFF_SHIFT	0
+#define CHV_PLL_DW6(ch) _PIPE(ch, _CHV_PLL_DW6_CH0, _CHV_PLL_DW6_CH1)
+
+#define _CHV_CMN_DW13_CH0		0x8134
+#define _CHV_CMN_DW0_CH1		0x8080
+#define   DPIO_CHV_S1_DIV_SHIFT		21
+#define   DPIO_CHV_P1_DIV_SHIFT		13 /* 3 bits */
+#define   DPIO_CHV_P2_DIV_SHIFT		8  /* 5 bits */
+#define   DPIO_CHV_K_DIV_SHIFT		4
+#define   DPIO_PLL_FREQLOCK		(1 << 1)
+#define   DPIO_PLL_LOCK			(1 << 0)
+#define CHV_CMN_DW13(ch) _PIPE(ch, _CHV_CMN_DW13_CH0, _CHV_CMN_DW0_CH1)
+
+#define _CHV_CMN_DW14_CH0		0x8138
+#define _CHV_CMN_DW1_CH1		0x8084
+#define   DPIO_AFC_RECAL		(1 << 14)
+#define   DPIO_DCLKP_EN			(1 << 13)
+#define CHV_CMN_DW14(ch) _PIPE(ch, _CHV_CMN_DW14_CH0, _CHV_CMN_DW1_CH1)
+
+#define CHV_CMN_DW30			0x8178
+#define   DPIO_LRC_BYPASS		(1 << 3)
+
+#define _TXLANE(ch, lane, offset) ((ch ? 0x2400 : 0) + \
+					(lane) * 0x200 + (offset))
+
+#define CHV_TX_DW0(ch, lane) _TXLANE(ch, lane, 0x80)
+#define CHV_TX_DW1(ch, lane) _TXLANE(ch, lane, 0x84)
+#define CHV_TX_DW2(ch, lane) _TXLANE(ch, lane, 0x88)
+#define CHV_TX_DW3(ch, lane) _TXLANE(ch, lane, 0x8c)
+#define CHV_TX_DW4(ch, lane) _TXLANE(ch, lane, 0x90)
+#define CHV_TX_DW5(ch, lane) _TXLANE(ch, lane, 0x94)
+#define CHV_TX_DW6(ch, lane) _TXLANE(ch, lane, 0x98)
+#define CHV_TX_DW7(ch, lane) _TXLANE(ch, lane, 0x9c)
+#define CHV_TX_DW8(ch, lane) _TXLANE(ch, lane, 0xa0)
+#define CHV_TX_DW9(ch, lane) _TXLANE(ch, lane, 0xa4)
+#define CHV_TX_DW10(ch, lane) _TXLANE(ch, lane, 0xa8)
+#define CHV_TX_DW11(ch, lane) _TXLANE(ch, lane, 0xac)
+#define   DPIO_FRC_LATENCY_SHFIT	8
+#define CHV_TX_DW14(ch, lane) _TXLANE(ch, lane, 0xb8)
+#define   DPIO_UPAR_SHIFT		30
 /*
  * Fence registers
  */
@@ -663,6 +946,7 @@ enum punit_power_well {
 #define RENDER_RING_BASE	0x02000
 #define BSD_RING_BASE		0x04000
 #define GEN6_BSD_RING_BASE	0x12000
+#define GEN8_BSD2_RING_BASE	0x1c000
 #define VEBOX_RING_BASE		0x1a000
 #define BLT_RING_BASE		0x22000
 #define RING_TAIL(base)		((base)+0x30)
@@ -688,9 +972,20 @@ enum punit_power_well {
 #define RING_MAX_IDLE(base)	((base)+0x54)
 #define RING_HWS_PGA(base)	((base)+0x80)
 #define RING_HWS_PGA_GEN6(base)	((base)+0x2080)
-#define ARB_MODE		0x04030
+
+#define GEN7_WR_WATERMARK	0x4028
+#define GEN7_GFX_PRIO_CTRL	0x402C
+#define ARB_MODE		0x4030
 #define   ARB_MODE_SWIZZLE_SNB	(1<<4)
 #define   ARB_MODE_SWIZZLE_IVB	(1<<5)
+#define GEN7_GFX_PEND_TLB0	0x4034
+#define GEN7_GFX_PEND_TLB1	0x4038
+/* L3, CVS, ZTLB, RCC, CASC LRA min, max values */
+#define GEN7_LRA_LIMITS_BASE	0x403C
+#define GEN7_LRA_LIMITS_REG_NUM	13
+#define GEN7_MEDIA_MAX_REQ_COUNT	0x4070
+#define GEN7_GFX_MAX_REQ_COUNT		0x4074
+
 #define GAMTARBMODE		0x04a08
 #define   ARB_MODE_BWGTLB_DISABLE (1<<9)
 #define   ARB_MODE_SWIZZLE_BDW	(1<<1)
@@ -725,6 +1020,9 @@ enum punit_power_well {
 #define   RING_WAIT_I8XX	(1<<0) /* gen2, PRBx_HEAD */
 #define   RING_WAIT		(1<<11) /* gen3+, PRBx_CTL */
 #define   RING_WAIT_SEMAPHORE	(1<<10) /* gen6+ */
+
+#define GEN7_TLB_RD_ADDR	0x4700
+
 #if 0
 #define PRB0_TAIL	0x02030
 #define PRB0_HEAD	0x02034
@@ -748,6 +1046,7 @@ enum punit_power_well {
 #define RING_INSTDONE(base)	((base)+0x6c)
 #define RING_INSTPS(base)	((base)+0x70)
 #define RING_DMA_FADD(base)	((base)+0x78)
+#define RING_DMA_FADD_UDW(base)	((base)+0x60) /* gen8+ */
 #define RING_INSTPM(base)	((base)+0xc0)
 #define RING_MI_MODE(base)	((base)+0x9c)
 #define INSTPS		0x02070 /* 965+ only */
@@ -842,21 +1141,26 @@ enum punit_power_well {
 #define GFX_MODE_GEN7	0x0229c
 #define RING_MODE_GEN7(ring)	((ring)->mmio_base+0x29c)
 #define   GFX_RUN_LIST_ENABLE		(1<<15)
-#define   GFX_TLB_INVALIDATE_ALWAYS	(1<<13)
+#define   GFX_TLB_INVALIDATE_EXPLICIT	(1<<13)
 #define   GFX_SURFACE_FAULT_ENABLE	(1<<12)
 #define   GFX_REPLAY_MODE		(1<<11)
 #define   GFX_PSMI_GRANULARITY		(1<<10)
 #define   GFX_PPGTT_ENABLE		(1<<9)
 
 #define VLV_DISPLAY_BASE 0x180000
+#define VLV_MIPI_BASE VLV_DISPLAY_BASE
 
+#define VLV_GU_CTL0	(VLV_DISPLAY_BASE + 0x2030)
+#define VLV_GU_CTL1	(VLV_DISPLAY_BASE + 0x2034)
 #define SCPD0		0x0209c /* 915+ only */
 #define IER		0x020a0
 #define IIR		0x020a4
 #define IMR		0x020a8
 #define ISR		0x020ac
 #define VLV_GUNIT_CLOCK_GATE	(VLV_DISPLAY_BASE + 0x2060)
+#define   GINT_DIS		(1<<22)
 #define   GCFG_DIS		(1<<8)
+#define VLV_GUNIT_CLOCK_GATE2	(VLV_DISPLAY_BASE + 0x2064)
 #define VLV_IIR_RW	(VLV_DISPLAY_BASE + 0x2084)
 #define VLV_IER		(VLV_DISPLAY_BASE + 0x20a0)
 #define VLV_IIR		(VLV_DISPLAY_BASE + 0x20a4)
@@ -875,7 +1179,7 @@ enum punit_power_well {
 #define   I915_ERROR_INSTRUCTION			(1<<0)
 #define INSTPM	        0x020c0
 #define   INSTPM_SELF_EN (1<<12) /* 915GM only */
-#define   INSTPM_AGPBUSY_DIS (1<<11) /* gen3: when disabled, pending interrupts
+#define   INSTPM_AGPBUSY_INT_EN (1<<11) /* gen3: when disabled, pending interrupts
 					will not assert AGPBUSY# and will only
 					be delivered when out of C3. */
 #define   INSTPM_FORCE_ORDERING				(1<<7) /* GEN6+ */
@@ -956,6 +1260,10 @@ enum punit_power_well {
 #define   MI_ARB_DISPLAY_PRIORITY_A_B		(0 << 0)	/* display A > display B */
 #define   MI_ARB_DISPLAY_PRIORITY_B_A		(1 << 0)	/* display B > display A */
 
+#define MI_STATE	0x020e4 /* gen2 only */
+#define   MI_AGPBUSY_INT_EN			(1 << 1) /* 85x only */
+#define   MI_AGPBUSY_830_MODE			(1 << 0) /* 85x only */
+
 #define CACHE_MODE_0	0x02120 /* 915+ only */
 #define   CM0_PIPELINED_RENDER_FLUSH_DISABLE (1<<8)
 #define   CM0_IZ_OPT_DISABLE      (1<<6)
@@ -973,6 +1281,7 @@ enum punit_power_well {
 #define   ECO_FLIP_DONE		(1<<0)
 
 #define CACHE_MODE_0_GEN7	0x7000 /* IVB+ */
+#define RC_OP_FLUSH_ENABLE (1<<0)
 #define   HIZ_RAW_STALL_OPT_DISABLE (1<<2)
 #define CACHE_MODE_1		0x7004 /* IVB+ */
 #define   PIXEL_SUBSPAN_COLLECT_OPT_DISABLE	(1<<6)
@@ -984,6 +1293,7 @@ enum punit_power_well {
 
 #define GEN6_RC_SLEEP_PSMI_CONTROL	0x2050
 #define   GEN8_RC_SEMA_IDLE_MSG_DISABLE	(1 << 12)
+#define   GEN8_FF_DOP_CLOCK_GATE_DISABLE	(1<<10)
 
 #define GEN6_BSD_SLEEP_PSMI_CONTROL	0x12050
 #define   GEN6_BSD_SLEEP_MSG_DISABLE	(1 << 0)
@@ -1024,24 +1334,43 @@ enum punit_power_well {
 
 /* These are all the "old" interrupts */
 #define ILK_BSD_USER_INTERRUPT				(1<<5)
+
+#define I915_PM_INTERRUPT				(1<<31)
+#define I915_ISP_INTERRUPT				(1<<22)
+#define I915_LPE_PIPE_B_INTERRUPT			(1<<21)
+#define I915_LPE_PIPE_A_INTERRUPT			(1<<20)
+#define I915_MIPIB_INTERRUPT				(1<<19)
+#define I915_MIPIA_INTERRUPT				(1<<18)
 #define I915_PIPE_CONTROL_NOTIFY_INTERRUPT		(1<<18)
 #define I915_DISPLAY_PORT_INTERRUPT			(1<<17)
+#define I915_DISPLAY_PIPE_C_HBLANK_INTERRUPT		(1<<16)
+#define I915_MASTER_ERROR_INTERRUPT			(1<<15)
 #define I915_RENDER_COMMAND_PARSER_ERROR_INTERRUPT	(1<<15)
+#define I915_DISPLAY_PIPE_B_HBLANK_INTERRUPT		(1<<14)
 #define I915_GMCH_THERMAL_SENSOR_EVENT_INTERRUPT	(1<<14) /* p-state */
+#define I915_DISPLAY_PIPE_A_HBLANK_INTERRUPT		(1<<13)
 #define I915_HWB_OOM_INTERRUPT				(1<<13)
+#define I915_LPE_PIPE_C_INTERRUPT			(1<<12)
 #define I915_SYNC_STATUS_INTERRUPT			(1<<12)
+#define I915_MISC_INTERRUPT				(1<<11)
 #define I915_DISPLAY_PLANE_A_FLIP_PENDING_INTERRUPT	(1<<11)
+#define I915_DISPLAY_PIPE_C_VBLANK_INTERRUPT		(1<<10)
 #define I915_DISPLAY_PLANE_B_FLIP_PENDING_INTERRUPT	(1<<10)
+#define I915_DISPLAY_PIPE_C_EVENT_INTERRUPT		(1<<9)
 #define I915_OVERLAY_PLANE_FLIP_PENDING_INTERRUPT	(1<<9)
+#define I915_DISPLAY_PIPE_C_DPBM_INTERRUPT		(1<<8)
 #define I915_DISPLAY_PLANE_C_FLIP_PENDING_INTERRUPT	(1<<8)
 #define I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT		(1<<7)
 #define I915_DISPLAY_PIPE_A_EVENT_INTERRUPT		(1<<6)
 #define I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT		(1<<5)
 #define I915_DISPLAY_PIPE_B_EVENT_INTERRUPT		(1<<4)
+#define I915_DISPLAY_PIPE_A_DPBM_INTERRUPT		(1<<3)
+#define I915_DISPLAY_PIPE_B_DPBM_INTERRUPT		(1<<2)
 #define I915_DEBUG_INTERRUPT				(1<<2)
+#define I915_WINVALID_INTERRUPT				(1<<1)
 #define I915_USER_INTERRUPT				(1<<1)
 #define I915_ASLE_INTERRUPT				(1<<0)
-#define I915_BSD_USER_INTERRUPT				(1 << 25)
+#define I915_BSD_USER_INTERRUPT				(1<<25)
 
 #define GEN6_BSD_RNCID			0x12198
 
@@ -1198,6 +1527,7 @@ enum punit_power_well {
 #define   GMBUS_PORT_SSC	1
 #define   GMBUS_PORT_VGADDC	2
 #define   GMBUS_PORT_PANEL	3
+#define   GMBUS_PORT_DPD_CHV	3 /* HDMID_CHV */
 #define   GMBUS_PORT_DPC	4 /* HDMIC */
 #define   GMBUS_PORT_DPB	5 /* SDVO, HDMIB */
 #define   GMBUS_PORT_DPD	6 /* HDMID */
@@ -1239,6 +1569,7 @@ enum punit_power_well {
  */
 #define DPLL_A_OFFSET 0x6014
 #define DPLL_B_OFFSET 0x6018
+#define CHV_DPLL_C_OFFSET 0x6030
 #define DPLL(pipe) (dev_priv->info.dpll_offsets[pipe] + \
 		    dev_priv->info.display_mmio_offset)
 
@@ -1273,10 +1604,23 @@ enum punit_power_well {
 #define   DPLL_LOCK_VLV			(1<<15)
 #define   DPLL_INTEGRATED_CRI_CLK_VLV	(1<<14)
 #define   DPLL_INTEGRATED_CLOCK_VLV	(1<<13)
+#define   DPLL_SSC_REF_CLOCK_CHV	(1<<13)
 #define   DPLL_PORTC_READY_MASK		(0xf << 4)
 #define   DPLL_PORTB_READY_MASK		(0xf)
 
 #define   DPLL_FPA01_P1_POST_DIV_MASK_I830	0x001f0000
+
+/* Additional CHV pll/phy registers */
+#define DPIO_PHY_STATUS			(VLV_DISPLAY_BASE + 0x6240)
+#define   DPLL_PORTD_READY_MASK		(0xf)
+#define DISPLAY_PHY_CONTROL (VLV_DISPLAY_BASE + 0x60100)
+#define   PHY_COM_LANE_RESET_DEASSERT(phy, val) \
+				((phy == DPIO_PHY0) ? (val | 1) : (val | 2))
+#define   PHY_COM_LANE_RESET_ASSERT(phy, val) \
+				((phy == DPIO_PHY0) ? (val & ~1) : (val & ~2))
+#define DISPLAY_PHY_STATUS (VLV_DISPLAY_BASE + 0x60104)
+#define   PHY_POWERGOOD(phy)	((phy == DPIO_PHY0) ? (1<<31) : (1<<30))
+
 /*
  * The i830 generation, in LVDS mode, defines P1 as the bit number set within
  * this field (only one bit may be set).
@@ -1317,6 +1661,7 @@ enum punit_power_well {
 
 #define DPLL_A_MD_OFFSET 0x601c /* 965+ only */
 #define DPLL_B_MD_OFFSET 0x6020 /* 965+ only */
+#define CHV_DPLL_C_MD_OFFSET 0x603c
 #define DPLL_MD(pipe) (dev_priv->info.dpll_md_offsets[pipe] + \
 		       dev_priv->info.display_mmio_offset)
 
@@ -1416,7 +1761,7 @@ enum punit_power_well {
 # define DPIOUNIT_CLOCK_GATE_DISABLE		(1 << 6) /* 915-945 */
 # define OVFUNIT_CLOCK_GATE_DISABLE		(1 << 5)
 # define OVBUNIT_CLOCK_GATE_DISABLE		(1 << 4)
-/**
+/*
  * This bit must be set on the 830 to prevent hangs when turning off the
  * overlay scaler.
  */
@@ -1436,12 +1781,12 @@ enum punit_power_well {
 # define COLOR_CALCULATOR_CLOCK_GATE_DISABLE	(1 << 7)
 # define MOTION_COMP_CLOCK_GATE_DISABLE		(1 << 6)
 # define MAG_CLOCK_GATE_DISABLE			(1 << 5)
-/** This bit must be unset on 855,865 */
+/* This bit must be unset on 855,865 */
 # define MECI_CLOCK_GATE_DISABLE		(1 << 4)
 # define DCMP_CLOCK_GATE_DISABLE		(1 << 3)
 # define MEC_CLOCK_GATE_DISABLE			(1 << 2)
 # define MECO_CLOCK_GATE_DISABLE		(1 << 1)
-/** This bit must be set on 855,865. */
+/* This bit must be set on 855,865. */
 # define SV_CLOCK_GATE_DISABLE			(1 << 0)
 # define I915_MPEG_CLOCK_GATE_DISABLE		(1 << 16)
 # define I915_VLD_IP_PR_CLOCK_GATE_DISABLE	(1 << 15)
@@ -1462,14 +1807,14 @@ enum punit_power_well {
 # define I915_BY_CLOCK_GATE_DISABLE		(1 << 0)
 
 # define I965_RCZ_CLOCK_GATE_DISABLE		(1 << 30)
-/** This bit must always be set on 965G/965GM */
+/* This bit must always be set on 965G/965GM */
 # define I965_RCC_CLOCK_GATE_DISABLE		(1 << 29)
 # define I965_RCPB_CLOCK_GATE_DISABLE		(1 << 28)
 # define I965_DAP_CLOCK_GATE_DISABLE		(1 << 27)
 # define I965_ROC_CLOCK_GATE_DISABLE		(1 << 26)
 # define I965_GW_CLOCK_GATE_DISABLE		(1 << 25)
 # define I965_TD_CLOCK_GATE_DISABLE		(1 << 24)
-/** This bit must always be set on 965G */
+/* This bit must always be set on 965G */
 # define I965_ISC_CLOCK_GATE_DISABLE		(1 << 23)
 # define I965_IC_CLOCK_GATE_DISABLE		(1 << 22)
 # define I965_EU_CLOCK_GATE_DISABLE		(1 << 21)
@@ -1494,6 +1839,10 @@ enum punit_power_well {
 #define VF_UNIT_CLOCK_GATE_DISABLE		(1 << 9)
 #define GS_UNIT_CLOCK_GATE_DISABLE		(1 << 7)
 #define CL_UNIT_CLOCK_GATE_DISABLE		(1 << 6)
+
+#define VDECCLK_GATE_D		0x620C		/* g4x only */
+#define  VCP_UNIT_CLOCK_GATE_DISABLE		(1 << 4)
+
 #define RAMCLK_GATE_D		0x6210		/* CRL only */
 #define DEUC			0x6214          /* CRL only */
 
@@ -1513,6 +1862,7 @@ enum punit_power_well {
  */
 #define PALETTE_A_OFFSET 0xa000
 #define PALETTE_B_OFFSET 0xa800
+#define CHV_PALETTE_C_OFFSET 0xc000
 #define PALETTE(pipe) (dev_priv->info.palette_offsets[pipe] + \
 		       dev_priv->info.display_mmio_offset)
 
@@ -1535,7 +1885,7 @@ enum punit_power_well {
 /* Memory controller frequency in MCHBAR for Haswell (possible SNB+) */
 #define DCLK (MCHBAR_MIRROR_BASE_SNB + 0x5e04)
 
-/** 915-945 and GM965 MCH register controlling DRAM channel access */
+/* 915-945 and GM965 MCH register controlling DRAM channel access */
 #define DCC			0x10200
 #define DCC_ADDRESSING_MODE_SINGLE_CHANNEL		(0 << 0)
 #define DCC_ADDRESSING_MODE_DUAL_CHANNEL_ASYMMETRIC	(1 << 0)
@@ -1544,15 +1894,15 @@ enum punit_power_well {
 #define DCC_CHANNEL_XOR_DISABLE				(1 << 10)
 #define DCC_CHANNEL_XOR_BIT_17				(1 << 9)
 
-/** Pineview MCH register contains DDR3 setting */
+/* Pineview MCH register contains DDR3 setting */
 #define CSHRDDR3CTL            0x101a8
 #define CSHRDDR3CTL_DDR3       (1 << 2)
 
-/** 965 MCH register controlling DRAM channel configuration */
+/* 965 MCH register controlling DRAM channel configuration */
 #define C0DRB3			0x10206
 #define C1DRB3			0x10606
 
-/** snb MCH registers for reading the DRAM channel configuration */
+/* snb MCH registers for reading the DRAM channel configuration */
 #define MAD_DIMM_C0			(MCHBAR_MIRROR_BASE_SNB + 0x5004)
 #define MAD_DIMM_C1			(MCHBAR_MIRROR_BASE_SNB + 0x5008)
 #define MAD_DIMM_C2			(MCHBAR_MIRROR_BASE_SNB + 0x500C)
@@ -1574,7 +1924,7 @@ enum punit_power_well {
 #define   MAD_DIMM_A_SIZE_SHIFT		0
 #define   MAD_DIMM_A_SIZE_MASK		(0xff << MAD_DIMM_A_SIZE_SHIFT)
 
-/** snb MCH registers for priority tuning */
+/* snb MCH registers for priority tuning */
 #define MCH_SSKPD			(MCHBAR_MIRROR_BASE_SNB + 0x5d10)
 #define   MCH_SSKPD_WM0_MASK		0x3f
 #define   MCH_SSKPD_WM0_VAL		0xc
@@ -2002,6 +2352,7 @@ enum punit_power_well {
 #define TRANSCODER_A_OFFSET 0x60000
 #define TRANSCODER_B_OFFSET 0x61000
 #define TRANSCODER_C_OFFSET 0x62000
+#define CHV_TRANSCODER_C_OFFSET 0x63000
 #define TRANSCODER_EDP_OFFSET 0x6f000
 
 #define _TRANSCODER2(pipe, reg) (dev_priv->info.trans_offsets[(pipe)] - \
@@ -2226,6 +2577,7 @@ enum punit_power_well {
 #define GEN3_SDVOC	0x61160
 #define GEN4_HDMIB	GEN3_SDVOB
 #define GEN4_HDMIC	GEN3_SDVOC
+#define CHV_HDMID	0x6116C
 #define PCH_SDVOB	0xe1140
 #define PCH_HDMIB	PCH_SDVOB
 #define PCH_HDMIC	0xe1150
@@ -2246,7 +2598,7 @@ enum punit_power_well {
 #define   SDVO_PIPE_B_SELECT			(1 << 30)
 #define   SDVO_STALL_SELECT			(1 << 29)
 #define   SDVO_INTERRUPT_ENABLE			(1 << 26)
-/**
+/*
  * 915G/GM SDVO pixel multiplier.
  * Programmed value is multiplier - 1, up to 5x.
  * \sa DPLL_MD_UDI_MULTIPLIER_MASK
@@ -2286,6 +2638,10 @@ enum punit_power_well {
 #define   SDVO_PIPE_SEL_CPT(pipe)		((pipe) << 29)
 #define   SDVO_PIPE_SEL_MASK_CPT		(3 << 29)
 
+/* CHV SDVO/HDMI bits: */
+#define   SDVO_PIPE_SEL_CHV(pipe)		((pipe) << 24)
+#define   SDVO_PIPE_SEL_MASK_CHV		(3 << 24)
+
 
 /* DVO port control */
 #define DVOA			0x61120
@@ -2556,65 +2912,65 @@ enum punit_power_well {
 
 /* TV port control */
 #define TV_CTL			0x68000
-/** Enables the TV encoder */
+/* Enables the TV encoder */
 # define TV_ENC_ENABLE			(1 << 31)
-/** Sources the TV encoder input from pipe B instead of A. */
+/* Sources the TV encoder input from pipe B instead of A. */
 # define TV_ENC_PIPEB_SELECT		(1 << 30)
-/** Outputs composite video (DAC A only) */
+/* Outputs composite video (DAC A only) */
 # define TV_ENC_OUTPUT_COMPOSITE	(0 << 28)
-/** Outputs SVideo video (DAC B/C) */
+/* Outputs SVideo video (DAC B/C) */
 # define TV_ENC_OUTPUT_SVIDEO		(1 << 28)
-/** Outputs Component video (DAC A/B/C) */
+/* Outputs Component video (DAC A/B/C) */
 # define TV_ENC_OUTPUT_COMPONENT	(2 << 28)
-/** Outputs Composite and SVideo (DAC A/B/C) */
+/* Outputs Composite and SVideo (DAC A/B/C) */
 # define TV_ENC_OUTPUT_SVIDEO_COMPOSITE	(3 << 28)
 # define TV_TRILEVEL_SYNC		(1 << 21)
-/** Enables slow sync generation (945GM only) */
+/* Enables slow sync generation (945GM only) */
 # define TV_SLOW_SYNC			(1 << 20)
-/** Selects 4x oversampling for 480i and 576p */
+/* Selects 4x oversampling for 480i and 576p */
 # define TV_OVERSAMPLE_4X		(0 << 18)
-/** Selects 2x oversampling for 720p and 1080i */
+/* Selects 2x oversampling for 720p and 1080i */
 # define TV_OVERSAMPLE_2X		(1 << 18)
-/** Selects no oversampling for 1080p */
+/* Selects no oversampling for 1080p */
 # define TV_OVERSAMPLE_NONE		(2 << 18)
-/** Selects 8x oversampling */
+/* Selects 8x oversampling */
 # define TV_OVERSAMPLE_8X		(3 << 18)
-/** Selects progressive mode rather than interlaced */
+/* Selects progressive mode rather than interlaced */
 # define TV_PROGRESSIVE			(1 << 17)
-/** Sets the colorburst to PAL mode.  Required for non-M PAL modes. */
+/* Sets the colorburst to PAL mode.  Required for non-M PAL modes. */
 # define TV_PAL_BURST			(1 << 16)
-/** Field for setting delay of Y compared to C */
+/* Field for setting delay of Y compared to C */
 # define TV_YC_SKEW_MASK		(7 << 12)
-/** Enables a fix for 480p/576p standard definition modes on the 915GM only */
+/* Enables a fix for 480p/576p standard definition modes on the 915GM only */
 # define TV_ENC_SDP_FIX			(1 << 11)
-/**
+/*
  * Enables a fix for the 915GM only.
  *
  * Not sure what it does.
  */
 # define TV_ENC_C0_FIX			(1 << 10)
-/** Bits that must be preserved by software */
+/* Bits that must be preserved by software */
 # define TV_CTL_SAVE			((1 << 11) | (3 << 9) | (7 << 6) | 0xf)
 # define TV_FUSE_STATE_MASK		(3 << 4)
-/** Read-only state that reports all features enabled */
+/* Read-only state that reports all features enabled */
 # define TV_FUSE_STATE_ENABLED		(0 << 4)
-/** Read-only state that reports that Macrovision is disabled in hardware*/
+/* Read-only state that reports that Macrovision is disabled in hardware*/
 # define TV_FUSE_STATE_NO_MACROVISION	(1 << 4)
-/** Read-only state that reports that TV-out is disabled in hardware. */
+/* Read-only state that reports that TV-out is disabled in hardware. */
 # define TV_FUSE_STATE_DISABLED		(2 << 4)
-/** Normal operation */
+/* Normal operation */
 # define TV_TEST_MODE_NORMAL		(0 << 0)
-/** Encoder test pattern 1 - combo pattern */
+/* Encoder test pattern 1 - combo pattern */
 # define TV_TEST_MODE_PATTERN_1		(1 << 0)
-/** Encoder test pattern 2 - full screen vertical 75% color bars */
+/* Encoder test pattern 2 - full screen vertical 75% color bars */
 # define TV_TEST_MODE_PATTERN_2		(2 << 0)
-/** Encoder test pattern 3 - full screen horizontal 75% color bars */
+/* Encoder test pattern 3 - full screen horizontal 75% color bars */
 # define TV_TEST_MODE_PATTERN_3		(3 << 0)
-/** Encoder test pattern 4 - random noise */
+/* Encoder test pattern 4 - random noise */
 # define TV_TEST_MODE_PATTERN_4		(4 << 0)
-/** Encoder test pattern 5 - linear color ramps */
+/* Encoder test pattern 5 - linear color ramps */
 # define TV_TEST_MODE_PATTERN_5		(5 << 0)
-/**
+/*
  * This test mode forces the DACs to 50% of full output.
  *
  * This is used for load detection in combination with TVDAC_SENSE_MASK
@@ -2624,35 +2980,35 @@ enum punit_power_well {
 
 #define TV_DAC			0x68004
 # define TV_DAC_SAVE		0x00ffff00
-/**
+/*
  * Reports that DAC state change logic has reported change (RO).
  *
  * This gets cleared when TV_DAC_STATE_EN is cleared
 */
 # define TVDAC_STATE_CHG		(1 << 31)
 # define TVDAC_SENSE_MASK		(7 << 28)
-/** Reports that DAC A voltage is above the detect threshold */
+/* Reports that DAC A voltage is above the detect threshold */
 # define TVDAC_A_SENSE			(1 << 30)
-/** Reports that DAC B voltage is above the detect threshold */
+/* Reports that DAC B voltage is above the detect threshold */
 # define TVDAC_B_SENSE			(1 << 29)
-/** Reports that DAC C voltage is above the detect threshold */
+/* Reports that DAC C voltage is above the detect threshold */
 # define TVDAC_C_SENSE			(1 << 28)
-/**
+/*
  * Enables DAC state detection logic, for load-based TV detection.
  *
  * The PLL of the chosen pipe (in TV_CTL) must be running, and the encoder set
  * to off, for load detection to work.
  */
 # define TVDAC_STATE_CHG_EN		(1 << 27)
-/** Sets the DAC A sense value to high */
+/* Sets the DAC A sense value to high */
 # define TVDAC_A_SENSE_CTL		(1 << 26)
-/** Sets the DAC B sense value to high */
+/* Sets the DAC B sense value to high */
 # define TVDAC_B_SENSE_CTL		(1 << 25)
-/** Sets the DAC C sense value to high */
+/* Sets the DAC C sense value to high */
 # define TVDAC_C_SENSE_CTL		(1 << 24)
-/** Overrides the ENC_ENABLE and DAC voltage levels */
+/* Overrides the ENC_ENABLE and DAC voltage levels */
 # define DAC_CTL_OVERRIDE		(1 << 7)
-/** Sets the slew rate.  Must be preserved in software */
+/* Sets the slew rate.  Must be preserved in software */
 # define ENC_TVDAC_SLEW_FAST		(1 << 6)
 # define DAC_A_1_3_V			(0 << 4)
 # define DAC_A_1_1_V			(1 << 4)
@@ -2667,7 +3023,7 @@ enum punit_power_well {
 # define DAC_C_0_7_V			(2 << 0)
 # define DAC_C_MASK			(3 << 0)
 
-/**
+/*
  * CSC coefficients are stored in a floating point format with 9 bits of
  * mantissa and 2 or 3 bits of exponent.  The exponent is represented as 2**-n,
  * where 2-bit exponents are unsigned n, and 3-bit exponents are signed n with
@@ -2682,7 +3038,7 @@ enum punit_power_well {
 #define TV_CSC_Y2		0x68014
 # define TV_BY_MASK			0x07ff0000
 # define TV_BY_SHIFT			16
-/**
+/*
  * Y attenuation for component video.
  *
  * Stored in 1.9 fixed point.
@@ -2699,7 +3055,7 @@ enum punit_power_well {
 #define TV_CSC_U2		0x6801c
 # define TV_BU_MASK			0x07ff0000
 # define TV_BU_SHIFT			16
-/**
+/*
  * U attenuation for component video.
  *
  * Stored in 1.9 fixed point.
@@ -2716,7 +3072,7 @@ enum punit_power_well {
 #define TV_CSC_V2		0x68024
 # define TV_BV_MASK			0x07ff0000
 # define TV_BV_SHIFT			16
-/**
+/*
  * V attenuation for component video.
  *
  * Stored in 1.9 fixed point.
@@ -2725,74 +3081,74 @@ enum punit_power_well {
 # define TV_AV_SHIFT			0
 
 #define TV_CLR_KNOBS		0x68028
-/** 2s-complement brightness adjustment */
+/* 2s-complement brightness adjustment */
 # define TV_BRIGHTNESS_MASK		0xff000000
 # define TV_BRIGHTNESS_SHIFT		24
-/** Contrast adjustment, as a 2.6 unsigned floating point number */
+/* Contrast adjustment, as a 2.6 unsigned floating point number */
 # define TV_CONTRAST_MASK		0x00ff0000
 # define TV_CONTRAST_SHIFT		16
-/** Saturation adjustment, as a 2.6 unsigned floating point number */
+/* Saturation adjustment, as a 2.6 unsigned floating point number */
 # define TV_SATURATION_MASK		0x0000ff00
 # define TV_SATURATION_SHIFT		8
-/** Hue adjustment, as an integer phase angle in degrees */
+/* Hue adjustment, as an integer phase angle in degrees */
 # define TV_HUE_MASK			0x000000ff
 # define TV_HUE_SHIFT			0
 
 #define TV_CLR_LEVEL		0x6802c
-/** Controls the DAC level for black */
+/* Controls the DAC level for black */
 # define TV_BLACK_LEVEL_MASK		0x01ff0000
 # define TV_BLACK_LEVEL_SHIFT		16
-/** Controls the DAC level for blanking */
+/* Controls the DAC level for blanking */
 # define TV_BLANK_LEVEL_MASK		0x000001ff
 # define TV_BLANK_LEVEL_SHIFT		0
 
 #define TV_H_CTL_1		0x68030
-/** Number of pixels in the hsync. */
+/* Number of pixels in the hsync. */
 # define TV_HSYNC_END_MASK		0x1fff0000
 # define TV_HSYNC_END_SHIFT		16
-/** Total number of pixels minus one in the line (display and blanking). */
+/* Total number of pixels minus one in the line (display and blanking). */
 # define TV_HTOTAL_MASK			0x00001fff
 # define TV_HTOTAL_SHIFT		0
 
 #define TV_H_CTL_2		0x68034
-/** Enables the colorburst (needed for non-component color) */
+/* Enables the colorburst (needed for non-component color) */
 # define TV_BURST_ENA			(1 << 31)
-/** Offset of the colorburst from the start of hsync, in pixels minus one. */
+/* Offset of the colorburst from the start of hsync, in pixels minus one. */
 # define TV_HBURST_START_SHIFT		16
 # define TV_HBURST_START_MASK		0x1fff0000
-/** Length of the colorburst */
+/* Length of the colorburst */
 # define TV_HBURST_LEN_SHIFT		0
 # define TV_HBURST_LEN_MASK		0x0001fff
 
 #define TV_H_CTL_3		0x68038
-/** End of hblank, measured in pixels minus one from start of hsync */
+/* End of hblank, measured in pixels minus one from start of hsync */
 # define TV_HBLANK_END_SHIFT		16
 # define TV_HBLANK_END_MASK		0x1fff0000
-/** Start of hblank, measured in pixels minus one from start of hsync */
+/* Start of hblank, measured in pixels minus one from start of hsync */
 # define TV_HBLANK_START_SHIFT		0
 # define TV_HBLANK_START_MASK		0x0001fff
 
 #define TV_V_CTL_1		0x6803c
-/** XXX */
+/* XXX */
 # define TV_NBR_END_SHIFT		16
 # define TV_NBR_END_MASK		0x07ff0000
-/** XXX */
+/* XXX */
 # define TV_VI_END_F1_SHIFT		8
 # define TV_VI_END_F1_MASK		0x00003f00
-/** XXX */
+/* XXX */
 # define TV_VI_END_F2_SHIFT		0
 # define TV_VI_END_F2_MASK		0x0000003f
 
 #define TV_V_CTL_2		0x68040
-/** Length of vsync, in half lines */
+/* Length of vsync, in half lines */
 # define TV_VSYNC_LEN_MASK		0x07ff0000
 # define TV_VSYNC_LEN_SHIFT		16
-/** Offset of the start of vsync in field 1, measured in one less than the
+/* Offset of the start of vsync in field 1, measured in one less than the
  * number of half lines.
  */
 # define TV_VSYNC_START_F1_MASK		0x00007f00
 # define TV_VSYNC_START_F1_SHIFT	8
-/**
+/*
  * Offset of the start of vsync in field 2, measured in one less than the
  * number of half lines.
  */
@@ -2800,17 +3156,17 @@ enum punit_power_well {
 # define TV_VSYNC_START_F2_SHIFT	0
 
 #define TV_V_CTL_3		0x68044
-/** Enables generation of the equalization signal */
+/* Enables generation of the equalization signal */
 # define TV_EQUAL_ENA			(1 << 31)
-/** Length of vsync, in half lines */
+/* Length of vsync, in half lines */
 # define TV_VEQ_LEN_MASK		0x007f0000
 # define TV_VEQ_LEN_SHIFT		16
-/** Offset of the start of equalization in field 1, measured in one less than
+/* Offset of the start of equalization in field 1, measured in one less than
  * the number of half lines.
  */
 # define TV_VEQ_START_F1_MASK		0x0007f00
 # define TV_VEQ_START_F1_SHIFT		8
-/**
+/*
  * Offset of the start of equalization in field 2, measured in one less than
  * the number of half lines.
  */
@@ -2818,13 +3174,13 @@ enum punit_power_well {
 # define TV_VEQ_START_F2_SHIFT		0
 
 #define TV_V_CTL_4		0x68048
-/**
+/*
  * Offset to start of vertical colorburst, measured in one less than the
  * number of lines from vertical start.
  */
 # define TV_VBURST_START_F1_MASK	0x003f0000
 # define TV_VBURST_START_F1_SHIFT	16
-/**
+/*
  * Offset to the end of vertical colorburst, measured in one less than the
  * number of lines from the start of NBR.
  */
@@ -2832,13 +3188,13 @@ enum punit_power_well {
 # define TV_VBURST_END_F1_SHIFT		0
 
 #define TV_V_CTL_5		0x6804c
-/**
+/*
  * Offset to start of vertical colorburst, measured in one less than the
  * number of lines from vertical start.
  */
 # define TV_VBURST_START_F2_MASK	0x003f0000
 # define TV_VBURST_START_F2_SHIFT	16
-/**
+/*
  * Offset to the end of vertical colorburst, measured in one less than the
  * number of lines from the start of NBR.
  */
@@ -2846,13 +3202,13 @@ enum punit_power_well {
 # define TV_VBURST_END_F2_SHIFT		0
 
 #define TV_V_CTL_6		0x68050
-/**
+/*
  * Offset to start of vertical colorburst, measured in one less than the
  * number of lines from vertical start.
  */
 # define TV_VBURST_START_F3_MASK	0x003f0000
 # define TV_VBURST_START_F3_SHIFT	16
-/**
+/*
  * Offset to the end of vertical colorburst, measured in one less than the
  * number of lines from the start of NBR.
  */
@@ -2860,13 +3216,13 @@ enum punit_power_well {
 # define TV_VBURST_END_F3_SHIFT		0
 
 #define TV_V_CTL_7		0x68054
-/**
+/*
  * Offset to start of vertical colorburst, measured in one less than the
  * number of lines from vertical start.
  */
 # define TV_VBURST_START_F4_MASK	0x003f0000
 # define TV_VBURST_START_F4_SHIFT	16
-/**
+/*
  * Offset to the end of vertical colorburst, measured in one less than the
  * number of lines from the start of NBR.
  */
@@ -2874,56 +3230,56 @@ enum punit_power_well {
 # define TV_VBURST_END_F4_SHIFT		0
 
 #define TV_SC_CTL_1		0x68060
-/** Turns on the first subcarrier phase generation DDA */
+/* Turns on the first subcarrier phase generation DDA */
 # define TV_SC_DDA1_EN			(1 << 31)
-/** Turns on the first subcarrier phase generation DDA */
+/* Turns on the first subcarrier phase generation DDA */
 # define TV_SC_DDA2_EN			(1 << 30)
-/** Turns on the first subcarrier phase generation DDA */
+/* Turns on the first subcarrier phase generation DDA */
 # define TV_SC_DDA3_EN			(1 << 29)
-/** Sets the subcarrier DDA to reset frequency every other field */
+/* Sets the subcarrier DDA to reset frequency every other field */
 # define TV_SC_RESET_EVERY_2		(0 << 24)
-/** Sets the subcarrier DDA to reset frequency every fourth field */
+/* Sets the subcarrier DDA to reset frequency every fourth field */
 # define TV_SC_RESET_EVERY_4		(1 << 24)
-/** Sets the subcarrier DDA to reset frequency every eighth field */
+/* Sets the subcarrier DDA to reset frequency every eighth field */
 # define TV_SC_RESET_EVERY_8		(2 << 24)
-/** Sets the subcarrier DDA to never reset the frequency */
+/* Sets the subcarrier DDA to never reset the frequency */
 # define TV_SC_RESET_NEVER		(3 << 24)
-/** Sets the peak amplitude of the colorburst.*/
+/* Sets the peak amplitude of the colorburst.*/
 # define TV_BURST_LEVEL_MASK		0x00ff0000
 # define TV_BURST_LEVEL_SHIFT		16
-/** Sets the increment of the first subcarrier phase generation DDA */
+/* Sets the increment of the first subcarrier phase generation DDA */
 # define TV_SCDDA1_INC_MASK		0x00000fff
 # define TV_SCDDA1_INC_SHIFT		0
 
 #define TV_SC_CTL_2		0x68064
-/** Sets the rollover for the second subcarrier phase generation DDA */
+/* Sets the rollover for the second subcarrier phase generation DDA */
 # define TV_SCDDA2_SIZE_MASK		0x7fff0000
 # define TV_SCDDA2_SIZE_SHIFT		16
-/** Sets the increent of the second subcarrier phase generation DDA */
+/* Sets the increent of the second subcarrier phase generation DDA */
 # define TV_SCDDA2_INC_MASK		0x00007fff
 # define TV_SCDDA2_INC_SHIFT		0
 
 #define TV_SC_CTL_3		0x68068
-/** Sets the rollover for the third subcarrier phase generation DDA */
+/* Sets the rollover for the third subcarrier phase generation DDA */
 # define TV_SCDDA3_SIZE_MASK		0x7fff0000
 # define TV_SCDDA3_SIZE_SHIFT		16
-/** Sets the increent of the third subcarrier phase generation DDA */
+/* Sets the increent of the third subcarrier phase generation DDA */
 # define TV_SCDDA3_INC_MASK		0x00007fff
 # define TV_SCDDA3_INC_SHIFT		0
 
 #define TV_WIN_POS		0x68070
-/** X coordinate of the display from the start of horizontal active */
+/* X coordinate of the display from the start of horizontal active */
 # define TV_XPOS_MASK			0x1fff0000
 # define TV_XPOS_SHIFT			16
-/** Y coordinate of the display from the start of vertical active (NBR) */
+/* Y coordinate of the display from the start of vertical active (NBR) */
 # define TV_YPOS_MASK			0x00000fff
 # define TV_YPOS_SHIFT			0
 
 #define TV_WIN_SIZE		0x68074
-/** Horizontal size of the display window, measured in pixels*/
+/* Horizontal size of the display window, measured in pixels*/
 # define TV_XSIZE_MASK			0x1fff0000
 # define TV_XSIZE_SHIFT			16
-/**
+/*
  * Vertical size of the display window, measured in pixels.
  *
  * Must be even for interlaced modes.
@@ -2932,28 +3288,28 @@ enum punit_power_well {
 # define TV_YSIZE_SHIFT			0
 
 #define TV_FILTER_CTL_1		0x68080
-/**
+/*
  * Enables automatic scaling calculation.
  *
  * If set, the rest of the registers are ignored, and the calculated values can
  * be read back from the register.
  */
 # define TV_AUTO_SCALE			(1 << 31)
-/**
+/*
  * Disables the vertical filter.
  *
  * This is required on modes more than 1024 pixels wide */
 # define TV_V_FILTER_BYPASS		(1 << 29)
-/** Enables adaptive vertical filtering */
+/* Enables adaptive vertical filtering */
 # define TV_VADAPT			(1 << 28)
 # define TV_VADAPT_MODE_MASK		(3 << 26)
-/** Selects the least adaptive vertical filtering mode */
+/* Selects the least adaptive vertical filtering mode */
 # define TV_VADAPT_MODE_LEAST		(0 << 26)
-/** Selects the moderately adaptive vertical filtering mode */
+/* Selects the moderately adaptive vertical filtering mode */
 # define TV_VADAPT_MODE_MODERATE	(1 << 26)
-/** Selects the most adaptive vertical filtering mode */
+/* Selects the most adaptive vertical filtering mode */
 # define TV_VADAPT_MODE_MOST		(3 << 26)
-/**
+/*
  * Sets the horizontal scaling factor.
  *
  * This should be the fractional part of the horizontal scaling factor divided
@@ -2965,14 +3321,14 @@ enum punit_power_well {
 # define TV_HSCALE_FRAC_SHIFT		0
 
 #define TV_FILTER_CTL_2		0x68084
-/**
+/*
  * Sets the integer part of the 3.15 fixed-point vertical scaling factor.
  *
  * TV_VSCALE should be (src height - 1) / ((interlace * dest height) - 1)
  */
 # define TV_VSCALE_INT_MASK		0x00038000
 # define TV_VSCALE_INT_SHIFT		15
-/**
+/*
  * Sets the fractional part of the 3.15 fixed-point vertical scaling factor.
  *
  * \sa TV_VSCALE_INT_MASK
@@ -2981,7 +3337,7 @@ enum punit_power_well {
 # define TV_VSCALE_FRAC_SHIFT		0
 
 #define TV_FILTER_CTL_3		0x68088
-/**
+/*
  * Sets the integer part of the 3.15 fixed-point vertical scaling factor.
  *
  * TV_VSCALE should be (src height - 1) / (1/4 * (dest height - 1))
@@ -2990,7 +3346,7 @@ enum punit_power_well {
  */
 # define TV_VSCALE_IP_INT_MASK		0x00038000
 # define TV_VSCALE_IP_INT_SHIFT		15
-/**
+/*
  * Sets the fractional part of the 3.15 fixed-point vertical scaling factor.
  *
  * For progressive modes, TV_VSCALE_IP_INT should be set to zeroes.
@@ -3002,26 +3358,26 @@ enum punit_power_well {
 
 #define TV_CC_CONTROL		0x68090
 # define TV_CC_ENABLE			(1 << 31)
-/**
+/*
  * Specifies which field to send the CC data in.
  *
  * CC data is usually sent in field 0.
  */
 # define TV_CC_FID_MASK			(1 << 27)
 # define TV_CC_FID_SHIFT		27
-/** Sets the horizontal position of the CC data.  Usually 135. */
+/* Sets the horizontal position of the CC data.  Usually 135. */
 # define TV_CC_HOFF_MASK		0x03ff0000
 # define TV_CC_HOFF_SHIFT		16
-/** Sets the vertical position of the CC data.  Usually 21 */
+/* Sets the vertical position of the CC data.  Usually 21 */
 # define TV_CC_LINE_MASK		0x0000003f
 # define TV_CC_LINE_SHIFT		0
 
 #define TV_CC_DATA		0x68094
 # define TV_CC_RDY			(1 << 31)
-/** Second word of CC data to be transmitted. */
+/* Second word of CC data to be transmitted. */
 # define TV_CC_DATA_2_MASK		0x007f0000
 # define TV_CC_DATA_2_SHIFT		16
-/** First word of CC data to be transmitted. */
+/* First word of CC data to be transmitted. */
 # define TV_CC_DATA_1_MASK		0x0000007f
 # define TV_CC_DATA_1_SHIFT		0
 
@@ -3043,6 +3399,8 @@ enum punit_power_well {
 #define   DP_PORT_EN			(1 << 31)
 #define   DP_PIPEB_SELECT		(1 << 30)
 #define   DP_PIPE_MASK			(1 << 30)
+#define   DP_PIPE_SELECT_CHV(pipe)	((pipe) << 16)
+#define   DP_PIPE_MASK_CHV		(3 << 16)
 
 /* Link training mode - select a suitable mode for each stage */
 #define   DP_LINK_TRAIN_PAT_1		(0 << 28)
@@ -3090,32 +3448,32 @@ enum punit_power_well {
 #define   DP_PLL_FREQ_160MHZ		(1 << 16)
 #define   DP_PLL_FREQ_MASK		(3 << 16)
 
-/** locked once port is enabled */
+/* locked once port is enabled */
 #define   DP_PORT_REVERSAL		(1 << 15)
 
 /* eDP */
 #define   DP_PLL_ENABLE			(1 << 14)
 
-/** sends the clock on lane 15 of the PEG for debug */
+/* sends the clock on lane 15 of the PEG for debug */
 #define   DP_CLOCK_OUTPUT_ENABLE	(1 << 13)
 
 #define   DP_SCRAMBLING_DISABLE		(1 << 12)
 #define   DP_SCRAMBLING_DISABLE_IRONLAKE	(1 << 7)
 
-/** limit RGB values to avoid confusing TVs */
+/* limit RGB values to avoid confusing TVs */
 #define   DP_COLOR_RANGE_16_235		(1 << 8)
 
-/** Turn on the audio link */
+/* Turn on the audio link */
 #define   DP_AUDIO_OUTPUT_ENABLE	(1 << 6)
 
-/** vs and hs sync polarity */
+/* vs and hs sync polarity */
 #define   DP_SYNC_VS_HIGH		(1 << 4)
 #define   DP_SYNC_HS_HIGH		(1 << 3)
 
-/** A fantasy */
+/* A fantasy */
 #define   DP_DETECTED			(1 << 2)
 
-/** The aux channel provides a way to talk to the
+/* The aux channel provides a way to talk to the
  * signal sink for DDC etc. Max packet size supported
  * is 20 bytes in each direction, hence the 5 fixed
  * data registers
@@ -3258,6 +3616,7 @@ enum punit_power_well {
 #define   PIPECONF_INTERLACED_DBL_ILK		(4 << 21) /* ilk/snb only */
 #define   PIPECONF_PFIT_PF_INTERLACED_DBL_ILK	(5 << 21) /* ilk/snb only */
 #define   PIPECONF_INTERLACE_MODE_MASK		(7 << 21)
+#define   PIPECONF_EDP_RR_MODE_SWITCH		(1 << 20)
 #define   PIPECONF_CXSR_DOWNCLOCK	(1<<16)
 #define   PIPECONF_COLOR_RANGE_SELECT	(1 << 13)
 #define   PIPECONF_BPC_MASK	(0x7 << 5)
@@ -3276,6 +3635,7 @@ enum punit_power_well {
 #define   SPRITE1_FLIP_DONE_INT_EN_VLV		(1UL<<30)
 #define   PIPE_CRC_ERROR_ENABLE			(1UL<<29)
 #define   PIPE_CRC_DONE_ENABLE			(1UL<<28)
+#define   PERF_COUNTER2_INTERRUPT_EN		(1UL<<27)
 #define   PIPE_GMBUS_EVENT_ENABLE		(1UL<<27)
 #define   PLANE_FLIP_DONE_INT_EN_VLV		(1UL<<26)
 #define   PIPE_HOTPLUG_INTERRUPT_ENABLE		(1UL<<26)
@@ -3287,8 +3647,10 @@ enum punit_power_well {
 #define   PIPE_ODD_FIELD_INTERRUPT_ENABLE	(1UL<<21)
 #define   PIPE_EVEN_FIELD_INTERRUPT_ENABLE	(1UL<<20)
 #define   PIPE_B_PSR_INTERRUPT_ENABLE_VLV	(1UL<<19)
+#define   PERF_COUNTER_INTERRUPT_EN		(1UL<<19)
 #define   PIPE_HOTPLUG_TV_INTERRUPT_ENABLE	(1UL<<18) /* pre-965 */
 #define   PIPE_START_VBLANK_INTERRUPT_ENABLE	(1UL<<18) /* 965 or later */
+#define   PIPE_FRAMESTART_INTERRUPT_ENABLE	(1UL<<17)
 #define   PIPE_VBLANK_INTERRUPT_ENABLE		(1UL<<17)
 #define   PIPEA_HBLANK_INT_EN_VLV		(1UL<<16)
 #define   PIPE_OVERLAY_UPDATED_ENABLE		(1UL<<16)
@@ -3296,6 +3658,7 @@ enum punit_power_well {
 #define   SPRITE0_FLIP_DONE_INT_STATUS_VLV	(1UL<<14)
 #define   PIPE_CRC_ERROR_INTERRUPT_STATUS	(1UL<<13)
 #define   PIPE_CRC_DONE_INTERRUPT_STATUS	(1UL<<12)
+#define   PERF_COUNTER2_INTERRUPT_STATUS	(1UL<<11)
 #define   PIPE_GMBUS_INTERRUPT_STATUS		(1UL<<11)
 #define   PLANE_FLIP_DONE_INT_STATUS_VLV	(1UL<<10)
 #define   PIPE_HOTPLUG_INTERRUPT_STATUS		(1UL<<10)
@@ -3304,20 +3667,25 @@ enum punit_power_well {
 #define   PIPE_DPST_EVENT_STATUS		(1UL<<7)
 #define   PIPE_LEGACY_BLC_EVENT_STATUS		(1UL<<6)
 #define   PIPE_A_PSR_STATUS_VLV			(1UL<<6)
+#define   PIPE_LEGACY_BLC_EVENT_STATUS		(1UL<<6)
 #define   PIPE_ODD_FIELD_INTERRUPT_STATUS	(1UL<<5)
 #define   PIPE_EVEN_FIELD_INTERRUPT_STATUS	(1UL<<4)
 #define   PIPE_B_PSR_STATUS_VLV			(1UL<<3)
+#define   PERF_COUNTER_INTERRUPT_STATUS		(1UL<<3)
 #define   PIPE_HOTPLUG_TV_INTERRUPT_STATUS	(1UL<<2) /* pre-965 */
 #define   PIPE_START_VBLANK_INTERRUPT_STATUS	(1UL<<2) /* 965 or later */
+#define   PIPE_FRAMESTART_INTERRUPT_STATUS	(1UL<<1)
 #define   PIPE_VBLANK_INTERRUPT_STATUS		(1UL<<1)
+#define   PIPE_HBLANK_INT_STATUS		(1UL<<0)
 #define   PIPE_OVERLAY_UPDATED_STATUS		(1UL<<0)
 
 #define PIPESTAT_INT_ENABLE_MASK		0x7fff0000
 #define PIPESTAT_INT_STATUS_MASK		0x0000ffff
 
-#define PIPE_A_OFFSET	0x70000
-#define PIPE_B_OFFSET	0x71000
-#define PIPE_C_OFFSET	0x72000
+#define PIPE_A_OFFSET		0x70000
+#define PIPE_B_OFFSET		0x71000
+#define PIPE_C_OFFSET		0x72000
+#define CHV_PIPE_C_OFFSET	0x74000
 /*
  * There's actually no pipe EDP. Some pipe registers have
  * simply shifted from the pipe to the transcoder, while
@@ -3355,14 +3723,25 @@ enum punit_power_well {
 #define   SPRITED_FLIP_DONE_INT_EN		(1<<26)
 #define   SPRITEC_FLIP_DONE_INT_EN		(1<<25)
 #define   PLANEB_FLIP_DONE_INT_EN		(1<<24)
+#define   PIPE_PSR_INT_EN			(1<<22)
 #define   PIPEA_LINE_COMPARE_INT_EN		(1<<21)
 #define   PIPEA_HLINE_INT_EN			(1<<20)
 #define   PIPEA_VBLANK_INT_EN			(1<<19)
 #define   SPRITEB_FLIP_DONE_INT_EN		(1<<18)
 #define   SPRITEA_FLIP_DONE_INT_EN		(1<<17)
 #define   PLANEA_FLIPDONE_INT_EN		(1<<16)
-
-#define DPINVGTT				(VLV_DISPLAY_BASE + 0x7002c) /* VLV only */
+#define   PIPEC_LINE_COMPARE_INT_EN		(1<<13)
+#define   PIPEC_HLINE_INT_EN			(1<<12)
+#define   PIPEC_VBLANK_INT_EN			(1<<11)
+#define   SPRITEF_FLIPDONE_INT_EN		(1<<10)
+#define   SPRITEE_FLIPDONE_INT_EN		(1<<9)
+#define   PLANEC_FLIPDONE_INT_EN		(1<<8)
+
+#define DPINVGTT				(VLV_DISPLAY_BASE + 0x7002c) /* VLV/CHV only */
+#define   SPRITEF_INVALID_GTT_INT_EN		(1<<27)
+#define   SPRITEE_INVALID_GTT_INT_EN		(1<<26)
+#define   PLANEC_INVALID_GTT_INT_EN		(1<<25)
+#define   CURSORC_INVALID_GTT_INT_EN		(1<<24)
 #define   CURSORB_INVALID_GTT_INT_EN		(1<<23)
 #define   CURSORA_INVALID_GTT_INT_EN		(1<<22)
 #define   SPRITED_INVALID_GTT_INT_EN		(1<<21)
@@ -3372,6 +3751,11 @@ enum punit_power_well {
 #define   SPRITEA_INVALID_GTT_INT_EN		(1<<17)
 #define   PLANEA_INVALID_GTT_INT_EN		(1<<16)
 #define   DPINVGTT_EN_MASK			0xff0000
+#define   DPINVGTT_EN_MASK_CHV			0xfff0000
+#define   SPRITEF_INVALID_GTT_STATUS		(1<<11)
+#define   SPRITEE_INVALID_GTT_STATUS		(1<<10)
+#define   PLANEC_INVALID_GTT_STATUS		(1<<9)
+#define   CURSORC_INVALID_GTT_STATUS		(1<<8)
 #define   CURSORB_INVALID_GTT_STATUS		(1<<7)
 #define   CURSORA_INVALID_GTT_STATUS		(1<<6)
 #define   SPRITED_INVALID_GTT_STATUS		(1<<5)
@@ -3381,6 +3765,7 @@ enum punit_power_well {
 #define   SPRITEA_INVALID_GTT_STATUS		(1<<1)
 #define   PLANEA_INVALID_GTT_STATUS		(1<<0)
 #define   DPINVGTT_STATUS_MASK			0xff
+#define   DPINVGTT_STATUS_MASK_CHV		0xfff
 
 #define DSPARB			0x70030
 #define   DSPARB_CSTART_MASK	(0x7f << 7)
@@ -3420,14 +3805,43 @@ enum punit_power_well {
 #define DDL_CURSORA_PRECISION_32	(1<<31)
 #define DDL_CURSORA_PRECISION_16	(0<<31)
 #define DDL_CURSORA_SHIFT		24
+#define DDL_SPRITEB_PRECISION_32	(1<<23)
+#define DDL_SPRITEB_PRECISION_16	(0<<23)
+#define DDL_SPRITEB_SHIFT		16
+#define DDL_SPRITEA_PRECISION_32	(1<<15)
+#define DDL_SPRITEA_PRECISION_16	(0<<15)
+#define DDL_SPRITEA_SHIFT		8
 #define DDL_PLANEA_PRECISION_32		(1<<7)
 #define DDL_PLANEA_PRECISION_16		(0<<7)
+#define DDL_PLANEA_SHIFT		0
+
 #define VLV_DDL2			(VLV_DISPLAY_BASE + 0x70054)
 #define DDL_CURSORB_PRECISION_32	(1<<31)
 #define DDL_CURSORB_PRECISION_16	(0<<31)
 #define DDL_CURSORB_SHIFT		24
+#define DDL_SPRITED_PRECISION_32	(1<<23)
+#define DDL_SPRITED_PRECISION_16	(0<<23)
+#define DDL_SPRITED_SHIFT		16
+#define DDL_SPRITEC_PRECISION_32	(1<<15)
+#define DDL_SPRITEC_PRECISION_16	(0<<15)
+#define DDL_SPRITEC_SHIFT		8
 #define DDL_PLANEB_PRECISION_32		(1<<7)
 #define DDL_PLANEB_PRECISION_16		(0<<7)
+#define DDL_PLANEB_SHIFT		0
+
+#define VLV_DDL3			(VLV_DISPLAY_BASE + 0x70058)
+#define DDL_CURSORC_PRECISION_32	(1<<31)
+#define DDL_CURSORC_PRECISION_16	(0<<31)
+#define DDL_CURSORC_SHIFT		24
+#define DDL_SPRITEF_PRECISION_32	(1<<23)
+#define DDL_SPRITEF_PRECISION_16	(0<<23)
+#define DDL_SPRITEF_SHIFT		16
+#define DDL_SPRITEE_PRECISION_32	(1<<15)
+#define DDL_SPRITEE_PRECISION_16	(0<<15)
+#define DDL_SPRITEE_SHIFT		8
+#define DDL_PLANEC_PRECISION_32		(1<<7)
+#define DDL_PLANEC_PRECISION_16		(0<<7)
+#define DDL_PLANEC_SHIFT		0
 
 /* FIFO watermark sizes etc */
 #define G4X_FIFO_LINE_SIZE	64
@@ -3535,12 +3949,13 @@ enum punit_power_well {
 #define   PIPE_PIXEL_MASK         0x00ffffff
 #define   PIPE_PIXEL_SHIFT        0
 /* GM45+ just has to be different */
-#define _PIPEA_FRMCOUNT_GM45	(dev_priv->info.display_mmio_offset + 0x70040)
-#define _PIPEA_FLIPCOUNT_GM45	(dev_priv->info.display_mmio_offset + 0x70044)
-#define PIPE_FRMCOUNT_GM45(pipe) _PIPE(pipe, _PIPEA_FRMCOUNT_GM45, _PIPEB_FRMCOUNT_GM45)
+#define _PIPEA_FRMCOUNT_GM45	0x70040
+#define _PIPEA_FLIPCOUNT_GM45	0x70044
+#define PIPE_FRMCOUNT_GM45(pipe) _PIPE2(pipe, _PIPEA_FRMCOUNT_GM45)
+#define PIPE_FLIPCOUNT_GM45(pipe) _PIPE2(pipe, _PIPEA_FLIPCOUNT_GM45)
 
 /* Cursor A & B regs */
-#define _CURACNTR		(dev_priv->info.display_mmio_offset + 0x70080)
+#define _CURACNTR		0x70080
 /* Old style CUR*CNTR flags (desktop 8xx) */
 #define   CURSOR_ENABLE		0x80000000
 #define   CURSOR_GAMMA_ENABLE	0x40000000
@@ -3567,28 +3982,34 @@ enum punit_power_well {
 #define   MCURSOR_PIPE_B	(1 << 28)
 #define   MCURSOR_GAMMA_ENABLE  (1 << 26)
 #define   CURSOR_TRICKLE_FEED_DISABLE	(1 << 14)
-#define _CURABASE		(dev_priv->info.display_mmio_offset + 0x70084)
-#define _CURAPOS		(dev_priv->info.display_mmio_offset + 0x70088)
+#define _CURABASE		0x70084
+#define _CURAPOS		0x70088
 #define   CURSOR_POS_MASK       0x007FF
 #define   CURSOR_POS_SIGN       0x8000
 #define   CURSOR_X_SHIFT        0
 #define   CURSOR_Y_SHIFT        16
 #define CURSIZE			0x700a0
-#define _CURBCNTR		(dev_priv->info.display_mmio_offset + 0x700c0)
-#define _CURBBASE		(dev_priv->info.display_mmio_offset + 0x700c4)
-#define _CURBPOS		(dev_priv->info.display_mmio_offset + 0x700c8)
+#define _CURBCNTR		0x700c0
+#define _CURBBASE		0x700c4
+#define _CURBPOS		0x700c8
 
 #define _CURBCNTR_IVB		0x71080
 #define _CURBBASE_IVB		0x71084
 #define _CURBPOS_IVB		0x71088
 
-#define CURCNTR(pipe) _PIPE(pipe, _CURACNTR, _CURBCNTR)
-#define CURBASE(pipe) _PIPE(pipe, _CURABASE, _CURBBASE)
-#define CURPOS(pipe) _PIPE(pipe, _CURAPOS, _CURBPOS)
+#define _CURSOR2(pipe, reg) (dev_priv->info.cursor_offsets[(pipe)] - \
+	dev_priv->info.cursor_offsets[PIPE_A] + (reg) + \
+	dev_priv->info.display_mmio_offset)
+
+#define CURCNTR(pipe) _CURSOR2(pipe, _CURACNTR)
+#define CURBASE(pipe) _CURSOR2(pipe, _CURABASE)
+#define CURPOS(pipe) _CURSOR2(pipe, _CURAPOS)
 
-#define CURCNTR_IVB(pipe) _PIPE(pipe, _CURACNTR, _CURBCNTR_IVB)
-#define CURBASE_IVB(pipe) _PIPE(pipe, _CURABASE, _CURBBASE_IVB)
-#define CURPOS_IVB(pipe) _PIPE(pipe, _CURAPOS, _CURBPOS_IVB)
+#define CURSOR_A_OFFSET 0x70080
+#define CURSOR_B_OFFSET 0x700c0
+#define CHV_CURSOR_C_OFFSET 0x700e0
+#define IVB_CURSOR_B_OFFSET 0x71080
+#define IVB_CURSOR_C_OFFSET 0x72080
 
 /* Display A control */
 #define _DSPACNTR				0x70180
@@ -4093,6 +4514,7 @@ enum punit_power_well {
 #define  GEN8_DE_PIPE_A_IRQ		(1<<16)
 #define  GEN8_DE_PIPE_IRQ(pipe)		(1<<(16+pipe))
 #define  GEN8_GT_VECS_IRQ		(1<<6)
+#define  GEN8_GT_PM_IRQ			(1<<4)
 #define  GEN8_GT_VCS2_IRQ		(1<<3)
 #define  GEN8_GT_VCS1_IRQ		(1<<2)
 #define  GEN8_GT_BCS_IRQ		(1<<1)
@@ -4120,7 +4542,7 @@ enum punit_power_well {
 #define  GEN8_PIPE_SPRITE_FAULT		(1 << 9)
 #define  GEN8_PIPE_PRIMARY_FAULT	(1 << 8)
 #define  GEN8_PIPE_SPRITE_FLIP_DONE	(1 << 5)
-#define  GEN8_PIPE_FLIP_DONE		(1 << 4)
+#define  GEN8_PIPE_PRIMARY_FLIP_DONE	(1 << 4)
 #define  GEN8_PIPE_SCAN_LINE_EVENT	(1 << 2)
 #define  GEN8_PIPE_VSYNC		(1 << 1)
 #define  GEN8_PIPE_VBLANK		(1 << 0)
@@ -4832,6 +5254,8 @@ enum punit_power_well {
 #define  PORT_TRANS_SEL_CPT(pipe)	((pipe) << 29)
 #define  PORT_TO_PIPE(val)	(((val) & (1<<30)) >> 30)
 #define  PORT_TO_PIPE_CPT(val)	(((val) & PORT_TRANS_SEL_MASK) >> 29)
+#define  SDVO_PORT_TO_PIPE_CHV(val)	(((val) & (3<<24)) >> 24)
+#define  DP_PORT_TO_PIPE_CHV(val)	(((val) & (3<<16)) >> 16)
 
 #define TRANS_DP_CTL_A		0xe0300
 #define TRANS_DP_CTL_B		0xe1300
@@ -4888,6 +5312,8 @@ enum punit_power_well {
 
 #define  EDP_LINK_TRAIN_VOL_EMP_MASK_IVB	(0x3f<<22)
 
+#define  VLV_PMWGICZ				0x1300a4
+
 #define  FORCEWAKE				0xA18C
 #define  FORCEWAKE_VLV				0x1300b0
 #define  FORCEWAKE_ACK_VLV			0x1300b4
@@ -4896,15 +5322,22 @@ enum punit_power_well {
 #define  FORCEWAKE_ACK_HSW			0x130044
 #define  FORCEWAKE_ACK				0x130090
 #define  VLV_GTLC_WAKE_CTRL			0x130090
+#define   VLV_GTLC_RENDER_CTX_EXISTS		(1 << 25)
+#define   VLV_GTLC_MEDIA_CTX_EXISTS		(1 << 24)
+#define   VLV_GTLC_ALLOWWAKEREQ			(1 << 0)
+
 #define  VLV_GTLC_PW_STATUS			0x130094
-#define VLV_GTLC_PW_RENDER_STATUS_MASK		0x80
-#define VLV_GTLC_PW_MEDIA_STATUS_MASK		0x20
+#define   VLV_GTLC_ALLOWWAKEACK			(1 << 0)
+#define   VLV_GTLC_ALLOWWAKEERR			(1 << 1)
+#define   VLV_GTLC_PW_MEDIA_STATUS_MASK		(1 << 5)
+#define   VLV_GTLC_PW_RENDER_STATUS_MASK	(1 << 7)
 #define  FORCEWAKE_MT				0xa188 /* multi-threaded */
 #define   FORCEWAKE_KERNEL			0x1
 #define   FORCEWAKE_USER			0x2
 #define  FORCEWAKE_MT_ACK			0x130040
 #define  ECOBUS					0xa180
 #define    FORCEWAKE_MT_ENABLE			(1<<5)
+#define  VLV_SPAREG2H				0xA194
 
 #define  GTFIFODBG				0x120000
 #define    GT_FIFO_SBDROPERR			(1<<6)
@@ -4924,6 +5357,7 @@ enum punit_power_well {
 #define  HSW_EDRAM_PRESENT			0x120010
 
 #define GEN6_UCGCTL1				0x9400
+# define GEN6_EU_TCUNIT_CLOCK_GATE_DISABLE		(1 << 16)
 # define GEN6_BLBUNIT_CLOCK_GATE_DISABLE		(1 << 5)
 # define GEN6_CSUNIT_CLOCK_GATE_DISABLE			(1 << 7)
 
@@ -4934,12 +5368,19 @@ enum punit_power_well {
 # define GEN6_RCPBUNIT_CLOCK_GATE_DISABLE		(1 << 12)
 # define GEN6_RCCUNIT_CLOCK_GATE_DISABLE		(1 << 11)
 
+#define GEN6_UCGCTL3				0x9408
+
 #define GEN7_UCGCTL4				0x940c
 #define  GEN7_L3BANK2X_CLOCK_GATE_DISABLE	(1<<25)
 
+#define GEN6_RCGCTL1				0x9410
+#define GEN6_RCGCTL2				0x9414
+#define GEN6_RSTCTL				0x9420
+
 #define GEN8_UCGCTL6				0x9430
 #define   GEN8_SDEUNIT_CLOCK_GATE_DISABLE	(1<<14)
 
+#define GEN6_GFXPAUSE				0xA000
 #define GEN6_RPNSWREQ				0xA008
 #define   GEN6_TURBO_DISABLE			(1<<31)
 #define   GEN6_FREQUENCY(x)			((x)<<25)
@@ -4992,6 +5433,9 @@ enum punit_power_well {
 #define GEN6_RP_UP_EI				0xA068
 #define GEN6_RP_DOWN_EI				0xA06C
 #define GEN6_RP_IDLE_HYSTERSIS			0xA070
+#define GEN6_RPDEUHWTC				0xA080
+#define GEN6_RPDEUC				0xA084
+#define GEN6_RPDEUCSW				0xA088
 #define GEN6_RC_STATE				0xA094
 #define GEN6_RC1_WAKE_RATE_LIMIT		0xA098
 #define GEN6_RC6_WAKE_RATE_LIMIT		0xA09C
@@ -4999,11 +5443,15 @@ enum punit_power_well {
 #define GEN6_RC_EVALUATION_INTERVAL		0xA0A8
 #define GEN6_RC_IDLE_HYSTERSIS			0xA0AC
 #define GEN6_RC_SLEEP				0xA0B0
+#define GEN6_RCUBMABDTMR			0xA0B0
 #define GEN6_RC1e_THRESHOLD			0xA0B4
 #define GEN6_RC6_THRESHOLD			0xA0B8
 #define GEN6_RC6p_THRESHOLD			0xA0BC
+#define VLV_RCEDATA				0xA0BC
 #define GEN6_RC6pp_THRESHOLD			0xA0C0
 #define GEN6_PMINTRMSK				0xA168
+#define GEN8_PMINTR_REDIRECT_TO_NON_DISP	(1<<31)
+#define VLV_PWRDWNUPCTL				0xA294
 
 #define GEN6_PMISR				0x44020
 #define GEN6_PMIMR				0x44024 /* rps_lock */
@@ -5020,6 +5468,9 @@ enum punit_power_well {
 						 GEN6_PM_RP_DOWN_THRESHOLD | \
 						 GEN6_PM_RP_DOWN_TIMEOUT)
 
+#define GEN7_GT_SCRATCH_BASE			0x4F100
+#define GEN7_GT_SCRATCH_REG_NUM			8
+
 #define VLV_GTLC_SURVIVABILITY_REG              0x130098
 #define VLV_GFX_CLK_STATUS_BIT			(1<<3)
 #define VLV_GFX_CLK_FORCE_ON_BIT		(1<<2)
@@ -5030,6 +5481,9 @@ enum punit_power_well {
 #define   VLV_MEDIA_RC6_COUNT_EN		(1<<1)
 #define   VLV_RENDER_RC6_COUNT_EN		(1<<0)
 #define GEN6_GT_GFX_RC6				0x138108
+#define VLV_GT_RENDER_RC6			0x138108
+#define VLV_GT_MEDIA_RC6			0x13810C
+
 #define GEN6_GT_GFX_RC6p			0x13810C
 #define GEN6_GT_GFX_RC6pp			0x138110
 
diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c
index 56785e8fb2eb..043123c77a1f 100644
--- a/drivers/gpu/drm/i915/i915_suspend.c
+++ b/drivers/gpu/drm/i915/i915_suspend.c
@@ -328,8 +328,6 @@ int i915_save_state(struct drm_device *dev)
 		}
 	}
 
-	intel_disable_gt_powersave(dev);
-
 	/* Cache mode state */
 	if (INTEL_INFO(dev)->gen < 7)
 		dev_priv->regfile.saveCACHE_MODE_0 = I915_READ(CACHE_MODE_0);
diff --git a/drivers/gpu/drm/i915/i915_sysfs.c b/drivers/gpu/drm/i915/i915_sysfs.c
index 9c57029f6f4b..86ce39aad0ff 100644
--- a/drivers/gpu/drm/i915/i915_sysfs.c
+++ b/drivers/gpu/drm/i915/i915_sysfs.c
@@ -186,7 +186,7 @@ i915_l3_write(struct file *filp, struct kobject *kobj,
 	struct drm_minor *dminor = dev_to_drm_minor(dev);
 	struct drm_device *drm_dev = dminor->dev;
 	struct drm_i915_private *dev_priv = drm_dev->dev_private;
-	struct i915_hw_context *ctx;
+	struct intel_context *ctx;
 	u32 *temp = NULL; /* Just here to make handling failures easy */
 	int slice = (int)(uintptr_t)attr->private;
 	int ret;
@@ -263,6 +263,8 @@ static ssize_t gt_cur_freq_mhz_show(struct device *kdev,
 
 	flush_delayed_work(&dev_priv->rps.delayed_resume_work);
 
+	intel_runtime_pm_get(dev_priv);
+
 	mutex_lock(&dev_priv->rps.hw_lock);
 	if (IS_VALLEYVIEW(dev_priv->dev)) {
 		u32 freq;
@@ -273,6 +275,8 @@ static ssize_t gt_cur_freq_mhz_show(struct device *kdev,
 	}
 	mutex_unlock(&dev_priv->rps.hw_lock);
 
+	intel_runtime_pm_put(dev_priv);
+
 	return snprintf(buf, PAGE_SIZE, "%d\n", ret);
 }
 
diff --git a/drivers/gpu/drm/i915/i915_trace.h b/drivers/gpu/drm/i915/i915_trace.h
index 23c26f1f8b37..f5aa0067755a 100644
--- a/drivers/gpu/drm/i915/i915_trace.h
+++ b/drivers/gpu/drm/i915/i915_trace.h
@@ -7,6 +7,7 @@
 
 #include <drm/drmP.h>
 #include "i915_drv.h"
+#include "intel_drv.h"
 #include "intel_ringbuffer.h"
 
 #undef TRACE_SYSTEM
@@ -14,6 +15,80 @@
 #define TRACE_SYSTEM_STRING __stringify(TRACE_SYSTEM)
 #define TRACE_INCLUDE_FILE i915_trace
 
+/* pipe updates */
+
+TRACE_EVENT(i915_pipe_update_start,
+	    TP_PROTO(struct intel_crtc *crtc, u32 min, u32 max),
+	    TP_ARGS(crtc, min, max),
+
+	    TP_STRUCT__entry(
+			     __field(enum pipe, pipe)
+			     __field(u32, frame)
+			     __field(u32, scanline)
+			     __field(u32, min)
+			     __field(u32, max)
+			     ),
+
+	    TP_fast_assign(
+			   __entry->pipe = crtc->pipe;
+			   __entry->frame = crtc->base.dev->driver->get_vblank_counter(crtc->base.dev,
+										       crtc->pipe);
+			   __entry->scanline = intel_get_crtc_scanline(crtc);
+			   __entry->min = min;
+			   __entry->max = max;
+			   ),
+
+	    TP_printk("pipe %c, frame=%u, scanline=%u, min=%u, max=%u",
+		      pipe_name(__entry->pipe), __entry->frame,
+		       __entry->scanline, __entry->min, __entry->max)
+);
+
+TRACE_EVENT(i915_pipe_update_vblank_evaded,
+	    TP_PROTO(struct intel_crtc *crtc, u32 min, u32 max, u32 frame),
+	    TP_ARGS(crtc, min, max, frame),
+
+	    TP_STRUCT__entry(
+			     __field(enum pipe, pipe)
+			     __field(u32, frame)
+			     __field(u32, scanline)
+			     __field(u32, min)
+			     __field(u32, max)
+			     ),
+
+	    TP_fast_assign(
+			   __entry->pipe = crtc->pipe;
+			   __entry->frame = frame;
+			   __entry->scanline = intel_get_crtc_scanline(crtc);
+			   __entry->min = min;
+			   __entry->max = max;
+			   ),
+
+	    TP_printk("pipe %c, frame=%u, scanline=%u, min=%u, max=%u",
+		      pipe_name(__entry->pipe), __entry->frame,
+		       __entry->scanline, __entry->min, __entry->max)
+);
+
+TRACE_EVENT(i915_pipe_update_end,
+	    TP_PROTO(struct intel_crtc *crtc, u32 frame),
+	    TP_ARGS(crtc, frame),
+
+	    TP_STRUCT__entry(
+			     __field(enum pipe, pipe)
+			     __field(u32, frame)
+			     __field(u32, scanline)
+			     ),
+
+	    TP_fast_assign(
+			   __entry->pipe = crtc->pipe;
+			   __entry->frame = frame;
+			   __entry->scanline = intel_get_crtc_scanline(crtc);
+			   ),
+
+	    TP_printk("pipe %c, frame=%u, scanline=%u",
+		      pipe_name(__entry->pipe), __entry->frame,
+		      __entry->scanline)
+);
+
 /* object tracking */
 
 TRACE_EVENT(i915_gem_object_create,
@@ -251,8 +326,8 @@ TRACE_EVENT(i915_gem_evict_vm,
 );
 
 TRACE_EVENT(i915_gem_ring_sync_to,
-	    TP_PROTO(struct intel_ring_buffer *from,
-		     struct intel_ring_buffer *to,
+	    TP_PROTO(struct intel_engine_cs *from,
+		     struct intel_engine_cs *to,
 		     u32 seqno),
 	    TP_ARGS(from, to, seqno),
 
@@ -277,7 +352,7 @@ TRACE_EVENT(i915_gem_ring_sync_to,
 );
 
 TRACE_EVENT(i915_gem_ring_dispatch,
-	    TP_PROTO(struct intel_ring_buffer *ring, u32 seqno, u32 flags),
+	    TP_PROTO(struct intel_engine_cs *ring, u32 seqno, u32 flags),
 	    TP_ARGS(ring, seqno, flags),
 
 	    TP_STRUCT__entry(
@@ -300,7 +375,7 @@ TRACE_EVENT(i915_gem_ring_dispatch,
 );
 
 TRACE_EVENT(i915_gem_ring_flush,
-	    TP_PROTO(struct intel_ring_buffer *ring, u32 invalidate, u32 flush),
+	    TP_PROTO(struct intel_engine_cs *ring, u32 invalidate, u32 flush),
 	    TP_ARGS(ring, invalidate, flush),
 
 	    TP_STRUCT__entry(
@@ -323,7 +398,7 @@ TRACE_EVENT(i915_gem_ring_flush,
 );
 
 DECLARE_EVENT_CLASS(i915_gem_request,
-	    TP_PROTO(struct intel_ring_buffer *ring, u32 seqno),
+	    TP_PROTO(struct intel_engine_cs *ring, u32 seqno),
 	    TP_ARGS(ring, seqno),
 
 	    TP_STRUCT__entry(
@@ -343,12 +418,12 @@ DECLARE_EVENT_CLASS(i915_gem_request,
 );
 
 DEFINE_EVENT(i915_gem_request, i915_gem_request_add,
-	    TP_PROTO(struct intel_ring_buffer *ring, u32 seqno),
+	    TP_PROTO(struct intel_engine_cs *ring, u32 seqno),
 	    TP_ARGS(ring, seqno)
 );
 
 TRACE_EVENT(i915_gem_request_complete,
-	    TP_PROTO(struct intel_ring_buffer *ring),
+	    TP_PROTO(struct intel_engine_cs *ring),
 	    TP_ARGS(ring),
 
 	    TP_STRUCT__entry(
@@ -368,12 +443,12 @@ TRACE_EVENT(i915_gem_request_complete,
 );
 
 DEFINE_EVENT(i915_gem_request, i915_gem_request_retire,
-	    TP_PROTO(struct intel_ring_buffer *ring, u32 seqno),
+	    TP_PROTO(struct intel_engine_cs *ring, u32 seqno),
 	    TP_ARGS(ring, seqno)
 );
 
 TRACE_EVENT(i915_gem_request_wait_begin,
-	    TP_PROTO(struct intel_ring_buffer *ring, u32 seqno),
+	    TP_PROTO(struct intel_engine_cs *ring, u32 seqno),
 	    TP_ARGS(ring, seqno),
 
 	    TP_STRUCT__entry(
@@ -402,12 +477,12 @@ TRACE_EVENT(i915_gem_request_wait_begin,
 );
 
 DEFINE_EVENT(i915_gem_request, i915_gem_request_wait_end,
-	    TP_PROTO(struct intel_ring_buffer *ring, u32 seqno),
+	    TP_PROTO(struct intel_engine_cs *ring, u32 seqno),
 	    TP_ARGS(ring, seqno)
 );
 
 DECLARE_EVENT_CLASS(i915_ring,
-	    TP_PROTO(struct intel_ring_buffer *ring),
+	    TP_PROTO(struct intel_engine_cs *ring),
 	    TP_ARGS(ring),
 
 	    TP_STRUCT__entry(
@@ -424,12 +499,12 @@ DECLARE_EVENT_CLASS(i915_ring,
 );
 
 DEFINE_EVENT(i915_ring, i915_ring_wait_begin,
-	    TP_PROTO(struct intel_ring_buffer *ring),
+	    TP_PROTO(struct intel_engine_cs *ring),
 	    TP_ARGS(ring)
 );
 
 DEFINE_EVENT(i915_ring, i915_ring_wait_end,
-	    TP_PROTO(struct intel_ring_buffer *ring),
+	    TP_PROTO(struct intel_engine_cs *ring),
 	    TP_ARGS(ring)
 );
 
diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c
index aff4a113cda3..1ee98f121a00 100644
--- a/drivers/gpu/drm/i915/intel_bios.c
+++ b/drivers/gpu/drm/i915/intel_bios.c
@@ -49,13 +49,19 @@ find_section(struct bdb_header *bdb, int section_id)
 	total = bdb->bdb_size;
 
 	/* walk the sections looking for section_id */
-	while (index < total) {
+	while (index + 3 < total) {
 		current_id = *(base + index);
 		index++;
+
 		current_size = *((u16 *)(base + index));
 		index += 2;
+
+		if (index + current_size > total)
+			return NULL;
+
 		if (current_id == section_id)
 			return base + index;
+
 		index += current_size;
 	}
 
@@ -206,7 +212,7 @@ parse_lfp_panel_data(struct drm_i915_private *dev_priv,
 	const struct lvds_dvo_timing *panel_dvo_timing;
 	const struct lvds_fp_timing *fp_timing;
 	struct drm_display_mode *panel_fixed_mode;
-	int i, downclock;
+	int i, downclock, drrs_mode;
 
 	lvds_options = find_section(bdb, BDB_LVDS_OPTIONS);
 	if (!lvds_options)
@@ -218,6 +224,28 @@ parse_lfp_panel_data(struct drm_i915_private *dev_priv,
 
 	panel_type = lvds_options->panel_type;
 
+	drrs_mode = (lvds_options->dps_panel_type_bits
+				>> (panel_type * 2)) & MODE_MASK;
+	/*
+	 * VBT has static DRRS = 0 and seamless DRRS = 2.
+	 * The below piece of code is required to adjust vbt.drrs_type
+	 * to match the enum drrs_support_type.
+	 */
+	switch (drrs_mode) {
+	case 0:
+		dev_priv->vbt.drrs_type = STATIC_DRRS_SUPPORT;
+		DRM_DEBUG_KMS("DRRS supported mode is static\n");
+		break;
+	case 2:
+		dev_priv->vbt.drrs_type = SEAMLESS_DRRS_SUPPORT;
+		DRM_DEBUG_KMS("DRRS supported mode is seamless\n");
+		break;
+	default:
+		dev_priv->vbt.drrs_type = DRRS_NOT_SUPPORTED;
+		DRM_DEBUG_KMS("DRRS not supported (VBT input)\n");
+		break;
+	}
+
 	lvds_lfp_data = find_section(bdb, BDB_LVDS_LFP_DATA);
 	if (!lvds_lfp_data)
 		return;
@@ -526,6 +554,16 @@ parse_driver_features(struct drm_i915_private *dev_priv,
 
 	if (driver->dual_frequency)
 		dev_priv->render_reclock_avail = true;
+
+	DRM_DEBUG_KMS("DRRS State Enabled:%d\n", driver->drrs_enabled);
+	/*
+	 * If DRRS is not supported, drrs_type has to be set to 0.
+	 * This is because, VBT is configured in such a way that
+	 * static DRRS is 0 and DRRS not supported is represented by
+	 * driver->drrs_enabled=false
+	 */
+	if (!driver->drrs_enabled)
+		dev_priv->vbt.drrs_type = DRRS_NOT_SUPPORTED;
 }
 
 static void
@@ -628,19 +666,221 @@ parse_edp(struct drm_i915_private *dev_priv, struct bdb_header *bdb)
 	}
 }
 
+static u8 *goto_next_sequence(u8 *data, int *size)
+{
+	u16 len;
+	int tmp = *size;
+
+	if (--tmp < 0)
+		return NULL;
+
+	/* goto first element */
+	data++;
+	while (1) {
+		switch (*data) {
+		case MIPI_SEQ_ELEM_SEND_PKT:
+			/*
+			 * skip by this element payload size
+			 * skip elem id, command flag and data type
+			 */
+			tmp -= 5;
+			if (tmp < 0)
+				return NULL;
+
+			data += 3;
+			len = *((u16 *)data);
+
+			tmp -= len;
+			if (tmp < 0)
+				return NULL;
+
+			/* skip by len */
+			data = data + 2 + len;
+			break;
+		case MIPI_SEQ_ELEM_DELAY:
+			/* skip by elem id, and delay is 4 bytes */
+			tmp -= 5;
+			if (tmp < 0)
+				return NULL;
+
+			data += 5;
+			break;
+		case MIPI_SEQ_ELEM_GPIO:
+			tmp -= 3;
+			if (tmp < 0)
+				return NULL;
+
+			data += 3;
+			break;
+		default:
+			DRM_ERROR("Unknown element\n");
+			return NULL;
+		}
+
+		/* end of sequence ? */
+		if (*data == 0)
+			break;
+	}
+
+	/* goto next sequence or end of block byte */
+	if (--tmp < 0)
+		return NULL;
+
+	data++;
+
+	/* update amount of data left for the sequence block to be parsed */
+	*size = tmp;
+	return data;
+}
+
 static void
 parse_mipi(struct drm_i915_private *dev_priv, struct bdb_header *bdb)
 {
-	struct bdb_mipi *mipi;
+	struct bdb_mipi_config *start;
+	struct bdb_mipi_sequence *sequence;
+	struct mipi_config *config;
+	struct mipi_pps_data *pps;
+	u8 *data, *seq_data;
+	int i, panel_id, seq_size;
+	u16 block_size;
+
+	/* parse MIPI blocks only if LFP type is MIPI */
+	if (!dev_priv->vbt.has_mipi)
+		return;
+
+	/* Initialize this to undefined indicating no generic MIPI support */
+	dev_priv->vbt.dsi.panel_id = MIPI_DSI_UNDEFINED_PANEL_ID;
+
+	/* Block #40 is already parsed and panel_fixed_mode is
+	 * stored in dev_priv->lfp_lvds_vbt_mode
+	 * resuse this when needed
+	 */
+
+	/* Parse #52 for panel index used from panel_type already
+	 * parsed
+	 */
+	start = find_section(bdb, BDB_MIPI_CONFIG);
+	if (!start) {
+		DRM_DEBUG_KMS("No MIPI config BDB found");
+		return;
+	}
+
+	DRM_DEBUG_DRIVER("Found MIPI Config block, panel index = %d\n",
+								panel_type);
 
-	mipi = find_section(bdb, BDB_MIPI_CONFIG);
-	if (!mipi) {
-		DRM_DEBUG_KMS("No MIPI BDB found");
+	/*
+	 * get hold of the correct configuration block and pps data as per
+	 * the panel_type as index
+	 */
+	config = &start->config[panel_type];
+	pps = &start->pps[panel_type];
+
+	/* store as of now full data. Trim when we realise all is not needed */
+	dev_priv->vbt.dsi.config = kmemdup(config, sizeof(struct mipi_config), GFP_KERNEL);
+	if (!dev_priv->vbt.dsi.config)
+		return;
+
+	dev_priv->vbt.dsi.pps = kmemdup(pps, sizeof(struct mipi_pps_data), GFP_KERNEL);
+	if (!dev_priv->vbt.dsi.pps) {
+		kfree(dev_priv->vbt.dsi.config);
 		return;
 	}
 
-	/* XXX: add more info */
+	/* We have mandatory mipi config blocks. Initialize as generic panel */
 	dev_priv->vbt.dsi.panel_id = MIPI_DSI_GENERIC_PANEL_ID;
+
+	/* Check if we have sequence block as well */
+	sequence = find_section(bdb, BDB_MIPI_SEQUENCE);
+	if (!sequence) {
+		DRM_DEBUG_KMS("No MIPI Sequence found, parsing complete\n");
+		return;
+	}
+
+	DRM_DEBUG_DRIVER("Found MIPI sequence block\n");
+
+	block_size = get_blocksize(sequence);
+
+	/*
+	 * parse the sequence block for individual sequences
+	 */
+	dev_priv->vbt.dsi.seq_version = sequence->version;
+
+	seq_data = &sequence->data[0];
+
+	/*
+	 * sequence block is variable length and hence we need to parse and
+	 * get the sequence data for specific panel id
+	 */
+	for (i = 0; i < MAX_MIPI_CONFIGURATIONS; i++) {
+		panel_id = *seq_data;
+		seq_size = *((u16 *) (seq_data + 1));
+		if (panel_id == panel_type)
+			break;
+
+		/* skip the sequence including seq header of 3 bytes */
+		seq_data = seq_data + 3 + seq_size;
+		if ((seq_data - &sequence->data[0]) > block_size) {
+			DRM_ERROR("Sequence start is beyond sequence block size, corrupted sequence block\n");
+			return;
+		}
+	}
+
+	if (i == MAX_MIPI_CONFIGURATIONS) {
+		DRM_ERROR("Sequence block detected but no valid configuration\n");
+		return;
+	}
+
+	/* check if found sequence is completely within the sequence block
+	 * just being paranoid */
+	if (seq_size > block_size) {
+		DRM_ERROR("Corrupted sequence/size, bailing out\n");
+		return;
+	}
+
+	/* skip the panel id(1 byte) and seq size(2 bytes) */
+	dev_priv->vbt.dsi.data = kmemdup(seq_data + 3, seq_size, GFP_KERNEL);
+	if (!dev_priv->vbt.dsi.data)
+		return;
+
+	/*
+	 * loop into the sequence data and split into multiple sequneces
+	 * There are only 5 types of sequences as of now
+	 */
+	data = dev_priv->vbt.dsi.data;
+	dev_priv->vbt.dsi.size = seq_size;
+
+	/* two consecutive 0x00 indicate end of all sequences */
+	while (1) {
+		int seq_id = *data;
+		if (MIPI_SEQ_MAX > seq_id && seq_id > MIPI_SEQ_UNDEFINED) {
+			dev_priv->vbt.dsi.sequence[seq_id] = data;
+			DRM_DEBUG_DRIVER("Found mipi sequence - %d\n", seq_id);
+		} else {
+			DRM_ERROR("undefined sequence\n");
+			goto err;
+		}
+
+		/* partial parsing to skip elements */
+		data = goto_next_sequence(data, &seq_size);
+
+		if (data == NULL) {
+			DRM_ERROR("Sequence elements going beyond block itself. Sequence block parsing failed\n");
+			goto err;
+		}
+
+		if (*data == 0)
+			break; /* end of sequence reached */
+	}
+
+	DRM_DEBUG_DRIVER("MIPI related vbt parsing complete\n");
+	return;
+err:
+	kfree(dev_priv->vbt.dsi.data);
+	dev_priv->vbt.dsi.data = NULL;
+
+	/* error during parsing so set all pointers to null
+	 * because of partial parsing */
+	memset(dev_priv->vbt.dsi.sequence, 0, MIPI_SEQ_MAX);
 }
 
 static void parse_ddi_port(struct drm_i915_private *dev_priv, enum port port,
@@ -823,6 +1063,15 @@ parse_device_mapping(struct drm_i915_private *dev_priv,
 			/* skip the device block if device type is invalid */
 			continue;
 		}
+
+		if (p_child->common.dvo_port >= DVO_PORT_MIPIA
+		    && p_child->common.dvo_port <= DVO_PORT_MIPID
+		    &&p_child->common.device_type & DEVICE_TYPE_MIPI_OUTPUT) {
+			DRM_DEBUG_KMS("Found MIPI as LFP\n");
+			dev_priv->vbt.has_mipi = 1;
+			dev_priv->vbt.dsi.port = p_child->common.dvo_port;
+		}
+
 		child_dev_ptr = dev_priv->vbt.child_dev + count;
 		count++;
 		memcpy((void *)child_dev_ptr, (void *)p_child,
@@ -893,6 +1142,46 @@ static const struct dmi_system_id intel_no_opregion_vbt[] = {
 	{ }
 };
 
+static struct bdb_header *validate_vbt(char *base, size_t size,
+				       struct vbt_header *vbt,
+				       const char *source)
+{
+	size_t offset;
+	struct bdb_header *bdb;
+
+	if (vbt == NULL) {
+		DRM_DEBUG_DRIVER("VBT signature missing\n");
+		return NULL;
+	}
+
+	offset = (char *)vbt - base;
+	if (offset + sizeof(struct vbt_header) > size) {
+		DRM_DEBUG_DRIVER("VBT header incomplete\n");
+		return NULL;
+	}
+
+	if (memcmp(vbt->signature, "$VBT", 4)) {
+		DRM_DEBUG_DRIVER("VBT invalid signature\n");
+		return NULL;
+	}
+
+	offset += vbt->bdb_offset;
+	if (offset + sizeof(struct bdb_header) > size) {
+		DRM_DEBUG_DRIVER("BDB header incomplete\n");
+		return NULL;
+	}
+
+	bdb = (struct bdb_header *)(base + offset);
+	if (offset + bdb->bdb_size > size) {
+		DRM_DEBUG_DRIVER("BDB incomplete\n");
+		return NULL;
+	}
+
+	DRM_DEBUG_KMS("Using VBT from %s: %20s\n",
+		      source, vbt->signature);
+	return bdb;
+}
+
 /**
  * intel_parse_bios - find VBT and initialize settings from the BIOS
  * @dev: DRM device
@@ -916,20 +1205,13 @@ intel_parse_bios(struct drm_device *dev)
 	init_vbt_defaults(dev_priv);
 
 	/* XXX Should this validation be moved to intel_opregion.c? */
-	if (!dmi_check_system(intel_no_opregion_vbt) && dev_priv->opregion.vbt) {
-		struct vbt_header *vbt = dev_priv->opregion.vbt;
-		if (memcmp(vbt->signature, "$VBT", 4) == 0) {
-			DRM_DEBUG_KMS("Using VBT from OpRegion: %20s\n",
-					 vbt->signature);
-			bdb = (struct bdb_header *)((char *)vbt + vbt->bdb_offset);
-		} else
-			dev_priv->opregion.vbt = NULL;
-	}
+	if (!dmi_check_system(intel_no_opregion_vbt) && dev_priv->opregion.vbt)
+		bdb = validate_vbt((char *)dev_priv->opregion.header, OPREGION_SIZE,
+				   (struct vbt_header *)dev_priv->opregion.vbt,
+				   "OpRegion");
 
 	if (bdb == NULL) {
-		struct vbt_header *vbt = NULL;
-		size_t size;
-		int i;
+		size_t i, size;
 
 		bios = pci_map_rom(pdev, &size);
 		if (!bios)
@@ -937,19 +1219,18 @@ intel_parse_bios(struct drm_device *dev)
 
 		/* Scour memory looking for the VBT signature */
 		for (i = 0; i + 4 < size; i++) {
-			if (!memcmp(bios + i, "$VBT", 4)) {
-				vbt = (struct vbt_header *)(bios + i);
+			if (memcmp(bios + i, "$VBT", 4) == 0) {
+				bdb = validate_vbt(bios, size,
+						   (struct vbt_header *)(bios + i),
+						   "PCI ROM");
 				break;
 			}
 		}
 
-		if (!vbt) {
-			DRM_DEBUG_DRIVER("VBT signature missing\n");
+		if (!bdb) {
 			pci_unmap_rom(pdev, bios);
 			return -1;
 		}
-
-		bdb = (struct bdb_header *)(bios + i + vbt->bdb_offset);
 	}
 
 	/* Grab useful general definitions */
diff --git a/drivers/gpu/drm/i915/intel_bios.h b/drivers/gpu/drm/i915/intel_bios.h
index f27f7b282465..b98667796337 100644
--- a/drivers/gpu/drm/i915/intel_bios.h
+++ b/drivers/gpu/drm/i915/intel_bios.h
@@ -282,6 +282,9 @@ struct bdb_general_definitions {
 	union child_device_config devices[0];
 } __packed;
 
+/* Mask for DRRS / Panel Channel / SSC / BLT control bits extraction */
+#define MODE_MASK		0x3
+
 struct bdb_lvds_options {
 	u8 panel_type;
 	u8 rsvd1;
@@ -294,6 +297,18 @@ struct bdb_lvds_options {
 	u8 lvds_edid:1;
 	u8 rsvd2:1;
 	u8 rsvd4;
+	/* LVDS Panel channel bits stored here */
+	u32 lvds_panel_channel_bits;
+	/* LVDS SSC (Spread Spectrum Clock) bits stored here. */
+	u16 ssc_bits;
+	u16 ssc_freq;
+	u16 ssc_ddt;
+	/* Panel color depth defined here */
+	u16 panel_color_depth;
+	/* LVDS panel type bits stored here */
+	u32 dps_panel_type_bits;
+	/* LVDS backlight control type bits stored here */
+	u32 blt_control_type_bits;
 } __packed;
 
 /* LFP pointer table contains entries to the struct below */
@@ -482,6 +497,20 @@ struct bdb_driver_features {
 
 	u8 hdmi_termination;
 	u8 custom_vbt_version;
+	/* Driver features data block */
+	u16 rmpm_enabled:1;
+	u16 s2ddt_enabled:1;
+	u16 dpst_enabled:1;
+	u16 bltclt_enabled:1;
+	u16 adb_enabled:1;
+	u16 drrs_enabled:1;
+	u16 grs_enabled:1;
+	u16 gpmt_enabled:1;
+	u16 tbt_enabled:1;
+	u16 psr_enabled:1;
+	u16 ips_enabled:1;
+	u16 reserved3:4;
+	u16 pc_feature_valid:1;
 } __packed;
 
 #define EDP_18BPP	0
@@ -714,6 +743,10 @@ int intel_parse_bios(struct drm_device *dev);
 #define DVO_PORT_DPC	8
 #define DVO_PORT_DPD	9
 #define DVO_PORT_DPA	10
+#define DVO_PORT_MIPIA	21
+#define DVO_PORT_MIPIB	22
+#define DVO_PORT_MIPIC	23
+#define DVO_PORT_MIPID	24
 
 /* Block 52 contains MIPI Panel info
  * 6 such enteries will there. Index into correct
@@ -870,4 +903,35 @@ struct bdb_mipi_sequence {
 	u8 data[0];
 };
 
+/* MIPI Sequnece Block definitions */
+enum mipi_seq {
+	MIPI_SEQ_UNDEFINED = 0,
+	MIPI_SEQ_ASSERT_RESET,
+	MIPI_SEQ_INIT_OTP,
+	MIPI_SEQ_DISPLAY_ON,
+	MIPI_SEQ_DISPLAY_OFF,
+	MIPI_SEQ_DEASSERT_RESET,
+	MIPI_SEQ_MAX
+};
+
+enum mipi_seq_element {
+	MIPI_SEQ_ELEM_UNDEFINED = 0,
+	MIPI_SEQ_ELEM_SEND_PKT,
+	MIPI_SEQ_ELEM_DELAY,
+	MIPI_SEQ_ELEM_GPIO,
+	MIPI_SEQ_ELEM_STATUS,
+	MIPI_SEQ_ELEM_MAX
+};
+
+enum mipi_gpio_pin_index {
+	MIPI_GPIO_UNDEFINED = 0,
+	MIPI_GPIO_PANEL_ENABLE,
+	MIPI_GPIO_BL_ENABLE,
+	MIPI_GPIO_PWM_ENABLE,
+	MIPI_GPIO_RESET_N,
+	MIPI_GPIO_PWR_DOWN_R,
+	MIPI_GPIO_STDBY_RST_N,
+	MIPI_GPIO_MAX
+};
+
 #endif /* _I830_BIOS_H_ */
diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c
index aa5a3dc43342..5a045d3bd77e 100644
--- a/drivers/gpu/drm/i915/intel_crt.c
+++ b/drivers/gpu/drm/i915/intel_crt.c
@@ -144,28 +144,49 @@ static void intel_crt_set_dpms(struct intel_encoder *encoder, int mode)
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crt *crt = intel_encoder_to_crt(encoder);
-	u32 temp;
+	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
+	struct drm_display_mode *adjusted_mode = &crtc->config.adjusted_mode;
+	u32 adpa;
+
+	if (INTEL_INFO(dev)->gen >= 5)
+		adpa = ADPA_HOTPLUG_BITS;
+	else
+		adpa = 0;
 
-	temp = I915_READ(crt->adpa_reg);
-	temp &= ~(ADPA_HSYNC_CNTL_DISABLE | ADPA_VSYNC_CNTL_DISABLE);
-	temp &= ~ADPA_DAC_ENABLE;
+	if (adjusted_mode->flags & DRM_MODE_FLAG_PHSYNC)
+		adpa |= ADPA_HSYNC_ACTIVE_HIGH;
+	if (adjusted_mode->flags & DRM_MODE_FLAG_PVSYNC)
+		adpa |= ADPA_VSYNC_ACTIVE_HIGH;
+
+	/* For CPT allow 3 pipe config, for others just use A or B */
+	if (HAS_PCH_LPT(dev))
+		; /* Those bits don't exist here */
+	else if (HAS_PCH_CPT(dev))
+		adpa |= PORT_TRANS_SEL_CPT(crtc->pipe);
+	else if (crtc->pipe == 0)
+		adpa |= ADPA_PIPE_A_SELECT;
+	else
+		adpa |= ADPA_PIPE_B_SELECT;
+
+	if (!HAS_PCH_SPLIT(dev))
+		I915_WRITE(BCLRPAT(crtc->pipe), 0);
 
 	switch (mode) {
 	case DRM_MODE_DPMS_ON:
-		temp |= ADPA_DAC_ENABLE;
+		adpa |= ADPA_DAC_ENABLE;
 		break;
 	case DRM_MODE_DPMS_STANDBY:
-		temp |= ADPA_DAC_ENABLE | ADPA_HSYNC_CNTL_DISABLE;
+		adpa |= ADPA_DAC_ENABLE | ADPA_HSYNC_CNTL_DISABLE;
 		break;
 	case DRM_MODE_DPMS_SUSPEND:
-		temp |= ADPA_DAC_ENABLE | ADPA_VSYNC_CNTL_DISABLE;
+		adpa |= ADPA_DAC_ENABLE | ADPA_VSYNC_CNTL_DISABLE;
 		break;
 	case DRM_MODE_DPMS_OFF:
-		temp |= ADPA_HSYNC_CNTL_DISABLE | ADPA_VSYNC_CNTL_DISABLE;
+		adpa |= ADPA_HSYNC_CNTL_DISABLE | ADPA_VSYNC_CNTL_DISABLE;
 		break;
 	}
 
-	I915_WRITE(crt->adpa_reg, temp);
+	I915_WRITE(crt->adpa_reg, adpa);
 }
 
 static void intel_disable_crt(struct intel_encoder *encoder)
@@ -274,42 +295,6 @@ static bool intel_crt_compute_config(struct intel_encoder *encoder,
 	return true;
 }
 
-static void intel_crt_mode_set(struct intel_encoder *encoder)
-{
-
-	struct drm_device *dev = encoder->base.dev;
-	struct intel_crt *crt = intel_encoder_to_crt(encoder);
-	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct drm_display_mode *adjusted_mode = &crtc->config.adjusted_mode;
-	u32 adpa;
-
-	if (INTEL_INFO(dev)->gen >= 5)
-		adpa = ADPA_HOTPLUG_BITS;
-	else
-		adpa = 0;
-
-	if (adjusted_mode->flags & DRM_MODE_FLAG_PHSYNC)
-		adpa |= ADPA_HSYNC_ACTIVE_HIGH;
-	if (adjusted_mode->flags & DRM_MODE_FLAG_PVSYNC)
-		adpa |= ADPA_VSYNC_ACTIVE_HIGH;
-
-	/* For CPT allow 3 pipe config, for others just use A or B */
-	if (HAS_PCH_LPT(dev))
-		; /* Those bits don't exist here */
-	else if (HAS_PCH_CPT(dev))
-		adpa |= PORT_TRANS_SEL_CPT(crtc->pipe);
-	else if (crtc->pipe == 0)
-		adpa |= ADPA_PIPE_A_SELECT;
-	else
-		adpa |= ADPA_PIPE_B_SELECT;
-
-	if (!HAS_PCH_SPLIT(dev))
-		I915_WRITE(BCLRPAT(crtc->pipe), 0);
-
-	I915_WRITE(crt->adpa_reg, adpa);
-}
-
 static bool intel_ironlake_crt_detect_hotplug(struct drm_connector *connector)
 {
 	struct drm_device *dev = connector->dev;
@@ -645,11 +630,12 @@ intel_crt_detect(struct drm_connector *connector, bool force)
 	enum intel_display_power_domain power_domain;
 	enum drm_connector_status status;
 	struct intel_load_detect_pipe tmp;
+	struct drm_modeset_acquire_ctx ctx;
 
 	intel_runtime_pm_get(dev_priv);
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s] force=%d\n",
-		      connector->base.id, drm_get_connector_name(connector),
+		      connector->base.id, connector->name,
 		      force);
 
 	power_domain = intel_display_port_power_domain(intel_encoder);
@@ -688,12 +674,12 @@ intel_crt_detect(struct drm_connector *connector, bool force)
 	}
 
 	/* for pre-945g platforms use load detect */
-	if (intel_get_load_detect_pipe(connector, NULL, &tmp)) {
+	if (intel_get_load_detect_pipe(connector, NULL, &tmp, &ctx)) {
 		if (intel_crt_detect_ddc(connector))
 			status = connector_status_connected;
 		else
 			status = intel_crt_load_detect(crt);
-		intel_release_load_detect_pipe(connector, &tmp);
+		intel_release_load_detect_pipe(connector, &tmp, &ctx);
 	} else
 		status = connector_status_unknown;
 
@@ -867,7 +853,6 @@ void intel_crt_init(struct drm_device *dev)
 		crt->adpa_reg = ADPA;
 
 	crt->base.compute_config = intel_crt_compute_config;
-	crt->base.mode_set = intel_crt_mode_set;
 	crt->base.disable = intel_disable_crt;
 	crt->base.enable = intel_enable_crt;
 	if (I915_HAS_HOTPLUG(dev))
diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 0ad4e9600063..b17b9c7c769f 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -364,55 +364,6 @@ void hsw_fdi_link_train(struct drm_crtc *crtc)
 	DRM_ERROR("FDI link training failed!\n");
 }
 
-static void intel_ddi_mode_set(struct intel_encoder *encoder)
-{
-	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
-	int port = intel_ddi_get_encoder_port(encoder);
-	int pipe = crtc->pipe;
-	int type = encoder->type;
-	struct drm_display_mode *adjusted_mode = &crtc->config.adjusted_mode;
-
-	DRM_DEBUG_KMS("Preparing DDI mode on port %c, pipe %c\n",
-		      port_name(port), pipe_name(pipe));
-
-	crtc->eld_vld = false;
-	if (type == INTEL_OUTPUT_DISPLAYPORT || type == INTEL_OUTPUT_EDP) {
-		struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
-		struct intel_digital_port *intel_dig_port =
-			enc_to_dig_port(&encoder->base);
-
-		intel_dp->DP = intel_dig_port->saved_port_bits |
-			       DDI_BUF_CTL_ENABLE | DDI_BUF_EMP_400MV_0DB_HSW;
-		intel_dp->DP |= DDI_PORT_WIDTH(intel_dp->lane_count);
-
-		if (intel_dp->has_audio) {
-			DRM_DEBUG_DRIVER("DP audio on pipe %c on DDI\n",
-					 pipe_name(crtc->pipe));
-
-			/* write eld */
-			DRM_DEBUG_DRIVER("DP audio: write eld information\n");
-			intel_write_eld(&encoder->base, adjusted_mode);
-		}
-	} else if (type == INTEL_OUTPUT_HDMI) {
-		struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&encoder->base);
-
-		if (intel_hdmi->has_audio) {
-			/* Proper support for digital audio needs a new logic
-			 * and a new set of registers, so we leave it for future
-			 * patch bombing.
-			 */
-			DRM_DEBUG_DRIVER("HDMI audio on pipe %c on DDI\n",
-					 pipe_name(crtc->pipe));
-
-			/* write eld */
-			DRM_DEBUG_DRIVER("HDMI audio: write eld information\n");
-			intel_write_eld(&encoder->base, adjusted_mode);
-		}
-
-		intel_hdmi->set_infoframes(&encoder->base, adjusted_mode);
-	}
-}
-
 static struct intel_encoder *
 intel_ddi_get_crtc_encoder(struct drm_crtc *crtc)
 {
@@ -1062,9 +1013,7 @@ void intel_ddi_enable_transcoder_func(struct drm_crtc *crtc)
 	}
 
 	if (type == INTEL_OUTPUT_HDMI) {
-		struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
-
-		if (intel_hdmi->has_hdmi_sink)
+		if (intel_crtc->config.has_hdmi_sink)
 			temp |= TRANS_DDI_MODE_SELECT_HDMI;
 		else
 			temp |= TRANS_DDI_MODE_SELECT_DVI;
@@ -1293,28 +1242,48 @@ void intel_ddi_disable_pipe_clock(struct intel_crtc *intel_crtc)
 static void intel_ddi_pre_enable(struct intel_encoder *intel_encoder)
 {
 	struct drm_encoder *encoder = &intel_encoder->base;
-	struct drm_crtc *crtc = encoder->crtc;
 	struct drm_i915_private *dev_priv = encoder->dev->dev_private;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	struct intel_crtc *crtc = to_intel_crtc(encoder->crtc);
 	enum port port = intel_ddi_get_encoder_port(intel_encoder);
 	int type = intel_encoder->type;
 
+	if (crtc->config.has_audio) {
+		DRM_DEBUG_DRIVER("Audio on pipe %c on DDI\n",
+				 pipe_name(crtc->pipe));
+
+		/* write eld */
+		DRM_DEBUG_DRIVER("DDI audio: write eld information\n");
+		intel_write_eld(encoder, &crtc->config.adjusted_mode);
+	}
+
 	if (type == INTEL_OUTPUT_EDP) {
 		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 		intel_edp_panel_on(intel_dp);
 	}
 
-	WARN_ON(intel_crtc->ddi_pll_sel == PORT_CLK_SEL_NONE);
-	I915_WRITE(PORT_CLK_SEL(port), intel_crtc->ddi_pll_sel);
+	WARN_ON(crtc->ddi_pll_sel == PORT_CLK_SEL_NONE);
+	I915_WRITE(PORT_CLK_SEL(port), crtc->ddi_pll_sel);
 
 	if (type == INTEL_OUTPUT_DISPLAYPORT || type == INTEL_OUTPUT_EDP) {
 		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
+		struct intel_digital_port *intel_dig_port =
+			enc_to_dig_port(encoder);
+
+		intel_dp->DP = intel_dig_port->saved_port_bits |
+			       DDI_BUF_CTL_ENABLE | DDI_BUF_EMP_400MV_0DB_HSW;
+		intel_dp->DP |= DDI_PORT_WIDTH(intel_dp->lane_count);
 
 		intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_ON);
 		intel_dp_start_link_train(intel_dp);
 		intel_dp_complete_link_train(intel_dp);
 		if (port != PORT_A)
 			intel_dp_stop_link_train(intel_dp);
+	} else if (type == INTEL_OUTPUT_HDMI) {
+		struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
+
+		intel_hdmi->set_infoframes(encoder,
+					   crtc->config.has_hdmi_sink,
+					   &crtc->config.adjusted_mode);
 	}
 }
 
@@ -1385,7 +1354,8 @@ static void intel_enable_ddi(struct intel_encoder *intel_encoder)
 		intel_edp_psr_enable(intel_dp);
 	}
 
-	if (intel_crtc->eld_vld && type != INTEL_OUTPUT_EDP) {
+	if (intel_crtc->config.has_audio) {
+		intel_display_power_get(dev_priv, POWER_DOMAIN_AUDIO);
 		tmp = I915_READ(HSW_AUD_PIN_ELD_CP_VLD);
 		tmp |= ((AUDIO_OUTPUT_ENABLE_A | AUDIO_ELD_VALID_A) << (pipe * 4));
 		I915_WRITE(HSW_AUD_PIN_ELD_CP_VLD, tmp);
@@ -1403,11 +1373,14 @@ static void intel_disable_ddi(struct intel_encoder *intel_encoder)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	uint32_t tmp;
 
-	if (intel_crtc->eld_vld && type != INTEL_OUTPUT_EDP) {
+	/* We can't touch HSW_AUD_PIN_ELD_CP_VLD uncionditionally because this
+	 * register is part of the power well on Haswell. */
+	if (intel_crtc->config.has_audio) {
 		tmp = I915_READ(HSW_AUD_PIN_ELD_CP_VLD);
 		tmp &= ~((AUDIO_OUTPUT_ENABLE_A | AUDIO_ELD_VALID_A) <<
 			 (pipe * 4));
 		I915_WRITE(HSW_AUD_PIN_ELD_CP_VLD, tmp);
+		intel_display_power_put(dev_priv, POWER_DOMAIN_AUDIO);
 	}
 
 	if (type == INTEL_OUTPUT_EDP) {
@@ -1580,6 +1553,7 @@ void intel_ddi_get_config(struct intel_encoder *encoder,
 
 	switch (temp & TRANS_DDI_MODE_SELECT_MASK) {
 	case TRANS_DDI_MODE_SELECT_HDMI:
+		pipe_config->has_hdmi_sink = true;
 	case TRANS_DDI_MODE_SELECT_DVI:
 	case TRANS_DDI_MODE_SELECT_FDI:
 		break;
@@ -1592,6 +1566,12 @@ void intel_ddi_get_config(struct intel_encoder *encoder,
 		break;
 	}
 
+	if (intel_display_power_enabled(dev_priv, POWER_DOMAIN_AUDIO)) {
+		temp = I915_READ(HSW_AUD_PIN_ELD_CP_VLD);
+		if (temp & (AUDIO_OUTPUT_ENABLE_A << (intel_crtc->pipe * 4)))
+			pipe_config->has_audio = true;
+	}
+
 	if (encoder->type == INTEL_OUTPUT_EDP && dev_priv->vbt.edp_bpp &&
 	    pipe_config->pipe_bpp > dev_priv->vbt.edp_bpp) {
 		/*
@@ -1708,7 +1688,6 @@ void intel_ddi_init(struct drm_device *dev, enum port port)
 			 DRM_MODE_ENCODER_TMDS);
 
 	intel_encoder->compute_config = intel_ddi_compute_config;
-	intel_encoder->mode_set = intel_ddi_mode_set;
 	intel_encoder->enable = intel_enable_ddi;
 	intel_encoder->pre_enable = intel_ddi_pre_enable;
 	intel_encoder->disable = intel_disable_ddi;
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 5b60e25baa32..efd3cf50cb0f 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -41,6 +41,9 @@
 #include <drm/drm_crtc_helper.h>
 #include <linux/dma_remapping.h>
 
+#define DIV_ROUND_CLOSEST_ULL(ll, d)	\
+	({ unsigned long long _tmp = (ll)+(d)/2; do_div(_tmp, d); _tmp; })
+
 static void intel_increase_pllclock(struct drm_crtc *crtc);
 static void intel_crtc_update_cursor(struct drm_crtc *crtc, bool on);
 
@@ -55,6 +58,15 @@ static int intel_framebuffer_init(struct drm_device *dev,
 				  struct intel_framebuffer *ifb,
 				  struct drm_mode_fb_cmd2 *mode_cmd,
 				  struct drm_i915_gem_object *obj);
+static void intel_dp_set_m_n(struct intel_crtc *crtc);
+static void i9xx_set_pipeconf(struct intel_crtc *intel_crtc);
+static void intel_set_pipe_timings(struct intel_crtc *intel_crtc);
+static void intel_cpu_transcoder_set_m_n(struct intel_crtc *crtc,
+					 struct intel_link_m_n *m_n);
+static void ironlake_set_pipeconf(struct drm_crtc *crtc);
+static void haswell_set_pipeconf(struct drm_crtc *crtc);
+static void intel_set_pipe_csc(struct drm_crtc *crtc);
+static void vlv_prepare_pll(struct intel_crtc *crtc);
 
 typedef struct {
 	int	min, max;
@@ -328,6 +340,22 @@ static const intel_limit_t intel_limits_vlv = {
 	.p2 = { .p2_slow = 2, .p2_fast = 20 }, /* slow=min, fast=max */
 };
 
+static const intel_limit_t intel_limits_chv = {
+	/*
+	 * These are the data rate limits (measured in fast clocks)
+	 * since those are the strictest limits we have.  The fast
+	 * clock and actual rate limits are more relaxed, so checking
+	 * them would make no difference.
+	 */
+	.dot = { .min = 25000 * 5, .max = 540000 * 5},
+	.vco = { .min = 4860000, .max = 6700000 },
+	.n = { .min = 1, .max = 1 },
+	.m1 = { .min = 2, .max = 2 },
+	.m2 = { .min = 24 << 22, .max = 175 << 22 },
+	.p1 = { .min = 2, .max = 4 },
+	.p2 = {	.p2_slow = 1, .p2_fast = 14 },
+};
+
 static void vlv_clock(int refclk, intel_clock_t *clock)
 {
 	clock->m = clock->m1 * clock->m2;
@@ -412,6 +440,8 @@ static const intel_limit_t *intel_limit(struct drm_crtc *crtc, int refclk)
 			limit = &intel_limits_pineview_lvds;
 		else
 			limit = &intel_limits_pineview_sdvo;
+	} else if (IS_CHERRYVIEW(dev)) {
+		limit = &intel_limits_chv;
 	} else if (IS_VALLEYVIEW(dev)) {
 		limit = &intel_limits_vlv;
 	} else if (!IS_GEN2(dev)) {
@@ -456,6 +486,17 @@ static void i9xx_clock(int refclk, intel_clock_t *clock)
 	clock->dot = DIV_ROUND_CLOSEST(clock->vco, clock->p);
 }
 
+static void chv_clock(int refclk, intel_clock_t *clock)
+{
+	clock->m = clock->m1 * clock->m2;
+	clock->p = clock->p1 * clock->p2;
+	if (WARN_ON(clock->n == 0 || clock->p == 0))
+		return;
+	clock->vco = DIV_ROUND_CLOSEST_ULL((uint64_t)refclk * clock->m,
+			clock->n << 22);
+	clock->dot = DIV_ROUND_CLOSEST(clock->vco, clock->p);
+}
+
 #define INTELPllInvalid(s)   do { /* DRM_DEBUG(s); */ return false; } while (0)
 /**
  * Returns whether the given set of divisors are valid for a given refclk with
@@ -731,6 +772,58 @@ vlv_find_best_dpll(const intel_limit_t *limit, struct drm_crtc *crtc,
 	return found;
 }
 
+static bool
+chv_find_best_dpll(const intel_limit_t *limit, struct drm_crtc *crtc,
+		   int target, int refclk, intel_clock_t *match_clock,
+		   intel_clock_t *best_clock)
+{
+	struct drm_device *dev = crtc->dev;
+	intel_clock_t clock;
+	uint64_t m2;
+	int found = false;
+
+	memset(best_clock, 0, sizeof(*best_clock));
+
+	/*
+	 * Based on hardware doc, the n always set to 1, and m1 always
+	 * set to 2.  If requires to support 200Mhz refclk, we need to
+	 * revisit this because n may not 1 anymore.
+	 */
+	clock.n = 1, clock.m1 = 2;
+	target *= 5;	/* fast clock */
+
+	for (clock.p1 = limit->p1.max; clock.p1 >= limit->p1.min; clock.p1--) {
+		for (clock.p2 = limit->p2.p2_fast;
+				clock.p2 >= limit->p2.p2_slow;
+				clock.p2 -= clock.p2 > 10 ? 2 : 1) {
+
+			clock.p = clock.p1 * clock.p2;
+
+			m2 = DIV_ROUND_CLOSEST_ULL(((uint64_t)target * clock.p *
+					clock.n) << 22, refclk * clock.m1);
+
+			if (m2 > INT_MAX/clock.m1)
+				continue;
+
+			clock.m2 = m2;
+
+			chv_clock(refclk, &clock);
+
+			if (!intel_PLL_is_valid(dev, limit, &clock))
+				continue;
+
+			/* based on hardware requirement, prefer bigger p
+			 */
+			if (clock.p > best_clock->p) {
+				*best_clock = clock;
+				found = true;
+			}
+		}
+	}
+
+	return found;
+}
+
 bool intel_crtc_active(struct drm_crtc *crtc)
 {
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
@@ -765,7 +858,7 @@ static void g4x_wait_for_vblank(struct drm_device *dev, int pipe)
 	frame = I915_READ(frame_reg);
 
 	if (wait_for(I915_READ_NOTRACE(frame_reg) != frame, 50))
-		DRM_DEBUG_KMS("vblank wait timed out\n");
+		WARN(1, "vblank wait timed out\n");
 }
 
 /**
@@ -878,7 +971,7 @@ bool ibx_digital_port_connected(struct drm_i915_private *dev_priv,
 	u32 bit;
 
 	if (HAS_PCH_IBX(dev_priv->dev)) {
-		switch(port->port) {
+		switch (port->port) {
 		case PORT_B:
 			bit = SDE_PORTB_HOTPLUG;
 			break;
@@ -892,7 +985,7 @@ bool ibx_digital_port_connected(struct drm_i915_private *dev_priv,
 			return true;
 		}
 	} else {
-		switch(port->port) {
+		switch (port->port) {
 		case PORT_B:
 			bit = SDE_PORTB_HOTPLUG_CPT;
 			break;
@@ -1097,10 +1190,8 @@ static void assert_cursor(struct drm_i915_private *dev_priv,
 
 	if (IS_845G(dev) || IS_I865G(dev))
 		cur_state = I915_READ(_CURACNTR) & CURSOR_ENABLE;
-	else if (INTEL_INFO(dev)->gen <= 6 || IS_VALLEYVIEW(dev))
-		cur_state = I915_READ(CURCNTR(pipe)) & CURSOR_MODE;
 	else
-		cur_state = I915_READ(CURCNTR_IVB(pipe)) & CURSOR_MODE;
+		cur_state = I915_READ(CURCNTR(pipe)) & CURSOR_MODE;
 
 	WARN(cur_state != state,
 	     "cursor on pipe %c assertion failure (expected %s, current %s)\n",
@@ -1253,6 +1344,9 @@ static bool dp_pipe_enabled(struct drm_i915_private *dev_priv,
 		u32	trans_dp_ctl = I915_READ(trans_dp_ctl_reg);
 		if ((trans_dp_ctl & TRANS_DP_PORT_SEL_MASK) != port_sel)
 			return false;
+	} else if (IS_CHERRYVIEW(dev_priv->dev)) {
+		if ((val & DP_PIPE_MASK_CHV) != DP_PIPE_SELECT_CHV(pipe))
+			return false;
 	} else {
 		if ((val & DP_PIPE_MASK) != (pipe << 30))
 			return false;
@@ -1269,6 +1363,9 @@ static bool hdmi_pipe_enabled(struct drm_i915_private *dev_priv,
 	if (HAS_PCH_CPT(dev_priv->dev)) {
 		if ((val & SDVO_PIPE_SEL_MASK_CPT) != SDVO_PIPE_SEL_CPT(pipe))
 			return false;
+	} else if (IS_CHERRYVIEW(dev_priv->dev)) {
+		if ((val & SDVO_PIPE_SEL_MASK_CHV) != SDVO_PIPE_SEL_CHV(pipe))
+			return false;
 	} else {
 		if ((val & SDVO_PIPE_SEL_MASK) != SDVO_PIPE_SEL(pipe))
 			return false;
@@ -1367,7 +1464,17 @@ static void intel_init_dpio(struct drm_device *dev)
 	if (!IS_VALLEYVIEW(dev))
 		return;
 
-	DPIO_PHY_IOSF_PORT(DPIO_PHY0) = IOSF_PORT_DPIO;
+	/*
+	 * IOSF_PORT_DPIO is used for VLV x2 PHY (DP/HDMI B and C),
+	 * CHV x1 PHY (DP/HDMI D)
+	 * IOSF_PORT_DPIO_2 is used for CHV x2 PHY (DP/HDMI B and C)
+	 */
+	if (IS_CHERRYVIEW(dev)) {
+		DPIO_PHY_IOSF_PORT(DPIO_PHY0) = IOSF_PORT_DPIO_2;
+		DPIO_PHY_IOSF_PORT(DPIO_PHY1) = IOSF_PORT_DPIO;
+	} else {
+		DPIO_PHY_IOSF_PORT(DPIO_PHY0) = IOSF_PORT_DPIO;
+	}
 }
 
 static void intel_reset_dpio(struct drm_device *dev)
@@ -1377,25 +1484,48 @@ static void intel_reset_dpio(struct drm_device *dev)
 	if (!IS_VALLEYVIEW(dev))
 		return;
 
-	/*
-	 * Enable the CRI clock source so we can get at the display and the
-	 * reference clock for VGA hotplug / manual detection.
-	 */
-	I915_WRITE(DPLL(PIPE_B), I915_READ(DPLL(PIPE_B)) |
-		   DPLL_REFA_CLK_ENABLE_VLV |
-		   DPLL_INTEGRATED_CRI_CLK_VLV);
+	if (IS_CHERRYVIEW(dev)) {
+		enum dpio_phy phy;
+		u32 val;
 
-	/*
-	 * From VLV2A0_DP_eDP_DPIO_driver_vbios_notes_10.docx -
-	 *  6.	De-assert cmn_reset/side_reset. Same as VLV X0.
-	 *   a.	GUnit 0x2110 bit[0] set to 1 (def 0)
-	 *   b.	The other bits such as sfr settings / modesel may all be set
-	 *      to 0.
-	 *
-	 * This should only be done on init and resume from S3 with both
-	 * PLLs disabled, or we risk losing DPIO and PLL synchronization.
-	 */
-	I915_WRITE(DPIO_CTL, I915_READ(DPIO_CTL) | DPIO_CMNRST);
+		for (phy = DPIO_PHY0; phy < I915_NUM_PHYS_VLV; phy++) {
+			/* Poll for phypwrgood signal */
+			if (wait_for(I915_READ(DISPLAY_PHY_STATUS) &
+						PHY_POWERGOOD(phy), 1))
+				DRM_ERROR("Display PHY %d is not power up\n", phy);
+
+			/*
+			 * Deassert common lane reset for PHY.
+			 *
+			 * This should only be done on init and resume from S3
+			 * with both PLLs disabled, or we risk losing DPIO and
+			 * PLL synchronization.
+			 */
+			val = I915_READ(DISPLAY_PHY_CONTROL);
+			I915_WRITE(DISPLAY_PHY_CONTROL,
+				PHY_COM_LANE_RESET_DEASSERT(phy, val));
+		}
+
+	} else {
+		/*
+		 * If DPIO has already been reset, e.g. by BIOS, just skip all
+		 * this.
+		 */
+		if (I915_READ(DPIO_CTL) & DPIO_CMNRST)
+			return;
+
+		/*
+		 * From VLV2A0_DP_eDP_HDMI_DPIO_driver_vbios_notes_11.docx:
+		 * Need to assert and de-assert PHY SB reset by gating the
+		 * common lane power, then un-gating it.
+		 * Simply ungating isn't enough to reset the PHY enough to get
+		 * ports and lanes running.
+		 */
+		__vlv_set_power_well(dev_priv, PUNIT_POWER_WELL_DPIO_CMN_BC,
+				     false);
+		__vlv_set_power_well(dev_priv, PUNIT_POWER_WELL_DPIO_CMN_BC,
+				     true);
+	}
 }
 
 static void vlv_enable_pll(struct intel_crtc *crtc)
@@ -1436,6 +1566,44 @@ static void vlv_enable_pll(struct intel_crtc *crtc)
 	udelay(150); /* wait for warmup */
 }
 
+static void chv_enable_pll(struct intel_crtc *crtc)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int pipe = crtc->pipe;
+	enum dpio_channel port = vlv_pipe_to_channel(pipe);
+	u32 tmp;
+
+	assert_pipe_disabled(dev_priv, crtc->pipe);
+
+	BUG_ON(!IS_CHERRYVIEW(dev_priv->dev));
+
+	mutex_lock(&dev_priv->dpio_lock);
+
+	/* Enable back the 10bit clock to display controller */
+	tmp = vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW14(port));
+	tmp |= DPIO_DCLKP_EN;
+	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW14(port), tmp);
+
+	/*
+	 * Need to wait > 100ns between dclkp clock enable bit and PLL enable.
+	 */
+	udelay(1);
+
+	/* Enable PLL */
+	I915_WRITE(DPLL(pipe), crtc->config.dpll_hw_state.dpll);
+
+	/* Check PLL is locked */
+	if (wait_for(((I915_READ(DPLL(pipe)) & DPLL_LOCK_VLV) == DPLL_LOCK_VLV), 1))
+		DRM_ERROR("PLL %d failed to lock\n", pipe);
+
+	/* not sure when this should be written */
+	I915_WRITE(DPLL_MD(pipe), crtc->config.dpll_hw_state.dpll_md);
+	POSTING_READ(DPLL_MD(pipe));
+
+	mutex_unlock(&dev_priv->dpio_lock);
+}
+
 static void i9xx_enable_pll(struct intel_crtc *crtc)
 {
 	struct drm_device *dev = crtc->base.dev;
@@ -1519,45 +1687,92 @@ static void vlv_disable_pll(struct drm_i915_private *dev_priv, enum pipe pipe)
 		val = DPLL_INTEGRATED_CRI_CLK_VLV | DPLL_REFA_CLK_ENABLE_VLV;
 	I915_WRITE(DPLL(pipe), val);
 	POSTING_READ(DPLL(pipe));
+
+}
+
+static void chv_disable_pll(struct drm_i915_private *dev_priv, enum pipe pipe)
+{
+	enum dpio_channel port = vlv_pipe_to_channel(pipe);
+	u32 val;
+
+	/* Make sure the pipe isn't still relying on us */
+	assert_pipe_disabled(dev_priv, pipe);
+
+	/* Set PLL en = 0 */
+	val = DPLL_SSC_REF_CLOCK_CHV;
+	if (pipe != PIPE_A)
+		val |= DPLL_INTEGRATED_CRI_CLK_VLV;
+	I915_WRITE(DPLL(pipe), val);
+	POSTING_READ(DPLL(pipe));
+
+	mutex_lock(&dev_priv->dpio_lock);
+
+	/* Disable 10bit clock to display controller */
+	val = vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW14(port));
+	val &= ~DPIO_DCLKP_EN;
+	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW14(port), val);
+
+	mutex_unlock(&dev_priv->dpio_lock);
 }
 
 void vlv_wait_port_ready(struct drm_i915_private *dev_priv,
 		struct intel_digital_port *dport)
 {
 	u32 port_mask;
+	int dpll_reg;
 
 	switch (dport->port) {
 	case PORT_B:
 		port_mask = DPLL_PORTB_READY_MASK;
+		dpll_reg = DPLL(0);
 		break;
 	case PORT_C:
 		port_mask = DPLL_PORTC_READY_MASK;
+		dpll_reg = DPLL(0);
+		break;
+	case PORT_D:
+		port_mask = DPLL_PORTD_READY_MASK;
+		dpll_reg = DPIO_PHY_STATUS;
 		break;
 	default:
 		BUG();
 	}
 
-	if (wait_for((I915_READ(DPLL(0)) & port_mask) == 0, 1000))
+	if (wait_for((I915_READ(dpll_reg) & port_mask) == 0, 1000))
 		WARN(1, "timed out waiting for port %c ready: 0x%08x\n",
-		     port_name(dport->port), I915_READ(DPLL(0)));
+		     port_name(dport->port), I915_READ(dpll_reg));
+}
+
+static void intel_prepare_shared_dpll(struct intel_crtc *crtc)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_shared_dpll *pll = intel_crtc_to_shared_dpll(crtc);
+
+	WARN_ON(!pll->refcount);
+	if (pll->active == 0) {
+		DRM_DEBUG_DRIVER("setting up %s\n", pll->name);
+		WARN_ON(pll->on);
+		assert_shared_dpll_disabled(dev_priv, pll);
+
+		pll->mode_set(dev_priv, pll);
+	}
 }
 
 /**
- * ironlake_enable_shared_dpll - enable PCH PLL
+ * intel_enable_shared_dpll - enable PCH PLL
  * @dev_priv: i915 private structure
  * @pipe: pipe PLL to enable
  *
  * The PCH PLL needs to be enabled before the PCH transcoder, since it
  * drives the transcoder clock.
  */
-static void ironlake_enable_shared_dpll(struct intel_crtc *crtc)
+static void intel_enable_shared_dpll(struct intel_crtc *crtc)
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_shared_dpll *pll = intel_crtc_to_shared_dpll(crtc);
 
-	/* PCH PLLs only available on ILK, SNB and IVB */
-	BUG_ON(INTEL_INFO(dev)->gen < 5);
 	if (WARN_ON(pll == NULL))
 		return;
 
@@ -1804,16 +2019,6 @@ static void intel_enable_pipe(struct intel_crtc *crtc)
 
 	I915_WRITE(reg, val | PIPECONF_ENABLE);
 	POSTING_READ(reg);
-
-	/*
-	 * There's no guarantee the pipe will really start running now. It
-	 * depends on the Gen, the output type and the relative order between
-	 * pipe and plane enabling. Avoid waiting on HSW+ since it's not
-	 * necessary.
-	 * TODO: audit the previous gens.
-	 */
-	if (INTEL_INFO(dev)->gen <= 7 && !IS_HASWELL(dev))
-		intel_wait_for_vblank(dev_priv->dev, pipe);
 }
 
 /**
@@ -1890,18 +2095,17 @@ static void intel_enable_primary_hw_plane(struct drm_i915_private *dev_priv,
 	/* If the pipe isn't enabled, we can't pump pixels and may hang */
 	assert_pipe_enabled(dev_priv, pipe);
 
-	WARN(intel_crtc->primary_enabled, "Primary plane already enabled\n");
+	if (intel_crtc->primary_enabled)
+		return;
 
 	intel_crtc->primary_enabled = true;
 
 	reg = DSPCNTR(plane);
 	val = I915_READ(reg);
-	if (val & DISPLAY_PLANE_ENABLE)
-		return;
+	WARN_ON(val & DISPLAY_PLANE_ENABLE);
 
 	I915_WRITE(reg, val | DISPLAY_PLANE_ENABLE);
 	intel_flush_primary_plane(dev_priv, plane);
-	intel_wait_for_vblank(dev_priv->dev, pipe);
 }
 
 /**
@@ -1920,18 +2124,17 @@ static void intel_disable_primary_hw_plane(struct drm_i915_private *dev_priv,
 	int reg;
 	u32 val;
 
-	WARN(!intel_crtc->primary_enabled, "Primary plane already disabled\n");
+	if (!intel_crtc->primary_enabled)
+		return;
 
 	intel_crtc->primary_enabled = false;
 
 	reg = DSPCNTR(plane);
 	val = I915_READ(reg);
-	if ((val & DISPLAY_PLANE_ENABLE) == 0)
-		return;
+	WARN_ON((val & DISPLAY_PLANE_ENABLE) == 0);
 
 	I915_WRITE(reg, val & ~DISPLAY_PLANE_ENABLE);
 	intel_flush_primary_plane(dev_priv, plane);
-	intel_wait_for_vblank(dev_priv->dev, pipe);
 }
 
 static bool need_vtd_wa(struct drm_device *dev)
@@ -1954,7 +2157,7 @@ static int intel_align_height(struct drm_device *dev, int height, bool tiled)
 int
 intel_pin_and_fence_fb_obj(struct drm_device *dev,
 			   struct drm_i915_gem_object *obj,
-			   struct intel_ring_buffer *pipelined)
+			   struct intel_engine_cs *pipelined)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 alignment;
@@ -2134,7 +2337,7 @@ static void intel_find_plane_obj(struct intel_crtc *intel_crtc,
 	 * Failed to alloc the obj, check to see if we should share
 	 * an fb with another CRTC instead
 	 */
-	list_for_each_entry(c, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, c) {
 		i = to_intel_crtc(c);
 
 		if (c == &intel_crtc->base)
@@ -2152,9 +2355,9 @@ static void intel_find_plane_obj(struct intel_crtc *intel_crtc,
 	}
 }
 
-static int i9xx_update_primary_plane(struct drm_crtc *crtc,
-				     struct drm_framebuffer *fb,
-				     int x, int y)
+static void i9xx_update_primary_plane(struct drm_crtc *crtc,
+				      struct drm_framebuffer *fb,
+				      int x, int y)
 {
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -2166,15 +2369,6 @@ static int i9xx_update_primary_plane(struct drm_crtc *crtc,
 	u32 dspcntr;
 	u32 reg;
 
-	switch (plane) {
-	case 0:
-	case 1:
-		break;
-	default:
-		DRM_ERROR("Can't update plane %c in SAREA\n", plane_name(plane));
-		return -EINVAL;
-	}
-
 	intel_fb = to_intel_framebuffer(fb);
 	obj = intel_fb->obj;
 
@@ -2249,13 +2443,11 @@ static int i9xx_update_primary_plane(struct drm_crtc *crtc,
 	} else
 		I915_WRITE(DSPADDR(plane), i915_gem_obj_ggtt_offset(obj) + linear_offset);
 	POSTING_READ(reg);
-
-	return 0;
 }
 
-static int ironlake_update_primary_plane(struct drm_crtc *crtc,
-					 struct drm_framebuffer *fb,
-					 int x, int y)
+static void ironlake_update_primary_plane(struct drm_crtc *crtc,
+					  struct drm_framebuffer *fb,
+					  int x, int y)
 {
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -2267,16 +2459,6 @@ static int ironlake_update_primary_plane(struct drm_crtc *crtc,
 	u32 dspcntr;
 	u32 reg;
 
-	switch (plane) {
-	case 0:
-	case 1:
-	case 2:
-		break;
-	default:
-		DRM_ERROR("Can't update plane %c in SAREA\n", plane_name(plane));
-		return -EINVAL;
-	}
-
 	intel_fb = to_intel_framebuffer(fb);
 	obj = intel_fb->obj;
 
@@ -2343,8 +2525,6 @@ static int ironlake_update_primary_plane(struct drm_crtc *crtc,
 		I915_WRITE(DSPLINOFF(plane), linear_offset);
 	}
 	POSTING_READ(reg);
-
-	return 0;
 }
 
 /* Assume fb object is pinned & idle & fenced and just update base pointers */
@@ -2359,7 +2539,9 @@ intel_pipe_set_base_atomic(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 		dev_priv->display.disable_fbc(dev);
 	intel_increase_pllclock(crtc);
 
-	return dev_priv->display.update_primary_plane(crtc, fb, x, y);
+	dev_priv->display.update_primary_plane(crtc, fb, x, y);
+
+	return 0;
 }
 
 void intel_display_handle_reset(struct drm_device *dev)
@@ -2381,7 +2563,7 @@ void intel_display_handle_reset(struct drm_device *dev)
 	 * pending_flip_queue really got woken up.
 	 */
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, crtc) {
 		struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 		enum plane plane = intel_crtc->plane;
 
@@ -2389,10 +2571,10 @@ void intel_display_handle_reset(struct drm_device *dev)
 		intel_finish_page_flip_plane(dev, plane);
 	}
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, crtc) {
 		struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 
-		mutex_lock(&crtc->mutex);
+		drm_modeset_lock(&crtc->mutex, NULL);
 		/*
 		 * FIXME: Once we have proper support for primary planes (and
 		 * disabling them without disabling the entire crtc) allow again
@@ -2403,7 +2585,7 @@ void intel_display_handle_reset(struct drm_device *dev)
 							       crtc->primary->fb,
 							       crtc->x,
 							       crtc->y);
-		mutex_unlock(&crtc->mutex);
+		drm_modeset_unlock(&crtc->mutex);
 	}
 }
 
@@ -2518,14 +2700,7 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y,
 		intel_crtc->config.pipe_src_h = adjusted_mode->crtc_vdisplay;
 	}
 
-	ret = dev_priv->display.update_primary_plane(crtc, fb, x, y);
-	if (ret) {
-		mutex_lock(&dev->struct_mutex);
-		intel_unpin_fb_obj(to_intel_framebuffer(fb)->obj);
-		mutex_unlock(&dev->struct_mutex);
-		DRM_ERROR("failed to update base address\n");
-		return ret;
-	}
+	dev_priv->display.update_primary_plane(crtc, fb, x, y);
 
 	old_fb = crtc->primary->fb;
 	crtc->primary->fb = fb;
@@ -2628,12 +2803,10 @@ static void ironlake_fdi_link_train(struct drm_crtc *crtc)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_crtc->pipe;
-	int plane = intel_crtc->plane;
 	u32 reg, temp, tries;
 
-	/* FDI needs bits from pipe & plane first */
+	/* FDI needs bits from pipe first */
 	assert_pipe_enabled(dev_priv, pipe);
-	assert_plane_enabled(dev_priv, plane);
 
 	/* Train 1: umask FDI RX Interrupt symbol_lock and bit_lock bit
 	   for train result */
@@ -3064,9 +3237,8 @@ static void ironlake_fdi_disable(struct drm_crtc *crtc)
 	udelay(100);
 
 	/* Ironlake workaround, disable clock pointer after downing FDI */
-	if (HAS_PCH_IBX(dev)) {
+	if (HAS_PCH_IBX(dev))
 		I915_WRITE(FDI_RX_CHICKEN(pipe), FDI_RX_PHASE_SYNC_POINTER_OVR);
-	}
 
 	/* still set train pattern 1 */
 	reg = FDI_TX_CTL(pipe);
@@ -3104,7 +3276,7 @@ bool intel_has_pending_fb_unpin(struct drm_device *dev)
 	 * cannot claim and pin a new fb without at least acquring the
 	 * struct_mutex and so serialising with us.
 	 */
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		if (atomic_read(&crtc->unpin_work_count) == 0)
 			continue;
 
@@ -3117,7 +3289,7 @@ bool intel_has_pending_fb_unpin(struct drm_device *dev)
 	return false;
 }
 
-static void intel_crtc_wait_for_pending_flips(struct drm_crtc *crtc)
+void intel_crtc_wait_for_pending_flips(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -3127,8 +3299,9 @@ static void intel_crtc_wait_for_pending_flips(struct drm_crtc *crtc)
 
 	WARN_ON(waitqueue_active(&dev_priv->pending_flip_queue));
 
-	wait_event(dev_priv->pending_flip_queue,
-		   !intel_crtc_has_pending_flip(crtc));
+	WARN_ON(wait_event_timeout(dev_priv->pending_flip_queue,
+				   !intel_crtc_has_pending_flip(crtc),
+				   60*HZ) == 0);
 
 	mutex_lock(&dev->struct_mutex);
 	intel_finish_fb(crtc->primary->fb);
@@ -3341,7 +3514,7 @@ static void ironlake_pch_enable(struct drm_crtc *crtc)
 	 * Note that enable_shared_dpll tries to do the right thing, but
 	 * get_shared_dpll unconditionally resets the pll - we need that to have
 	 * the right LVDS enable sequence. */
-	ironlake_enable_shared_dpll(intel_crtc);
+	intel_enable_shared_dpll(intel_crtc);
 
 	/* set transcoder timing, panel must allow it */
 	assert_panel_unlocked(dev_priv, pipe);
@@ -3445,6 +3618,8 @@ static struct intel_shared_dpll *intel_get_shared_dpll(struct intel_crtc *crtc)
 		DRM_DEBUG_KMS("CRTC:%d using pre-allocated %s\n",
 			      crtc->base.base.id, pll->name);
 
+		WARN_ON(pll->refcount);
+
 		goto found;
 	}
 
@@ -3478,20 +3653,13 @@ static struct intel_shared_dpll *intel_get_shared_dpll(struct intel_crtc *crtc)
 	return NULL;
 
 found:
+	if (pll->refcount == 0)
+		pll->hw_state = crtc->config.dpll_hw_state;
+
 	crtc->config.shared_dpll = i;
 	DRM_DEBUG_DRIVER("using %s for pipe %c\n", pll->name,
 			 pipe_name(crtc->pipe));
 
-	if (pll->active == 0) {
-		memcpy(&pll->hw_state, &crtc->config.dpll_hw_state,
-		       sizeof(pll->hw_state));
-
-		DRM_DEBUG_DRIVER("setting up %s\n", pll->name);
-		WARN_ON(pll->on);
-		assert_shared_dpll_disabled(dev_priv, pll);
-
-		pll->mode_set(dev_priv, pll);
-	}
 	pll->refcount++;
 
 	return pll;
@@ -3562,17 +3730,17 @@ static void intel_disable_planes(struct drm_crtc *crtc)
 
 void hsw_enable_ips(struct intel_crtc *crtc)
 {
-	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	if (!crtc->config.ips_enabled)
 		return;
 
-	/* We can only enable IPS after we enable a plane and wait for a vblank.
-	 * We guarantee that the plane is enabled by calling intel_enable_ips
-	 * only after intel_enable_plane. And intel_enable_plane already waits
-	 * for a vblank, so all we need to do here is to enable the IPS bit. */
+	/* We can only enable IPS after we enable a plane and wait for a vblank */
+	intel_wait_for_vblank(dev, crtc->pipe);
+
 	assert_plane_enabled(dev_priv, crtc->plane);
-	if (IS_BROADWELL(crtc->base.dev)) {
+	if (IS_BROADWELL(dev)) {
 		mutex_lock(&dev_priv->rps.hw_lock);
 		WARN_ON(sandybridge_pcode_write(dev_priv, DISPLAY_IPS_CONTROL, 0xc0000000));
 		mutex_unlock(&dev_priv->rps.hw_lock);
@@ -3602,10 +3770,13 @@ void hsw_disable_ips(struct intel_crtc *crtc)
 		return;
 
 	assert_plane_enabled(dev_priv, crtc->plane);
-	if (IS_BROADWELL(crtc->base.dev)) {
+	if (IS_BROADWELL(dev)) {
 		mutex_lock(&dev_priv->rps.hw_lock);
 		WARN_ON(sandybridge_pcode_write(dev_priv, DISPLAY_IPS_CONTROL, 0));
 		mutex_unlock(&dev_priv->rps.hw_lock);
+		/* wait for pcode to finish disabling IPS, which may take up to 42ms */
+		if (wait_for((I915_READ(IPS_CTL) & IPS_ENABLE) == 0, 42))
+			DRM_ERROR("Timed out waiting for IPS disable\n");
 	} else {
 		I915_WRITE(IPS_CTL, 0);
 		POSTING_READ(IPS_CTL);
@@ -3662,6 +3833,94 @@ static void intel_crtc_load_lut(struct drm_crtc *crtc)
 		hsw_enable_ips(intel_crtc);
 }
 
+static void intel_crtc_dpms_overlay(struct intel_crtc *intel_crtc, bool enable)
+{
+	if (!enable && intel_crtc->overlay) {
+		struct drm_device *dev = intel_crtc->base.dev;
+		struct drm_i915_private *dev_priv = dev->dev_private;
+
+		mutex_lock(&dev->struct_mutex);
+		dev_priv->mm.interruptible = false;
+		(void) intel_overlay_switch_off(intel_crtc->overlay);
+		dev_priv->mm.interruptible = true;
+		mutex_unlock(&dev->struct_mutex);
+	}
+
+	/* Let userspace switch the overlay on again. In most cases userspace
+	 * has to recompute where to put it anyway.
+	 */
+}
+
+/**
+ * i9xx_fixup_plane - ugly workaround for G45 to fire up the hardware
+ * cursor plane briefly if not already running after enabling the display
+ * plane.
+ * This workaround avoids occasional blank screens when self refresh is
+ * enabled.
+ */
+static void
+g4x_fixup_plane(struct drm_i915_private *dev_priv, enum pipe pipe)
+{
+	u32 cntl = I915_READ(CURCNTR(pipe));
+
+	if ((cntl & CURSOR_MODE) == 0) {
+		u32 fw_bcl_self = I915_READ(FW_BLC_SELF);
+
+		I915_WRITE(FW_BLC_SELF, fw_bcl_self & ~FW_BLC_SELF_EN);
+		I915_WRITE(CURCNTR(pipe), CURSOR_MODE_64_ARGB_AX);
+		intel_wait_for_vblank(dev_priv->dev, pipe);
+		I915_WRITE(CURCNTR(pipe), cntl);
+		I915_WRITE(CURBASE(pipe), I915_READ(CURBASE(pipe)));
+		I915_WRITE(FW_BLC_SELF, fw_bcl_self);
+	}
+}
+
+static void intel_crtc_enable_planes(struct drm_crtc *crtc)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	int pipe = intel_crtc->pipe;
+	int plane = intel_crtc->plane;
+
+	intel_enable_primary_hw_plane(dev_priv, plane, pipe);
+	intel_enable_planes(crtc);
+	/* The fixup needs to happen before cursor is enabled */
+	if (IS_G4X(dev))
+		g4x_fixup_plane(dev_priv, pipe);
+	intel_crtc_update_cursor(crtc, true);
+	intel_crtc_dpms_overlay(intel_crtc, true);
+
+	hsw_enable_ips(intel_crtc);
+
+	mutex_lock(&dev->struct_mutex);
+	intel_update_fbc(dev);
+	intel_edp_psr_update(dev);
+	mutex_unlock(&dev->struct_mutex);
+}
+
+static void intel_crtc_disable_planes(struct drm_crtc *crtc)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	int pipe = intel_crtc->pipe;
+	int plane = intel_crtc->plane;
+
+	intel_crtc_wait_for_pending_flips(crtc);
+	drm_crtc_vblank_off(crtc);
+
+	if (dev_priv->fbc.plane == plane)
+		intel_disable_fbc(dev);
+
+	hsw_disable_ips(intel_crtc);
+
+	intel_crtc_dpms_overlay(intel_crtc, false);
+	intel_crtc_update_cursor(crtc, false);
+	intel_disable_planes(crtc);
+	intel_disable_primary_hw_plane(dev_priv, plane, pipe);
+}
+
 static void ironlake_crtc_enable(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
@@ -3669,13 +3928,35 @@ static void ironlake_crtc_enable(struct drm_crtc *crtc)
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_encoder *encoder;
 	int pipe = intel_crtc->pipe;
-	int plane = intel_crtc->plane;
+	enum plane plane = intel_crtc->plane;
 
 	WARN_ON(!crtc->enabled);
 
 	if (intel_crtc->active)
 		return;
 
+	if (intel_crtc->config.has_pch_encoder)
+		intel_prepare_shared_dpll(intel_crtc);
+
+	if (intel_crtc->config.has_dp_encoder)
+		intel_dp_set_m_n(intel_crtc);
+
+	intel_set_pipe_timings(intel_crtc);
+
+	if (intel_crtc->config.has_pch_encoder) {
+		intel_cpu_transcoder_set_m_n(intel_crtc,
+					     &intel_crtc->config.fdi_m_n);
+	}
+
+	ironlake_set_pipeconf(crtc);
+
+	/* Set up the display plane register */
+	I915_WRITE(DSPCNTR(plane), DISPPLANE_GAMMA_ENABLE);
+	POSTING_READ(DSPCNTR(plane));
+
+	dev_priv->display.update_primary_plane(crtc, crtc->primary->fb,
+					       crtc->x, crtc->y);
+
 	intel_crtc->active = true;
 
 	intel_set_cpu_fifo_underrun_reporting(dev, pipe, true);
@@ -3705,32 +3986,19 @@ static void ironlake_crtc_enable(struct drm_crtc *crtc)
 
 	intel_update_watermarks(crtc);
 	intel_enable_pipe(intel_crtc);
-	intel_enable_primary_hw_plane(dev_priv, plane, pipe);
-	intel_enable_planes(crtc);
-	intel_crtc_update_cursor(crtc, true);
 
 	if (intel_crtc->config.has_pch_encoder)
 		ironlake_pch_enable(crtc);
 
-	mutex_lock(&dev->struct_mutex);
-	intel_update_fbc(dev);
-	mutex_unlock(&dev->struct_mutex);
-
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		encoder->enable(encoder);
 
 	if (HAS_PCH_CPT(dev))
 		cpt_verify_modeset(dev, intel_crtc->pipe);
 
-	/*
-	 * There seems to be a race in PCH platform hw (at least on some
-	 * outputs) where an enabled pipe still completes any pageflip right
-	 * away (as if the pipe is off) instead of waiting for vblank. As soon
-	 * as the first vblank happend, everything works as expected. Hence just
-	 * wait for one vblank before returning to avoid strange things
-	 * happening.
-	 */
-	intel_wait_for_vblank(dev, intel_crtc->pipe);
+	intel_crtc_enable_planes(crtc);
+
+	drm_crtc_vblank_on(crtc);
 }
 
 /* IPS only exists on ULT machines and is tied to pipe A. */
@@ -3739,47 +4007,6 @@ static bool hsw_crtc_supports_ips(struct intel_crtc *crtc)
 	return HAS_IPS(crtc->base.dev) && crtc->pipe == PIPE_A;
 }
 
-static void haswell_crtc_enable_planes(struct drm_crtc *crtc)
-{
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	int pipe = intel_crtc->pipe;
-	int plane = intel_crtc->plane;
-
-	intel_enable_primary_hw_plane(dev_priv, plane, pipe);
-	intel_enable_planes(crtc);
-	intel_crtc_update_cursor(crtc, true);
-
-	hsw_enable_ips(intel_crtc);
-
-	mutex_lock(&dev->struct_mutex);
-	intel_update_fbc(dev);
-	mutex_unlock(&dev->struct_mutex);
-}
-
-static void haswell_crtc_disable_planes(struct drm_crtc *crtc)
-{
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	int pipe = intel_crtc->pipe;
-	int plane = intel_crtc->plane;
-
-	intel_crtc_wait_for_pending_flips(crtc);
-	drm_vblank_off(dev, pipe);
-
-	/* FBC must be disabled before disabling the plane on HSW. */
-	if (dev_priv->fbc.plane == plane)
-		intel_disable_fbc(dev);
-
-	hsw_disable_ips(intel_crtc);
-
-	intel_crtc_update_cursor(crtc, false);
-	intel_disable_planes(crtc);
-	intel_disable_primary_hw_plane(dev_priv, plane, pipe);
-}
-
 /*
  * This implements the workaround described in the "notes" section of the mode
  * set sequence documentation. When going from no pipes or single pipe to
@@ -3793,7 +4020,7 @@ static void haswell_mode_set_planes_workaround(struct intel_crtc *crtc)
 
 	/* We want to get the other_active_crtc only if there's only 1 other
 	 * active crtc. */
-	list_for_each_entry(crtc_it, &dev->mode_config.crtc_list, base.head) {
+	for_each_intel_crtc(dev, crtc_it) {
 		if (!crtc_it->active || crtc_it == crtc)
 			continue;
 
@@ -3816,12 +4043,34 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_encoder *encoder;
 	int pipe = intel_crtc->pipe;
+	enum plane plane = intel_crtc->plane;
 
 	WARN_ON(!crtc->enabled);
 
 	if (intel_crtc->active)
 		return;
 
+	if (intel_crtc->config.has_dp_encoder)
+		intel_dp_set_m_n(intel_crtc);
+
+	intel_set_pipe_timings(intel_crtc);
+
+	if (intel_crtc->config.has_pch_encoder) {
+		intel_cpu_transcoder_set_m_n(intel_crtc,
+					     &intel_crtc->config.fdi_m_n);
+	}
+
+	haswell_set_pipeconf(crtc);
+
+	intel_set_pipe_csc(crtc);
+
+	/* Set up the display plane register */
+	I915_WRITE(DSPCNTR(plane), DISPPLANE_GAMMA_ENABLE | DISPPLANE_PIPE_CSC_ENABLE);
+	POSTING_READ(DSPCNTR(plane));
+
+	dev_priv->display.update_primary_plane(crtc, crtc->primary->fb,
+					       crtc->x, crtc->y);
+
 	intel_crtc->active = true;
 
 	intel_set_cpu_fifo_underrun_reporting(dev, pipe, true);
@@ -3862,7 +4111,9 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 	/* If we change the relative order between pipe/planes enabling, we need
 	 * to change the workaround. */
 	haswell_mode_set_planes_workaround(intel_crtc);
-	haswell_crtc_enable_planes(crtc);
+	intel_crtc_enable_planes(crtc);
+
+	drm_crtc_vblank_on(crtc);
 }
 
 static void ironlake_pfit_disable(struct intel_crtc *crtc)
@@ -3887,26 +4138,16 @@ static void ironlake_crtc_disable(struct drm_crtc *crtc)
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_encoder *encoder;
 	int pipe = intel_crtc->pipe;
-	int plane = intel_crtc->plane;
 	u32 reg, temp;
 
-
 	if (!intel_crtc->active)
 		return;
 
+	intel_crtc_disable_planes(crtc);
+
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		encoder->disable(encoder);
 
-	intel_crtc_wait_for_pending_flips(crtc);
-	drm_vblank_off(dev, pipe);
-
-	if (dev_priv->fbc.plane == plane)
-		intel_disable_fbc(dev);
-
-	intel_crtc_update_cursor(crtc, false);
-	intel_disable_planes(crtc);
-	intel_disable_primary_hw_plane(dev_priv, plane, pipe);
-
 	if (intel_crtc->config.has_pch_encoder)
 		intel_set_pch_fifo_underrun_reporting(dev, pipe, false);
 
@@ -3950,6 +4191,7 @@ static void ironlake_crtc_disable(struct drm_crtc *crtc)
 
 	mutex_lock(&dev->struct_mutex);
 	intel_update_fbc(dev);
+	intel_edp_psr_update(dev);
 	mutex_unlock(&dev->struct_mutex);
 }
 
@@ -3965,7 +4207,7 @@ static void haswell_crtc_disable(struct drm_crtc *crtc)
 	if (!intel_crtc->active)
 		return;
 
-	haswell_crtc_disable_planes(crtc);
+	intel_crtc_disable_planes(crtc);
 
 	for_each_encoder_on_crtc(dev, crtc, encoder) {
 		intel_opregion_notify_encoder(encoder, false);
@@ -3997,6 +4239,7 @@ static void haswell_crtc_disable(struct drm_crtc *crtc)
 
 	mutex_lock(&dev->struct_mutex);
 	intel_update_fbc(dev);
+	intel_edp_psr_update(dev);
 	mutex_unlock(&dev->struct_mutex);
 }
 
@@ -4011,48 +4254,6 @@ static void haswell_crtc_off(struct drm_crtc *crtc)
 	intel_ddi_put_crtc_pll(crtc);
 }
 
-static void intel_crtc_dpms_overlay(struct intel_crtc *intel_crtc, bool enable)
-{
-	if (!enable && intel_crtc->overlay) {
-		struct drm_device *dev = intel_crtc->base.dev;
-		struct drm_i915_private *dev_priv = dev->dev_private;
-
-		mutex_lock(&dev->struct_mutex);
-		dev_priv->mm.interruptible = false;
-		(void) intel_overlay_switch_off(intel_crtc->overlay);
-		dev_priv->mm.interruptible = true;
-		mutex_unlock(&dev->struct_mutex);
-	}
-
-	/* Let userspace switch the overlay on again. In most cases userspace
-	 * has to recompute where to put it anyway.
-	 */
-}
-
-/**
- * i9xx_fixup_plane - ugly workaround for G45 to fire up the hardware
- * cursor plane briefly if not already running after enabling the display
- * plane.
- * This workaround avoids occasional blank screens when self refresh is
- * enabled.
- */
-static void
-g4x_fixup_plane(struct drm_i915_private *dev_priv, enum pipe pipe)
-{
-	u32 cntl = I915_READ(CURCNTR(pipe));
-
-	if ((cntl & CURSOR_MODE) == 0) {
-		u32 fw_bcl_self = I915_READ(FW_BLC_SELF);
-
-		I915_WRITE(FW_BLC_SELF, fw_bcl_self & ~FW_BLC_SELF_EN);
-		I915_WRITE(CURCNTR(pipe), CURSOR_MODE_64_ARGB_AX);
-		intel_wait_for_vblank(dev_priv->dev, pipe);
-		I915_WRITE(CURCNTR(pipe), cntl);
-		I915_WRITE(CURBASE(pipe), I915_READ(CURBASE(pipe)));
-		I915_WRITE(FW_BLC_SELF, fw_bcl_self);
-	}
-}
-
 static void i9xx_pfit_enable(struct intel_crtc *crtc)
 {
 	struct drm_device *dev = crtc->base.dev;
@@ -4164,7 +4365,7 @@ static void modeset_update_crtc_power_domains(struct drm_device *dev)
 	 * First get all needed power domains, then put all unneeded, to avoid
 	 * any unnecessary toggling of the power wells.
 	 */
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		enum intel_display_power_domain domain;
 
 		if (!crtc->base.enabled)
@@ -4176,7 +4377,7 @@ static void modeset_update_crtc_power_domains(struct drm_device *dev)
 			intel_display_power_get(dev_priv, domain);
 	}
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		enum intel_display_power_domain domain;
 
 		for_each_power_domain(domain, crtc->enabled_power_domains)
@@ -4207,6 +4408,9 @@ static void valleyview_set_cdclk(struct drm_device *dev, int cdclk)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 val, cmd;
 
+	WARN_ON(valleyview_cur_cdclk(dev_priv) != dev_priv->vlv_cdclk_freq);
+	dev_priv->vlv_cdclk_freq = cdclk;
+
 	if (cdclk >= 320) /* jump to highest voltage for 400MHz too */
 		cmd = 2;
 	else if (cdclk == 266)
@@ -4261,7 +4465,7 @@ static void valleyview_set_cdclk(struct drm_device *dev, int cdclk)
 	intel_i2c_reset(dev);
 }
 
-static int valleyview_cur_cdclk(struct drm_i915_private *dev_priv)
+int valleyview_cur_cdclk(struct drm_i915_private *dev_priv)
 {
 	int cur_cdclk, vco;
 	int divider;
@@ -4282,10 +4486,6 @@ static int valleyview_cur_cdclk(struct drm_i915_private *dev_priv)
 static int valleyview_calc_cdclk(struct drm_i915_private *dev_priv,
 				 int max_pixclk)
 {
-	int cur_cdclk;
-
-	cur_cdclk = valleyview_cur_cdclk(dev_priv);
-
 	/*
 	 * Really only a few cases to deal with, as only 4 CDclks are supported:
 	 *   200MHz
@@ -4311,8 +4511,7 @@ static int intel_mode_max_pixclk(struct drm_i915_private *dev_priv)
 	struct intel_crtc *intel_crtc;
 	int max_pixclk = 0;
 
-	list_for_each_entry(intel_crtc, &dev->mode_config.crtc_list,
-			    base.head) {
+	for_each_intel_crtc(dev, intel_crtc) {
 		if (intel_crtc->new_enabled)
 			max_pixclk = max(max_pixclk,
 					 intel_crtc->new_config->adjusted_mode.crtc_clock);
@@ -4327,14 +4526,13 @@ static void valleyview_modeset_global_pipes(struct drm_device *dev,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc;
 	int max_pixclk = intel_mode_max_pixclk(dev_priv);
-	int cur_cdclk = valleyview_cur_cdclk(dev_priv);
 
-	if (valleyview_calc_cdclk(dev_priv, max_pixclk) == cur_cdclk)
+	if (valleyview_calc_cdclk(dev_priv, max_pixclk) ==
+	    dev_priv->vlv_cdclk_freq)
 		return;
 
 	/* disable/enable all currently active pipes while we change cdclk */
-	list_for_each_entry(intel_crtc, &dev->mode_config.crtc_list,
-			    base.head)
+	for_each_intel_crtc(dev, intel_crtc)
 		if (intel_crtc->base.enabled)
 			*prepare_pipes |= (1 << intel_crtc->pipe);
 }
@@ -4343,10 +4541,9 @@ static void valleyview_modeset_global_resources(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int max_pixclk = intel_mode_max_pixclk(dev_priv);
-	int cur_cdclk = valleyview_cur_cdclk(dev_priv);
 	int req_cdclk = valleyview_calc_cdclk(dev_priv, max_pixclk);
 
-	if (req_cdclk != cur_cdclk)
+	if (req_cdclk != dev_priv->vlv_cdclk_freq)
 		valleyview_set_cdclk(dev, req_cdclk);
 	modeset_update_crtc_power_domains(dev);
 }
@@ -4360,22 +4557,55 @@ static void valleyview_crtc_enable(struct drm_crtc *crtc)
 	int pipe = intel_crtc->pipe;
 	int plane = intel_crtc->plane;
 	bool is_dsi;
+	u32 dspcntr;
 
 	WARN_ON(!crtc->enabled);
 
 	if (intel_crtc->active)
 		return;
 
+	vlv_prepare_pll(intel_crtc);
+
+	/* Set up the display plane register */
+	dspcntr = DISPPLANE_GAMMA_ENABLE;
+
+	if (intel_crtc->config.has_dp_encoder)
+		intel_dp_set_m_n(intel_crtc);
+
+	intel_set_pipe_timings(intel_crtc);
+
+	/* pipesrc and dspsize control the size that is scaled from,
+	 * which should always be the user's requested size.
+	 */
+	I915_WRITE(DSPSIZE(plane),
+		   ((intel_crtc->config.pipe_src_h - 1) << 16) |
+		   (intel_crtc->config.pipe_src_w - 1));
+	I915_WRITE(DSPPOS(plane), 0);
+
+	i9xx_set_pipeconf(intel_crtc);
+
+	I915_WRITE(DSPCNTR(plane), dspcntr);
+	POSTING_READ(DSPCNTR(plane));
+
+	dev_priv->display.update_primary_plane(crtc, crtc->primary->fb,
+					       crtc->x, crtc->y);
+
 	intel_crtc->active = true;
 
+	intel_set_cpu_fifo_underrun_reporting(dev, pipe, true);
+
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		if (encoder->pre_pll_enable)
 			encoder->pre_pll_enable(encoder);
 
 	is_dsi = intel_pipe_has_type(crtc, INTEL_OUTPUT_DSI);
 
-	if (!is_dsi)
-		vlv_enable_pll(intel_crtc);
+	if (!is_dsi) {
+		if (IS_CHERRYVIEW(dev))
+			chv_enable_pll(intel_crtc);
+		else
+			vlv_enable_pll(intel_crtc);
+	}
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		if (encoder->pre_enable)
@@ -4387,15 +4617,25 @@ static void valleyview_crtc_enable(struct drm_crtc *crtc)
 
 	intel_update_watermarks(crtc);
 	intel_enable_pipe(intel_crtc);
-	intel_set_cpu_fifo_underrun_reporting(dev, pipe, true);
-	intel_enable_primary_hw_plane(dev_priv, plane, pipe);
-	intel_enable_planes(crtc);
-	intel_crtc_update_cursor(crtc, true);
-
-	intel_update_fbc(dev);
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		encoder->enable(encoder);
+
+	intel_crtc_enable_planes(crtc);
+
+	drm_crtc_vblank_on(crtc);
+
+	/* Underruns don't raise interrupts, so check manually. */
+	i9xx_check_fifo_underruns(dev);
+}
+
+static void i9xx_set_pll_dividers(struct intel_crtc *crtc)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	I915_WRITE(FP0(crtc->pipe), crtc->config.dpll_hw_state.fp0);
+	I915_WRITE(FP1(crtc->pipe), crtc->config.dpll_hw_state.fp1);
 }
 
 static void i9xx_crtc_enable(struct drm_crtc *crtc)
@@ -4406,14 +4646,49 @@ static void i9xx_crtc_enable(struct drm_crtc *crtc)
 	struct intel_encoder *encoder;
 	int pipe = intel_crtc->pipe;
 	int plane = intel_crtc->plane;
+	u32 dspcntr;
 
 	WARN_ON(!crtc->enabled);
 
 	if (intel_crtc->active)
 		return;
 
+	i9xx_set_pll_dividers(intel_crtc);
+
+	/* Set up the display plane register */
+	dspcntr = DISPPLANE_GAMMA_ENABLE;
+
+	if (pipe == 0)
+		dspcntr &= ~DISPPLANE_SEL_PIPE_MASK;
+	else
+		dspcntr |= DISPPLANE_SEL_PIPE_B;
+
+	if (intel_crtc->config.has_dp_encoder)
+		intel_dp_set_m_n(intel_crtc);
+
+	intel_set_pipe_timings(intel_crtc);
+
+	/* pipesrc and dspsize control the size that is scaled from,
+	 * which should always be the user's requested size.
+	 */
+	I915_WRITE(DSPSIZE(plane),
+		   ((intel_crtc->config.pipe_src_h - 1) << 16) |
+		   (intel_crtc->config.pipe_src_w - 1));
+	I915_WRITE(DSPPOS(plane), 0);
+
+	i9xx_set_pipeconf(intel_crtc);
+
+	I915_WRITE(DSPCNTR(plane), dspcntr);
+	POSTING_READ(DSPCNTR(plane));
+
+	dev_priv->display.update_primary_plane(crtc, crtc->primary->fb,
+					       crtc->x, crtc->y);
+
 	intel_crtc->active = true;
 
+	if (!IS_GEN2(dev))
+		intel_set_cpu_fifo_underrun_reporting(dev, pipe, true);
+
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		if (encoder->pre_enable)
 			encoder->pre_enable(encoder);
@@ -4426,21 +4701,26 @@ static void i9xx_crtc_enable(struct drm_crtc *crtc)
 
 	intel_update_watermarks(crtc);
 	intel_enable_pipe(intel_crtc);
-	intel_set_cpu_fifo_underrun_reporting(dev, pipe, true);
-	intel_enable_primary_hw_plane(dev_priv, plane, pipe);
-	intel_enable_planes(crtc);
-	/* The fixup needs to happen before cursor is enabled */
-	if (IS_G4X(dev))
-		g4x_fixup_plane(dev_priv, pipe);
-	intel_crtc_update_cursor(crtc, true);
-
-	/* Give the overlay scaler a chance to enable if it's on this pipe */
-	intel_crtc_dpms_overlay(intel_crtc, true);
-
-	intel_update_fbc(dev);
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		encoder->enable(encoder);
+
+	intel_crtc_enable_planes(crtc);
+
+	/*
+	 * Gen2 reports pipe underruns whenever all planes are disabled.
+	 * So don't enable underrun reporting before at least some planes
+	 * are enabled.
+	 * FIXME: Need to fix the logic to work when we turn off all planes
+	 * but leave the pipe running.
+	 */
+	if (IS_GEN2(dev))
+		intel_set_cpu_fifo_underrun_reporting(dev, pipe, true);
+
+	drm_crtc_vblank_on(crtc);
+
+	/* Underruns don't raise interrupts, so check manually. */
+	i9xx_check_fifo_underruns(dev);
 }
 
 static void i9xx_pfit_disable(struct intel_crtc *crtc)
@@ -4465,27 +4745,31 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_encoder *encoder;
 	int pipe = intel_crtc->pipe;
-	int plane = intel_crtc->plane;
 
 	if (!intel_crtc->active)
 		return;
 
-	for_each_encoder_on_crtc(dev, crtc, encoder)
-		encoder->disable(encoder);
+	/*
+	 * Gen2 reports pipe underruns whenever all planes are disabled.
+	 * So diasble underrun reporting before all the planes get disabled.
+	 * FIXME: Need to fix the logic to work when we turn off all planes
+	 * but leave the pipe running.
+	 */
+	if (IS_GEN2(dev))
+		intel_set_cpu_fifo_underrun_reporting(dev, pipe, false);
 
-	/* Give the overlay scaler a chance to disable if it's on this pipe */
-	intel_crtc_wait_for_pending_flips(crtc);
-	drm_vblank_off(dev, pipe);
+	intel_crtc_disable_planes(crtc);
 
-	if (dev_priv->fbc.plane == plane)
-		intel_disable_fbc(dev);
+	for_each_encoder_on_crtc(dev, crtc, encoder)
+		encoder->disable(encoder);
 
-	intel_crtc_dpms_overlay(intel_crtc, false);
-	intel_crtc_update_cursor(crtc, false);
-	intel_disable_planes(crtc);
-	intel_disable_primary_hw_plane(dev_priv, plane, pipe);
+	/*
+	 * On gen2 planes are double buffered but the pipe isn't, so we must
+	 * wait for planes to fully turn off before disabling the pipe.
+	 */
+	if (IS_GEN2(dev))
+		intel_wait_for_vblank(dev, pipe);
 
-	intel_set_cpu_fifo_underrun_reporting(dev, pipe, false);
 	intel_disable_pipe(dev_priv, pipe);
 
 	i9xx_pfit_disable(intel_crtc);
@@ -4494,15 +4778,25 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 		if (encoder->post_disable)
 			encoder->post_disable(encoder);
 
-	if (IS_VALLEYVIEW(dev) && !intel_pipe_has_type(crtc, INTEL_OUTPUT_DSI))
-		vlv_disable_pll(dev_priv, pipe);
-	else if (!IS_VALLEYVIEW(dev))
-		i9xx_disable_pll(dev_priv, pipe);
+	if (!intel_pipe_has_type(crtc, INTEL_OUTPUT_DSI)) {
+		if (IS_CHERRYVIEW(dev))
+			chv_disable_pll(dev_priv, pipe);
+		else if (IS_VALLEYVIEW(dev))
+			vlv_disable_pll(dev_priv, pipe);
+		else
+			i9xx_disable_pll(dev_priv, pipe);
+	}
+
+	if (!IS_GEN2(dev))
+		intel_set_cpu_fifo_underrun_reporting(dev, pipe, false);
 
 	intel_crtc->active = false;
 	intel_update_watermarks(crtc);
 
+	mutex_lock(&dev->struct_mutex);
 	intel_update_fbc(dev);
+	intel_edp_psr_update(dev);
+	mutex_unlock(&dev->struct_mutex);
 }
 
 static void i9xx_crtc_off(struct drm_crtc *crtc)
@@ -4565,13 +4859,11 @@ static void intel_crtc_disable(struct drm_crtc *crtc)
 	struct drm_device *dev = crtc->dev;
 	struct drm_connector *connector;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 
 	/* crtc should still be enabled when we disable it. */
 	WARN_ON(!crtc->enabled);
 
 	dev_priv->display.crtc_disable(crtc);
-	intel_crtc->eld_vld = false;
 	intel_crtc_update_sarea(crtc, false);
 	dev_priv->display.off(crtc);
 
@@ -4635,7 +4927,7 @@ static void intel_connector_check_state(struct intel_connector *connector)
 
 		DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
 			      connector->base.base.id,
-			      drm_get_connector_name(&connector->base));
+			      connector->base.name);
 
 		WARN(connector->base.dpms == DRM_MODE_DPMS_OFF,
 		     "wrong connector dpms state\n");
@@ -5039,8 +5331,6 @@ static void i9xx_update_pll_dividers(struct intel_crtc *crtc,
 				     intel_clock_t *reduced_clock)
 {
 	struct drm_device *dev = crtc->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	int pipe = crtc->pipe;
 	u32 fp, fp2 = 0;
 
 	if (IS_PINEVIEW(dev)) {
@@ -5053,17 +5343,14 @@ static void i9xx_update_pll_dividers(struct intel_crtc *crtc,
 			fp2 = i9xx_dpll_compute_fp(reduced_clock);
 	}
 
-	I915_WRITE(FP0(pipe), fp);
 	crtc->config.dpll_hw_state.fp0 = fp;
 
 	crtc->lowfreq_avail = false;
 	if (intel_pipe_has_type(&crtc->base, INTEL_OUTPUT_LVDS) &&
 	    reduced_clock && i915.powersave) {
-		I915_WRITE(FP1(pipe), fp2);
 		crtc->config.dpll_hw_state.fp1 = fp2;
 		crtc->lowfreq_avail = true;
 	} else {
-		I915_WRITE(FP1(pipe), fp);
 		crtc->config.dpll_hw_state.fp1 = fp;
 	}
 }
@@ -5141,12 +5428,34 @@ static void intel_dp_set_m_n(struct intel_crtc *crtc)
 
 static void vlv_update_pll(struct intel_crtc *crtc)
 {
+	u32 dpll, dpll_md;
+
+	/*
+	 * Enable DPIO clock input. We should never disable the reference
+	 * clock for pipe B, since VGA hotplug / manual detection depends
+	 * on it.
+	 */
+	dpll = DPLL_EXT_BUFFER_ENABLE_VLV | DPLL_REFA_CLK_ENABLE_VLV |
+		DPLL_VGA_MODE_DIS | DPLL_INTEGRATED_CLOCK_VLV;
+	/* We should never disable this, set it here for state tracking */
+	if (crtc->pipe == PIPE_B)
+		dpll |= DPLL_INTEGRATED_CRI_CLK_VLV;
+	dpll |= DPLL_VCO_ENABLE;
+	crtc->config.dpll_hw_state.dpll = dpll;
+
+	dpll_md = (crtc->config.pixel_multiplier - 1)
+		<< DPLL_MD_UDI_MULTIPLIER_SHIFT;
+	crtc->config.dpll_hw_state.dpll_md = dpll_md;
+}
+
+static void vlv_prepare_pll(struct intel_crtc *crtc)
+{
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int pipe = crtc->pipe;
-	u32 dpll, mdiv;
+	u32 mdiv;
 	u32 bestn, bestm1, bestm2, bestp1, bestp2;
-	u32 coreclk, reg_val, dpll_md;
+	u32 coreclk, reg_val;
 
 	mutex_lock(&dev_priv->dpio_lock);
 
@@ -5159,7 +5468,7 @@ static void vlv_update_pll(struct intel_crtc *crtc)
 	/* See eDP HDMI DPIO driver vbios notes doc */
 
 	/* PLL B needs special handling */
-	if (pipe)
+	if (pipe == PIPE_B)
 		vlv_pllb_recal_opamp(dev_priv, pipe);
 
 	/* Set up Tx target for periodic Rcomp update */
@@ -5203,7 +5512,7 @@ static void vlv_update_pll(struct intel_crtc *crtc)
 	if (intel_pipe_has_type(&crtc->base, INTEL_OUTPUT_EDP) ||
 	    intel_pipe_has_type(&crtc->base, INTEL_OUTPUT_DISPLAYPORT)) {
 		/* Use SSC source */
-		if (!pipe)
+		if (pipe == PIPE_A)
 			vlv_dpio_write(dev_priv, pipe, VLV_PLL_DW5(pipe),
 					 0x0df40000);
 		else
@@ -5211,7 +5520,7 @@ static void vlv_update_pll(struct intel_crtc *crtc)
 					 0x0df70000);
 	} else { /* HDMI or VGA */
 		/* Use bend source */
-		if (!pipe)
+		if (pipe == PIPE_A)
 			vlv_dpio_write(dev_priv, pipe, VLV_PLL_DW5(pipe),
 					 0x0df70000);
 		else
@@ -5227,26 +5536,84 @@ static void vlv_update_pll(struct intel_crtc *crtc)
 	vlv_dpio_write(dev_priv, pipe, VLV_PLL_DW7(pipe), coreclk);
 
 	vlv_dpio_write(dev_priv, pipe, VLV_PLL_DW11(pipe), 0x87871000);
+	mutex_unlock(&dev_priv->dpio_lock);
+}
+
+static void chv_update_pll(struct intel_crtc *crtc)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int pipe = crtc->pipe;
+	int dpll_reg = DPLL(crtc->pipe);
+	enum dpio_channel port = vlv_pipe_to_channel(pipe);
+	u32 loopfilter, intcoeff;
+	u32 bestn, bestm1, bestm2, bestp1, bestp2, bestm2_frac;
+	int refclk;
+
+	crtc->config.dpll_hw_state.dpll = DPLL_SSC_REF_CLOCK_CHV |
+		DPLL_REFA_CLK_ENABLE_VLV | DPLL_VGA_MODE_DIS |
+		DPLL_VCO_ENABLE;
+	if (pipe != PIPE_A)
+		crtc->config.dpll_hw_state.dpll |= DPLL_INTEGRATED_CRI_CLK_VLV;
+
+	crtc->config.dpll_hw_state.dpll_md =
+		(crtc->config.pixel_multiplier - 1) << DPLL_MD_UDI_MULTIPLIER_SHIFT;
+
+	bestn = crtc->config.dpll.n;
+	bestm2_frac = crtc->config.dpll.m2 & 0x3fffff;
+	bestm1 = crtc->config.dpll.m1;
+	bestm2 = crtc->config.dpll.m2 >> 22;
+	bestp1 = crtc->config.dpll.p1;
+	bestp2 = crtc->config.dpll.p2;
 
 	/*
-	 * Enable DPIO clock input. We should never disable the reference
-	 * clock for pipe B, since VGA hotplug / manual detection depends
-	 * on it.
+	 * Enable Refclk and SSC
 	 */
-	dpll = DPLL_EXT_BUFFER_ENABLE_VLV | DPLL_REFA_CLK_ENABLE_VLV |
-		DPLL_VGA_MODE_DIS | DPLL_INTEGRATED_CLOCK_VLV;
-	/* We should never disable this, set it here for state tracking */
-	if (pipe == PIPE_B)
-		dpll |= DPLL_INTEGRATED_CRI_CLK_VLV;
-	dpll |= DPLL_VCO_ENABLE;
-	crtc->config.dpll_hw_state.dpll = dpll;
+	I915_WRITE(dpll_reg,
+		   crtc->config.dpll_hw_state.dpll & ~DPLL_VCO_ENABLE);
 
-	dpll_md = (crtc->config.pixel_multiplier - 1)
-		<< DPLL_MD_UDI_MULTIPLIER_SHIFT;
-	crtc->config.dpll_hw_state.dpll_md = dpll_md;
+	mutex_lock(&dev_priv->dpio_lock);
+
+	/* p1 and p2 divider */
+	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW13(port),
+			5 << DPIO_CHV_S1_DIV_SHIFT |
+			bestp1 << DPIO_CHV_P1_DIV_SHIFT |
+			bestp2 << DPIO_CHV_P2_DIV_SHIFT |
+			1 << DPIO_CHV_K_DIV_SHIFT);
+
+	/* Feedback post-divider - m2 */
+	vlv_dpio_write(dev_priv, pipe, CHV_PLL_DW0(port), bestm2);
+
+	/* Feedback refclk divider - n and m1 */
+	vlv_dpio_write(dev_priv, pipe, CHV_PLL_DW1(port),
+			DPIO_CHV_M1_DIV_BY_2 |
+			1 << DPIO_CHV_N_DIV_SHIFT);
+
+	/* M2 fraction division */
+	vlv_dpio_write(dev_priv, pipe, CHV_PLL_DW2(port), bestm2_frac);
+
+	/* M2 fraction division enable */
+	vlv_dpio_write(dev_priv, pipe, CHV_PLL_DW3(port),
+		       DPIO_CHV_FRAC_DIV_EN |
+		       (2 << DPIO_CHV_FEEDFWD_GAIN_SHIFT));
+
+	/* Loop filter */
+	refclk = i9xx_get_refclk(&crtc->base, 0);
+	loopfilter = 5 << DPIO_CHV_PROP_COEFF_SHIFT |
+		2 << DPIO_CHV_GAIN_CTRL_SHIFT;
+	if (refclk == 100000)
+		intcoeff = 11;
+	else if (refclk == 38400)
+		intcoeff = 10;
+	else
+		intcoeff = 9;
+	loopfilter |= intcoeff << DPIO_CHV_INT_COEFF_SHIFT;
+	vlv_dpio_write(dev_priv, pipe, CHV_PLL_DW6(port), loopfilter);
 
-	if (crtc->config.has_dp_encoder)
-		intel_dp_set_m_n(crtc);
+	/* AFC Recal */
+	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW14(port),
+			vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW14(port)) |
+			DPIO_AFC_RECAL);
 
 	mutex_unlock(&dev_priv->dpio_lock);
 }
@@ -5325,9 +5692,6 @@ static void i9xx_update_pll(struct intel_crtc *crtc,
 			<< DPLL_MD_UDI_MULTIPLIER_SHIFT;
 		crtc->config.dpll_hw_state.dpll_md = dpll_md;
 	}
-
-	if (crtc->config.has_dp_encoder)
-		intel_dp_set_m_n(crtc);
 }
 
 static void i8xx_update_pll(struct intel_crtc *crtc,
@@ -5567,16 +5931,12 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc,
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	int pipe = intel_crtc->pipe;
-	int plane = intel_crtc->plane;
 	int refclk, num_connectors = 0;
 	intel_clock_t clock, reduced_clock;
-	u32 dspcntr;
 	bool ok, has_reduced_clock = false;
 	bool is_lvds = false, is_dsi = false;
 	struct intel_encoder *encoder;
 	const intel_limit_t *limit;
-	int ret;
 
 	for_each_encoder_on_crtc(dev, crtc, encoder) {
 		switch (encoder->type) {
@@ -5592,7 +5952,7 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc,
 	}
 
 	if (is_dsi)
-		goto skip_dpll;
+		return 0;
 
 	if (!intel_crtc->config.clock_set) {
 		refclk = i9xx_get_refclk(crtc, num_connectors);
@@ -5637,43 +5997,17 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc,
 		i8xx_update_pll(intel_crtc,
 				has_reduced_clock ? &reduced_clock : NULL,
 				num_connectors);
+	} else if (IS_CHERRYVIEW(dev)) {
+		chv_update_pll(intel_crtc);
 	} else if (IS_VALLEYVIEW(dev)) {
 		vlv_update_pll(intel_crtc);
 	} else {
 		i9xx_update_pll(intel_crtc,
 				has_reduced_clock ? &reduced_clock : NULL,
-                                num_connectors);
-	}
-
-skip_dpll:
-	/* Set up the display plane register */
-	dspcntr = DISPPLANE_GAMMA_ENABLE;
-
-	if (!IS_VALLEYVIEW(dev)) {
-		if (pipe == 0)
-			dspcntr &= ~DISPPLANE_SEL_PIPE_MASK;
-		else
-			dspcntr |= DISPPLANE_SEL_PIPE_B;
+				num_connectors);
 	}
 
-	intel_set_pipe_timings(intel_crtc);
-
-	/* pipesrc and dspsize control the size that is scaled from,
-	 * which should always be the user's requested size.
-	 */
-	I915_WRITE(DSPSIZE(plane),
-		   ((intel_crtc->config.pipe_src_h - 1) << 16) |
-		   (intel_crtc->config.pipe_src_w - 1));
-	I915_WRITE(DSPPOS(plane), 0);
-
-	i9xx_set_pipeconf(intel_crtc);
-
-	I915_WRITE(DSPCNTR(plane), dspcntr);
-	POSTING_READ(DSPCNTR(plane));
-
-	ret = intel_pipe_set_base(crtc, x, y, fb);
-
-	return ret;
+	return 0;
 }
 
 static void i9xx_get_pfit_config(struct intel_crtc *crtc,
@@ -5793,6 +6127,36 @@ static void i9xx_get_plane_config(struct intel_crtc *crtc,
 
 }
 
+static void chv_crtc_clock_get(struct intel_crtc *crtc,
+			       struct intel_crtc_config *pipe_config)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int pipe = pipe_config->cpu_transcoder;
+	enum dpio_channel port = vlv_pipe_to_channel(pipe);
+	intel_clock_t clock;
+	u32 cmn_dw13, pll_dw0, pll_dw1, pll_dw2;
+	int refclk = 100000;
+
+	mutex_lock(&dev_priv->dpio_lock);
+	cmn_dw13 = vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW13(port));
+	pll_dw0 = vlv_dpio_read(dev_priv, pipe, CHV_PLL_DW0(port));
+	pll_dw1 = vlv_dpio_read(dev_priv, pipe, CHV_PLL_DW1(port));
+	pll_dw2 = vlv_dpio_read(dev_priv, pipe, CHV_PLL_DW2(port));
+	mutex_unlock(&dev_priv->dpio_lock);
+
+	clock.m1 = (pll_dw1 & 0x7) == DPIO_CHV_M1_DIV_BY_2 ? 2 : 0;
+	clock.m2 = ((pll_dw0 & 0xff) << 22) | (pll_dw2 & 0x3fffff);
+	clock.n = (pll_dw1 >> DPIO_CHV_N_DIV_SHIFT) & 0xf;
+	clock.p1 = (cmn_dw13 >> DPIO_CHV_P1_DIV_SHIFT) & 0x7;
+	clock.p2 = (cmn_dw13 >> DPIO_CHV_P2_DIV_SHIFT) & 0x1f;
+
+	chv_clock(refclk, &clock);
+
+	/* clock.dot is the fast clock */
+	pipe_config->port_clock = clock.dot / 5;
+}
+
 static bool i9xx_get_pipe_config(struct intel_crtc *crtc,
 				 struct intel_crtc_config *pipe_config)
 {
@@ -5827,6 +6191,9 @@ static bool i9xx_get_pipe_config(struct intel_crtc *crtc,
 		}
 	}
 
+	if (IS_VALLEYVIEW(dev) && (tmp & PIPECONF_COLOR_RANGE_SELECT))
+		pipe_config->limited_color_range = true;
+
 	if (INTEL_INFO(dev)->gen < 4)
 		pipe_config->double_wide = tmp & PIPECONF_DOUBLE_WIDE;
 
@@ -5862,7 +6229,9 @@ static bool i9xx_get_pipe_config(struct intel_crtc *crtc,
 						     DPLL_PORTB_READY_MASK);
 	}
 
-	if (IS_VALLEYVIEW(dev))
+	if (IS_CHERRYVIEW(dev))
+		chv_crtc_clock_get(crtc, pipe_config);
+	else if (IS_VALLEYVIEW(dev))
 		vlv_crtc_clock_get(crtc, pipe_config);
 	else
 		i9xx_crtc_clock_get(crtc, pipe_config);
@@ -5983,8 +6352,7 @@ static void ironlake_init_pch_refclk(struct drm_device *dev)
 			if (intel_panel_use_ssc(dev_priv) && can_ssc) {
 				DRM_DEBUG_KMS("Using SSC on eDP\n");
 				val |= DREF_CPU_SOURCE_OUTPUT_DOWNSPREAD;
-			}
-			else
+			} else
 				val |= DREF_CPU_SOURCE_OUTPUT_NONSPREAD;
 		} else
 			val |= DREF_CPU_SOURCE_OUTPUT_DISABLE;
@@ -6563,10 +6931,7 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
 				  struct drm_framebuffer *fb)
 {
 	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	int pipe = intel_crtc->pipe;
-	int plane = intel_crtc->plane;
 	int num_connectors = 0;
 	intel_clock_t clock, reduced_clock;
 	u32 dpll = 0, fp = 0, fp2 = 0;
@@ -6574,7 +6939,6 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
 	bool is_lvds = false;
 	struct intel_encoder *encoder;
 	struct intel_shared_dpll *pll;
-	int ret;
 
 	for_each_encoder_on_crtc(dev, crtc, encoder) {
 		switch (encoder->type) {
@@ -6624,36 +6988,18 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
 		pll = intel_get_shared_dpll(intel_crtc);
 		if (pll == NULL) {
 			DRM_DEBUG_DRIVER("failed to find PLL for pipe %c\n",
-					 pipe_name(pipe));
+					 pipe_name(intel_crtc->pipe));
 			return -EINVAL;
 		}
 	} else
 		intel_put_shared_dpll(intel_crtc);
 
-	if (intel_crtc->config.has_dp_encoder)
-		intel_dp_set_m_n(intel_crtc);
-
 	if (is_lvds && has_reduced_clock && i915.powersave)
 		intel_crtc->lowfreq_avail = true;
 	else
 		intel_crtc->lowfreq_avail = false;
 
-	intel_set_pipe_timings(intel_crtc);
-
-	if (intel_crtc->config.has_pch_encoder) {
-		intel_cpu_transcoder_set_m_n(intel_crtc,
-					     &intel_crtc->config.fdi_m_n);
-	}
-
-	ironlake_set_pipeconf(crtc);
-
-	/* Set up the display plane register */
-	I915_WRITE(DSPCNTR(plane), DISPPLANE_GAMMA_ENABLE);
-	POSTING_READ(DSPCNTR(plane));
-
-	ret = intel_pipe_set_base(crtc, x, y, fb);
-
-	return ret;
+	return 0;
 }
 
 static void intel_pch_transcoder_get_m_n(struct intel_crtc *crtc,
@@ -6831,6 +7177,9 @@ static bool ironlake_get_pipe_config(struct intel_crtc *crtc,
 		break;
 	}
 
+	if (tmp & PIPECONF_COLOR_RANGE_SELECT)
+		pipe_config->limited_color_range = true;
+
 	if (I915_READ(PCH_TRANSCONF(crtc->pipe)) & TRANS_ENABLE) {
 		struct intel_shared_dpll *pll;
 
@@ -6880,10 +7229,8 @@ static void assert_can_disable_lcpll(struct drm_i915_private *dev_priv)
 	struct drm_device *dev = dev_priv->dev;
 	struct intel_ddi_plls *plls = &dev_priv->ddi_plls;
 	struct intel_crtc *crtc;
-	unsigned long irqflags;
-	uint32_t val;
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head)
+	for_each_intel_crtc(dev, crtc)
 		WARN(crtc->active, "CRTC for pipe %c enabled\n",
 		     pipe_name(crtc->pipe));
 
@@ -6902,14 +7249,29 @@ static void assert_can_disable_lcpll(struct drm_i915_private *dev_priv)
 	     "Utility pin enabled\n");
 	WARN(I915_READ(PCH_GTC_CTL) & PCH_GTC_ENABLE, "PCH GTC enabled\n");
 
-	spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-	val = I915_READ(DEIMR);
-	WARN((val | DE_PCH_EVENT_IVB) != 0xffffffff,
-	     "Unexpected DEIMR bits enabled: 0x%x\n", val);
-	val = I915_READ(SDEIMR);
-	WARN((val | SDE_HOTPLUG_MASK_CPT) != 0xffffffff,
-	     "Unexpected SDEIMR bits enabled: 0x%x\n", val);
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+	/*
+	 * In theory we can still leave IRQs enabled, as long as only the HPD
+	 * interrupts remain enabled. We used to check for that, but since it's
+	 * gen-specific and since we only disable LCPLL after we fully disable
+	 * the interrupts, the check below should be enough.
+	 */
+	WARN(!dev_priv->pm.irqs_disabled, "IRQs enabled\n");
+}
+
+static void hsw_write_dcomp(struct drm_i915_private *dev_priv, uint32_t val)
+{
+	struct drm_device *dev = dev_priv->dev;
+
+	if (IS_HASWELL(dev)) {
+		mutex_lock(&dev_priv->rps.hw_lock);
+		if (sandybridge_pcode_write(dev_priv, GEN6_PCODE_WRITE_D_COMP,
+					    val))
+			DRM_ERROR("Failed to disable D_COMP\n");
+		mutex_unlock(&dev_priv->rps.hw_lock);
+	} else {
+		I915_WRITE(D_COMP, val);
+	}
+	POSTING_READ(D_COMP);
 }
 
 /*
@@ -6949,11 +7311,7 @@ static void hsw_disable_lcpll(struct drm_i915_private *dev_priv,
 
 	val = I915_READ(D_COMP);
 	val |= D_COMP_COMP_DISABLE;
-	mutex_lock(&dev_priv->rps.hw_lock);
-	if (sandybridge_pcode_write(dev_priv, GEN6_PCODE_WRITE_D_COMP, val))
-		DRM_ERROR("Failed to disable D_COMP\n");
-	mutex_unlock(&dev_priv->rps.hw_lock);
-	POSTING_READ(D_COMP);
+	hsw_write_dcomp(dev_priv, val);
 	ndelay(100);
 
 	if (wait_for((I915_READ(D_COMP) & D_COMP_RCOMP_IN_PROGRESS) == 0, 1))
@@ -7008,11 +7366,7 @@ static void hsw_restore_lcpll(struct drm_i915_private *dev_priv)
 	val = I915_READ(D_COMP);
 	val |= D_COMP_COMP_FORCE;
 	val &= ~D_COMP_COMP_DISABLE;
-	mutex_lock(&dev_priv->rps.hw_lock);
-	if (sandybridge_pcode_write(dev_priv, GEN6_PCODE_WRITE_D_COMP, val))
-		DRM_ERROR("Failed to enable D_COMP\n");
-	mutex_unlock(&dev_priv->rps.hw_lock);
-	POSTING_READ(D_COMP);
+	hsw_write_dcomp(dev_priv, val);
 
 	val = I915_READ(LCPLL_CTL);
 	val &= ~LCPLL_PLL_DISABLE;
@@ -7066,8 +7420,6 @@ void hsw_enable_pc8(struct drm_i915_private *dev_priv)
 	struct drm_device *dev = dev_priv->dev;
 	uint32_t val;
 
-	WARN_ON(!HAS_PC8(dev));
-
 	DRM_DEBUG_KMS("Enabling package C8+\n");
 
 	if (dev_priv->pch_id == INTEL_PCH_LPT_LP_DEVICE_ID_TYPE) {
@@ -7077,7 +7429,6 @@ void hsw_enable_pc8(struct drm_i915_private *dev_priv)
 	}
 
 	lpt_disable_clkout_dp(dev);
-	hsw_runtime_pm_disable_interrupts(dev);
 	hsw_disable_lcpll(dev_priv, true, true);
 }
 
@@ -7086,12 +7437,9 @@ void hsw_disable_pc8(struct drm_i915_private *dev_priv)
 	struct drm_device *dev = dev_priv->dev;
 	uint32_t val;
 
-	WARN_ON(!HAS_PC8(dev));
-
 	DRM_DEBUG_KMS("Disabling package C8+\n");
 
 	hsw_restore_lcpll(dev_priv);
-	hsw_runtime_pm_restore_interrupts(dev);
 	lpt_init_pch_refclk(dev);
 
 	if (dev_priv->pch_id == INTEL_PCH_LPT_LP_DEVICE_ID_TYPE) {
@@ -7101,10 +7449,11 @@ void hsw_disable_pc8(struct drm_i915_private *dev_priv)
 	}
 
 	intel_prepare_ddi(dev);
-	i915_gem_init_swizzling(dev);
-	mutex_lock(&dev_priv->rps.hw_lock);
-	gen6_update_ring_freq(dev);
-	mutex_unlock(&dev_priv->rps.hw_lock);
+}
+
+static void snb_modeset_global_resources(struct drm_device *dev)
+{
+	modeset_update_crtc_power_domains(dev);
 }
 
 static void haswell_modeset_global_resources(struct drm_device *dev)
@@ -7116,39 +7465,15 @@ static int haswell_crtc_mode_set(struct drm_crtc *crtc,
 				 int x, int y,
 				 struct drm_framebuffer *fb)
 {
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	int plane = intel_crtc->plane;
-	int ret;
 
 	if (!intel_ddi_pll_select(intel_crtc))
 		return -EINVAL;
 	intel_ddi_pll_enable(intel_crtc);
 
-	if (intel_crtc->config.has_dp_encoder)
-		intel_dp_set_m_n(intel_crtc);
-
 	intel_crtc->lowfreq_avail = false;
 
-	intel_set_pipe_timings(intel_crtc);
-
-	if (intel_crtc->config.has_pch_encoder) {
-		intel_cpu_transcoder_set_m_n(intel_crtc,
-					     &intel_crtc->config.fdi_m_n);
-	}
-
-	haswell_set_pipeconf(crtc);
-
-	intel_set_pipe_csc(crtc);
-
-	/* Set up the display plane register */
-	I915_WRITE(DSPCNTR(plane), DISPPLANE_GAMMA_ENABLE | DISPPLANE_PIPE_CSC_ENABLE);
-	POSTING_READ(DSPCNTR(plane));
-
-	ret = intel_pipe_set_base(crtc, x, y, fb);
-
-	return ret;
+	return 0;
 }
 
 static bool haswell_get_pipe_config(struct intel_crtc *crtc,
@@ -7228,38 +7553,6 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc,
 	return true;
 }
 
-static int intel_crtc_mode_set(struct drm_crtc *crtc,
-			       int x, int y,
-			       struct drm_framebuffer *fb)
-{
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_encoder *encoder;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct drm_display_mode *mode = &intel_crtc->config.requested_mode;
-	int pipe = intel_crtc->pipe;
-	int ret;
-
-	drm_vblank_pre_modeset(dev, pipe);
-
-	ret = dev_priv->display.crtc_mode_set(crtc, x, y, fb);
-
-	drm_vblank_post_modeset(dev, pipe);
-
-	if (ret != 0)
-		return ret;
-
-	for_each_encoder_on_crtc(dev, crtc, encoder) {
-		DRM_DEBUG_KMS("[ENCODER:%d:%s] set [MODE:%d:%s]\n",
-			encoder->base.base.id,
-			drm_get_encoder_name(&encoder->base),
-			mode->base.id, mode->name);
-		encoder->mode_set(encoder);
-	}
-
-	return 0;
-}
-
 static struct {
 	int clock;
 	u32 config;
@@ -7374,8 +7667,6 @@ static void haswell_write_eld(struct drm_connector *connector,
 {
 	struct drm_i915_private *dev_priv = connector->dev->dev_private;
 	uint8_t *eld = connector->eld;
-	struct drm_device *dev = crtc->dev;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	uint32_t eldv;
 	uint32_t i;
 	int len;
@@ -7387,17 +7678,14 @@ static void haswell_write_eld(struct drm_connector *connector,
 	int aud_config = HSW_AUD_CFG(pipe);
 	int aud_cntrl_st2 = HSW_AUD_PIN_ELD_CP_VLD;
 
-
-	DRM_DEBUG_DRIVER("HDMI: Haswell Audio initialize....\n");
-
 	/* Audio output enable */
 	DRM_DEBUG_DRIVER("HDMI audio: enable codec\n");
 	tmp = I915_READ(aud_cntrl_st2);
 	tmp |= (AUDIO_OUTPUT_ENABLE_A << (pipe * 4));
 	I915_WRITE(aud_cntrl_st2, tmp);
+	POSTING_READ(aud_cntrl_st2);
 
-	/* Wait for 1 vertical blank */
-	intel_wait_for_vblank(dev, pipe);
+	assert_pipe_disabled(dev_priv, to_intel_crtc(crtc)->pipe);
 
 	/* Set ELD valid state */
 	tmp = I915_READ(aud_cntrl_st2);
@@ -7417,7 +7705,6 @@ static void haswell_write_eld(struct drm_connector *connector,
 	DRM_DEBUG_DRIVER("ELD on pipe %c\n", pipe_name(pipe));
 
 	eldv = AUDIO_ELD_VALID_A << (pipe * 4);
-	intel_crtc->eld_vld = true;
 
 	if (intel_pipe_has_type(crtc, INTEL_OUTPUT_DISPLAYPORT)) {
 		DRM_DEBUG_DRIVER("ELD: DisplayPort detected\n");
@@ -7564,9 +7851,9 @@ void intel_write_eld(struct drm_encoder *encoder,
 
 	DRM_DEBUG_DRIVER("ELD on [CONNECTOR:%d:%s], [ENCODER:%d:%s]\n",
 			 connector->base.id,
-			 drm_get_connector_name(connector),
+			 connector->name,
 			 connector->encoder->base.id,
-			 drm_get_encoder_name(connector->encoder));
+			 connector->encoder->name);
 
 	connector->eld[6] = drm_av_sync_delay(connector, mode) / 2;
 
@@ -7579,29 +7866,33 @@ static void i845_update_cursor(struct drm_crtc *crtc, u32 base)
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	bool visible = base != 0;
-	u32 cntl;
+	uint32_t cntl;
 
-	if (intel_crtc->cursor_visible == visible)
-		return;
-
-	cntl = I915_READ(_CURACNTR);
-	if (visible) {
+	if (base != intel_crtc->cursor_base) {
 		/* On these chipsets we can only modify the base whilst
 		 * the cursor is disabled.
 		 */
+		if (intel_crtc->cursor_cntl) {
+			I915_WRITE(_CURACNTR, 0);
+			POSTING_READ(_CURACNTR);
+			intel_crtc->cursor_cntl = 0;
+		}
+
 		I915_WRITE(_CURABASE, base);
+		POSTING_READ(_CURABASE);
+	}
 
-		cntl &= ~(CURSOR_FORMAT_MASK);
-		/* XXX width must be 64, stride 256 => 0x00 << 28 */
-		cntl |= CURSOR_ENABLE |
+	/* XXX width must be 64, stride 256 => 0x00 << 28 */
+	cntl = 0;
+	if (base)
+		cntl = (CURSOR_ENABLE |
 			CURSOR_GAMMA_ENABLE |
-			CURSOR_FORMAT_ARGB;
-	} else
-		cntl &= ~(CURSOR_ENABLE | CURSOR_GAMMA_ENABLE);
-	I915_WRITE(_CURACNTR, cntl);
-
-	intel_crtc->cursor_visible = visible;
+			CURSOR_FORMAT_ARGB);
+	if (intel_crtc->cursor_cntl != cntl) {
+		I915_WRITE(_CURACNTR, cntl);
+		POSTING_READ(_CURACNTR);
+		intel_crtc->cursor_cntl = cntl;
+	}
 }
 
 static void i9xx_update_cursor(struct drm_crtc *crtc, u32 base)
@@ -7610,16 +7901,12 @@ static void i9xx_update_cursor(struct drm_crtc *crtc, u32 base)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_crtc->pipe;
-	bool visible = base != 0;
+	uint32_t cntl;
 
-	if (intel_crtc->cursor_visible != visible) {
-		int16_t width = intel_crtc->cursor_width;
-		uint32_t cntl = I915_READ(CURCNTR(pipe));
-		if (base) {
-			cntl &= ~(CURSOR_MODE | MCURSOR_PIPE_SELECT);
-			cntl |= MCURSOR_GAMMA_ENABLE;
-
-			switch (width) {
+	cntl = 0;
+	if (base) {
+		cntl = MCURSOR_GAMMA_ENABLE;
+		switch (intel_crtc->cursor_width) {
 			case 64:
 				cntl |= CURSOR_MODE_64_ARGB_AX;
 				break;
@@ -7632,18 +7919,16 @@ static void i9xx_update_cursor(struct drm_crtc *crtc, u32 base)
 			default:
 				WARN_ON(1);
 				return;
-			}
-			cntl |= pipe << 28; /* Connect to correct pipe */
-		} else {
-			cntl &= ~(CURSOR_MODE | MCURSOR_GAMMA_ENABLE);
-			cntl |= CURSOR_MODE_DISABLE;
 		}
+		cntl |= pipe << 28; /* Connect to correct pipe */
+	}
+	if (intel_crtc->cursor_cntl != cntl) {
 		I915_WRITE(CURCNTR(pipe), cntl);
-
-		intel_crtc->cursor_visible = visible;
+		POSTING_READ(CURCNTR(pipe));
+		intel_crtc->cursor_cntl = cntl;
 	}
+
 	/* and commit changes on next vblank */
-	POSTING_READ(CURCNTR(pipe));
 	I915_WRITE(CURBASE(pipe), base);
 	POSTING_READ(CURBASE(pipe));
 }
@@ -7654,15 +7939,12 @@ static void ivb_update_cursor(struct drm_crtc *crtc, u32 base)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_crtc->pipe;
-	bool visible = base != 0;
-
-	if (intel_crtc->cursor_visible != visible) {
-		int16_t width = intel_crtc->cursor_width;
-		uint32_t cntl = I915_READ(CURCNTR_IVB(pipe));
-		if (base) {
-			cntl &= ~CURSOR_MODE;
-			cntl |= MCURSOR_GAMMA_ENABLE;
-			switch (width) {
+	uint32_t cntl;
+
+	cntl = 0;
+	if (base) {
+		cntl = MCURSOR_GAMMA_ENABLE;
+		switch (intel_crtc->cursor_width) {
 			case 64:
 				cntl |= CURSOR_MODE_64_ARGB_AX;
 				break;
@@ -7675,23 +7957,20 @@ static void ivb_update_cursor(struct drm_crtc *crtc, u32 base)
 			default:
 				WARN_ON(1);
 				return;
-			}
-		} else {
-			cntl &= ~(CURSOR_MODE | MCURSOR_GAMMA_ENABLE);
-			cntl |= CURSOR_MODE_DISABLE;
-		}
-		if (IS_HASWELL(dev) || IS_BROADWELL(dev)) {
-			cntl |= CURSOR_PIPE_CSC_ENABLE;
-			cntl &= ~CURSOR_TRICKLE_FEED_DISABLE;
 		}
-		I915_WRITE(CURCNTR_IVB(pipe), cntl);
+	}
+	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
+		cntl |= CURSOR_PIPE_CSC_ENABLE;
 
-		intel_crtc->cursor_visible = visible;
+	if (intel_crtc->cursor_cntl != cntl) {
+		I915_WRITE(CURCNTR(pipe), cntl);
+		POSTING_READ(CURCNTR(pipe));
+		intel_crtc->cursor_cntl = cntl;
 	}
+
 	/* and commit changes on next vblank */
-	POSTING_READ(CURCNTR_IVB(pipe));
-	I915_WRITE(CURBASE_IVB(pipe), base);
-	POSTING_READ(CURBASE_IVB(pipe));
+	I915_WRITE(CURBASE(pipe), base);
+	POSTING_READ(CURBASE(pipe));
 }
 
 /* If no-part of the cursor is visible on the framebuffer, then the GPU may hang... */
@@ -7705,7 +7984,6 @@ static void intel_crtc_update_cursor(struct drm_crtc *crtc,
 	int x = intel_crtc->cursor_x;
 	int y = intel_crtc->cursor_y;
 	u32 base = 0, pos = 0;
-	bool visible;
 
 	if (on)
 		base = intel_crtc->cursor_addr;
@@ -7734,20 +8012,18 @@ static void intel_crtc_update_cursor(struct drm_crtc *crtc,
 	}
 	pos |= y << CURSOR_Y_SHIFT;
 
-	visible = base != 0;
-	if (!visible && !intel_crtc->cursor_visible)
+	if (base == 0 && intel_crtc->cursor_base == 0)
 		return;
 
-	if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev) || IS_BROADWELL(dev)) {
-		I915_WRITE(CURPOS_IVB(pipe), pos);
+	I915_WRITE(CURPOS(pipe), pos);
+
+	if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev) || IS_BROADWELL(dev))
 		ivb_update_cursor(crtc, base);
-	} else {
-		I915_WRITE(CURPOS(pipe), pos);
-		if (IS_845G(dev) || IS_I865G(dev))
-			i845_update_cursor(crtc, base);
-		else
-			i9xx_update_cursor(crtc, base);
-	}
+	else if (IS_845G(dev) || IS_I865G(dev))
+		i845_update_cursor(crtc, base);
+	else
+		i9xx_update_cursor(crtc, base);
+	intel_crtc->cursor_base = base;
 }
 
 static int intel_crtc_cursor_set(struct drm_crtc *crtc,
@@ -8015,7 +8291,8 @@ mode_fits_in_fbdev(struct drm_device *dev,
 
 bool intel_get_load_detect_pipe(struct drm_connector *connector,
 				struct drm_display_mode *mode,
-				struct intel_load_detect_pipe *old)
+				struct intel_load_detect_pipe *old,
+				struct drm_modeset_acquire_ctx *ctx)
 {
 	struct intel_crtc *intel_crtc;
 	struct intel_encoder *intel_encoder =
@@ -8025,11 +8302,19 @@ bool intel_get_load_detect_pipe(struct drm_connector *connector,
 	struct drm_crtc *crtc = NULL;
 	struct drm_device *dev = encoder->dev;
 	struct drm_framebuffer *fb;
-	int i = -1;
+	struct drm_mode_config *config = &dev->mode_config;
+	int ret, i = -1;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s], [ENCODER:%d:%s]\n",
-		      connector->base.id, drm_get_connector_name(connector),
-		      encoder->base.id, drm_get_encoder_name(encoder));
+		      connector->base.id, connector->name,
+		      encoder->base.id, encoder->name);
+
+	drm_modeset_acquire_init(ctx, 0);
+
+retry:
+	ret = drm_modeset_lock(&config->connection_mutex, ctx);
+	if (ret)
+		goto fail_unlock;
 
 	/*
 	 * Algorithm gets a little messy:
@@ -8045,7 +8330,9 @@ bool intel_get_load_detect_pipe(struct drm_connector *connector,
 	if (encoder->crtc) {
 		crtc = encoder->crtc;
 
-		mutex_lock(&crtc->mutex);
+		ret = drm_modeset_lock(&crtc->mutex, ctx);
+		if (ret)
+			goto fail_unlock;
 
 		old->dpms_mode = connector->dpms;
 		old->load_detect_temp = false;
@@ -8058,7 +8345,7 @@ bool intel_get_load_detect_pipe(struct drm_connector *connector,
 	}
 
 	/* Find an unused one (if possible) */
-	list_for_each_entry(possible_crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, possible_crtc) {
 		i++;
 		if (!(encoder->possible_crtcs & (1 << i)))
 			continue;
@@ -8073,10 +8360,12 @@ bool intel_get_load_detect_pipe(struct drm_connector *connector,
 	 */
 	if (!crtc) {
 		DRM_DEBUG_KMS("no pipe available for load-detect\n");
-		return false;
+		goto fail_unlock;
 	}
 
-	mutex_lock(&crtc->mutex);
+	ret = drm_modeset_lock(&crtc->mutex, ctx);
+	if (ret)
+		goto fail_unlock;
 	intel_encoder->new_crtc = to_intel_crtc(crtc);
 	to_intel_connector(connector)->new_encoder = intel_encoder;
 
@@ -8126,12 +8415,21 @@ bool intel_get_load_detect_pipe(struct drm_connector *connector,
 		intel_crtc->new_config = &intel_crtc->config;
 	else
 		intel_crtc->new_config = NULL;
-	mutex_unlock(&crtc->mutex);
+fail_unlock:
+	if (ret == -EDEADLK) {
+		drm_modeset_backoff(ctx);
+		goto retry;
+	}
+
+	drm_modeset_drop_locks(ctx);
+	drm_modeset_acquire_fini(ctx);
+
 	return false;
 }
 
 void intel_release_load_detect_pipe(struct drm_connector *connector,
-				    struct intel_load_detect_pipe *old)
+				    struct intel_load_detect_pipe *old,
+				    struct drm_modeset_acquire_ctx *ctx)
 {
 	struct intel_encoder *intel_encoder =
 		intel_attached_encoder(connector);
@@ -8140,8 +8438,8 @@ void intel_release_load_detect_pipe(struct drm_connector *connector,
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s], [ENCODER:%d:%s]\n",
-		      connector->base.id, drm_get_connector_name(connector),
-		      encoder->base.id, drm_get_encoder_name(encoder));
+		      connector->base.id, connector->name,
+		      encoder->base.id, encoder->name);
 
 	if (old->load_detect_temp) {
 		to_intel_connector(connector)->new_encoder = NULL;
@@ -8155,7 +8453,7 @@ void intel_release_load_detect_pipe(struct drm_connector *connector,
 			drm_framebuffer_unreference(old->release_fb);
 		}
 
-		mutex_unlock(&crtc->mutex);
+		goto unlock;
 		return;
 	}
 
@@ -8163,7 +8461,9 @@ void intel_release_load_detect_pipe(struct drm_connector *connector,
 	if (old->dpms_mode != DRM_MODE_DPMS_ON)
 		connector->funcs->dpms(connector, old->dpms_mode);
 
-	mutex_unlock(&crtc->mutex);
+unlock:
+	drm_modeset_drop_locks(ctx);
+	drm_modeset_acquire_fini(ctx);
 }
 
 static int i9xx_pll_refclk(struct drm_device *dev,
@@ -8449,7 +8749,7 @@ void intel_mark_idle(struct drm_device *dev)
 	if (!i915.powersave)
 		goto out;
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, crtc) {
 		if (!crtc->primary->fb)
 			continue;
 
@@ -8464,7 +8764,7 @@ out:
 }
 
 void intel_mark_fb_busy(struct drm_i915_gem_object *obj,
-			struct intel_ring_buffer *ring)
+			struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = obj->base.dev;
 	struct drm_crtc *crtc;
@@ -8472,7 +8772,7 @@ void intel_mark_fb_busy(struct drm_i915_gem_object *obj,
 	if (!i915.powersave)
 		return;
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, crtc) {
 		if (!crtc->primary->fb)
 			continue;
 
@@ -8560,7 +8860,7 @@ static void do_intel_finish_page_flip(struct drm_device *dev,
 	if (work->event)
 		drm_send_vblank_event(dev, intel_crtc->pipe, work->event);
 
-	drm_vblank_put(dev, intel_crtc->pipe);
+	drm_crtc_vblank_put(crtc);
 
 	spin_unlock_irqrestore(&dev->event_lock, flags);
 
@@ -8587,6 +8887,48 @@ void intel_finish_page_flip_plane(struct drm_device *dev, int plane)
 	do_intel_finish_page_flip(dev, crtc);
 }
 
+/* Is 'a' after or equal to 'b'? */
+static bool g4x_flip_count_after_eq(u32 a, u32 b)
+{
+	return !((a - b) & 0x80000000);
+}
+
+static bool page_flip_finished(struct intel_crtc *crtc)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	/*
+	 * The relevant registers doen't exist on pre-ctg.
+	 * As the flip done interrupt doesn't trigger for mmio
+	 * flips on gmch platforms, a flip count check isn't
+	 * really needed there. But since ctg has the registers,
+	 * include it in the check anyway.
+	 */
+	if (INTEL_INFO(dev)->gen < 5 && !IS_G4X(dev))
+		return true;
+
+	/*
+	 * A DSPSURFLIVE check isn't enough in case the mmio and CS flips
+	 * used the same base address. In that case the mmio flip might
+	 * have completed, but the CS hasn't even executed the flip yet.
+	 *
+	 * A flip count check isn't enough as the CS might have updated
+	 * the base address just after start of vblank, but before we
+	 * managed to process the interrupt. This means we'd complete the
+	 * CS flip too soon.
+	 *
+	 * Combining both checks should get us a good enough result. It may
+	 * still happen that the CS flip has been executed, but has not
+	 * yet actually completed. But in case the base address is the same
+	 * anyway, we don't really care.
+	 */
+	return (I915_READ(DSPSURFLIVE(crtc->plane)) & ~0xfff) ==
+		crtc->unpin_work->gtt_offset &&
+		g4x_flip_count_after_eq(I915_READ(PIPE_FLIPCOUNT_GM45(crtc->pipe)),
+				    crtc->unpin_work->flip_count);
+}
+
 void intel_prepare_page_flip(struct drm_device *dev, int plane)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -8599,12 +8941,12 @@ void intel_prepare_page_flip(struct drm_device *dev, int plane)
 	 * is also accompanied by a spurious intel_prepare_page_flip().
 	 */
 	spin_lock_irqsave(&dev->event_lock, flags);
-	if (intel_crtc->unpin_work)
+	if (intel_crtc->unpin_work && page_flip_finished(intel_crtc))
 		atomic_inc_not_zero(&intel_crtc->unpin_work->pending);
 	spin_unlock_irqrestore(&dev->event_lock, flags);
 }
 
-inline static void intel_mark_page_flip_active(struct intel_crtc *intel_crtc)
+static inline void intel_mark_page_flip_active(struct intel_crtc *intel_crtc)
 {
 	/* Ensure that the work item is consistent when activating it ... */
 	smp_wmb();
@@ -8617,21 +8959,16 @@ static int intel_gen2_queue_flip(struct drm_device *dev,
 				 struct drm_crtc *crtc,
 				 struct drm_framebuffer *fb,
 				 struct drm_i915_gem_object *obj,
+				 struct intel_engine_cs *ring,
 				 uint32_t flags)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	u32 flip_mask;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
 	int ret;
 
-	ret = intel_pin_and_fence_fb_obj(dev, obj, ring);
-	if (ret)
-		goto err;
-
 	ret = intel_ring_begin(ring, 6);
 	if (ret)
-		goto err_unpin;
+		return ret;
 
 	/* Can't queue multiple flips, so wait for the previous
 	 * one to finish before executing the next.
@@ -8645,38 +8982,28 @@ static int intel_gen2_queue_flip(struct drm_device *dev,
 	intel_ring_emit(ring, MI_DISPLAY_FLIP |
 			MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
 	intel_ring_emit(ring, fb->pitches[0]);
-	intel_ring_emit(ring, i915_gem_obj_ggtt_offset(obj) + intel_crtc->dspaddr_offset);
+	intel_ring_emit(ring, intel_crtc->unpin_work->gtt_offset);
 	intel_ring_emit(ring, 0); /* aux display base address, unused */
 
 	intel_mark_page_flip_active(intel_crtc);
 	__intel_ring_advance(ring);
 	return 0;
-
-err_unpin:
-	intel_unpin_fb_obj(obj);
-err:
-	return ret;
 }
 
 static int intel_gen3_queue_flip(struct drm_device *dev,
 				 struct drm_crtc *crtc,
 				 struct drm_framebuffer *fb,
 				 struct drm_i915_gem_object *obj,
+				 struct intel_engine_cs *ring,
 				 uint32_t flags)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	u32 flip_mask;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
 	int ret;
 
-	ret = intel_pin_and_fence_fb_obj(dev, obj, ring);
-	if (ret)
-		goto err;
-
 	ret = intel_ring_begin(ring, 6);
 	if (ret)
-		goto err_unpin;
+		return ret;
 
 	if (intel_crtc->plane)
 		flip_mask = MI_WAIT_FOR_PLANE_B_FLIP;
@@ -8687,38 +9014,29 @@ static int intel_gen3_queue_flip(struct drm_device *dev,
 	intel_ring_emit(ring, MI_DISPLAY_FLIP_I915 |
 			MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
 	intel_ring_emit(ring, fb->pitches[0]);
-	intel_ring_emit(ring, i915_gem_obj_ggtt_offset(obj) + intel_crtc->dspaddr_offset);
+	intel_ring_emit(ring, intel_crtc->unpin_work->gtt_offset);
 	intel_ring_emit(ring, MI_NOOP);
 
 	intel_mark_page_flip_active(intel_crtc);
 	__intel_ring_advance(ring);
 	return 0;
-
-err_unpin:
-	intel_unpin_fb_obj(obj);
-err:
-	return ret;
 }
 
 static int intel_gen4_queue_flip(struct drm_device *dev,
 				 struct drm_crtc *crtc,
 				 struct drm_framebuffer *fb,
 				 struct drm_i915_gem_object *obj,
+				 struct intel_engine_cs *ring,
 				 uint32_t flags)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	uint32_t pf, pipesrc;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
 	int ret;
 
-	ret = intel_pin_and_fence_fb_obj(dev, obj, ring);
-	if (ret)
-		goto err;
-
 	ret = intel_ring_begin(ring, 4);
 	if (ret)
-		goto err_unpin;
+		return ret;
 
 	/* i965+ uses the linear or tiled offsets from the
 	 * Display Registers (which do not change across a page-flip)
@@ -8727,8 +9045,7 @@ static int intel_gen4_queue_flip(struct drm_device *dev,
 	intel_ring_emit(ring, MI_DISPLAY_FLIP |
 			MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
 	intel_ring_emit(ring, fb->pitches[0]);
-	intel_ring_emit(ring,
-			(i915_gem_obj_ggtt_offset(obj) + intel_crtc->dspaddr_offset) |
+	intel_ring_emit(ring, intel_crtc->unpin_work->gtt_offset |
 			obj->tiling_mode);
 
 	/* XXX Enabling the panel-fitter across page-flip is so far
@@ -8742,37 +9059,28 @@ static int intel_gen4_queue_flip(struct drm_device *dev,
 	intel_mark_page_flip_active(intel_crtc);
 	__intel_ring_advance(ring);
 	return 0;
-
-err_unpin:
-	intel_unpin_fb_obj(obj);
-err:
-	return ret;
 }
 
 static int intel_gen6_queue_flip(struct drm_device *dev,
 				 struct drm_crtc *crtc,
 				 struct drm_framebuffer *fb,
 				 struct drm_i915_gem_object *obj,
+				 struct intel_engine_cs *ring,
 				 uint32_t flags)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
 	uint32_t pf, pipesrc;
 	int ret;
 
-	ret = intel_pin_and_fence_fb_obj(dev, obj, ring);
-	if (ret)
-		goto err;
-
 	ret = intel_ring_begin(ring, 4);
 	if (ret)
-		goto err_unpin;
+		return ret;
 
 	intel_ring_emit(ring, MI_DISPLAY_FLIP |
 			MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
 	intel_ring_emit(ring, fb->pitches[0] | obj->tiling_mode);
-	intel_ring_emit(ring, i915_gem_obj_ggtt_offset(obj) + intel_crtc->dspaddr_offset);
+	intel_ring_emit(ring, intel_crtc->unpin_work->gtt_offset);
 
 	/* Contrary to the suggestions in the documentation,
 	 * "Enable Panel Fitter" does not seem to be required when page
@@ -8787,34 +9095,20 @@ static int intel_gen6_queue_flip(struct drm_device *dev,
 	intel_mark_page_flip_active(intel_crtc);
 	__intel_ring_advance(ring);
 	return 0;
-
-err_unpin:
-	intel_unpin_fb_obj(obj);
-err:
-	return ret;
 }
 
 static int intel_gen7_queue_flip(struct drm_device *dev,
 				 struct drm_crtc *crtc,
 				 struct drm_framebuffer *fb,
 				 struct drm_i915_gem_object *obj,
+				 struct intel_engine_cs *ring,
 				 uint32_t flags)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct intel_ring_buffer *ring;
 	uint32_t plane_bit = 0;
 	int len, ret;
 
-	ring = obj->ring;
-	if (IS_VALLEYVIEW(dev) || ring == NULL || ring->id != RCS)
-		ring = &dev_priv->ring[BCS];
-
-	ret = intel_pin_and_fence_fb_obj(dev, obj, ring);
-	if (ret)
-		goto err;
-
-	switch(intel_crtc->plane) {
+	switch (intel_crtc->plane) {
 	case PLANE_A:
 		plane_bit = MI_DISPLAY_FLIP_IVB_PLANE_A;
 		break;
@@ -8826,13 +9120,20 @@ static int intel_gen7_queue_flip(struct drm_device *dev,
 		break;
 	default:
 		WARN_ONCE(1, "unknown plane in flip command\n");
-		ret = -ENODEV;
-		goto err_unpin;
+		return -ENODEV;
 	}
 
 	len = 4;
-	if (ring->id == RCS)
+	if (ring->id == RCS) {
 		len += 6;
+		/*
+		 * On Gen 8, SRM is now taking an extra dword to accommodate
+		 * 48bits addresses, and we need a NOOP for the batch size to
+		 * stay even.
+		 */
+		if (IS_GEN8(dev))
+			len += 2;
+	}
 
 	/*
 	 * BSpec MI_DISPLAY_FLIP for IVB:
@@ -8846,11 +9147,11 @@ static int intel_gen7_queue_flip(struct drm_device *dev,
 	 */
 	ret = intel_ring_cacheline_align(ring);
 	if (ret)
-		goto err_unpin;
+		return ret;
 
 	ret = intel_ring_begin(ring, len);
 	if (ret)
-		goto err_unpin;
+		return ret;
 
 	/* Unmask the flip-done completion message. Note that the bspec says that
 	 * we should do this for both the BCS and RCS, and that we must not unmask
@@ -8867,31 +9168,35 @@ static int intel_gen7_queue_flip(struct drm_device *dev,
 		intel_ring_emit(ring, ~(DERRMR_PIPEA_PRI_FLIP_DONE |
 					DERRMR_PIPEB_PRI_FLIP_DONE |
 					DERRMR_PIPEC_PRI_FLIP_DONE));
-		intel_ring_emit(ring, MI_STORE_REGISTER_MEM(1) |
-				MI_SRM_LRM_GLOBAL_GTT);
+		if (IS_GEN8(dev))
+			intel_ring_emit(ring, MI_STORE_REGISTER_MEM_GEN8(1) |
+					      MI_SRM_LRM_GLOBAL_GTT);
+		else
+			intel_ring_emit(ring, MI_STORE_REGISTER_MEM(1) |
+					      MI_SRM_LRM_GLOBAL_GTT);
 		intel_ring_emit(ring, DERRMR);
 		intel_ring_emit(ring, ring->scratch.gtt_offset + 256);
+		if (IS_GEN8(dev)) {
+			intel_ring_emit(ring, 0);
+			intel_ring_emit(ring, MI_NOOP);
+		}
 	}
 
 	intel_ring_emit(ring, MI_DISPLAY_FLIP_I915 | plane_bit);
 	intel_ring_emit(ring, (fb->pitches[0] | obj->tiling_mode));
-	intel_ring_emit(ring, i915_gem_obj_ggtt_offset(obj) + intel_crtc->dspaddr_offset);
+	intel_ring_emit(ring, intel_crtc->unpin_work->gtt_offset);
 	intel_ring_emit(ring, (MI_NOOP));
 
 	intel_mark_page_flip_active(intel_crtc);
 	__intel_ring_advance(ring);
 	return 0;
-
-err_unpin:
-	intel_unpin_fb_obj(obj);
-err:
-	return ret;
 }
 
 static int intel_default_queue_flip(struct drm_device *dev,
 				    struct drm_crtc *crtc,
 				    struct drm_framebuffer *fb,
 				    struct drm_i915_gem_object *obj,
+				    struct intel_engine_cs *ring,
 				    uint32_t flags)
 {
 	return -ENODEV;
@@ -8908,6 +9213,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	struct drm_i915_gem_object *obj = to_intel_framebuffer(fb)->obj;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_unpin_work *work;
+	struct intel_engine_cs *ring;
 	unsigned long flags;
 	int ret;
 
@@ -8936,7 +9242,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	work->old_fb_obj = to_intel_framebuffer(old_fb)->obj;
 	INIT_WORK(&work->work, intel_unpin_work_fn);
 
-	ret = drm_vblank_get(dev, intel_crtc->pipe);
+	ret = drm_crtc_vblank_get(crtc);
 	if (ret)
 		goto free_work;
 
@@ -8945,7 +9251,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	if (intel_crtc->unpin_work) {
 		spin_unlock_irqrestore(&dev->event_lock, flags);
 		kfree(work);
-		drm_vblank_put(dev, intel_crtc->pipe);
+		drm_crtc_vblank_put(crtc);
 
 		DRM_DEBUG_DRIVER("flip queue: crtc already busy\n");
 		return -EBUSY;
@@ -8973,10 +9279,30 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	atomic_inc(&intel_crtc->unpin_work_count);
 	intel_crtc->reset_counter = atomic_read(&dev_priv->gpu_error.reset_counter);
 
-	ret = dev_priv->display.queue_flip(dev, crtc, fb, obj, page_flip_flags);
+	if (INTEL_INFO(dev)->gen >= 5 || IS_G4X(dev))
+		work->flip_count = I915_READ(PIPE_FLIPCOUNT_GM45(intel_crtc->pipe)) + 1;
+
+	if (IS_VALLEYVIEW(dev)) {
+		ring = &dev_priv->ring[BCS];
+	} else if (INTEL_INFO(dev)->gen >= 7) {
+		ring = obj->ring;
+		if (ring == NULL || ring->id != RCS)
+			ring = &dev_priv->ring[BCS];
+	} else {
+		ring = &dev_priv->ring[RCS];
+	}
+
+	ret = intel_pin_and_fence_fb_obj(dev, obj, ring);
 	if (ret)
 		goto cleanup_pending;
 
+	work->gtt_offset =
+		i915_gem_obj_ggtt_offset(obj) + intel_crtc->dspaddr_offset;
+
+	ret = dev_priv->display.queue_flip(dev, crtc, fb, obj, ring, page_flip_flags);
+	if (ret)
+		goto cleanup_unpin;
+
 	intel_disable_fbc(dev);
 	intel_mark_fb_busy(obj, NULL);
 	mutex_unlock(&dev->struct_mutex);
@@ -8985,6 +9311,8 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 
 	return 0;
 
+cleanup_unpin:
+	intel_unpin_fb_obj(obj);
 cleanup_pending:
 	atomic_dec(&intel_crtc->unpin_work_count);
 	crtc->primary->fb = old_fb;
@@ -8997,7 +9325,7 @@ cleanup:
 	intel_crtc->unpin_work = NULL;
 	spin_unlock_irqrestore(&dev->event_lock, flags);
 
-	drm_vblank_put(dev, intel_crtc->pipe);
+	drm_crtc_vblank_put(crtc);
 free_work:
 	kfree(work);
 
@@ -9040,8 +9368,7 @@ static void intel_modeset_update_staged_output_state(struct drm_device *dev)
 			to_intel_crtc(encoder->base.crtc);
 	}
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list,
-			    base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		crtc->new_enabled = crtc->base.enabled;
 
 		if (crtc->new_enabled)
@@ -9072,21 +9399,20 @@ static void intel_modeset_commit_output_state(struct drm_device *dev)
 		encoder->base.crtc = &encoder->new_crtc->base;
 	}
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list,
-			    base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		crtc->base.enabled = crtc->new_enabled;
 	}
 }
 
 static void
-connected_sink_compute_bpp(struct intel_connector * connector,
+connected_sink_compute_bpp(struct intel_connector *connector,
 			   struct intel_crtc_config *pipe_config)
 {
 	int bpp = pipe_config->pipe_bpp;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s] checking for sink bpp constrains\n",
 		connector->base.base.id,
-		drm_get_connector_name(&connector->base));
+		connector->base.name);
 
 	/* Don't use an invalid EDID bpc value */
 	if (connector->base.display_info.bpc &&
@@ -9427,8 +9753,7 @@ intel_modeset_affected_pipes(struct drm_crtc *crtc, unsigned *modeset_pipes,
 	}
 
 	/* Check for pipes that will be enabled/disabled ... */
-	list_for_each_entry(intel_crtc, &dev->mode_config.crtc_list,
-			    base.head) {
+	for_each_intel_crtc(dev, intel_crtc) {
 		if (intel_crtc->base.enabled == intel_crtc->new_enabled)
 			continue;
 
@@ -9501,8 +9826,7 @@ intel_modeset_update_state(struct drm_device *dev, unsigned prepare_pipes)
 	intel_modeset_commit_output_state(dev);
 
 	/* Double check state. */
-	list_for_each_entry(intel_crtc, &dev->mode_config.crtc_list,
-			    base.head) {
+	for_each_intel_crtc(dev, intel_crtc) {
 		WARN_ON(intel_crtc->base.enabled != intel_crtc_in_use(&intel_crtc->base));
 		WARN_ON(intel_crtc->new_config &&
 			intel_crtc->new_config != &intel_crtc->config);
@@ -9631,6 +9955,12 @@ intel_pipe_config_compare(struct drm_device *dev,
 	PIPE_CONF_CHECK_I(adjusted_mode.crtc_vsync_end);
 
 	PIPE_CONF_CHECK_I(pixel_multiplier);
+	PIPE_CONF_CHECK_I(has_hdmi_sink);
+	if ((INTEL_INFO(dev)->gen < 8 && !IS_HASWELL(dev)) ||
+	    IS_VALLEYVIEW(dev))
+		PIPE_CONF_CHECK_I(limited_color_range);
+
+	PIPE_CONF_CHECK_I(has_audio);
 
 	PIPE_CONF_CHECK_FLAGS(adjusted_mode.flags,
 			      DRM_MODE_FLAG_INTERLACE);
@@ -9728,7 +10058,7 @@ check_encoder_state(struct drm_device *dev)
 
 		DRM_DEBUG_KMS("[ENCODER:%d:%s]\n",
 			      encoder->base.base.id,
-			      drm_get_encoder_name(&encoder->base));
+			      encoder->base.name);
 
 		WARN(&encoder->new_crtc->base != encoder->base.crtc,
 		     "encoder's stage crtc doesn't match current crtc\n");
@@ -9780,8 +10110,7 @@ check_crtc_state(struct drm_device *dev)
 	struct intel_encoder *encoder;
 	struct intel_crtc_config pipe_config;
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list,
-			    base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		bool enabled = false;
 		bool active = false;
 
@@ -9870,8 +10199,7 @@ check_shared_dpll_state(struct drm_device *dev)
 		     "pll on state mismatch (expected %i, found %i)\n",
 		     pll->on, active);
 
-		list_for_each_entry(crtc, &dev->mode_config.crtc_list,
-				    base.head) {
+		for_each_intel_crtc(dev, crtc) {
 			if (crtc->base.enabled && intel_crtc_to_shared_dpll(crtc) == pll)
 				enabled_crtcs++;
 			if (crtc->active && intel_crtc_to_shared_dpll(crtc) == pll)
@@ -9911,6 +10239,44 @@ void ironlake_check_encoder_dotclock(const struct intel_crtc_config *pipe_config
 	     pipe_config->adjusted_mode.crtc_clock, dotclock);
 }
 
+static void update_scanline_offset(struct intel_crtc *crtc)
+{
+	struct drm_device *dev = crtc->base.dev;
+
+	/*
+	 * The scanline counter increments at the leading edge of hsync.
+	 *
+	 * On most platforms it starts counting from vtotal-1 on the
+	 * first active line. That means the scanline counter value is
+	 * always one less than what we would expect. Ie. just after
+	 * start of vblank, which also occurs at start of hsync (on the
+	 * last active line), the scanline counter will read vblank_start-1.
+	 *
+	 * On gen2 the scanline counter starts counting from 1 instead
+	 * of vtotal-1, so we have to subtract one (or rather add vtotal-1
+	 * to keep the value positive), instead of adding one.
+	 *
+	 * On HSW+ the behaviour of the scanline counter depends on the output
+	 * type. For DP ports it behaves like most other platforms, but on HDMI
+	 * there's an extra 1 line difference. So we need to add two instead of
+	 * one to the value.
+	 */
+	if (IS_GEN2(dev)) {
+		const struct drm_display_mode *mode = &crtc->config.adjusted_mode;
+		int vtotal;
+
+		vtotal = mode->crtc_vtotal;
+		if (mode->flags & DRM_MODE_FLAG_INTERLACE)
+			vtotal /= 2;
+
+		crtc->scanline_offset = vtotal - 1;
+	} else if (HAS_DDI(dev) &&
+		   intel_pipe_has_type(&crtc->base, INTEL_OUTPUT_HDMI)) {
+		crtc->scanline_offset = 2;
+	} else
+		crtc->scanline_offset = 1;
+}
+
 static int __intel_set_mode(struct drm_crtc *crtc,
 			    struct drm_display_mode *mode,
 			    int x, int y, struct drm_framebuffer *fb)
@@ -10002,15 +10368,38 @@ static int __intel_set_mode(struct drm_crtc *crtc,
 	 * on the DPLL.
 	 */
 	for_each_intel_crtc_masked(dev, modeset_pipes, intel_crtc) {
-		ret = intel_crtc_mode_set(&intel_crtc->base,
-					  x, y, fb);
+		struct drm_framebuffer *old_fb;
+
+		mutex_lock(&dev->struct_mutex);
+		ret = intel_pin_and_fence_fb_obj(dev,
+						 to_intel_framebuffer(fb)->obj,
+						 NULL);
+		if (ret != 0) {
+			DRM_ERROR("pin & fence failed\n");
+			mutex_unlock(&dev->struct_mutex);
+			goto done;
+		}
+		old_fb = crtc->primary->fb;
+		if (old_fb)
+			intel_unpin_fb_obj(to_intel_framebuffer(old_fb)->obj);
+		mutex_unlock(&dev->struct_mutex);
+
+		crtc->primary->fb = fb;
+		crtc->x = x;
+		crtc->y = y;
+
+		ret = dev_priv->display.crtc_mode_set(&intel_crtc->base,
+						      x, y, fb);
 		if (ret)
 			goto done;
 	}
 
 	/* Now enable the clocks, plane, pipe, and connectors that we set up. */
-	for_each_intel_crtc_masked(dev, prepare_pipes, intel_crtc)
+	for_each_intel_crtc_masked(dev, prepare_pipes, intel_crtc) {
+		update_scanline_offset(intel_crtc);
+
 		dev_priv->display.crtc_enable(&intel_crtc->base);
+	}
 
 	/* FIXME: add subpixel order */
 done:
@@ -10086,7 +10475,7 @@ static int intel_set_config_save_state(struct drm_device *dev,
 	 * restored, not the drivers personal bookkeeping.
 	 */
 	count = 0;
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, crtc) {
 		config->save_crtc_enabled[count++] = crtc->enabled;
 	}
 
@@ -10112,7 +10501,7 @@ static void intel_set_config_restore_state(struct drm_device *dev,
 	int count;
 
 	count = 0;
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		crtc->new_enabled = config->save_crtc_enabled[count++];
 
 		if (crtc->new_enabled)
@@ -10236,7 +10625,7 @@ intel_modeset_stage_output_state(struct drm_device *dev,
 
 			DRM_DEBUG_KMS("[CONNECTOR:%d:%s] to [NOCRTC]\n",
 				connector->base.base.id,
-				drm_get_connector_name(&connector->base));
+				connector->base.name);
 		}
 
 
@@ -10271,7 +10660,7 @@ intel_modeset_stage_output_state(struct drm_device *dev,
 
 		DRM_DEBUG_KMS("[CONNECTOR:%d:%s] to [CRTC:%d]\n",
 			connector->base.base.id,
-			drm_get_connector_name(&connector->base),
+			connector->base.name,
 			new_crtc->base.id);
 	}
 
@@ -10302,8 +10691,7 @@ intel_modeset_stage_output_state(struct drm_device *dev,
 	}
 	/* Now we've also updated encoder->new_crtc for all encoders. */
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list,
-			    base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		crtc->new_enabled = false;
 
 		list_for_each_entry(encoder,
@@ -10516,7 +10904,7 @@ static void ibx_pch_dpll_disable(struct drm_i915_private *dev_priv,
 	struct intel_crtc *crtc;
 
 	/* Make sure no transcoder isn't still depending on us. */
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		if (intel_crtc_to_shared_dpll(crtc) == pll)
 			assert_pch_transcoder_disabled(dev_priv, crtc->pipe);
 	}
@@ -10573,16 +10961,6 @@ static void intel_crtc_init(struct drm_device *dev, int pipe)
 
 	drm_crtc_init(dev, &intel_crtc->base, &intel_crtc_funcs);
 
-	if (IS_GEN2(dev)) {
-		intel_crtc->max_cursor_width = GEN2_CURSOR_WIDTH;
-		intel_crtc->max_cursor_height = GEN2_CURSOR_HEIGHT;
-	} else {
-		intel_crtc->max_cursor_width = CURSOR_WIDTH;
-		intel_crtc->max_cursor_height = CURSOR_HEIGHT;
-	}
-	dev->mode_config.cursor_width = intel_crtc->max_cursor_width;
-	dev->mode_config.cursor_height = intel_crtc->max_cursor_height;
-
 	drm_mode_crtc_set_gamma_size(&intel_crtc->base, 256);
 	for (i = 0; i < 256; i++) {
 		intel_crtc->lut_r[i] = i;
@@ -10601,19 +10979,27 @@ static void intel_crtc_init(struct drm_device *dev, int pipe)
 		intel_crtc->plane = !pipe;
 	}
 
+	intel_crtc->cursor_base = ~0;
+	intel_crtc->cursor_cntl = ~0;
+
+	init_waitqueue_head(&intel_crtc->vbl_wait);
+
 	BUG_ON(pipe >= ARRAY_SIZE(dev_priv->plane_to_crtc_mapping) ||
 	       dev_priv->plane_to_crtc_mapping[intel_crtc->plane] != NULL);
 	dev_priv->plane_to_crtc_mapping[intel_crtc->plane] = &intel_crtc->base;
 	dev_priv->pipe_to_crtc_mapping[intel_crtc->pipe] = &intel_crtc->base;
 
 	drm_crtc_helper_add(&intel_crtc->base, &intel_helper_funcs);
+
+	WARN_ON(drm_crtc_index(&intel_crtc->base) != intel_crtc->pipe);
 }
 
 enum pipe intel_get_pipe_from_connector(struct intel_connector *connector)
 {
 	struct drm_encoder *encoder = connector->base.encoder;
+	struct drm_device *dev = connector->base.dev;
 
-	WARN_ON(!mutex_is_locked(&connector->base.dev->mode_config.mutex));
+	WARN_ON(!drm_modeset_is_locked(&dev->mode_config.connection_mutex));
 
 	if (!encoder)
 		return INVALID_PIPE;
@@ -10709,7 +11095,7 @@ static void intel_setup_outputs(struct drm_device *dev)
 
 	intel_lvds_init(dev);
 
-	if (!IS_ULT(dev))
+	if (!IS_ULT(dev) && !IS_CHERRYVIEW(dev) && dev_priv->vbt.int_crt_support)
 		intel_crt_init(dev);
 
 	if (HAS_DDI(dev)) {
@@ -10773,6 +11159,15 @@ static void intel_setup_outputs(struct drm_device *dev)
 				intel_dp_init(dev, VLV_DISPLAY_BASE + DP_C, PORT_C);
 		}
 
+		if (IS_CHERRYVIEW(dev)) {
+			if (I915_READ(VLV_DISPLAY_BASE + CHV_HDMID) & SDVO_DETECTED) {
+				intel_hdmi_init(dev, VLV_DISPLAY_BASE + CHV_HDMID,
+						PORT_D);
+				if (I915_READ(VLV_DISPLAY_BASE + DP_D) & DP_DETECTED)
+					intel_dp_init(dev, VLV_DISPLAY_BASE + DP_D, PORT_D);
+			}
+		}
+
 		intel_dsi_init(dev);
 	} else if (SUPPORTS_DIGITAL_OUTPUTS(dev)) {
 		bool found = false;
@@ -11002,6 +11397,8 @@ static void intel_init_display(struct drm_device *dev)
 
 	if (HAS_PCH_SPLIT(dev) || IS_G4X(dev))
 		dev_priv->display.find_dpll = g4x_find_best_dpll;
+	else if (IS_CHERRYVIEW(dev))
+		dev_priv->display.find_dpll = chv_find_best_dpll;
 	else if (IS_VALLEYVIEW(dev))
 		dev_priv->display.find_dpll = vlv_find_best_dpll;
 	else if (IS_PINEVIEW(dev))
@@ -11083,6 +11480,8 @@ static void intel_init_display(struct drm_device *dev)
 		} else if (IS_GEN6(dev)) {
 			dev_priv->display.fdi_link_train = gen6_fdi_link_train;
 			dev_priv->display.write_eld = ironlake_write_eld;
+			dev_priv->display.modeset_global_resources =
+				snb_modeset_global_resources;
 		} else if (IS_IVYBRIDGE(dev)) {
 			/* FIXME: detect B0+ stepping and use auto training */
 			dev_priv->display.fdi_link_train = ivb_manual_fdi_link_train;
@@ -11211,9 +11610,6 @@ static struct intel_quirk intel_quirks[] = {
 	/* ThinkPad T60 needs pipe A force quirk (bug #16494) */
 	{ 0x2782, 0x17aa, 0x201a, quirk_pipea_force },
 
-	/* 830 needs to leave pipe A & dpll A up */
-	{ 0x3577, PCI_ANY_ID, PCI_ANY_ID, quirk_pipea_force },
-
 	/* Lenovo U160 cannot use SSC on LVDS */
 	{ 0x0046, 0x17aa, 0x3920, quirk_ssc_force_disable },
 
@@ -11287,9 +11683,7 @@ void intel_modeset_init_hw(struct drm_device *dev)
 
 	intel_reset_dpio(dev);
 
-	mutex_lock(&dev->struct_mutex);
 	intel_enable_gt_powersave(dev);
-	mutex_unlock(&dev->struct_mutex);
 }
 
 void intel_modeset_suspend_hw(struct drm_device *dev)
@@ -11333,6 +11727,15 @@ void intel_modeset_init(struct drm_device *dev)
 		dev->mode_config.max_width = 8192;
 		dev->mode_config.max_height = 8192;
 	}
+
+	if (IS_GEN2(dev)) {
+		dev->mode_config.cursor_width = GEN2_CURSOR_WIDTH;
+		dev->mode_config.cursor_height = GEN2_CURSOR_HEIGHT;
+	} else {
+		dev->mode_config.cursor_width = MAX_CURSOR_WIDTH;
+		dev->mode_config.cursor_height = MAX_CURSOR_HEIGHT;
+	}
+
 	dev->mode_config.fb_base = dev_priv->gtt.mappable_base;
 
 	DRM_DEBUG_KMS("%d display pipe%s available.\n",
@@ -11362,12 +11765,11 @@ void intel_modeset_init(struct drm_device *dev)
 	/* Just in case the BIOS is doing something questionable. */
 	intel_disable_fbc(dev);
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock_all(dev);
 	intel_modeset_setup_hw_state(dev, false);
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock_all(dev);
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list,
-			    base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		if (!crtc->active)
 			continue;
 
@@ -11395,6 +11797,7 @@ static void intel_enable_pipe_a(struct drm_device *dev)
 	struct intel_connector *connector;
 	struct drm_connector *crt = NULL;
 	struct intel_load_detect_pipe load_detect_temp;
+	struct drm_modeset_acquire_ctx ctx;
 
 	/* We can't just switch on the pipe A, we need to set things up with a
 	 * proper mode and output configuration. As a gross hack, enable pipe A
@@ -11411,8 +11814,8 @@ static void intel_enable_pipe_a(struct drm_device *dev)
 	if (!crt)
 		return;
 
-	if (intel_get_load_detect_pipe(crt, NULL, &load_detect_temp))
-		intel_release_load_detect_pipe(crt, &load_detect_temp);
+	if (intel_get_load_detect_pipe(crt, NULL, &load_detect_temp, &ctx))
+		intel_release_load_detect_pipe(crt, &load_detect_temp, &ctx);
 
 
 }
@@ -11447,6 +11850,12 @@ static void intel_sanitize_crtc(struct intel_crtc *crtc)
 	reg = PIPECONF(crtc->config.cpu_transcoder);
 	I915_WRITE(reg, I915_READ(reg) & ~PIPECONF_FRAME_START_DELAY_MASK);
 
+	/* restore vblank interrupts to correct state */
+	if (crtc->active)
+		drm_vblank_on(dev, crtc->pipe);
+	else
+		drm_vblank_off(dev, crtc->pipe);
+
 	/* We need to sanitize the plane -> pipe mapping first because this will
 	 * disable the crtc (and hence change the state) if it is wrong. Note
 	 * that gen4+ has a fixed plane -> pipe mapping.  */
@@ -11525,16 +11934,25 @@ static void intel_sanitize_crtc(struct intel_crtc *crtc)
 			encoder->base.crtc = NULL;
 		}
 	}
-	if (crtc->active) {
+
+	if (crtc->active || IS_VALLEYVIEW(dev) || INTEL_INFO(dev)->gen < 5) {
 		/*
 		 * We start out with underrun reporting disabled to avoid races.
 		 * For correct bookkeeping mark this on active crtcs.
 		 *
+		 * Also on gmch platforms we dont have any hardware bits to
+		 * disable the underrun reporting. Which means we need to start
+		 * out with underrun reporting disabled also on inactive pipes,
+		 * since otherwise we'll complain about the garbage we read when
+		 * e.g. coming up after runtime pm.
+		 *
 		 * No protection against concurrent access is required - at
 		 * worst a fifo underrun happens which also sets this to false.
 		 */
 		crtc->cpu_fifo_underrun_disabled = true;
 		crtc->pch_fifo_underrun_disabled = true;
+
+		update_scanline_offset(crtc);
 	}
 }
 
@@ -11552,7 +11970,7 @@ static void intel_sanitize_encoder(struct intel_encoder *encoder)
 	if (encoder->connectors_active && !has_active_crtc) {
 		DRM_DEBUG_KMS("[ENCODER:%d:%s] has active connectors but no active pipe!\n",
 			      encoder->base.base.id,
-			      drm_get_encoder_name(&encoder->base));
+			      encoder->base.name);
 
 		/* Connector is active, but has no active pipe. This is
 		 * fallout from our resume register restoring. Disable
@@ -11560,7 +11978,7 @@ static void intel_sanitize_encoder(struct intel_encoder *encoder)
 		if (encoder->base.crtc) {
 			DRM_DEBUG_KMS("[ENCODER:%d:%s] manually disabled\n",
 				      encoder->base.base.id,
-				      drm_get_encoder_name(&encoder->base));
+				      encoder->base.name);
 			encoder->disable(encoder);
 		}
 		encoder->base.crtc = NULL;
@@ -11611,6 +12029,16 @@ void i915_redisable_vga(struct drm_device *dev)
 	i915_redisable_vga_power_on(dev);
 }
 
+static bool primary_get_hw_state(struct intel_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+
+	if (!crtc->active)
+		return false;
+
+	return I915_READ(DSPCNTR(crtc->plane)) & DISPLAY_PLANE_ENABLE;
+}
+
 static void intel_modeset_readout_hw_state(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -11620,8 +12048,7 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 	struct intel_connector *connector;
 	int i;
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list,
-			    base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		memset(&crtc->config, 0, sizeof(crtc->config));
 
 		crtc->config.quirks |= PIPE_CONFIG_QUIRK_INHERITED_MODE;
@@ -11630,7 +12057,7 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 								 &crtc->config);
 
 		crtc->base.enabled = crtc->active;
-		crtc->primary_enabled = crtc->active;
+		crtc->primary_enabled = primary_get_hw_state(crtc);
 
 		DRM_DEBUG_KMS("[CRTC:%d] hw state readout: %s\n",
 			      crtc->base.base.id,
@@ -11646,8 +12073,7 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 
 		pll->on = pll->get_hw_state(dev_priv, pll, &pll->hw_state);
 		pll->active = 0;
-		list_for_each_entry(crtc, &dev->mode_config.crtc_list,
-				    base.head) {
+		for_each_intel_crtc(dev, crtc) {
 			if (crtc->active && intel_crtc_to_shared_dpll(crtc) == pll)
 				pll->active++;
 		}
@@ -11672,7 +12098,7 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 		encoder->connectors_active = false;
 		DRM_DEBUG_KMS("[ENCODER:%d:%s] hw state readout: %s, pipe %c\n",
 			      encoder->base.base.id,
-			      drm_get_encoder_name(&encoder->base),
+			      encoder->base.name,
 			      encoder->base.crtc ? "enabled" : "disabled",
 			      pipe_name(pipe));
 	}
@@ -11689,7 +12115,7 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 		}
 		DRM_DEBUG_KMS("[CONNECTOR:%d:%s] hw state readout: %s\n",
 			      connector->base.base.id,
-			      drm_get_connector_name(&connector->base),
+			      connector->base.name,
 			      connector->base.encoder ? "enabled" : "disabled");
 	}
 }
@@ -11712,8 +12138,7 @@ void intel_modeset_setup_hw_state(struct drm_device *dev,
 	 * Note that this could go away if we move to using crtc_config
 	 * checking everywhere.
 	 */
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list,
-			    base.head) {
+	for_each_intel_crtc(dev, crtc) {
 		if (crtc->active && i915.fastboot) {
 			intel_mode_from_pipe_config(&crtc->base.mode, &crtc->config);
 			DRM_DEBUG_KMS("[CRTC:%d] found active mode: ",
@@ -11789,7 +12214,7 @@ void intel_modeset_gem_init(struct drm_device *dev)
 	 * for this.
 	 */
 	mutex_lock(&dev->struct_mutex);
-	list_for_each_entry(c, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, c) {
 		if (!c->primary->fb)
 			continue;
 
@@ -11835,7 +12260,7 @@ void intel_modeset_cleanup(struct drm_device *dev)
 
 	intel_unregister_dsm_handler();
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, crtc) {
 		/* Skip inactive CRTCs */
 		if (!crtc->primary->fb)
 			continue;
@@ -11933,6 +12358,7 @@ struct intel_display_error_state {
 	struct intel_pipe_error_state {
 		bool power_domain_on;
 		u32 source;
+		u32 stat;
 	} pipe[I915_MAX_PIPES];
 
 	struct intel_plane_error_state {
@@ -11990,15 +12416,9 @@ intel_display_capture_error_state(struct drm_device *dev)
 		if (!error->pipe[i].power_domain_on)
 			continue;
 
-		if (INTEL_INFO(dev)->gen <= 6 || IS_VALLEYVIEW(dev)) {
-			error->cursor[i].control = I915_READ(CURCNTR(i));
-			error->cursor[i].position = I915_READ(CURPOS(i));
-			error->cursor[i].base = I915_READ(CURBASE(i));
-		} else {
-			error->cursor[i].control = I915_READ(CURCNTR_IVB(i));
-			error->cursor[i].position = I915_READ(CURPOS_IVB(i));
-			error->cursor[i].base = I915_READ(CURBASE_IVB(i));
-		}
+		error->cursor[i].control = I915_READ(CURCNTR(i));
+		error->cursor[i].position = I915_READ(CURPOS(i));
+		error->cursor[i].base = I915_READ(CURBASE(i));
 
 		error->plane[i].control = I915_READ(DSPCNTR(i));
 		error->plane[i].stride = I915_READ(DSPSTRIDE(i));
@@ -12014,6 +12434,9 @@ intel_display_capture_error_state(struct drm_device *dev)
 		}
 
 		error->pipe[i].source = I915_READ(PIPESRC(i));
+
+		if (!HAS_PCH_SPLIT(dev))
+			error->pipe[i].stat = I915_READ(PIPESTAT(i));
 	}
 
 	error->num_transcoders = INTEL_INFO(dev)->num_pipes;
@@ -12064,6 +12487,7 @@ intel_display_print_error_state(struct drm_i915_error_state_buf *m,
 		err_printf(m, "  Power: %s\n",
 			   error->pipe[i].power_domain_on ? "on" : "off");
 		err_printf(m, "  SRC: %08x\n", error->pipe[i].source);
+		err_printf(m, "  STAT: %08x\n", error->pipe[i].stat);
 
 		err_printf(m, "Plane [%d]:\n", i);
 		err_printf(m, "  CNTR: %08x\n", error->plane[i].control);
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 2a00cb828d20..52fda950fd2a 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -64,6 +64,24 @@ static const struct dp_link_dpll vlv_dpll[] = {
 		{ .p1 = 2, .p2 = 2, .n = 1, .m1 = 2, .m2 = 27 } }
 };
 
+/*
+ * CHV supports eDP 1.4 that have  more link rates.
+ * Below only provides the fixed rate but exclude variable rate.
+ */
+static const struct dp_link_dpll chv_dpll[] = {
+	/*
+	 * CHV requires to program fractional division for m2.
+	 * m2 is stored in fixed point format using formula below
+	 * (m2_int << 22) | m2_fraction
+	 */
+	{ DP_LINK_BW_1_62,	/* m2_int = 32, m2_fraction = 1677722 */
+		{ .p1 = 4, .p2 = 2, .n = 1, .m1 = 2, .m2 = 0x819999a } },
+	{ DP_LINK_BW_2_7,	/* m2_int = 27, m2_fraction = 0 */
+		{ .p1 = 4, .p2 = 1, .n = 1, .m1 = 2, .m2 = 0x6c00000 } },
+	{ DP_LINK_BW_5_4,	/* m2_int = 27, m2_fraction = 0 */
+		{ .p1 = 2, .p2 = 1, .n = 1, .m1 = 2, .m2 = 0x6c00000 } }
+};
+
 /**
  * is_edp - is the given port attached to an eDP panel (either CPU or PCH)
  * @intel_dp: DP struct
@@ -330,8 +348,12 @@ static bool edp_have_panel_vdd(struct intel_dp *intel_dp)
 {
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
+	struct intel_encoder *intel_encoder = &intel_dig_port->base;
+	enum intel_display_power_domain power_domain;
 
-	return !dev_priv->pm.suspended &&
+	power_domain = intel_display_port_power_domain(intel_encoder);
+	return intel_display_power_enabled(dev_priv, power_domain) &&
 	       (I915_READ(_pp_ctrl_reg(intel_dp)) & EDP_FORCE_VDD) != 0;
 }
 
@@ -697,9 +719,9 @@ intel_dp_aux_init(struct intel_dp *intel_dp, struct intel_connector *connector)
 	DRM_DEBUG_KMS("registering %s bus for %s\n", name,
 		      connector->base.kdev->kobj.name);
 
-	ret = drm_dp_aux_register_i2c_bus(&intel_dp->aux);
+	ret = drm_dp_aux_register(&intel_dp->aux);
 	if (ret < 0) {
-		DRM_ERROR("drm_dp_aux_register_i2c_bus() for %s failed (%d)\n",
+		DRM_ERROR("drm_dp_aux_register() for %s failed (%d)\n",
 			  name, ret);
 		return;
 	}
@@ -709,7 +731,7 @@ intel_dp_aux_init(struct intel_dp *intel_dp, struct intel_connector *connector)
 				intel_dp->aux.ddc.dev.kobj.name);
 	if (ret < 0) {
 		DRM_ERROR("sysfs_create_link() for %s failed (%d)\n", name, ret);
-		drm_dp_aux_unregister_i2c_bus(&intel_dp->aux);
+		drm_dp_aux_unregister(&intel_dp->aux);
 	}
 }
 
@@ -739,6 +761,9 @@ intel_dp_set_clock(struct intel_encoder *encoder,
 	} else if (HAS_PCH_SPLIT(dev)) {
 		divisor = pch_dpll;
 		count = ARRAY_SIZE(pch_dpll);
+	} else if (IS_CHERRYVIEW(dev)) {
+		divisor = chv_dpll;
+		count = ARRAY_SIZE(chv_dpll);
 	} else if (IS_VALLEYVIEW(dev)) {
 		divisor = vlv_dpll;
 		count = ARRAY_SIZE(vlv_dpll);
@@ -755,6 +780,20 @@ intel_dp_set_clock(struct intel_encoder *encoder,
 	}
 }
 
+static void
+intel_dp_set_m2_n2(struct intel_crtc *crtc, struct intel_link_m_n *m_n)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	enum transcoder transcoder = crtc->config.cpu_transcoder;
+
+	I915_WRITE(PIPE_DATA_M2(transcoder),
+		TU_SIZE(m_n->tu) | m_n->gmch_m);
+	I915_WRITE(PIPE_DATA_N2(transcoder), m_n->gmch_n);
+	I915_WRITE(PIPE_LINK_M2(transcoder), m_n->link_m);
+	I915_WRITE(PIPE_LINK_N2(transcoder), m_n->link_n);
+}
+
 bool
 intel_dp_compute_config(struct intel_encoder *encoder,
 			struct intel_crtc_config *pipe_config)
@@ -780,6 +819,7 @@ intel_dp_compute_config(struct intel_encoder *encoder,
 		pipe_config->has_pch_encoder = true;
 
 	pipe_config->has_dp_encoder = true;
+	pipe_config->has_audio = intel_dp->has_audio;
 
 	if (is_edp(intel_dp) && intel_connector->panel.fixed_mode) {
 		intel_fixed_panel_mode(intel_connector->panel.fixed_mode,
@@ -880,6 +920,14 @@ found:
 			       pipe_config->port_clock,
 			       &pipe_config->dp_m_n);
 
+	if (intel_connector->panel.downclock_mode != NULL &&
+		intel_dp->drrs_state.type == SEAMLESS_DRRS_SUPPORT) {
+			intel_link_compute_m_n(bpp, lane_count,
+				intel_connector->panel.downclock_mode->clock,
+				pipe_config->port_clock,
+				&pipe_config->dp_m2_n2);
+	}
+
 	intel_dp_set_clock(encoder, pipe_config, intel_dp->link_bw);
 
 	return true;
@@ -915,7 +963,7 @@ static void ironlake_set_pll_cpu_edp(struct intel_dp *intel_dp)
 	udelay(500);
 }
 
-static void intel_dp_mode_set(struct intel_encoder *encoder)
+static void intel_dp_prepare(struct intel_encoder *encoder)
 {
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -950,7 +998,7 @@ static void intel_dp_mode_set(struct intel_encoder *encoder)
 	intel_dp->DP |= DP_VOLTAGE_0_4 | DP_PRE_EMPHASIS_0;
 	intel_dp->DP |= DP_PORT_WIDTH(intel_dp->lane_count);
 
-	if (intel_dp->has_audio) {
+	if (crtc->config.has_audio) {
 		DRM_DEBUG_DRIVER("Enabling DP audio on pipe %c\n",
 				 pipe_name(crtc->pipe));
 		intel_dp->DP |= DP_AUDIO_OUTPUT_ENABLE;
@@ -983,14 +1031,15 @@ static void intel_dp_mode_set(struct intel_encoder *encoder)
 		if (drm_dp_enhanced_frame_cap(intel_dp->dpcd))
 			intel_dp->DP |= DP_ENHANCED_FRAMING;
 
-		if (crtc->pipe == 1)
-			intel_dp->DP |= DP_PIPEB_SELECT;
+		if (!IS_CHERRYVIEW(dev)) {
+			if (crtc->pipe == 1)
+				intel_dp->DP |= DP_PIPEB_SELECT;
+		} else {
+			intel_dp->DP |= DP_PIPE_SELECT_CHV(crtc->pipe);
+		}
 	} else {
 		intel_dp->DP |= DP_LINK_TRAIN_OFF_CPT;
 	}
-
-	if (port == PORT_A && !IS_VALLEYVIEW(dev))
-		ironlake_set_pll_cpu_edp(intel_dp);
 }
 
 #define IDLE_ON_MASK		(PP_ON | PP_SEQUENCE_MASK | 0                     | PP_SEQUENCE_STATE_MASK)
@@ -1082,7 +1131,10 @@ static  u32 ironlake_get_pp_control(struct intel_dp *intel_dp)
 static bool _edp_panel_vdd_on(struct intel_dp *intel_dp)
 {
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
+	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
+	struct intel_encoder *intel_encoder = &intel_dig_port->base;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	enum intel_display_power_domain power_domain;
 	u32 pp;
 	u32 pp_stat_reg, pp_ctrl_reg;
 	bool need_to_disable = !intel_dp->want_panel_vdd;
@@ -1095,7 +1147,8 @@ static bool _edp_panel_vdd_on(struct intel_dp *intel_dp)
 	if (edp_have_panel_vdd(intel_dp))
 		return need_to_disable;
 
-	intel_runtime_pm_get(dev_priv);
+	power_domain = intel_display_port_power_domain(intel_encoder);
+	intel_display_power_get(dev_priv, power_domain);
 
 	DRM_DEBUG_KMS("Turning eDP VDD on\n");
 
@@ -1139,9 +1192,14 @@ static void edp_panel_vdd_off_sync(struct intel_dp *intel_dp)
 	u32 pp;
 	u32 pp_stat_reg, pp_ctrl_reg;
 
-	WARN_ON(!mutex_is_locked(&dev->mode_config.mutex));
+	WARN_ON(!drm_modeset_is_locked(&dev->mode_config.connection_mutex));
 
 	if (!intel_dp->want_panel_vdd && edp_have_panel_vdd(intel_dp)) {
+		struct intel_digital_port *intel_dig_port =
+						dp_to_dig_port(intel_dp);
+		struct intel_encoder *intel_encoder = &intel_dig_port->base;
+		enum intel_display_power_domain power_domain;
+
 		DRM_DEBUG_KMS("Turning eDP VDD off\n");
 
 		pp = ironlake_get_pp_control(intel_dp);
@@ -1160,7 +1218,8 @@ static void edp_panel_vdd_off_sync(struct intel_dp *intel_dp)
 		if ((pp & POWER_TARGET_ON) == 0)
 			intel_dp->last_power_cycle = jiffies;
 
-		intel_runtime_pm_put(dev_priv);
+		power_domain = intel_display_port_power_domain(intel_encoder);
+		intel_display_power_put(dev_priv, power_domain);
 	}
 }
 
@@ -1170,9 +1229,9 @@ static void edp_panel_vdd_work(struct work_struct *__work)
 						 struct intel_dp, panel_vdd_work);
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
 	edp_panel_vdd_off_sync(intel_dp);
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock(&dev->mode_config.connection_mutex);
 }
 
 static void edp_panel_vdd_off(struct intel_dp *intel_dp, bool sync)
@@ -1244,8 +1303,11 @@ void intel_edp_panel_on(struct intel_dp *intel_dp)
 
 void intel_edp_panel_off(struct intel_dp *intel_dp)
 {
+	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
+	struct intel_encoder *intel_encoder = &intel_dig_port->base;
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	enum intel_display_power_domain power_domain;
 	u32 pp;
 	u32 pp_ctrl_reg;
 
@@ -1275,7 +1337,8 @@ void intel_edp_panel_off(struct intel_dp *intel_dp)
 	wait_panel_off(intel_dp);
 
 	/* We got a reference when we enabled the VDD. */
-	intel_runtime_pm_put(dev_priv);
+	power_domain = intel_display_port_power_domain(intel_encoder);
+	intel_display_power_put(dev_priv, power_domain);
 }
 
 void intel_edp_backlight_on(struct intel_dp *intel_dp)
@@ -1432,6 +1495,8 @@ static bool intel_dp_get_hw_state(struct intel_encoder *encoder,
 
 	if (port == PORT_A && IS_GEN7(dev) && !IS_VALLEYVIEW(dev)) {
 		*pipe = PORT_TO_PIPE_CPT(tmp);
+	} else if (IS_CHERRYVIEW(dev)) {
+		*pipe = DP_PORT_TO_PIPE_CHV(tmp);
 	} else if (!HAS_PCH_CPT(dev) || port == PORT_A) {
 		*pipe = PORT_TO_PIPE(tmp);
 	} else {
@@ -1479,8 +1544,11 @@ static void intel_dp_get_config(struct intel_encoder *encoder,
 	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
 	int dotclock;
 
+	tmp = I915_READ(intel_dp->output_reg);
+	if (tmp & DP_AUDIO_OUTPUT_ENABLE)
+		pipe_config->has_audio = true;
+
 	if ((port == PORT_A) || !HAS_PCH_CPT(dev)) {
-		tmp = I915_READ(intel_dp->output_reg);
 		if (tmp & DP_SYNC_HS_HIGH)
 			flags |= DRM_MODE_FLAG_PHSYNC;
 		else
@@ -1816,17 +1884,59 @@ static void intel_disable_dp(struct intel_encoder *encoder)
 		intel_dp_link_down(intel_dp);
 }
 
-static void intel_post_disable_dp(struct intel_encoder *encoder)
+static void g4x_post_disable_dp(struct intel_encoder *encoder)
 {
 	struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
 	enum port port = dp_to_dig_port(intel_dp)->port;
+
+	if (port != PORT_A)
+		return;
+
+	intel_dp_link_down(intel_dp);
+	ironlake_edp_pll_off(intel_dp);
+}
+
+static void vlv_post_disable_dp(struct intel_encoder *encoder)
+{
+	struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
+
+	intel_dp_link_down(intel_dp);
+}
+
+static void chv_post_disable_dp(struct intel_encoder *encoder)
+{
+	struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
+	struct intel_digital_port *dport = dp_to_dig_port(intel_dp);
 	struct drm_device *dev = encoder->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc =
+		to_intel_crtc(encoder->base.crtc);
+	enum dpio_channel ch = vlv_dport_to_channel(dport);
+	enum pipe pipe = intel_crtc->pipe;
+	u32 val;
 
-	if (port == PORT_A || IS_VALLEYVIEW(dev)) {
-		intel_dp_link_down(intel_dp);
-		if (!IS_VALLEYVIEW(dev))
-			ironlake_edp_pll_off(intel_dp);
-	}
+	intel_dp_link_down(intel_dp);
+
+	mutex_lock(&dev_priv->dpio_lock);
+
+	/* Propagate soft reset to data lane reset */
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW1(ch));
+	val |= CHV_PCS_REQ_SOFTRESET_EN;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW1(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW1(ch));
+	val |= CHV_PCS_REQ_SOFTRESET_EN;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW1(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW0(ch));
+	val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW0(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW0(ch));
+	val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW0(ch), val);
+
+	mutex_unlock(&dev_priv->dpio_lock);
 }
 
 static void intel_enable_dp(struct intel_encoder *encoder)
@@ -1868,8 +1978,13 @@ static void g4x_pre_enable_dp(struct intel_encoder *encoder)
 	struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
 	struct intel_digital_port *dport = dp_to_dig_port(intel_dp);
 
-	if (dport->port == PORT_A)
+	intel_dp_prepare(encoder);
+
+	/* Only ilk+ has port A */
+	if (dport->port == PORT_A) {
+		ironlake_set_pll_cpu_edp(intel_dp);
 		ironlake_edp_pll_on(intel_dp);
+	}
 }
 
 static void vlv_pre_enable_dp(struct intel_encoder *encoder)
@@ -1921,6 +2036,8 @@ static void vlv_dp_pre_pll_enable(struct intel_encoder *encoder)
 	enum dpio_channel port = vlv_dport_to_channel(dport);
 	int pipe = intel_crtc->pipe;
 
+	intel_dp_prepare(encoder);
+
 	/* Program Tx lane resets to default */
 	mutex_lock(&dev_priv->dpio_lock);
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS_DW0(port),
@@ -1939,6 +2056,69 @@ static void vlv_dp_pre_pll_enable(struct intel_encoder *encoder)
 	mutex_unlock(&dev_priv->dpio_lock);
 }
 
+static void chv_pre_enable_dp(struct intel_encoder *encoder)
+{
+	struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
+	struct intel_digital_port *dport = dp_to_dig_port(intel_dp);
+	struct drm_device *dev = encoder->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct edp_power_seq power_seq;
+	struct intel_crtc *intel_crtc =
+		to_intel_crtc(encoder->base.crtc);
+	enum dpio_channel ch = vlv_dport_to_channel(dport);
+	int pipe = intel_crtc->pipe;
+	int data, i;
+	u32 val;
+
+	mutex_lock(&dev_priv->dpio_lock);
+
+	/* Deassert soft data lane reset*/
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW1(ch));
+	val |= CHV_PCS_REQ_SOFTRESET_EN;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW1(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW1(ch));
+	val |= CHV_PCS_REQ_SOFTRESET_EN;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW1(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW0(ch));
+	val |= (DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW0(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW0(ch));
+	val |= (DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW0(ch), val);
+
+	/* Program Tx lane latency optimal setting*/
+	for (i = 0; i < 4; i++) {
+		/* Set the latency optimal bit */
+		data = (i == 1) ? 0x0 : 0x6;
+		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW11(ch, i),
+				data << DPIO_FRC_LATENCY_SHFIT);
+
+		/* Set the upar bit */
+		data = (i == 1) ? 0x0 : 0x1;
+		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW14(ch, i),
+				data << DPIO_UPAR_SHIFT);
+	}
+
+	/* Data lane stagger programming */
+	/* FIXME: Fix up value only after power analysis */
+
+	mutex_unlock(&dev_priv->dpio_lock);
+
+	if (is_edp(intel_dp)) {
+		/* init power sequencer on this pipe and port */
+		intel_dp_init_panel_power_sequencer(dev, intel_dp, &power_seq);
+		intel_dp_init_panel_power_sequencer_registers(dev, intel_dp,
+							      &power_seq);
+	}
+
+	intel_enable_dp(encoder);
+
+	vlv_wait_port_ready(dev_priv, dport);
+}
+
 /*
  * Native read with retry for link status and receiver capability reads for
  * cases where the sink may still be asleep.
@@ -2163,6 +2343,166 @@ static uint32_t intel_vlv_signal_levels(struct intel_dp *intel_dp)
 	return 0;
 }
 
+static uint32_t intel_chv_signal_levels(struct intel_dp *intel_dp)
+{
+	struct drm_device *dev = intel_dp_to_dev(intel_dp);
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_digital_port *dport = dp_to_dig_port(intel_dp);
+	struct intel_crtc *intel_crtc = to_intel_crtc(dport->base.base.crtc);
+	u32 deemph_reg_value, margin_reg_value, val;
+	uint8_t train_set = intel_dp->train_set[0];
+	enum dpio_channel ch = vlv_dport_to_channel(dport);
+	enum pipe pipe = intel_crtc->pipe;
+	int i;
+
+	switch (train_set & DP_TRAIN_PRE_EMPHASIS_MASK) {
+	case DP_TRAIN_PRE_EMPHASIS_0:
+		switch (train_set & DP_TRAIN_VOLTAGE_SWING_MASK) {
+		case DP_TRAIN_VOLTAGE_SWING_400:
+			deemph_reg_value = 128;
+			margin_reg_value = 52;
+			break;
+		case DP_TRAIN_VOLTAGE_SWING_600:
+			deemph_reg_value = 128;
+			margin_reg_value = 77;
+			break;
+		case DP_TRAIN_VOLTAGE_SWING_800:
+			deemph_reg_value = 128;
+			margin_reg_value = 102;
+			break;
+		case DP_TRAIN_VOLTAGE_SWING_1200:
+			deemph_reg_value = 128;
+			margin_reg_value = 154;
+			/* FIXME extra to set for 1200 */
+			break;
+		default:
+			return 0;
+		}
+		break;
+	case DP_TRAIN_PRE_EMPHASIS_3_5:
+		switch (train_set & DP_TRAIN_VOLTAGE_SWING_MASK) {
+		case DP_TRAIN_VOLTAGE_SWING_400:
+			deemph_reg_value = 85;
+			margin_reg_value = 78;
+			break;
+		case DP_TRAIN_VOLTAGE_SWING_600:
+			deemph_reg_value = 85;
+			margin_reg_value = 116;
+			break;
+		case DP_TRAIN_VOLTAGE_SWING_800:
+			deemph_reg_value = 85;
+			margin_reg_value = 154;
+			break;
+		default:
+			return 0;
+		}
+		break;
+	case DP_TRAIN_PRE_EMPHASIS_6:
+		switch (train_set & DP_TRAIN_VOLTAGE_SWING_MASK) {
+		case DP_TRAIN_VOLTAGE_SWING_400:
+			deemph_reg_value = 64;
+			margin_reg_value = 104;
+			break;
+		case DP_TRAIN_VOLTAGE_SWING_600:
+			deemph_reg_value = 64;
+			margin_reg_value = 154;
+			break;
+		default:
+			return 0;
+		}
+		break;
+	case DP_TRAIN_PRE_EMPHASIS_9_5:
+		switch (train_set & DP_TRAIN_VOLTAGE_SWING_MASK) {
+		case DP_TRAIN_VOLTAGE_SWING_400:
+			deemph_reg_value = 43;
+			margin_reg_value = 154;
+			break;
+		default:
+			return 0;
+		}
+		break;
+	default:
+		return 0;
+	}
+
+	mutex_lock(&dev_priv->dpio_lock);
+
+	/* Clear calc init */
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW10(ch));
+	val &= ~(DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW10(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW10(ch));
+	val &= ~(DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW10(ch), val);
+
+	/* Program swing deemph */
+	for (i = 0; i < 4; i++) {
+		val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW4(ch, i));
+		val &= ~DPIO_SWING_DEEMPH9P5_MASK;
+		val |= deemph_reg_value << DPIO_SWING_DEEMPH9P5_SHIFT;
+		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW4(ch, i), val);
+	}
+
+	/* Program swing margin */
+	for (i = 0; i < 4; i++) {
+		val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW2(ch, i));
+		val &= ~DPIO_SWING_MARGIN_MASK;
+		val |= margin_reg_value << DPIO_SWING_MARGIN_SHIFT;
+		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW2(ch, i), val);
+	}
+
+	/* Disable unique transition scale */
+	for (i = 0; i < 4; i++) {
+		val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW3(ch, i));
+		val &= ~DPIO_TX_UNIQ_TRANS_SCALE_EN;
+		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW3(ch, i), val);
+	}
+
+	if (((train_set & DP_TRAIN_PRE_EMPHASIS_MASK)
+			== DP_TRAIN_PRE_EMPHASIS_0) &&
+		((train_set & DP_TRAIN_VOLTAGE_SWING_MASK)
+			== DP_TRAIN_VOLTAGE_SWING_1200)) {
+
+		/*
+		 * The document said it needs to set bit 27 for ch0 and bit 26
+		 * for ch1. Might be a typo in the doc.
+		 * For now, for this unique transition scale selection, set bit
+		 * 27 for ch0 and ch1.
+		 */
+		for (i = 0; i < 4; i++) {
+			val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW3(ch, i));
+			val |= DPIO_TX_UNIQ_TRANS_SCALE_EN;
+			vlv_dpio_write(dev_priv, pipe, CHV_TX_DW3(ch, i), val);
+		}
+
+		for (i = 0; i < 4; i++) {
+			val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW2(ch, i));
+			val &= ~(0xff << DPIO_UNIQ_TRANS_SCALE_SHIFT);
+			val |= (0x9a << DPIO_UNIQ_TRANS_SCALE_SHIFT);
+			vlv_dpio_write(dev_priv, pipe, CHV_TX_DW2(ch, i), val);
+		}
+	}
+
+	/* Start swing calculation */
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW10(ch));
+	val |= DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW10(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW10(ch));
+	val |= DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW10(ch), val);
+
+	/* LRC Bypass */
+	val = vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW30);
+	val |= DPIO_LRC_BYPASS;
+	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW30, val);
+
+	mutex_unlock(&dev_priv->dpio_lock);
+
+	return 0;
+}
+
 static void
 intel_get_adjust_train(struct intel_dp *intel_dp,
 		       const uint8_t link_status[DP_LINK_STATUS_SIZE])
@@ -2377,6 +2717,9 @@ intel_dp_set_signal_levels(struct intel_dp *intel_dp, uint32_t *DP)
 	} else if (IS_HASWELL(dev)) {
 		signal_levels = intel_hsw_signal_levels(train_set);
 		mask = DDI_BUF_EMP_MASK;
+	} else if (IS_CHERRYVIEW(dev)) {
+		signal_levels = intel_chv_signal_levels(intel_dp);
+		mask = 0;
 	} else if (IS_VALLEYVIEW(dev)) {
 		signal_levels = intel_vlv_signal_levels(intel_dp);
 		mask = 0;
@@ -2743,22 +3086,7 @@ intel_dp_link_down(struct intel_dp *intel_dp)
 		to_intel_crtc(intel_dig_port->base.base.crtc);
 	uint32_t DP = intel_dp->DP;
 
-	/*
-	 * DDI code has a strict mode set sequence and we should try to respect
-	 * it, otherwise we might hang the machine in many different ways. So we
-	 * really should be disabling the port only on a complete crtc_disable
-	 * sequence. This function is just called under two conditions on DDI
-	 * code:
-	 * - Link train failed while doing crtc_enable, and on this case we
-	 *   really should respect the mode set sequence and wait for a
-	 *   crtc_disable.
-	 * - Someone turned the monitor off and intel_dp_check_link_status
-	 *   called us. We don't need to disable the whole port on this case, so
-	 *   when someone turns the monitor on again,
-	 *   intel_ddi_prepare_link_retrain will take care of redoing the link
-	 *   train.
-	 */
-	if (HAS_DDI(dev))
+	if (WARN_ON(HAS_DDI(dev)))
 		return;
 
 	if (WARN_ON((I915_READ(intel_dp->output_reg) & DP_PORT_EN) == 0))
@@ -2775,9 +3103,6 @@ intel_dp_link_down(struct intel_dp *intel_dp)
 	}
 	POSTING_READ(intel_dp->output_reg);
 
-	/* We don't really know why we're doing this */
-	intel_wait_for_vblank(dev, intel_crtc->pipe);
-
 	if (HAS_PCH_IBX(dev) &&
 	    I915_READ(intel_dp->output_reg) & DP_PIPEB_SELECT) {
 		struct drm_crtc *crtc = intel_dig_port->base.base.crtc;
@@ -2948,6 +3273,7 @@ intel_dp_check_link_status(struct intel_dp *intel_dp)
 	u8 sink_irq_vector;
 	u8 link_status[DP_LINK_STATUS_SIZE];
 
+	/* FIXME: This access isn't protected by any locks. */
 	if (!intel_encoder->connectors_active)
 		return;
 
@@ -2980,7 +3306,7 @@ intel_dp_check_link_status(struct intel_dp *intel_dp)
 
 	if (!drm_dp_channel_eq_ok(link_status, intel_dp->lane_count)) {
 		DRM_DEBUG_KMS("%s: channel EQ not ok, retraining\n",
-			      drm_get_encoder_name(&intel_encoder->base));
+			      intel_encoder->base.name);
 		intel_dp_start_link_train(intel_dp);
 		intel_dp_complete_link_train(intel_dp);
 		intel_dp_stop_link_train(intel_dp);
@@ -3166,7 +3492,7 @@ intel_dp_detect(struct drm_connector *connector, bool force)
 	intel_display_power_get(dev_priv, power_domain);
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
-		      connector->base.id, drm_get_connector_name(connector));
+		      connector->base.id, connector->name);
 
 	intel_dp->has_audio = false;
 
@@ -3374,13 +3700,13 @@ void intel_dp_encoder_destroy(struct drm_encoder *encoder)
 	struct intel_dp *intel_dp = &intel_dig_port->dp;
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 
-	drm_dp_aux_unregister_i2c_bus(&intel_dp->aux);
+	drm_dp_aux_unregister(&intel_dp->aux);
 	drm_encoder_cleanup(encoder);
 	if (is_edp(intel_dp)) {
 		cancel_delayed_work_sync(&intel_dp->panel_vdd_work);
-		mutex_lock(&dev->mode_config.mutex);
+		drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
 		edp_panel_vdd_off_sync(intel_dp);
-		mutex_unlock(&dev->mode_config.mutex);
+		drm_modeset_unlock(&dev->mode_config.connection_mutex);
 	}
 	kfree(intel_dig_port);
 }
@@ -3651,6 +3977,130 @@ intel_dp_init_panel_power_sequencer_registers(struct drm_device *dev,
 		      I915_READ(pp_div_reg));
 }
 
+void intel_dp_set_drrs_state(struct drm_device *dev, int refresh_rate)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_encoder *encoder;
+	struct intel_dp *intel_dp = NULL;
+	struct intel_crtc_config *config = NULL;
+	struct intel_crtc *intel_crtc = NULL;
+	struct intel_connector *intel_connector = dev_priv->drrs.connector;
+	u32 reg, val;
+	enum edp_drrs_refresh_rate_type index = DRRS_HIGH_RR;
+
+	if (refresh_rate <= 0) {
+		DRM_DEBUG_KMS("Refresh rate should be positive non-zero.\n");
+		return;
+	}
+
+	if (intel_connector == NULL) {
+		DRM_DEBUG_KMS("DRRS supported for eDP only.\n");
+		return;
+	}
+
+	if (INTEL_INFO(dev)->gen < 8 && intel_edp_is_psr_enabled(dev)) {
+		DRM_DEBUG_KMS("DRRS is disabled as PSR is enabled\n");
+		return;
+	}
+
+	encoder = intel_attached_encoder(&intel_connector->base);
+	intel_dp = enc_to_intel_dp(&encoder->base);
+	intel_crtc = encoder->new_crtc;
+
+	if (!intel_crtc) {
+		DRM_DEBUG_KMS("DRRS: intel_crtc not initialized\n");
+		return;
+	}
+
+	config = &intel_crtc->config;
+
+	if (intel_dp->drrs_state.type < SEAMLESS_DRRS_SUPPORT) {
+		DRM_DEBUG_KMS("Only Seamless DRRS supported.\n");
+		return;
+	}
+
+	if (intel_connector->panel.downclock_mode->vrefresh == refresh_rate)
+		index = DRRS_LOW_RR;
+
+	if (index == intel_dp->drrs_state.refresh_rate_type) {
+		DRM_DEBUG_KMS(
+			"DRRS requested for previously set RR...ignoring\n");
+		return;
+	}
+
+	if (!intel_crtc->active) {
+		DRM_DEBUG_KMS("eDP encoder disabled. CRTC not Active\n");
+		return;
+	}
+
+	if (INTEL_INFO(dev)->gen > 6 && INTEL_INFO(dev)->gen < 8) {
+		reg = PIPECONF(intel_crtc->config.cpu_transcoder);
+		val = I915_READ(reg);
+		if (index > DRRS_HIGH_RR) {
+			val |= PIPECONF_EDP_RR_MODE_SWITCH;
+			intel_dp_set_m2_n2(intel_crtc, &config->dp_m2_n2);
+		} else {
+			val &= ~PIPECONF_EDP_RR_MODE_SWITCH;
+		}
+		I915_WRITE(reg, val);
+	}
+
+	/*
+	 * mutex taken to ensure that there is no race between differnt
+	 * drrs calls trying to update refresh rate. This scenario may occur
+	 * in future when idleness detection based DRRS in kernel and
+	 * possible calls from user space to set differnt RR are made.
+	 */
+
+	mutex_lock(&intel_dp->drrs_state.mutex);
+
+	intel_dp->drrs_state.refresh_rate_type = index;
+
+	mutex_unlock(&intel_dp->drrs_state.mutex);
+
+	DRM_DEBUG_KMS("eDP Refresh Rate set to : %dHz\n", refresh_rate);
+}
+
+static struct drm_display_mode *
+intel_dp_drrs_init(struct intel_digital_port *intel_dig_port,
+			struct intel_connector *intel_connector,
+			struct drm_display_mode *fixed_mode)
+{
+	struct drm_connector *connector = &intel_connector->base;
+	struct intel_dp *intel_dp = &intel_dig_port->dp;
+	struct drm_device *dev = intel_dig_port->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_display_mode *downclock_mode = NULL;
+
+	if (INTEL_INFO(dev)->gen <= 6) {
+		DRM_DEBUG_KMS("DRRS supported for Gen7 and above\n");
+		return NULL;
+	}
+
+	if (dev_priv->vbt.drrs_type != SEAMLESS_DRRS_SUPPORT) {
+		DRM_INFO("VBT doesn't support DRRS\n");
+		return NULL;
+	}
+
+	downclock_mode = intel_find_panel_downclock
+					(dev, fixed_mode, connector);
+
+	if (!downclock_mode) {
+		DRM_INFO("DRRS not supported\n");
+		return NULL;
+	}
+
+	dev_priv->drrs.connector = intel_connector;
+
+	mutex_init(&intel_dp->drrs_state.mutex);
+
+	intel_dp->drrs_state.type = dev_priv->vbt.drrs_type;
+
+	intel_dp->drrs_state.refresh_rate_type = DRRS_HIGH_RR;
+	DRM_INFO("seamless DRRS supported for eDP panel.\n");
+	return downclock_mode;
+}
+
 static bool intel_edp_init_connector(struct intel_dp *intel_dp,
 				     struct intel_connector *intel_connector,
 				     struct edp_power_seq *power_seq)
@@ -3661,10 +4111,13 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp,
 	struct drm_device *dev = intel_encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_display_mode *fixed_mode = NULL;
+	struct drm_display_mode *downclock_mode = NULL;
 	bool has_dpcd;
 	struct drm_display_mode *scan;
 	struct edid *edid;
 
+	intel_dp->drrs_state.type = DRRS_NOT_SUPPORTED;
+
 	if (!is_edp(intel_dp))
 		return true;
 
@@ -3715,6 +4168,9 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp,
 	list_for_each_entry(scan, &connector->probed_modes, head) {
 		if ((scan->type & DRM_MODE_TYPE_PREFERRED)) {
 			fixed_mode = drm_mode_duplicate(dev, scan);
+			downclock_mode = intel_dp_drrs_init(
+						intel_dig_port,
+						intel_connector, fixed_mode);
 			break;
 		}
 	}
@@ -3728,7 +4184,7 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp,
 	}
 	mutex_unlock(&dev->mode_config.mutex);
 
-	intel_panel_init(&intel_connector->panel, fixed_mode, NULL);
+	intel_panel_init(&intel_connector->panel, fixed_mode, downclock_mode);
 	intel_panel_setup_backlight(connector);
 
 	return true;
@@ -3826,12 +4282,12 @@ intel_dp_init_connector(struct intel_digital_port *intel_dig_port,
 	intel_dp->psr_setup_done = false;
 
 	if (!intel_edp_init_connector(intel_dp, intel_connector, &power_seq)) {
-		drm_dp_aux_unregister_i2c_bus(&intel_dp->aux);
+		drm_dp_aux_unregister(&intel_dp->aux);
 		if (is_edp(intel_dp)) {
 			cancel_delayed_work_sync(&intel_dp->panel_vdd_work);
-			mutex_lock(&dev->mode_config.mutex);
+			drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
 			edp_panel_vdd_off_sync(intel_dp);
-			mutex_unlock(&dev->mode_config.mutex);
+			drm_modeset_unlock(&dev->mode_config.connection_mutex);
 		}
 		drm_sysfs_connector_remove(connector);
 		drm_connector_cleanup(connector);
@@ -3877,25 +4333,36 @@ intel_dp_init(struct drm_device *dev, int output_reg, enum port port)
 			 DRM_MODE_ENCODER_TMDS);
 
 	intel_encoder->compute_config = intel_dp_compute_config;
-	intel_encoder->mode_set = intel_dp_mode_set;
 	intel_encoder->disable = intel_disable_dp;
-	intel_encoder->post_disable = intel_post_disable_dp;
 	intel_encoder->get_hw_state = intel_dp_get_hw_state;
 	intel_encoder->get_config = intel_dp_get_config;
-	if (IS_VALLEYVIEW(dev)) {
+	if (IS_CHERRYVIEW(dev)) {
+		intel_encoder->pre_enable = chv_pre_enable_dp;
+		intel_encoder->enable = vlv_enable_dp;
+		intel_encoder->post_disable = chv_post_disable_dp;
+	} else if (IS_VALLEYVIEW(dev)) {
 		intel_encoder->pre_pll_enable = vlv_dp_pre_pll_enable;
 		intel_encoder->pre_enable = vlv_pre_enable_dp;
 		intel_encoder->enable = vlv_enable_dp;
+		intel_encoder->post_disable = vlv_post_disable_dp;
 	} else {
 		intel_encoder->pre_enable = g4x_pre_enable_dp;
 		intel_encoder->enable = g4x_enable_dp;
+		intel_encoder->post_disable = g4x_post_disable_dp;
 	}
 
 	intel_dig_port->port = port;
 	intel_dig_port->dp.output_reg = output_reg;
 
 	intel_encoder->type = INTEL_OUTPUT_DISPLAYPORT;
-	intel_encoder->crtc_mask = (1 << 0) | (1 << 1) | (1 << 2);
+	if (IS_CHERRYVIEW(dev)) {
+		if (port == PORT_D)
+			intel_encoder->crtc_mask = 1 << 2;
+		else
+			intel_encoder->crtc_mask = (1 << 0) | (1 << 1);
+	} else {
+		intel_encoder->crtc_mask = (1 << 0) | (1 << 1) | (1 << 2);
+	}
 	intel_encoder->cloneable = 0;
 	intel_encoder->hot_plug = intel_dp_hot_plug;
 
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 328b1a70264b..bda0ae3d80cc 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -81,8 +81,8 @@
 /* Maximum cursor sizes */
 #define GEN2_CURSOR_WIDTH 64
 #define GEN2_CURSOR_HEIGHT 64
-#define CURSOR_WIDTH 256
-#define CURSOR_HEIGHT 256
+#define MAX_CURSOR_WIDTH 256
+#define MAX_CURSOR_HEIGHT 256
 
 #define INTEL_I2C_BUS_DVO 1
 #define INTEL_I2C_BUS_SDVO 2
@@ -106,8 +106,8 @@
 #define INTEL_DVO_CHIP_TMDS 2
 #define INTEL_DVO_CHIP_TVOUT 4
 
-#define INTEL_DSI_COMMAND_MODE	0
-#define INTEL_DSI_VIDEO_MODE	1
+#define INTEL_DSI_VIDEO_MODE	0
+#define INTEL_DSI_COMMAND_MODE	1
 
 struct intel_framebuffer {
 	struct drm_framebuffer base;
@@ -273,6 +273,13 @@ struct intel_crtc_config {
 	 * accordingly. */
 	bool has_dp_encoder;
 
+	/* Whether we should send NULL infoframes. Required for audio. */
+	bool has_hdmi_sink;
+
+	/* Audio enabled on this pipe. Only valid if either has_hdmi_sink or
+	 * has_dp_encoder is set. */
+	bool has_audio;
+
 	/*
 	 * Enable dithering, used when the selected pipe bpp doesn't match the
 	 * plane bpp.
@@ -306,6 +313,9 @@ struct intel_crtc_config {
 	int pipe_bpp;
 	struct intel_link_m_n dp_m_n;
 
+	/* m2_n2 for eDP downclock */
+	struct intel_link_m_n dp_m2_n2;
+
 	/*
 	 * Frequence the dpll for the port should run at. Differs from the
 	 * adjusted dotclock e.g. for DP or 12bpc hdmi mode. This is also
@@ -343,6 +353,9 @@ struct intel_pipe_wm {
 	struct intel_wm_level wm[5];
 	uint32_t linetime;
 	bool fbc_wm_enabled;
+	bool pipe_enabled;
+	bool sprites_enabled;
+	bool sprites_scaled;
 };
 
 struct intel_crtc {
@@ -357,7 +370,6 @@ struct intel_crtc {
 	 */
 	bool active;
 	unsigned long enabled_power_domains;
-	bool eld_vld;
 	bool primary_enabled; /* is the primary plane (partially) visible? */
 	bool lowfreq_avail;
 	struct intel_overlay *overlay;
@@ -374,8 +386,8 @@ struct intel_crtc {
 	uint32_t cursor_addr;
 	int16_t cursor_x, cursor_y;
 	int16_t cursor_width, cursor_height;
-	int16_t max_cursor_width, max_cursor_height;
-	bool cursor_visible;
+	uint32_t cursor_cntl;
+	uint32_t cursor_base;
 
 	struct intel_plane_config plane_config;
 	struct intel_crtc_config config;
@@ -396,6 +408,10 @@ struct intel_crtc {
 		/* watermarks currently being used  */
 		struct intel_pipe_wm active;
 	} wm;
+
+	wait_queue_head_t vbl_wait;
+
+	int scanline_offset;
 };
 
 struct intel_plane_wm_parameters {
@@ -479,11 +495,23 @@ struct intel_hdmi {
 				enum hdmi_infoframe_type type,
 				const void *frame, ssize_t len);
 	void (*set_infoframes)(struct drm_encoder *encoder,
+			       bool enable,
 			       struct drm_display_mode *adjusted_mode);
 };
 
 #define DP_MAX_DOWNSTREAM_PORTS		0x10
 
+/**
+ * HIGH_RR is the highest eDP panel refresh rate read from EDID
+ * LOW_RR is the lowest eDP panel refresh rate found from EDID
+ * parsing for same resolution.
+ */
+enum edp_drrs_refresh_rate_type {
+	DRRS_HIGH_RR,
+	DRRS_LOW_RR,
+	DRRS_MAX_RR, /* RR count */
+};
+
 struct intel_dp {
 	uint32_t output_reg;
 	uint32_t aux_ch_ctl_reg;
@@ -522,6 +550,12 @@ struct intel_dp {
 				     bool has_aux_irq,
 				     int send_bytes,
 				     uint32_t aux_clock_divider);
+	struct {
+		enum drrs_support_type type;
+		enum edp_drrs_refresh_rate_type refresh_rate_type;
+		struct mutex mutex;
+	} drrs_state;
+
 };
 
 struct intel_digital_port {
@@ -537,6 +571,7 @@ vlv_dport_to_channel(struct intel_digital_port *dport)
 {
 	switch (dport->port) {
 	case PORT_B:
+	case PORT_D:
 		return DPIO_CH0;
 	case PORT_C:
 		return DPIO_CH1;
@@ -545,6 +580,20 @@ vlv_dport_to_channel(struct intel_digital_port *dport)
 	}
 }
 
+static inline int
+vlv_pipe_to_channel(enum pipe pipe)
+{
+	switch (pipe) {
+	case PIPE_A:
+	case PIPE_C:
+		return DPIO_CH0;
+	case PIPE_B:
+		return DPIO_CH1;
+	default:
+		BUG();
+	}
+}
+
 static inline struct drm_crtc *
 intel_get_crtc_for_pipe(struct drm_device *dev, int pipe)
 {
@@ -569,6 +618,8 @@ struct intel_unpin_work {
 #define INTEL_FLIP_INACTIVE	0
 #define INTEL_FLIP_PENDING	1
 #define INTEL_FLIP_COMPLETE	2
+	u32 flip_count;
+	u32 gtt_offset;
 	bool enable_stall_check;
 };
 
@@ -620,8 +671,6 @@ hdmi_to_dig_port(struct intel_hdmi *intel_hdmi)
 /* i915_irq.c */
 bool intel_set_cpu_fifo_underrun_reporting(struct drm_device *dev,
 					   enum pipe pipe, bool enable);
-bool __intel_set_cpu_fifo_underrun_reporting(struct drm_device *dev,
-					     enum pipe pipe, bool enable);
 bool intel_set_pch_fifo_underrun_reporting(struct drm_device *dev,
 					   enum transcoder pch_transcoder,
 					   bool enable);
@@ -629,8 +678,12 @@ void ilk_enable_gt_irq(struct drm_i915_private *dev_priv, uint32_t mask);
 void ilk_disable_gt_irq(struct drm_i915_private *dev_priv, uint32_t mask);
 void snb_enable_pm_irq(struct drm_i915_private *dev_priv, uint32_t mask);
 void snb_disable_pm_irq(struct drm_i915_private *dev_priv, uint32_t mask);
-void hsw_runtime_pm_disable_interrupts(struct drm_device *dev);
-void hsw_runtime_pm_restore_interrupts(struct drm_device *dev);
+void bdw_enable_pm_irq(struct drm_i915_private *dev_priv, uint32_t mask);
+void bdw_disable_pm_irq(struct drm_i915_private *dev_priv, uint32_t mask);
+void intel_runtime_pm_disable_interrupts(struct drm_device *dev);
+void intel_runtime_pm_restore_interrupts(struct drm_device *dev);
+int intel_get_crtc_scanline(struct intel_crtc *crtc);
+void i9xx_check_fifo_underruns(struct drm_device *dev);
 
 
 /* intel_crt.c */
@@ -666,9 +719,10 @@ void intel_ddi_get_config(struct intel_encoder *encoder,
 const char *intel_output_name(int output);
 bool intel_has_pending_fb_unpin(struct drm_device *dev);
 int intel_pch_rawclk(struct drm_device *dev);
+int valleyview_cur_cdclk(struct drm_i915_private *dev_priv);
 void intel_mark_busy(struct drm_device *dev);
 void intel_mark_fb_busy(struct drm_i915_gem_object *obj,
-			struct intel_ring_buffer *ring);
+			struct intel_engine_cs *ring);
 void intel_mark_idle(struct drm_device *dev);
 void intel_crtc_restore_mode(struct drm_crtc *crtc);
 void intel_crtc_update_dpms(struct drm_crtc *crtc);
@@ -695,12 +749,14 @@ void vlv_wait_port_ready(struct drm_i915_private *dev_priv,
 			 struct intel_digital_port *dport);
 bool intel_get_load_detect_pipe(struct drm_connector *connector,
 				struct drm_display_mode *mode,
-				struct intel_load_detect_pipe *old);
+				struct intel_load_detect_pipe *old,
+				struct drm_modeset_acquire_ctx *ctx);
 void intel_release_load_detect_pipe(struct drm_connector *connector,
-				    struct intel_load_detect_pipe *old);
+				    struct intel_load_detect_pipe *old,
+				    struct drm_modeset_acquire_ctx *ctx);
 int intel_pin_and_fence_fb_obj(struct drm_device *dev,
 			       struct drm_i915_gem_object *obj,
-			       struct intel_ring_buffer *pipelined);
+			       struct intel_engine_cs *pipelined);
 void intel_unpin_fb_obj(struct drm_i915_gem_object *obj);
 struct drm_framebuffer *
 __intel_framebuffer_create(struct drm_device *dev,
@@ -751,6 +807,8 @@ int valleyview_get_vco(struct drm_i915_private *dev_priv);
 void intel_mode_from_pipe_config(struct drm_display_mode *mode,
 				 struct intel_crtc_config *pipe_config);
 int intel_format_to_fourcc(int format);
+void intel_crtc_wait_for_pending_flips(struct drm_crtc *crtc);
+
 
 /* intel_dp.c */
 void intel_dp_init(struct drm_device *dev, int output_reg, enum port port);
@@ -774,7 +832,7 @@ void intel_edp_panel_off(struct intel_dp *intel_dp);
 void intel_edp_psr_enable(struct intel_dp *intel_dp);
 void intel_edp_psr_disable(struct intel_dp *intel_dp);
 void intel_edp_psr_update(struct drm_device *dev);
-
+void intel_dp_set_drrs_state(struct drm_device *dev, int refresh_rate);
 
 /* intel_dsi.c */
 bool intel_dsi_init(struct drm_device *dev);
@@ -876,6 +934,7 @@ extern struct drm_display_mode *intel_find_panel_downclock(
 /* intel_pm.c */
 void intel_init_clock_gating(struct drm_device *dev);
 void intel_suspend_hw(struct drm_device *dev);
+int ilk_wm_max_level(const struct drm_device *dev);
 void intel_update_watermarks(struct drm_crtc *crtc);
 void intel_update_sprite_watermarks(struct drm_plane *plane,
 				    struct drm_crtc *crtc,
@@ -902,6 +961,7 @@ void intel_init_gt_powersave(struct drm_device *dev);
 void intel_cleanup_gt_powersave(struct drm_device *dev);
 void intel_enable_gt_powersave(struct drm_device *dev);
 void intel_disable_gt_powersave(struct drm_device *dev);
+void intel_reset_gt_powersave(struct drm_device *dev);
 void ironlake_teardown_rc6(struct drm_device *dev);
 void gen6_update_ring_freq(struct drm_device *dev);
 void gen6_rps_idle(struct drm_i915_private *dev_priv);
@@ -909,11 +969,13 @@ void gen6_rps_boost(struct drm_i915_private *dev_priv);
 void intel_aux_display_runtime_get(struct drm_i915_private *dev_priv);
 void intel_aux_display_runtime_put(struct drm_i915_private *dev_priv);
 void intel_runtime_pm_get(struct drm_i915_private *dev_priv);
+void intel_runtime_pm_get_noresume(struct drm_i915_private *dev_priv);
 void intel_runtime_pm_put(struct drm_i915_private *dev_priv);
 void intel_init_runtime_pm(struct drm_i915_private *dev_priv);
 void intel_fini_runtime_pm(struct drm_i915_private *dev_priv);
 void ilk_wm_get_hw_state(struct drm_device *dev);
-
+void __vlv_set_power_well(struct drm_i915_private *dev_priv,
+			  enum punit_power_well power_well_id, bool enable);
 
 /* intel_sdvo.c */
 bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg, bool is_sdvob);
diff --git a/drivers/gpu/drm/i915/intel_dsi.c b/drivers/gpu/drm/i915/intel_dsi.c
index 33656647f8bc..02f99d768d49 100644
--- a/drivers/gpu/drm/i915/intel_dsi.c
+++ b/drivers/gpu/drm/i915/intel_dsi.c
@@ -35,6 +35,11 @@
 
 /* the sub-encoders aka panel drivers */
 static const struct intel_dsi_device intel_dsi_devices[] = {
+	{
+		.panel_id = MIPI_DSI_GENERIC_PANEL_ID,
+		.name = "vbt-generic-dsi-vid-mode-display",
+		.dev_ops = &vbt_generic_dsi_display_ops,
+	},
 };
 
 static void band_gap_reset(struct drm_i915_private *dev_priv)
@@ -59,12 +64,12 @@ static struct intel_dsi *intel_attached_dsi(struct drm_connector *connector)
 
 static inline bool is_vid_mode(struct intel_dsi *intel_dsi)
 {
-	return intel_dsi->dev.type == INTEL_DSI_VIDEO_MODE;
+	return intel_dsi->operation_mode == INTEL_DSI_VIDEO_MODE;
 }
 
 static inline bool is_cmd_mode(struct intel_dsi *intel_dsi)
 {
-	return intel_dsi->dev.type == INTEL_DSI_COMMAND_MODE;
+	return intel_dsi->operation_mode == INTEL_DSI_COMMAND_MODE;
 }
 
 static void intel_dsi_hot_plug(struct intel_encoder *encoder)
@@ -94,13 +99,6 @@ static bool intel_dsi_compute_config(struct intel_encoder *encoder,
 	return true;
 }
 
-static void intel_dsi_pre_pll_enable(struct intel_encoder *encoder)
-{
-	DRM_DEBUG_KMS("\n");
-
-	vlv_enable_dsi_pll(encoder);
-}
-
 static void intel_dsi_device_ready(struct intel_encoder *encoder)
 {
 	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
@@ -110,6 +108,15 @@ static void intel_dsi_device_ready(struct intel_encoder *encoder)
 
 	DRM_DEBUG_KMS("\n");
 
+	mutex_lock(&dev_priv->dpio_lock);
+	/* program rcomp for compliance, reduce from 50 ohms to 45 ohms
+	 * needed everytime after power gate */
+	vlv_flisdsi_write(dev_priv, 0x04, 0x0004);
+	mutex_unlock(&dev_priv->dpio_lock);
+
+	/* bandgap reset is needed after everytime we do power gate */
+	band_gap_reset(dev_priv);
+
 	val = I915_READ(MIPI_PORT_CTRL(pipe));
 	I915_WRITE(MIPI_PORT_CTRL(pipe), val | LP_OUTPUT_HOLD);
 	usleep_range(1000, 1500);
@@ -122,21 +129,6 @@ static void intel_dsi_device_ready(struct intel_encoder *encoder)
 	I915_WRITE(MIPI_DEVICE_READY(pipe), DEVICE_READY);
 	usleep_range(2000, 2500);
 }
-static void intel_dsi_pre_enable(struct intel_encoder *encoder)
-{
-	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
-
-	DRM_DEBUG_KMS("\n");
-
-	if (intel_dsi->dev.dev_ops->panel_reset)
-		intel_dsi->dev.dev_ops->panel_reset(&intel_dsi->dev);
-
-	/* put device in ready state */
-	intel_dsi_device_ready(encoder);
-
-	if (intel_dsi->dev.dev_ops->send_otp_cmds)
-		intel_dsi->dev.dev_ops->send_otp_cmds(&intel_dsi->dev);
-}
 
 static void intel_dsi_enable(struct intel_encoder *encoder)
 {
@@ -153,18 +145,78 @@ static void intel_dsi_enable(struct intel_encoder *encoder)
 		I915_WRITE(MIPI_MAX_RETURN_PKT_SIZE(pipe), 8 * 4);
 	else {
 		msleep(20); /* XXX */
-		dpi_send_cmd(intel_dsi, TURN_ON);
+		dpi_send_cmd(intel_dsi, TURN_ON, DPI_LP_MODE_EN);
 		msleep(100);
 
+		if (intel_dsi->dev.dev_ops->enable)
+			intel_dsi->dev.dev_ops->enable(&intel_dsi->dev);
+
 		/* assert ip_tg_enable signal */
 		temp = I915_READ(MIPI_PORT_CTRL(pipe)) & ~LANE_CONFIGURATION_MASK;
 		temp = temp | intel_dsi->port_bits;
 		I915_WRITE(MIPI_PORT_CTRL(pipe), temp | DPI_ENABLE);
 		POSTING_READ(MIPI_PORT_CTRL(pipe));
 	}
+}
+
+static void intel_dsi_pre_enable(struct intel_encoder *encoder)
+{
+	struct drm_device *dev = encoder->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
+	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->base.crtc);
+	enum pipe pipe = intel_crtc->pipe;
+	u32 tmp;
+
+	DRM_DEBUG_KMS("\n");
+
+	/* Disable DPOunit clock gating, can stall pipe
+	 * and we need DPLL REFA always enabled */
+	tmp = I915_READ(DPLL(pipe));
+	tmp |= DPLL_REFA_CLK_ENABLE_VLV;
+	I915_WRITE(DPLL(pipe), tmp);
+
+	tmp = I915_READ(DSPCLK_GATE_D);
+	tmp |= DPOUNIT_CLOCK_GATE_DISABLE;
+	I915_WRITE(DSPCLK_GATE_D, tmp);
+
+	/* put device in ready state */
+	intel_dsi_device_ready(encoder);
+
+	msleep(intel_dsi->panel_on_delay);
+
+	if (intel_dsi->dev.dev_ops->panel_reset)
+		intel_dsi->dev.dev_ops->panel_reset(&intel_dsi->dev);
+
+	if (intel_dsi->dev.dev_ops->send_otp_cmds)
+		intel_dsi->dev.dev_ops->send_otp_cmds(&intel_dsi->dev);
 
-	if (intel_dsi->dev.dev_ops->enable)
-		intel_dsi->dev.dev_ops->enable(&intel_dsi->dev);
+	/* Enable port in pre-enable phase itself because as per hw team
+	 * recommendation, port should be enabled befor plane & pipe */
+	intel_dsi_enable(encoder);
+}
+
+static void intel_dsi_enable_nop(struct intel_encoder *encoder)
+{
+	DRM_DEBUG_KMS("\n");
+
+	/* for DSI port enable has to be done before pipe
+	 * and plane enable, so port enable is done in
+	 * pre_enable phase itself unlike other encoders
+	 */
+}
+
+static void intel_dsi_pre_disable(struct intel_encoder *encoder)
+{
+	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
+
+	DRM_DEBUG_KMS("\n");
+
+	if (is_vid_mode(intel_dsi)) {
+		/* Send Shutdown command to the panel in LP mode */
+		dpi_send_cmd(intel_dsi, SHUTDOWN, DPI_LP_MODE_EN);
+		msleep(10);
+	}
 }
 
 static void intel_dsi_disable(struct intel_encoder *encoder)
@@ -179,9 +231,6 @@ static void intel_dsi_disable(struct intel_encoder *encoder)
 	DRM_DEBUG_KMS("\n");
 
 	if (is_vid_mode(intel_dsi)) {
-		dpi_send_cmd(intel_dsi, SHUTDOWN);
-		msleep(10);
-
 		/* de-assert ip_tg_enable signal */
 		temp = I915_READ(MIPI_PORT_CTRL(pipe));
 		I915_WRITE(MIPI_PORT_CTRL(pipe), temp & ~DPI_ENABLE);
@@ -190,6 +239,23 @@ static void intel_dsi_disable(struct intel_encoder *encoder)
 		msleep(2);
 	}
 
+	/* Panel commands can be sent when clock is in LP11 */
+	I915_WRITE(MIPI_DEVICE_READY(pipe), 0x0);
+
+	temp = I915_READ(MIPI_CTRL(pipe));
+	temp &= ~ESCAPE_CLOCK_DIVIDER_MASK;
+	I915_WRITE(MIPI_CTRL(pipe), temp |
+			intel_dsi->escape_clk_div <<
+			ESCAPE_CLOCK_DIVIDER_SHIFT);
+
+	I915_WRITE(MIPI_EOT_DISABLE(pipe), CLOCKSTOP);
+
+	temp = I915_READ(MIPI_DSI_FUNC_PRG(pipe));
+	temp &= ~VID_MODE_FORMAT_MASK;
+	I915_WRITE(MIPI_DSI_FUNC_PRG(pipe), temp);
+
+	I915_WRITE(MIPI_DEVICE_READY(pipe), 0x1);
+
 	/* if disable packets are sent before sending shutdown packet then in
 	 * some next enable sequence send turn on packet error is observed */
 	if (intel_dsi->dev.dev_ops->disable)
@@ -227,16 +293,28 @@ static void intel_dsi_clear_device_ready(struct intel_encoder *encoder)
 
 	vlv_disable_dsi_pll(encoder);
 }
+
 static void intel_dsi_post_disable(struct intel_encoder *encoder)
 {
+	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
+	u32 val;
 
 	DRM_DEBUG_KMS("\n");
 
+	intel_dsi_disable(encoder);
+
 	intel_dsi_clear_device_ready(encoder);
 
+	val = I915_READ(DSPCLK_GATE_D);
+	val &= ~DPOUNIT_CLOCK_GATE_DISABLE;
+	I915_WRITE(DSPCLK_GATE_D, val);
+
 	if (intel_dsi->dev.dev_ops->disable_panel_power)
 		intel_dsi->dev.dev_ops->disable_panel_power(&intel_dsi->dev);
+
+	msleep(intel_dsi->panel_off_delay);
+	msleep(intel_dsi->panel_pwr_cycle_delay);
 }
 
 static bool intel_dsi_get_hw_state(struct intel_encoder *encoder,
@@ -364,7 +442,7 @@ static void set_dsi_timings(struct drm_encoder *encoder,
 	I915_WRITE(MIPI_VBP_COUNT(pipe), vbp);
 }
 
-static void intel_dsi_mode_set(struct intel_encoder *intel_encoder)
+static void intel_dsi_prepare(struct intel_encoder *intel_encoder)
 {
 	struct drm_encoder *encoder = &intel_encoder->base;
 	struct drm_device *dev = encoder->dev;
@@ -379,9 +457,6 @@ static void intel_dsi_mode_set(struct intel_encoder *intel_encoder)
 
 	DRM_DEBUG_KMS("pipe %c\n", pipe_name(pipe));
 
-	/* XXX: Location of the call */
-	band_gap_reset(dev_priv);
-
 	/* escape clock divider, 20MHz, shared for A and C. device ready must be
 	 * off when doing this! txclkesc? */
 	tmp = I915_READ(MIPI_CTRL(0));
@@ -452,10 +527,20 @@ static void intel_dsi_mode_set(struct intel_encoder *intel_encoder)
 	/* dphy stuff */
 
 	/* in terms of low power clock */
-	I915_WRITE(MIPI_INIT_COUNT(pipe), txclkesc(ESCAPE_CLOCK_DIVIDER_1, 100));
+	I915_WRITE(MIPI_INIT_COUNT(pipe), txclkesc(intel_dsi->escape_clk_div, 100));
+
+	val = 0;
+	if (intel_dsi->eotp_pkt == 0)
+		val |= EOT_DISABLE;
+
+	if (intel_dsi->clock_stop)
+		val |= CLOCKSTOP;
 
 	/* recovery disables */
-	I915_WRITE(MIPI_EOT_DISABLE(pipe), intel_dsi->eot_disable);
+	I915_WRITE(MIPI_EOT_DISABLE(pipe), val);
+
+	/* in terms of low power clock */
+	I915_WRITE(MIPI_INIT_COUNT(pipe), intel_dsi->init_count);
 
 	/* in terms of txbyteclkhs. actual high to low switch +
 	 * MIPI_STOP_STATE_STALL * MIPI_LP_BYTECLK.
@@ -484,9 +569,23 @@ static void intel_dsi_mode_set(struct intel_encoder *intel_encoder)
 		   intel_dsi->clk_hs_to_lp_count << HS_LP_PWR_SW_CNT_SHIFT);
 
 	if (is_vid_mode(intel_dsi))
+		/* Some panels might have resolution which is not a multiple of
+		 * 64 like 1366 x 768. Enable RANDOM resolution support for such
+		 * panels by default */
 		I915_WRITE(MIPI_VIDEO_MODE_FORMAT(pipe),
 				intel_dsi->video_frmt_cfg_bits |
-				intel_dsi->video_mode_format);
+				intel_dsi->video_mode_format |
+				IP_TG_CONFIG |
+				RANDOM_DPI_DISPLAY_RESOLUTION);
+}
+
+static void intel_dsi_pre_pll_enable(struct intel_encoder *encoder)
+{
+	DRM_DEBUG_KMS("\n");
+
+	intel_dsi_prepare(encoder);
+
+	vlv_enable_dsi_pll(encoder);
 }
 
 static enum drm_connector_status
@@ -566,11 +665,16 @@ bool intel_dsi_init(struct drm_device *dev)
 	struct intel_connector *intel_connector;
 	struct drm_connector *connector;
 	struct drm_display_mode *fixed_mode = NULL;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	const struct intel_dsi_device *dsi;
 	unsigned int i;
 
 	DRM_DEBUG_KMS("\n");
 
+	/* There is no detection method for MIPI so rely on VBT */
+	if (!dev_priv->vbt.has_mipi)
+		return false;
+
 	intel_dsi = kzalloc(sizeof(*intel_dsi), GFP_KERNEL);
 	if (!intel_dsi)
 		return false;
@@ -585,6 +689,13 @@ bool intel_dsi_init(struct drm_device *dev)
 	encoder = &intel_encoder->base;
 	intel_dsi->attached_connector = intel_connector;
 
+	if (IS_VALLEYVIEW(dev)) {
+		dev_priv->mipi_mmio_base = VLV_MIPI_BASE;
+	} else {
+		DRM_ERROR("Unsupported Mipi device to reg base");
+		return false;
+	}
+
 	connector = &intel_connector->base;
 
 	drm_encoder_init(dev, encoder, &intel_dsi_funcs, DRM_MODE_ENCODER_DSI);
@@ -594,9 +705,8 @@ bool intel_dsi_init(struct drm_device *dev)
 	intel_encoder->compute_config = intel_dsi_compute_config;
 	intel_encoder->pre_pll_enable = intel_dsi_pre_pll_enable;
 	intel_encoder->pre_enable = intel_dsi_pre_enable;
-	intel_encoder->enable = intel_dsi_enable;
-	intel_encoder->mode_set = intel_dsi_mode_set;
-	intel_encoder->disable = intel_dsi_disable;
+	intel_encoder->enable = intel_dsi_enable_nop;
+	intel_encoder->disable = intel_dsi_pre_disable;
 	intel_encoder->post_disable = intel_dsi_post_disable;
 	intel_encoder->get_hw_state = intel_dsi_get_hw_state;
 	intel_encoder->get_config = intel_dsi_get_config;
diff --git a/drivers/gpu/drm/i915/intel_dsi.h b/drivers/gpu/drm/i915/intel_dsi.h
index b4a27cec882f..31db33d3e5cc 100644
--- a/drivers/gpu/drm/i915/intel_dsi.h
+++ b/drivers/gpu/drm/i915/intel_dsi.h
@@ -31,7 +31,6 @@
 struct intel_dsi_device {
 	unsigned int panel_id;
 	const char *name;
-	int type;
 	const struct intel_dsi_dev_ops *dev_ops;
 	void *dev_priv;
 };
@@ -85,6 +84,9 @@ struct intel_dsi {
 	/* virtual channel */
 	int channel;
 
+	/* Video mode or command mode */
+	u16 operation_mode;
+
 	/* number of DSI lanes */
 	unsigned int lane_count;
 
@@ -95,8 +97,10 @@ struct intel_dsi {
 	u32 video_mode_format;
 
 	/* eot for MIPI_EOT_DISABLE register */
-	u32 eot_disable;
+	u8 eotp_pkt;
+	u8 clock_stop;
 
+	u8 escape_clk_div;
 	u32 port_bits;
 	u32 bw_timer;
 	u32 dphy_reg;
@@ -110,6 +114,15 @@ struct intel_dsi {
 	u16 hs_to_lp_count;
 	u16 clk_lp_to_hs_count;
 	u16 clk_hs_to_lp_count;
+
+	u16 init_count;
+
+	/* all delays in ms */
+	u16 backlight_off_delay;
+	u16 backlight_on_delay;
+	u16 panel_on_delay;
+	u16 panel_off_delay;
+	u16 panel_pwr_cycle_delay;
 };
 
 static inline struct intel_dsi *enc_to_intel_dsi(struct drm_encoder *encoder)
@@ -120,4 +133,6 @@ static inline struct intel_dsi *enc_to_intel_dsi(struct drm_encoder *encoder)
 extern void vlv_enable_dsi_pll(struct intel_encoder *encoder);
 extern void vlv_disable_dsi_pll(struct intel_encoder *encoder);
 
+extern struct intel_dsi_dev_ops vbt_generic_dsi_display_ops;
+
 #endif /* _INTEL_DSI_H */
diff --git a/drivers/gpu/drm/i915/intel_dsi_cmd.c b/drivers/gpu/drm/i915/intel_dsi_cmd.c
index 7c40f981d2c7..3eeb21b9fddf 100644
--- a/drivers/gpu/drm/i915/intel_dsi_cmd.c
+++ b/drivers/gpu/drm/i915/intel_dsi_cmd.c
@@ -389,7 +389,7 @@ int dsi_vc_generic_read(struct intel_dsi *intel_dsi, int channel,
  *
  * XXX: commands with data in MIPI_DPI_DATA?
  */
-int dpi_send_cmd(struct intel_dsi *intel_dsi, u32 cmd)
+int dpi_send_cmd(struct intel_dsi *intel_dsi, u32 cmd, bool hs)
 {
 	struct drm_encoder *encoder = &intel_dsi->base.base;
 	struct drm_device *dev = encoder->dev;
@@ -399,7 +399,7 @@ int dpi_send_cmd(struct intel_dsi *intel_dsi, u32 cmd)
 	u32 mask;
 
 	/* XXX: pipe, hs */
-	if (intel_dsi->hs)
+	if (hs)
 		cmd &= ~DPI_LP_MODE;
 	else
 		cmd |= DPI_LP_MODE;
diff --git a/drivers/gpu/drm/i915/intel_dsi_cmd.h b/drivers/gpu/drm/i915/intel_dsi_cmd.h
index 54c8a234a2e0..9a18cbfa5460 100644
--- a/drivers/gpu/drm/i915/intel_dsi_cmd.h
+++ b/drivers/gpu/drm/i915/intel_dsi_cmd.h
@@ -33,6 +33,9 @@
 #include "intel_drv.h"
 #include "intel_dsi.h"
 
+#define DPI_LP_MODE_EN	false
+#define DPI_HS_MODE_EN	true
+
 void dsi_hs_mode_enable(struct intel_dsi *intel_dsi, bool enable);
 
 int dsi_vc_dcs_write(struct intel_dsi *intel_dsi, int channel,
@@ -47,7 +50,7 @@ int dsi_vc_dcs_read(struct intel_dsi *intel_dsi, int channel, u8 dcs_cmd,
 int dsi_vc_generic_read(struct intel_dsi *intel_dsi, int channel,
 			u8 *reqdata, int reqlen, u8 *buf, int buflen);
 
-int dpi_send_cmd(struct intel_dsi *intel_dsi, u32 cmd);
+int dpi_send_cmd(struct intel_dsi *intel_dsi, u32 cmd, bool hs);
 
 /* XXX: questionable write helpers */
 static inline int dsi_vc_dcs_write_0(struct intel_dsi *intel_dsi,
diff --git a/drivers/gpu/drm/i915/intel_dsi_panel_vbt.c b/drivers/gpu/drm/i915/intel_dsi_panel_vbt.c
new file mode 100644
index 000000000000..21a0d348cedc
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_dsi_panel_vbt.c
@@ -0,0 +1,589 @@
+/*
+ * Copyright © 2014 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ *
+ * Author: Shobhit Kumar <shobhit.kumar@intel.com>
+ *
+ */
+
+#include <drm/drmP.h>
+#include <drm/drm_crtc.h>
+#include <drm/drm_edid.h>
+#include <drm/i915_drm.h>
+#include <linux/slab.h>
+#include <video/mipi_display.h>
+#include <asm/intel-mid.h>
+#include <video/mipi_display.h>
+#include "i915_drv.h"
+#include "intel_drv.h"
+#include "intel_dsi.h"
+#include "intel_dsi_cmd.h"
+
+#define MIPI_TRANSFER_MODE_SHIFT	0
+#define MIPI_VIRTUAL_CHANNEL_SHIFT	1
+#define MIPI_PORT_SHIFT			3
+
+#define PREPARE_CNT_MAX		0x3F
+#define EXIT_ZERO_CNT_MAX	0x3F
+#define CLK_ZERO_CNT_MAX	0xFF
+#define TRAIL_CNT_MAX		0x1F
+
+#define NS_KHZ_RATIO 1000000
+
+#define GPI0_NC_0_HV_DDI0_HPD           0x4130
+#define GPIO_NC_0_HV_DDI0_PAD           0x4138
+#define GPIO_NC_1_HV_DDI0_DDC_SDA       0x4120
+#define GPIO_NC_1_HV_DDI0_DDC_SDA_PAD   0x4128
+#define GPIO_NC_2_HV_DDI0_DDC_SCL       0x4110
+#define GPIO_NC_2_HV_DDI0_DDC_SCL_PAD   0x4118
+#define GPIO_NC_3_PANEL0_VDDEN          0x4140
+#define GPIO_NC_3_PANEL0_VDDEN_PAD      0x4148
+#define GPIO_NC_4_PANEL0_BLKEN          0x4150
+#define GPIO_NC_4_PANEL0_BLKEN_PAD      0x4158
+#define GPIO_NC_5_PANEL0_BLKCTL         0x4160
+#define GPIO_NC_5_PANEL0_BLKCTL_PAD     0x4168
+#define GPIO_NC_6_PCONF0                0x4180
+#define GPIO_NC_6_PAD                   0x4188
+#define GPIO_NC_7_PCONF0                0x4190
+#define GPIO_NC_7_PAD                   0x4198
+#define GPIO_NC_8_PCONF0                0x4170
+#define GPIO_NC_8_PAD                   0x4178
+#define GPIO_NC_9_PCONF0                0x4100
+#define GPIO_NC_9_PAD                   0x4108
+#define GPIO_NC_10_PCONF0               0x40E0
+#define GPIO_NC_10_PAD                  0x40E8
+#define GPIO_NC_11_PCONF0               0x40F0
+#define GPIO_NC_11_PAD                  0x40F8
+
+struct gpio_table {
+	u16 function_reg;
+	u16 pad_reg;
+	u8 init;
+};
+
+static struct gpio_table gtable[] = {
+	{ GPI0_NC_0_HV_DDI0_HPD, GPIO_NC_0_HV_DDI0_PAD, 0 },
+	{ GPIO_NC_1_HV_DDI0_DDC_SDA, GPIO_NC_1_HV_DDI0_DDC_SDA_PAD, 0 },
+	{ GPIO_NC_2_HV_DDI0_DDC_SCL, GPIO_NC_2_HV_DDI0_DDC_SCL_PAD, 0 },
+	{ GPIO_NC_3_PANEL0_VDDEN, GPIO_NC_3_PANEL0_VDDEN_PAD, 0 },
+	{ GPIO_NC_4_PANEL0_BLKEN, GPIO_NC_4_PANEL0_BLKEN_PAD, 0 },
+	{ GPIO_NC_5_PANEL0_BLKCTL, GPIO_NC_5_PANEL0_BLKCTL_PAD, 0 },
+	{ GPIO_NC_6_PCONF0, GPIO_NC_6_PAD, 0 },
+	{ GPIO_NC_7_PCONF0, GPIO_NC_7_PAD, 0 },
+	{ GPIO_NC_8_PCONF0, GPIO_NC_8_PAD, 0 },
+	{ GPIO_NC_9_PCONF0, GPIO_NC_9_PAD, 0 },
+	{ GPIO_NC_10_PCONF0, GPIO_NC_10_PAD, 0},
+	{ GPIO_NC_11_PCONF0, GPIO_NC_11_PAD, 0}
+};
+
+static u8 *mipi_exec_send_packet(struct intel_dsi *intel_dsi, u8 *data)
+{
+	u8 type, byte, mode, vc, port;
+	u16 len;
+
+	byte = *data++;
+	mode = (byte >> MIPI_TRANSFER_MODE_SHIFT) & 0x1;
+	vc = (byte >> MIPI_VIRTUAL_CHANNEL_SHIFT) & 0x3;
+	port = (byte >> MIPI_PORT_SHIFT) & 0x3;
+
+	/* LP or HS mode */
+	intel_dsi->hs = mode;
+
+	/* get packet type and increment the pointer */
+	type = *data++;
+
+	len = *((u16 *) data);
+	data += 2;
+
+	switch (type) {
+	case MIPI_DSI_GENERIC_SHORT_WRITE_0_PARAM:
+		dsi_vc_generic_write_0(intel_dsi, vc);
+		break;
+	case MIPI_DSI_GENERIC_SHORT_WRITE_1_PARAM:
+		dsi_vc_generic_write_1(intel_dsi, vc, *data);
+		break;
+	case MIPI_DSI_GENERIC_SHORT_WRITE_2_PARAM:
+		dsi_vc_generic_write_2(intel_dsi, vc, *data, *(data + 1));
+		break;
+	case MIPI_DSI_GENERIC_READ_REQUEST_0_PARAM:
+	case MIPI_DSI_GENERIC_READ_REQUEST_1_PARAM:
+	case MIPI_DSI_GENERIC_READ_REQUEST_2_PARAM:
+		DRM_DEBUG_DRIVER("Generic Read not yet implemented or used\n");
+		break;
+	case MIPI_DSI_GENERIC_LONG_WRITE:
+		dsi_vc_generic_write(intel_dsi, vc, data, len);
+		break;
+	case MIPI_DSI_DCS_SHORT_WRITE:
+		dsi_vc_dcs_write_0(intel_dsi, vc, *data);
+		break;
+	case MIPI_DSI_DCS_SHORT_WRITE_PARAM:
+		dsi_vc_dcs_write_1(intel_dsi, vc, *data, *(data + 1));
+		break;
+	case MIPI_DSI_DCS_READ:
+		DRM_DEBUG_DRIVER("DCS Read not yet implemented or used\n");
+		break;
+	case MIPI_DSI_DCS_LONG_WRITE:
+		dsi_vc_dcs_write(intel_dsi, vc, data, len);
+		break;
+	};
+
+	data += len;
+
+	return data;
+}
+
+static u8 *mipi_exec_delay(struct intel_dsi *intel_dsi, u8 *data)
+{
+	u32 delay = *((u32 *) data);
+
+	usleep_range(delay, delay + 10);
+	data += 4;
+
+	return data;
+}
+
+static u8 *mipi_exec_gpio(struct intel_dsi *intel_dsi, u8 *data)
+{
+	u8 gpio, action;
+	u16 function, pad;
+	u32 val;
+	struct drm_device *dev = intel_dsi->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	gpio = *data++;
+
+	/* pull up/down */
+	action = *data++;
+
+	function = gtable[gpio].function_reg;
+	pad = gtable[gpio].pad_reg;
+
+	mutex_lock(&dev_priv->dpio_lock);
+	if (!gtable[gpio].init) {
+		/* program the function */
+		/* FIXME: remove constant below */
+		vlv_gpio_nc_write(dev_priv, function, 0x2000CC00);
+		gtable[gpio].init = 1;
+	}
+
+	val = 0x4 | action;
+
+	/* pull up/down */
+	vlv_gpio_nc_write(dev_priv, pad, val);
+	mutex_unlock(&dev_priv->dpio_lock);
+
+	return data;
+}
+
+typedef u8 * (*fn_mipi_elem_exec)(struct intel_dsi *intel_dsi, u8 *data);
+static const fn_mipi_elem_exec exec_elem[] = {
+	NULL, /* reserved */
+	mipi_exec_send_packet,
+	mipi_exec_delay,
+	mipi_exec_gpio,
+	NULL, /* status read; later */
+};
+
+/*
+ * MIPI Sequence from VBT #53 parsing logic
+ * We have already separated each seqence during bios parsing
+ * Following is generic execution function for any sequence
+ */
+
+static const char * const seq_name[] = {
+	"UNDEFINED",
+	"MIPI_SEQ_ASSERT_RESET",
+	"MIPI_SEQ_INIT_OTP",
+	"MIPI_SEQ_DISPLAY_ON",
+	"MIPI_SEQ_DISPLAY_OFF",
+	"MIPI_SEQ_DEASSERT_RESET"
+};
+
+static void generic_exec_sequence(struct intel_dsi *intel_dsi, char *sequence)
+{
+	u8 *data = sequence;
+	fn_mipi_elem_exec mipi_elem_exec;
+	int index;
+
+	if (!sequence)
+		return;
+
+	DRM_DEBUG_DRIVER("Starting MIPI sequence - %s\n", seq_name[*data]);
+
+	/* go to the first element of the sequence */
+	data++;
+
+	/* parse each byte till we reach end of sequence byte - 0x00 */
+	while (1) {
+		index = *data;
+		mipi_elem_exec = exec_elem[index];
+		if (!mipi_elem_exec) {
+			DRM_ERROR("Unsupported MIPI element, skipping sequence execution\n");
+			return;
+		}
+
+		/* goto element payload */
+		data++;
+
+		/* execute the element specific rotines */
+		data = mipi_elem_exec(intel_dsi, data);
+
+		/*
+		 * After processing the element, data should point to
+		 * next element or end of sequence
+		 * check if have we reached end of sequence
+		 */
+		if (*data == 0x00)
+			break;
+	}
+}
+
+static bool generic_init(struct intel_dsi_device *dsi)
+{
+	struct intel_dsi *intel_dsi = container_of(dsi, struct intel_dsi, dev);
+	struct drm_device *dev = intel_dsi->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct mipi_config *mipi_config = dev_priv->vbt.dsi.config;
+	struct mipi_pps_data *pps = dev_priv->vbt.dsi.pps;
+	struct drm_display_mode *mode = dev_priv->vbt.lfp_lvds_vbt_mode;
+	u32 bits_per_pixel = 24;
+	u32 tlpx_ns, extra_byte_count, bitrate, tlpx_ui;
+	u32 ui_num, ui_den;
+	u32 prepare_cnt, exit_zero_cnt, clk_zero_cnt, trail_cnt;
+	u32 ths_prepare_ns, tclk_trail_ns;
+	u32 tclk_prepare_clkzero, ths_prepare_hszero;
+	u32 lp_to_hs_switch, hs_to_lp_switch;
+
+	DRM_DEBUG_KMS("\n");
+
+	intel_dsi->eotp_pkt = mipi_config->eot_pkt_disabled ? 0 : 1;
+	intel_dsi->clock_stop = mipi_config->enable_clk_stop ? 1 : 0;
+	intel_dsi->lane_count = mipi_config->lane_cnt + 1;
+	intel_dsi->pixel_format = mipi_config->videomode_color_format << 7;
+
+	if (intel_dsi->pixel_format == VID_MODE_FORMAT_RGB666)
+		bits_per_pixel = 18;
+	else if (intel_dsi->pixel_format == VID_MODE_FORMAT_RGB565)
+		bits_per_pixel = 16;
+
+	bitrate = (mode->clock * bits_per_pixel) / intel_dsi->lane_count;
+
+	intel_dsi->operation_mode = mipi_config->is_cmd_mode;
+	intel_dsi->video_mode_format = mipi_config->video_transfer_mode;
+	intel_dsi->escape_clk_div = mipi_config->byte_clk_sel;
+	intel_dsi->lp_rx_timeout = mipi_config->lp_rx_timeout;
+	intel_dsi->turn_arnd_val = mipi_config->turn_around_timeout;
+	intel_dsi->rst_timer_val = mipi_config->device_reset_timer;
+	intel_dsi->init_count = mipi_config->master_init_timer;
+	intel_dsi->bw_timer = mipi_config->dbi_bw_timer;
+	intel_dsi->video_frmt_cfg_bits = mipi_config->bta_enabled ? DISABLE_VIDEO_BTA : 0;
+
+	switch (intel_dsi->escape_clk_div) {
+	case 0:
+		tlpx_ns = 50;
+		break;
+	case 1:
+		tlpx_ns = 100;
+		break;
+
+	case 2:
+		tlpx_ns = 200;
+		break;
+	default:
+		tlpx_ns = 50;
+		break;
+	}
+
+	switch (intel_dsi->lane_count) {
+	case 1:
+	case 2:
+		extra_byte_count = 2;
+		break;
+	case 3:
+		extra_byte_count = 4;
+		break;
+	case 4:
+	default:
+		extra_byte_count = 3;
+		break;
+	}
+
+	/*
+	 * ui(s) = 1/f [f in hz]
+	 * ui(ns) = 10^9 / (f*10^6) [f in Mhz] -> 10^3/f(Mhz)
+	 */
+
+	/* in Kbps */
+	ui_num = NS_KHZ_RATIO;
+	ui_den = bitrate;
+
+	tclk_prepare_clkzero = mipi_config->tclk_prepare_clkzero;
+	ths_prepare_hszero = mipi_config->ths_prepare_hszero;
+
+	/*
+	 * B060
+	 * LP byte clock = TLPX/ (8UI)
+	 */
+	intel_dsi->lp_byte_clk = DIV_ROUND_UP(tlpx_ns * ui_den, 8 * ui_num);
+
+	/* count values in UI = (ns value) * (bitrate / (2 * 10^6))
+	 *
+	 * Since txddrclkhs_i is 2xUI, all the count values programmed in
+	 * DPHY param register are divided by 2
+	 *
+	 * prepare count
+	 */
+	ths_prepare_ns = max(mipi_config->ths_prepare, mipi_config->tclk_prepare);
+	prepare_cnt = DIV_ROUND_UP(ths_prepare_ns * ui_den, ui_num * 2);
+
+	/* exit zero count */
+	exit_zero_cnt = DIV_ROUND_UP(
+				(ths_prepare_hszero - ths_prepare_ns) * ui_den,
+				ui_num * 2
+				);
+
+	/*
+	 * Exit zero  is unified val ths_zero and ths_exit
+	 * minimum value for ths_exit = 110ns
+	 * min (exit_zero_cnt * 2) = 110/UI
+	 * exit_zero_cnt = 55/UI
+	 */
+	 if (exit_zero_cnt < (55 * ui_den / ui_num))
+		if ((55 * ui_den) % ui_num)
+			exit_zero_cnt += 1;
+
+	/* clk zero count */
+	clk_zero_cnt = DIV_ROUND_UP(
+			(tclk_prepare_clkzero -	ths_prepare_ns)
+			* ui_den, 2 * ui_num);
+
+	/* trail count */
+	tclk_trail_ns = max(mipi_config->tclk_trail, mipi_config->ths_trail);
+	trail_cnt = DIV_ROUND_UP(tclk_trail_ns * ui_den, 2 * ui_num);
+
+	if (prepare_cnt > PREPARE_CNT_MAX ||
+		exit_zero_cnt > EXIT_ZERO_CNT_MAX ||
+		clk_zero_cnt > CLK_ZERO_CNT_MAX ||
+		trail_cnt > TRAIL_CNT_MAX)
+		DRM_DEBUG_DRIVER("Values crossing maximum limits, restricting to max values\n");
+
+	if (prepare_cnt > PREPARE_CNT_MAX)
+		prepare_cnt = PREPARE_CNT_MAX;
+
+	if (exit_zero_cnt > EXIT_ZERO_CNT_MAX)
+		exit_zero_cnt = EXIT_ZERO_CNT_MAX;
+
+	if (clk_zero_cnt > CLK_ZERO_CNT_MAX)
+		clk_zero_cnt = CLK_ZERO_CNT_MAX;
+
+	if (trail_cnt > TRAIL_CNT_MAX)
+		trail_cnt = TRAIL_CNT_MAX;
+
+	/* B080 */
+	intel_dsi->dphy_reg = exit_zero_cnt << 24 | trail_cnt << 16 |
+						clk_zero_cnt << 8 | prepare_cnt;
+
+	/*
+	 * LP to HS switch count = 4TLPX + PREP_COUNT * 2 + EXIT_ZERO_COUNT * 2
+	 *					+ 10UI + Extra Byte Count
+	 *
+	 * HS to LP switch count = THS-TRAIL + 2TLPX + Extra Byte Count
+	 * Extra Byte Count is calculated according to number of lanes.
+	 * High Low Switch Count is the Max of LP to HS and
+	 * HS to LP switch count
+	 *
+	 */
+	tlpx_ui = DIV_ROUND_UP(tlpx_ns * ui_den, ui_num);
+
+	/* B044 */
+	/* FIXME:
+	 * The comment above does not match with the code */
+	lp_to_hs_switch = DIV_ROUND_UP(4 * tlpx_ui + prepare_cnt * 2 +
+						exit_zero_cnt * 2 + 10, 8);
+
+	hs_to_lp_switch = DIV_ROUND_UP(mipi_config->ths_trail + 2 * tlpx_ui, 8);
+
+	intel_dsi->hs_to_lp_count = max(lp_to_hs_switch, hs_to_lp_switch);
+	intel_dsi->hs_to_lp_count += extra_byte_count;
+
+	/* B088 */
+	/* LP -> HS for clock lanes
+	 * LP clk sync + LP11 + LP01 + tclk_prepare + tclk_zero +
+	 *						extra byte count
+	 * 2TPLX + 1TLPX + 1 TPLX(in ns) + prepare_cnt * 2 + clk_zero_cnt *
+	 *					2(in UI) + extra byte count
+	 * In byteclks = (4TLPX + prepare_cnt * 2 + clk_zero_cnt *2 (in UI)) /
+	 *					8 + extra byte count
+	 */
+	intel_dsi->clk_lp_to_hs_count =
+		DIV_ROUND_UP(
+			4 * tlpx_ui + prepare_cnt * 2 +
+			clk_zero_cnt * 2,
+			8);
+
+	intel_dsi->clk_lp_to_hs_count += extra_byte_count;
+
+	/* HS->LP for Clock Lanes
+	 * Low Power clock synchronisations + 1Tx byteclk + tclk_trail +
+	 *						Extra byte count
+	 * 2TLPX + 8UI + (trail_count*2)(in UI) + Extra byte count
+	 * In byteclks = (2*TLpx(in UI) + trail_count*2 +8)(in UI)/8 +
+	 *						Extra byte count
+	 */
+	intel_dsi->clk_hs_to_lp_count =
+		DIV_ROUND_UP(2 * tlpx_ui + trail_cnt * 2 + 8,
+			8);
+	intel_dsi->clk_hs_to_lp_count += extra_byte_count;
+
+	DRM_DEBUG_KMS("Eot %s\n", intel_dsi->eotp_pkt ? "enabled" : "disabled");
+	DRM_DEBUG_KMS("Clockstop %s\n", intel_dsi->clock_stop ?
+						"disabled" : "enabled");
+	DRM_DEBUG_KMS("Mode %s\n", intel_dsi->operation_mode ? "command" : "video");
+	DRM_DEBUG_KMS("Pixel Format %d\n", intel_dsi->pixel_format);
+	DRM_DEBUG_KMS("TLPX %d\n", intel_dsi->escape_clk_div);
+	DRM_DEBUG_KMS("LP RX Timeout 0x%x\n", intel_dsi->lp_rx_timeout);
+	DRM_DEBUG_KMS("Turnaround Timeout 0x%x\n", intel_dsi->turn_arnd_val);
+	DRM_DEBUG_KMS("Init Count 0x%x\n", intel_dsi->init_count);
+	DRM_DEBUG_KMS("HS to LP Count 0x%x\n", intel_dsi->hs_to_lp_count);
+	DRM_DEBUG_KMS("LP Byte Clock %d\n", intel_dsi->lp_byte_clk);
+	DRM_DEBUG_KMS("DBI BW Timer 0x%x\n", intel_dsi->bw_timer);
+	DRM_DEBUG_KMS("LP to HS Clock Count 0x%x\n", intel_dsi->clk_lp_to_hs_count);
+	DRM_DEBUG_KMS("HS to LP Clock Count 0x%x\n", intel_dsi->clk_hs_to_lp_count);
+	DRM_DEBUG_KMS("BTA %s\n",
+			intel_dsi->video_frmt_cfg_bits & DISABLE_VIDEO_BTA ?
+			"disabled" : "enabled");
+
+	/* delays in VBT are in unit of 100us, so need to convert
+	 * here in ms
+	 * Delay (100us) * 100 /1000 = Delay / 10 (ms) */
+	intel_dsi->backlight_off_delay = pps->bl_disable_delay / 10;
+	intel_dsi->backlight_on_delay = pps->bl_enable_delay / 10;
+	intel_dsi->panel_on_delay = pps->panel_on_delay / 10;
+	intel_dsi->panel_off_delay = pps->panel_off_delay / 10;
+	intel_dsi->panel_pwr_cycle_delay = pps->panel_power_cycle_delay / 10;
+
+	return true;
+}
+
+static int generic_mode_valid(struct intel_dsi_device *dsi,
+		   struct drm_display_mode *mode)
+{
+	return MODE_OK;
+}
+
+static bool generic_mode_fixup(struct intel_dsi_device *dsi,
+		    const struct drm_display_mode *mode,
+		    struct drm_display_mode *adjusted_mode) {
+	return true;
+}
+
+static void generic_panel_reset(struct intel_dsi_device *dsi)
+{
+	struct intel_dsi *intel_dsi = container_of(dsi, struct intel_dsi, dev);
+	struct drm_device *dev = intel_dsi->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	char *sequence = dev_priv->vbt.dsi.sequence[MIPI_SEQ_ASSERT_RESET];
+
+	generic_exec_sequence(intel_dsi, sequence);
+}
+
+static void generic_disable_panel_power(struct intel_dsi_device *dsi)
+{
+	struct intel_dsi *intel_dsi = container_of(dsi, struct intel_dsi, dev);
+	struct drm_device *dev = intel_dsi->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	char *sequence = dev_priv->vbt.dsi.sequence[MIPI_SEQ_DEASSERT_RESET];
+
+	generic_exec_sequence(intel_dsi, sequence);
+}
+
+static void generic_send_otp_cmds(struct intel_dsi_device *dsi)
+{
+	struct intel_dsi *intel_dsi = container_of(dsi, struct intel_dsi, dev);
+	struct drm_device *dev = intel_dsi->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	char *sequence = dev_priv->vbt.dsi.sequence[MIPI_SEQ_INIT_OTP];
+
+	generic_exec_sequence(intel_dsi, sequence);
+}
+
+static void generic_enable(struct intel_dsi_device *dsi)
+{
+	struct intel_dsi *intel_dsi = container_of(dsi, struct intel_dsi, dev);
+	struct drm_device *dev = intel_dsi->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	char *sequence = dev_priv->vbt.dsi.sequence[MIPI_SEQ_DISPLAY_ON];
+
+	generic_exec_sequence(intel_dsi, sequence);
+}
+
+static void generic_disable(struct intel_dsi_device *dsi)
+{
+	struct intel_dsi *intel_dsi = container_of(dsi, struct intel_dsi, dev);
+	struct drm_device *dev = intel_dsi->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	char *sequence = dev_priv->vbt.dsi.sequence[MIPI_SEQ_DISPLAY_OFF];
+
+	generic_exec_sequence(intel_dsi, sequence);
+}
+
+static enum drm_connector_status generic_detect(struct intel_dsi_device *dsi)
+{
+	return connector_status_connected;
+}
+
+static bool generic_get_hw_state(struct intel_dsi_device *dev)
+{
+	return true;
+}
+
+static struct drm_display_mode *generic_get_modes(struct intel_dsi_device *dsi)
+{
+	struct intel_dsi *intel_dsi = container_of(dsi, struct intel_dsi, dev);
+	struct drm_device *dev = intel_dsi->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	dev_priv->vbt.lfp_lvds_vbt_mode->type |= DRM_MODE_TYPE_PREFERRED;
+	return dev_priv->vbt.lfp_lvds_vbt_mode;
+}
+
+static void generic_destroy(struct intel_dsi_device *dsi) { }
+
+/* Callbacks. We might not need them all. */
+struct intel_dsi_dev_ops vbt_generic_dsi_display_ops = {
+	.init = generic_init,
+	.mode_valid = generic_mode_valid,
+	.mode_fixup = generic_mode_fixup,
+	.panel_reset = generic_panel_reset,
+	.disable_panel_power = generic_disable_panel_power,
+	.send_otp_cmds = generic_send_otp_cmds,
+	.enable = generic_enable,
+	.disable = generic_disable,
+	.detect = generic_detect,
+	.get_hw_state = generic_get_hw_state,
+	.get_modes = generic_get_modes,
+	.destroy = generic_destroy,
+};
diff --git a/drivers/gpu/drm/i915/intel_dvo.c b/drivers/gpu/drm/i915/intel_dvo.c
index 7fe3feedfe03..a3631c0a5c28 100644
--- a/drivers/gpu/drm/i915/intel_dvo.c
+++ b/drivers/gpu/drm/i915/intel_dvo.c
@@ -285,7 +285,7 @@ static bool intel_dvo_compute_config(struct intel_encoder *encoder,
 	return true;
 }
 
-static void intel_dvo_mode_set(struct intel_encoder *encoder)
+static void intel_dvo_pre_enable(struct intel_encoder *encoder)
 {
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -343,7 +343,7 @@ intel_dvo_detect(struct drm_connector *connector, bool force)
 {
 	struct intel_dvo *intel_dvo = intel_attached_dvo(connector);
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
-		      connector->base.id, drm_get_connector_name(connector));
+		      connector->base.id, connector->name);
 	return intel_dvo->dev.dev_ops->detect(&intel_dvo->dev);
 }
 
@@ -475,7 +475,7 @@ void intel_dvo_init(struct drm_device *dev)
 	intel_encoder->get_hw_state = intel_dvo_get_hw_state;
 	intel_encoder->get_config = intel_dvo_get_config;
 	intel_encoder->compute_config = intel_dvo_compute_config;
-	intel_encoder->mode_set = intel_dvo_mode_set;
+	intel_encoder->pre_enable = intel_dvo_pre_enable;
 	intel_connector->get_hw_state = intel_dvo_connector_get_hw_state;
 	intel_connector->unregister = intel_connector_unregister;
 
diff --git a/drivers/gpu/drm/i915/intel_fbdev.c b/drivers/gpu/drm/i915/intel_fbdev.c
index f73ba5e6b7a8..088fe9378a4c 100644
--- a/drivers/gpu/drm/i915/intel_fbdev.c
+++ b/drivers/gpu/drm/i915/intel_fbdev.c
@@ -343,15 +343,15 @@ static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper,
 			num_connectors_detected++;
 
 		if (!enabled[i]) {
-			DRM_DEBUG_KMS("connector %d not enabled, skipping\n",
-				      connector->base.id);
+			DRM_DEBUG_KMS("connector %s not enabled, skipping\n",
+				      connector->name);
 			continue;
 		}
 
 		encoder = connector->encoder;
 		if (!encoder || WARN_ON(!encoder->crtc)) {
-			DRM_DEBUG_KMS("connector %d has no encoder or crtc, skipping\n",
-				      connector->base.id);
+			DRM_DEBUG_KMS("connector %s has no encoder or crtc, skipping\n",
+				      connector->name);
 			enabled[i] = false;
 			continue;
 		}
@@ -373,16 +373,16 @@ static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper,
 			}
 		}
 
-		DRM_DEBUG_KMS("looking for cmdline mode on connector %d\n",
-			      fb_conn->connector->base.id);
+		DRM_DEBUG_KMS("looking for cmdline mode on connector %s\n",
+			      connector->name);
 
 		/* go for command line mode first */
 		modes[i] = drm_pick_cmdline_mode(fb_conn, width, height);
 
 		/* try for preferred next */
 		if (!modes[i]) {
-			DRM_DEBUG_KMS("looking for preferred mode on connector %d\n",
-				      fb_conn->connector->base.id);
+			DRM_DEBUG_KMS("looking for preferred mode on connector %s\n",
+				      connector->name);
 			modes[i] = drm_has_preferred_mode(fb_conn, width,
 							  height);
 		}
@@ -390,7 +390,7 @@ static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper,
 		/* No preferred mode marked by the EDID? Are there any modes? */
 		if (!modes[i] && !list_empty(&connector->modes)) {
 			DRM_DEBUG_KMS("using first mode listed on connector %s\n",
-				      drm_get_connector_name(connector));
+				      connector->name);
 			modes[i] = list_first_entry(&connector->modes,
 						    struct drm_display_mode,
 						    head);
@@ -409,16 +409,20 @@ static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper,
 			 * since the fb helper layer wants a pointer to
 			 * something we own.
 			 */
+			DRM_DEBUG_KMS("looking for current mode on connector %s\n",
+				      connector->name);
 			intel_mode_from_pipe_config(&encoder->crtc->hwmode,
 						    &to_intel_crtc(encoder->crtc)->config);
 			modes[i] = &encoder->crtc->hwmode;
 		}
 		crtcs[i] = new_crtc;
 
-		DRM_DEBUG_KMS("connector %s on crtc %d: %s\n",
-			      drm_get_connector_name(connector),
+		DRM_DEBUG_KMS("connector %s on pipe %d [CRTC:%d]: %dx%d%s\n",
+			      connector->name,
+			      pipe_name(to_intel_crtc(encoder->crtc)->pipe),
 			      encoder->crtc->base.id,
-			      modes[i]->name);
+			      modes[i]->hdisplay, modes[i]->vdisplay,
+			      modes[i]->flags & DRM_MODE_FLAG_INTERLACE ? "i" :"");
 
 		fallback = false;
 	}
@@ -497,7 +501,7 @@ static bool intel_fbdev_init_bios(struct drm_device *dev,
 		return false;
 
 	/* Find the largest fb */
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, crtc) {
 		intel_crtc = to_intel_crtc(crtc);
 
 		if (!intel_crtc->active || !crtc->primary->fb) {
@@ -521,7 +525,7 @@ static bool intel_fbdev_init_bios(struct drm_device *dev,
 	}
 
 	/* Now make sure all the pipes will fit into it */
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, crtc) {
 		unsigned int cur_size;
 
 		intel_crtc = to_intel_crtc(crtc);
@@ -586,7 +590,7 @@ static bool intel_fbdev_init_bios(struct drm_device *dev,
 	drm_framebuffer_reference(&ifbdev->fb->base);
 
 	/* Final pass to check if any active pipes don't have fbs */
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, crtc) {
 		intel_crtc = to_intel_crtc(crtc);
 
 		if (!intel_crtc->active)
@@ -692,11 +696,7 @@ void intel_fbdev_restore_mode(struct drm_device *dev)
 	if (!dev_priv->fbdev)
 		return;
 
-	drm_modeset_lock_all(dev);
-
-	ret = drm_fb_helper_restore_fbdev_mode(&dev_priv->fbdev->helper);
+	ret = drm_fb_helper_restore_fbdev_mode_unlocked(&dev_priv->fbdev->helper);
 	if (ret)
 		DRM_DEBUG("failed to restore crtc mode\n");
-
-	drm_modeset_unlock_all(dev);
 }
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 157267aa3561..eee2bbec2958 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -418,6 +418,7 @@ intel_hdmi_set_hdmi_infoframe(struct drm_encoder *encoder,
 }
 
 static void g4x_set_infoframes(struct drm_encoder *encoder,
+			       bool enable,
 			       struct drm_display_mode *adjusted_mode)
 {
 	struct drm_i915_private *dev_priv = encoder->dev->dev_private;
@@ -440,7 +441,7 @@ static void g4x_set_infoframes(struct drm_encoder *encoder,
 	 * either. */
 	val |= VIDEO_DIP_SELECT_AVI | VIDEO_DIP_FREQ_VSYNC;
 
-	if (!intel_hdmi->has_hdmi_sink) {
+	if (!enable) {
 		if (!(val & VIDEO_DIP_ENABLE))
 			return;
 		val &= ~VIDEO_DIP_ENABLE;
@@ -471,6 +472,7 @@ static void g4x_set_infoframes(struct drm_encoder *encoder,
 }
 
 static void ibx_set_infoframes(struct drm_encoder *encoder,
+			       bool enable,
 			       struct drm_display_mode *adjusted_mode)
 {
 	struct drm_i915_private *dev_priv = encoder->dev->dev_private;
@@ -486,7 +488,7 @@ static void ibx_set_infoframes(struct drm_encoder *encoder,
 	/* See the big comment in g4x_set_infoframes() */
 	val |= VIDEO_DIP_SELECT_AVI | VIDEO_DIP_FREQ_VSYNC;
 
-	if (!intel_hdmi->has_hdmi_sink) {
+	if (!enable) {
 		if (!(val & VIDEO_DIP_ENABLE))
 			return;
 		val &= ~VIDEO_DIP_ENABLE;
@@ -518,6 +520,7 @@ static void ibx_set_infoframes(struct drm_encoder *encoder,
 }
 
 static void cpt_set_infoframes(struct drm_encoder *encoder,
+			       bool enable,
 			       struct drm_display_mode *adjusted_mode)
 {
 	struct drm_i915_private *dev_priv = encoder->dev->dev_private;
@@ -531,7 +534,7 @@ static void cpt_set_infoframes(struct drm_encoder *encoder,
 	/* See the big comment in g4x_set_infoframes() */
 	val |= VIDEO_DIP_SELECT_AVI | VIDEO_DIP_FREQ_VSYNC;
 
-	if (!intel_hdmi->has_hdmi_sink) {
+	if (!enable) {
 		if (!(val & VIDEO_DIP_ENABLE))
 			return;
 		val &= ~(VIDEO_DIP_ENABLE | VIDEO_DIP_ENABLE_AVI);
@@ -554,20 +557,23 @@ static void cpt_set_infoframes(struct drm_encoder *encoder,
 }
 
 static void vlv_set_infoframes(struct drm_encoder *encoder,
+			       bool enable,
 			       struct drm_display_mode *adjusted_mode)
 {
 	struct drm_i915_private *dev_priv = encoder->dev->dev_private;
+	struct intel_digital_port *intel_dig_port = enc_to_dig_port(encoder);
 	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
 	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
 	u32 reg = VLV_TVIDEO_DIP_CTL(intel_crtc->pipe);
 	u32 val = I915_READ(reg);
+	u32 port = VIDEO_DIP_PORT(intel_dig_port->port);
 
 	assert_hdmi_port_disabled(intel_hdmi);
 
 	/* See the big comment in g4x_set_infoframes() */
 	val |= VIDEO_DIP_SELECT_AVI | VIDEO_DIP_FREQ_VSYNC;
 
-	if (!intel_hdmi->has_hdmi_sink) {
+	if (!enable) {
 		if (!(val & VIDEO_DIP_ENABLE))
 			return;
 		val &= ~VIDEO_DIP_ENABLE;
@@ -576,9 +582,19 @@ static void vlv_set_infoframes(struct drm_encoder *encoder,
 		return;
 	}
 
+	if (port != (val & VIDEO_DIP_PORT_MASK)) {
+		if (val & VIDEO_DIP_ENABLE) {
+			val &= ~VIDEO_DIP_ENABLE;
+			I915_WRITE(reg, val);
+			POSTING_READ(reg);
+		}
+		val &= ~VIDEO_DIP_PORT_MASK;
+		val |= port;
+	}
+
 	val |= VIDEO_DIP_ENABLE;
-	val &= ~(VIDEO_DIP_ENABLE_VENDOR | VIDEO_DIP_ENABLE_GAMUT |
-		 VIDEO_DIP_ENABLE_GCP);
+	val &= ~(VIDEO_DIP_ENABLE_AVI | VIDEO_DIP_ENABLE_VENDOR |
+		 VIDEO_DIP_ENABLE_GAMUT | VIDEO_DIP_ENABLE_GCP);
 
 	I915_WRITE(reg, val);
 	POSTING_READ(reg);
@@ -589,6 +605,7 @@ static void vlv_set_infoframes(struct drm_encoder *encoder,
 }
 
 static void hsw_set_infoframes(struct drm_encoder *encoder,
+			       bool enable,
 			       struct drm_display_mode *adjusted_mode)
 {
 	struct drm_i915_private *dev_priv = encoder->dev->dev_private;
@@ -599,7 +616,7 @@ static void hsw_set_infoframes(struct drm_encoder *encoder,
 
 	assert_hdmi_port_disabled(intel_hdmi);
 
-	if (!intel_hdmi->has_hdmi_sink) {
+	if (!enable) {
 		I915_WRITE(reg, 0);
 		POSTING_READ(reg);
 		return;
@@ -616,7 +633,7 @@ static void hsw_set_infoframes(struct drm_encoder *encoder,
 	intel_hdmi_set_hdmi_infoframe(encoder, adjusted_mode);
 }
 
-static void intel_hdmi_mode_set(struct intel_encoder *encoder)
+static void intel_hdmi_prepare(struct intel_encoder *encoder)
 {
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -638,27 +655,26 @@ static void intel_hdmi_mode_set(struct intel_encoder *encoder)
 	else
 		hdmi_val |= SDVO_COLOR_FORMAT_8bpc;
 
-	/* Required on CPT */
-	if (intel_hdmi->has_hdmi_sink && HAS_PCH_CPT(dev))
+	if (crtc->config.has_hdmi_sink)
 		hdmi_val |= HDMI_MODE_SELECT_HDMI;
 
-	if (intel_hdmi->has_audio) {
+	if (crtc->config.has_audio) {
+		WARN_ON(!crtc->config.has_hdmi_sink);
 		DRM_DEBUG_DRIVER("Enabling HDMI audio on pipe %c\n",
 				 pipe_name(crtc->pipe));
 		hdmi_val |= SDVO_AUDIO_ENABLE;
-		hdmi_val |= HDMI_MODE_SELECT_HDMI;
 		intel_write_eld(&encoder->base, adjusted_mode);
 	}
 
 	if (HAS_PCH_CPT(dev))
 		hdmi_val |= SDVO_PIPE_SEL_CPT(crtc->pipe);
+	else if (IS_CHERRYVIEW(dev))
+		hdmi_val |= SDVO_PIPE_SEL_CHV(crtc->pipe);
 	else
 		hdmi_val |= SDVO_PIPE_SEL(crtc->pipe);
 
 	I915_WRITE(intel_hdmi->hdmi_reg, hdmi_val);
 	POSTING_READ(intel_hdmi->hdmi_reg);
-
-	intel_hdmi->set_infoframes(&encoder->base, adjusted_mode);
 }
 
 static bool intel_hdmi_get_hw_state(struct intel_encoder *encoder,
@@ -681,6 +697,8 @@ static bool intel_hdmi_get_hw_state(struct intel_encoder *encoder,
 
 	if (HAS_PCH_CPT(dev))
 		*pipe = PORT_TO_PIPE_CPT(tmp);
+	else if (IS_CHERRYVIEW(dev))
+		*pipe = SDVO_PORT_TO_PIPE_CHV(tmp);
 	else
 		*pipe = PORT_TO_PIPE(tmp);
 
@@ -707,6 +725,12 @@ static void intel_hdmi_get_config(struct intel_encoder *encoder,
 	else
 		flags |= DRM_MODE_FLAG_NVSYNC;
 
+	if (tmp & HDMI_MODE_SELECT_HDMI)
+		pipe_config->has_hdmi_sink = true;
+
+	if (tmp & HDMI_MODE_SELECT_HDMI)
+		pipe_config->has_audio = true;
+
 	pipe_config->adjusted_mode.flags |= flags;
 
 	if ((tmp & SDVO_COLOR_FORMAT_MASK) == HDMI_COLOR_FORMAT_12bpc)
@@ -729,7 +753,7 @@ static void intel_enable_hdmi(struct intel_encoder *encoder)
 	u32 temp;
 	u32 enable_bits = SDVO_ENABLE;
 
-	if (intel_hdmi->has_audio)
+	if (intel_crtc->config.has_audio)
 		enable_bits |= SDVO_AUDIO_ENABLE;
 
 	temp = I915_READ(intel_hdmi->hdmi_reg);
@@ -883,9 +907,11 @@ bool intel_hdmi_compute_config(struct intel_encoder *encoder,
 	int portclock_limit = hdmi_portclock_limit(intel_hdmi, false);
 	int desired_bpp;
 
+	pipe_config->has_hdmi_sink = intel_hdmi->has_hdmi_sink;
+
 	if (intel_hdmi->color_range_auto) {
 		/* See CEA-861-E - 5.1 Default Encoding Parameters */
-		if (intel_hdmi->has_hdmi_sink &&
+		if (pipe_config->has_hdmi_sink &&
 		    drm_match_cea_mode(adjusted_mode) > 1)
 			intel_hdmi->color_range = HDMI_COLOR_RANGE_16_235;
 		else
@@ -898,13 +924,16 @@ bool intel_hdmi_compute_config(struct intel_encoder *encoder,
 	if (HAS_PCH_SPLIT(dev) && !HAS_DDI(dev))
 		pipe_config->has_pch_encoder = true;
 
+	if (pipe_config->has_hdmi_sink && intel_hdmi->has_audio)
+		pipe_config->has_audio = true;
+
 	/*
 	 * HDMI is either 12 or 8, so if the display lets 10bpc sneak
 	 * through, clamp it down. Note that g4x/vlv don't support 12bpc hdmi
 	 * outputs. We also need to check that the higher clock still fits
 	 * within limits.
 	 */
-	if (pipe_config->pipe_bpp > 8*3 && intel_hdmi->has_hdmi_sink &&
+	if (pipe_config->pipe_bpp > 8*3 && pipe_config->has_hdmi_sink &&
 	    clock_12bpc <= portclock_limit &&
 	    hdmi_12bpc_possible(encoder->new_crtc)) {
 		DRM_DEBUG_KMS("picking bpc to 12 for HDMI output\n");
@@ -944,7 +973,7 @@ intel_hdmi_detect(struct drm_connector *connector, bool force)
 	enum drm_connector_status status = connector_status_disconnected;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
-		      connector->base.id, drm_get_connector_name(connector));
+		      connector->base.id, connector->name);
 
 	power_domain = intel_display_port_power_domain(intel_encoder);
 	intel_display_power_get(dev_priv, power_domain);
@@ -1104,20 +1133,34 @@ done:
 	return 0;
 }
 
+static void intel_hdmi_pre_enable(struct intel_encoder *encoder)
+{
+	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&encoder->base);
+	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->base.crtc);
+	struct drm_display_mode *adjusted_mode =
+		&intel_crtc->config.adjusted_mode;
+
+	intel_hdmi_prepare(encoder);
+
+	intel_hdmi->set_infoframes(&encoder->base,
+				   intel_crtc->config.has_hdmi_sink,
+				   adjusted_mode);
+}
+
 static void vlv_hdmi_pre_enable(struct intel_encoder *encoder)
 {
 	struct intel_digital_port *dport = enc_to_dig_port(&encoder->base);
+	struct intel_hdmi *intel_hdmi = &dport->hdmi;
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc =
 		to_intel_crtc(encoder->base.crtc);
+	struct drm_display_mode *adjusted_mode =
+		&intel_crtc->config.adjusted_mode;
 	enum dpio_channel port = vlv_dport_to_channel(dport);
 	int pipe = intel_crtc->pipe;
 	u32 val;
 
-	if (!IS_VALLEYVIEW(dev))
-		return;
-
 	/* Enable clock channels for this port */
 	mutex_lock(&dev_priv->dpio_lock);
 	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW8(port));
@@ -1144,6 +1187,10 @@ static void vlv_hdmi_pre_enable(struct intel_encoder *encoder)
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS_DW23(port), 0x00400888);
 	mutex_unlock(&dev_priv->dpio_lock);
 
+	intel_hdmi->set_infoframes(&encoder->base,
+				   intel_crtc->config.has_hdmi_sink,
+				   adjusted_mode);
+
 	intel_enable_hdmi(encoder);
 
 	vlv_wait_port_ready(dev_priv, dport);
@@ -1159,8 +1206,7 @@ static void vlv_hdmi_pre_pll_enable(struct intel_encoder *encoder)
 	enum dpio_channel port = vlv_dport_to_channel(dport);
 	int pipe = intel_crtc->pipe;
 
-	if (!IS_VALLEYVIEW(dev))
-		return;
+	intel_hdmi_prepare(encoder);
 
 	/* Program Tx lane resets to default */
 	mutex_lock(&dev_priv->dpio_lock);
@@ -1199,6 +1245,152 @@ static void vlv_hdmi_post_disable(struct intel_encoder *encoder)
 	mutex_unlock(&dev_priv->dpio_lock);
 }
 
+static void chv_hdmi_post_disable(struct intel_encoder *encoder)
+{
+	struct intel_digital_port *dport = enc_to_dig_port(&encoder->base);
+	struct drm_device *dev = encoder->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc =
+		to_intel_crtc(encoder->base.crtc);
+	enum dpio_channel ch = vlv_dport_to_channel(dport);
+	enum pipe pipe = intel_crtc->pipe;
+	u32 val;
+
+	mutex_lock(&dev_priv->dpio_lock);
+
+	/* Propagate soft reset to data lane reset */
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW1(ch));
+	val |= CHV_PCS_REQ_SOFTRESET_EN;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW1(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW1(ch));
+	val |= CHV_PCS_REQ_SOFTRESET_EN;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW1(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW0(ch));
+	val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW0(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW0(ch));
+	val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW0(ch), val);
+
+	mutex_unlock(&dev_priv->dpio_lock);
+}
+
+static void chv_hdmi_pre_enable(struct intel_encoder *encoder)
+{
+	struct intel_digital_port *dport = enc_to_dig_port(&encoder->base);
+	struct drm_device *dev = encoder->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc =
+		to_intel_crtc(encoder->base.crtc);
+	enum dpio_channel ch = vlv_dport_to_channel(dport);
+	int pipe = intel_crtc->pipe;
+	int data, i;
+	u32 val;
+
+	mutex_lock(&dev_priv->dpio_lock);
+
+	/* Deassert soft data lane reset*/
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW1(ch));
+	val |= CHV_PCS_REQ_SOFTRESET_EN;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW1(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW1(ch));
+	val |= CHV_PCS_REQ_SOFTRESET_EN;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW1(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW0(ch));
+	val |= (DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW0(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW0(ch));
+	val |= (DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW0(ch), val);
+
+	/* Program Tx latency optimal setting */
+	for (i = 0; i < 4; i++) {
+		/* Set the latency optimal bit */
+		data = (i == 1) ? 0x0 : 0x6;
+		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW11(ch, i),
+				data << DPIO_FRC_LATENCY_SHFIT);
+
+		/* Set the upar bit */
+		data = (i == 1) ? 0x0 : 0x1;
+		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW14(ch, i),
+				data << DPIO_UPAR_SHIFT);
+	}
+
+	/* Data lane stagger programming */
+	/* FIXME: Fix up value only after power analysis */
+
+	/* Clear calc init */
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW10(ch));
+	val &= ~(DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW10(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW10(ch));
+	val &= ~(DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3);
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW10(ch), val);
+
+	/* FIXME: Program the support xxx V-dB */
+	/* Use 800mV-0dB */
+	for (i = 0; i < 4; i++) {
+		val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW4(ch, i));
+		val &= ~DPIO_SWING_DEEMPH9P5_MASK;
+		val |= 128 << DPIO_SWING_DEEMPH9P5_SHIFT;
+		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW4(ch, i), val);
+	}
+
+	for (i = 0; i < 4; i++) {
+		val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW2(ch, i));
+		val &= ~DPIO_SWING_MARGIN_MASK;
+		val |= 102 << DPIO_SWING_MARGIN_SHIFT;
+		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW2(ch, i), val);
+	}
+
+	/* Disable unique transition scale */
+	for (i = 0; i < 4; i++) {
+		val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW3(ch, i));
+		val &= ~DPIO_TX_UNIQ_TRANS_SCALE_EN;
+		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW3(ch, i), val);
+	}
+
+	/* Additional steps for 1200mV-0dB */
+#if 0
+	val = vlv_dpio_read(dev_priv, pipe, VLV_TX_DW3(ch));
+	if (ch)
+		val |= DPIO_TX_UNIQ_TRANS_SCALE_CH1;
+	else
+		val |= DPIO_TX_UNIQ_TRANS_SCALE_CH0;
+	vlv_dpio_write(dev_priv, pipe, VLV_TX_DW3(ch), val);
+
+	vlv_dpio_write(dev_priv, pipe, VLV_TX_DW2(ch),
+			vlv_dpio_read(dev_priv, pipe, VLV_TX_DW2(ch)) |
+				(0x9a << DPIO_UNIQ_TRANS_SCALE_SHIFT));
+#endif
+	/* Start swing calculation */
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW10(ch));
+	val |= DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW10(ch), val);
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW10(ch));
+	val |= DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW10(ch), val);
+
+	/* LRC Bypass */
+	val = vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW30);
+	val |= DPIO_LRC_BYPASS;
+	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW30, val);
+
+	mutex_unlock(&dev_priv->dpio_lock);
+
+	intel_enable_hdmi(encoder);
+
+	vlv_wait_port_ready(dev_priv, dport);
+}
+
 static void intel_hdmi_destroy(struct drm_connector *connector)
 {
 	drm_connector_cleanup(connector);
@@ -1259,7 +1451,10 @@ void intel_hdmi_init_connector(struct intel_digital_port *intel_dig_port,
 		intel_encoder->hpd_pin = HPD_PORT_C;
 		break;
 	case PORT_D:
-		intel_hdmi->ddc_bus = GMBUS_PORT_DPD;
+		if (IS_CHERRYVIEW(dev))
+			intel_hdmi->ddc_bus = GMBUS_PORT_DPD_CHV;
+		else
+			intel_hdmi->ddc_bus = GMBUS_PORT_DPD;
 		intel_encoder->hpd_pin = HPD_PORT_D;
 		break;
 	case PORT_A:
@@ -1329,21 +1524,32 @@ void intel_hdmi_init(struct drm_device *dev, int hdmi_reg, enum port port)
 			 DRM_MODE_ENCODER_TMDS);
 
 	intel_encoder->compute_config = intel_hdmi_compute_config;
-	intel_encoder->mode_set = intel_hdmi_mode_set;
 	intel_encoder->disable = intel_disable_hdmi;
 	intel_encoder->get_hw_state = intel_hdmi_get_hw_state;
 	intel_encoder->get_config = intel_hdmi_get_config;
-	if (IS_VALLEYVIEW(dev)) {
+	if (IS_CHERRYVIEW(dev)) {
+		intel_encoder->pre_enable = chv_hdmi_pre_enable;
+		intel_encoder->enable = vlv_enable_hdmi;
+		intel_encoder->post_disable = chv_hdmi_post_disable;
+	} else if (IS_VALLEYVIEW(dev)) {
 		intel_encoder->pre_pll_enable = vlv_hdmi_pre_pll_enable;
 		intel_encoder->pre_enable = vlv_hdmi_pre_enable;
 		intel_encoder->enable = vlv_enable_hdmi;
 		intel_encoder->post_disable = vlv_hdmi_post_disable;
 	} else {
+		intel_encoder->pre_enable = intel_hdmi_pre_enable;
 		intel_encoder->enable = intel_enable_hdmi;
 	}
 
 	intel_encoder->type = INTEL_OUTPUT_HDMI;
-	intel_encoder->crtc_mask = (1 << 0) | (1 << 1) | (1 << 2);
+	if (IS_CHERRYVIEW(dev)) {
+		if (port == PORT_D)
+			intel_encoder->crtc_mask = 1 << 2;
+		else
+			intel_encoder->crtc_mask = (1 << 0) | (1 << 1);
+	} else {
+		intel_encoder->crtc_mask = (1 << 0) | (1 << 1) | (1 << 2);
+	}
 	intel_encoder->cloneable = 1 << INTEL_OUTPUT_ANALOG;
 	/*
 	 * BSpec is unclear about HDMI+HDMI cloning on g4x, but it seems
diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index f1ecf916474a..23126023aeba 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -111,13 +111,6 @@ static void intel_lvds_get_config(struct intel_encoder *encoder,
 
 	pipe_config->adjusted_mode.flags |= flags;
 
-	/* gen2/3 store dither state in pfit control, needs to match */
-	if (INTEL_INFO(dev)->gen < 4) {
-		tmp = I915_READ(PFIT_CONTROL);
-
-		pipe_config->gmch_pfit.control |= tmp & PANEL_8TO6_DITHER_ENABLE;
-	}
-
 	dotclock = pipe_config->port_clock;
 
 	if (HAS_PCH_SPLIT(dev_priv->dev))
@@ -126,10 +119,6 @@ static void intel_lvds_get_config(struct intel_encoder *encoder,
 	pipe_config->adjusted_mode.crtc_clock = dotclock;
 }
 
-/* The LVDS pin pair needs to be on before the DPLLs are enabled.
- * This is an exception to the general rule that mode_set doesn't turn
- * things on.
- */
 static void intel_pre_enable_lvds(struct intel_encoder *encoder)
 {
 	struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(&encoder->base);
@@ -331,15 +320,6 @@ static bool intel_lvds_compute_config(struct intel_encoder *intel_encoder,
 	return true;
 }
 
-static void intel_lvds_mode_set(struct intel_encoder *encoder)
-{
-	/*
-	 * We don't do anything here, the LVDS port is fully set up in the pre
-	 * enable hook - the ordering constraints for enabling the lvds port vs.
-	 * enabling the display pll are too strict.
-	 */
-}
-
 /**
  * Detect the LVDS connection.
  *
@@ -354,7 +334,7 @@ intel_lvds_detect(struct drm_connector *connector, bool force)
 	enum drm_connector_status status;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
-		      connector->base.id, drm_get_connector_name(connector));
+		      connector->base.id, connector->name);
 
 	status = intel_panel_detect(dev);
 	if (status != connector_status_unknown)
@@ -953,7 +933,6 @@ void intel_lvds_init(struct drm_device *dev)
 	intel_encoder->enable = intel_enable_lvds;
 	intel_encoder->pre_enable = intel_pre_enable_lvds;
 	intel_encoder->compute_config = intel_lvds_compute_config;
-	intel_encoder->mode_set = intel_lvds_mode_set;
 	intel_encoder->disable = intel_disable_lvds;
 	intel_encoder->get_hw_state = intel_lvds_get_hw_state;
 	intel_encoder->get_config = intel_lvds_get_config;
diff --git a/drivers/gpu/drm/i915/intel_opregion.c b/drivers/gpu/drm/i915/intel_opregion.c
index acde2945eb8a..2e2c71fcc9ed 100644
--- a/drivers/gpu/drm/i915/intel_opregion.c
+++ b/drivers/gpu/drm/i915/intel_opregion.c
@@ -410,7 +410,7 @@ static u32 asle_set_backlight(struct drm_device *dev, u32 bclp)
 	if (bclp > 255)
 		return ASLC_BACKLIGHT_FAILED;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
 
 	/*
 	 * Update backlight on all connectors that support backlight (usually
@@ -421,7 +421,7 @@ static u32 asle_set_backlight(struct drm_device *dev, u32 bclp)
 		intel_panel_set_backlight(intel_connector, bclp, 255);
 	iowrite32(DIV_ROUND_UP(bclp * 100, 255) | ASLE_CBLV_VALID, &asle->cblv);
 
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock(&dev->mode_config.connection_mutex);
 
 
 	return 0;
diff --git a/drivers/gpu/drm/i915/intel_overlay.c b/drivers/gpu/drm/i915/intel_overlay.c
index 129db0c7d835..daa118978eec 100644
--- a/drivers/gpu/drm/i915/intel_overlay.c
+++ b/drivers/gpu/drm/i915/intel_overlay.c
@@ -213,7 +213,7 @@ static int intel_overlay_do_wait_request(struct intel_overlay *overlay,
 {
 	struct drm_device *dev = overlay->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
 	int ret;
 
 	BUG_ON(overlay->last_flip_req);
@@ -236,7 +236,7 @@ static int intel_overlay_on(struct intel_overlay *overlay)
 {
 	struct drm_device *dev = overlay->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
 	int ret;
 
 	BUG_ON(overlay->active);
@@ -263,7 +263,7 @@ static int intel_overlay_continue(struct intel_overlay *overlay,
 {
 	struct drm_device *dev = overlay->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
 	u32 flip_addr = overlay->flip_addr;
 	u32 tmp;
 	int ret;
@@ -320,7 +320,7 @@ static int intel_overlay_off(struct intel_overlay *overlay)
 {
 	struct drm_device *dev = overlay->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
 	u32 flip_addr = overlay->flip_addr;
 	int ret;
 
@@ -363,7 +363,7 @@ static int intel_overlay_recover_from_interrupt(struct intel_overlay *overlay)
 {
 	struct drm_device *dev = overlay->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
 	int ret;
 
 	if (overlay->last_flip_req == 0)
@@ -389,7 +389,7 @@ static int intel_overlay_release_old_vid(struct intel_overlay *overlay)
 {
 	struct drm_device *dev = overlay->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
 	int ret;
 
 	/* Only wait if there is actually an old frame to release to
@@ -688,7 +688,7 @@ static int intel_overlay_do_put_image(struct intel_overlay *overlay,
 	u32 swidth, swidthsw, sheight, ostride;
 
 	BUG_ON(!mutex_is_locked(&dev->struct_mutex));
-	BUG_ON(!mutex_is_locked(&dev->mode_config.mutex));
+	BUG_ON(!drm_modeset_is_locked(&dev->mode_config.connection_mutex));
 	BUG_ON(!overlay);
 
 	ret = intel_overlay_release_old_vid(overlay);
@@ -793,7 +793,7 @@ int intel_overlay_switch_off(struct intel_overlay *overlay)
 	int ret;
 
 	BUG_ON(!mutex_is_locked(&dev->struct_mutex));
-	BUG_ON(!mutex_is_locked(&dev->mode_config.mutex));
+	BUG_ON(!drm_modeset_is_locked(&dev->mode_config.connection_mutex));
 
 	ret = intel_overlay_recover_from_interrupt(overlay);
 	if (ret != 0)
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index cb8cfb7e0974..5e6c888b4928 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -42,6 +42,59 @@ intel_fixed_panel_mode(const struct drm_display_mode *fixed_mode,
 	drm_mode_set_crtcinfo(adjusted_mode, 0);
 }
 
+/**
+ * intel_find_panel_downclock - find the reduced downclock for LVDS in EDID
+ * @dev: drm device
+ * @fixed_mode : panel native mode
+ * @connector: LVDS/eDP connector
+ *
+ * Return downclock_avail
+ * Find the reduced downclock for LVDS/eDP in EDID.
+ */
+struct drm_display_mode *
+intel_find_panel_downclock(struct drm_device *dev,
+			struct drm_display_mode *fixed_mode,
+			struct drm_connector *connector)
+{
+	struct drm_display_mode *scan, *tmp_mode;
+	int temp_downclock;
+
+	temp_downclock = fixed_mode->clock;
+	tmp_mode = NULL;
+
+	list_for_each_entry(scan, &connector->probed_modes, head) {
+		/*
+		 * If one mode has the same resolution with the fixed_panel
+		 * mode while they have the different refresh rate, it means
+		 * that the reduced downclock is found. In such
+		 * case we can set the different FPx0/1 to dynamically select
+		 * between low and high frequency.
+		 */
+		if (scan->hdisplay == fixed_mode->hdisplay &&
+		    scan->hsync_start == fixed_mode->hsync_start &&
+		    scan->hsync_end == fixed_mode->hsync_end &&
+		    scan->htotal == fixed_mode->htotal &&
+		    scan->vdisplay == fixed_mode->vdisplay &&
+		    scan->vsync_start == fixed_mode->vsync_start &&
+		    scan->vsync_end == fixed_mode->vsync_end &&
+		    scan->vtotal == fixed_mode->vtotal) {
+			if (scan->clock < temp_downclock) {
+				/*
+				 * The downclock is already found. But we
+				 * expect to find the lower downclock.
+				 */
+				temp_downclock = scan->clock;
+				tmp_mode = scan;
+			}
+		}
+	}
+
+	if (temp_downclock < fixed_mode->clock)
+		return drm_mode_duplicate(dev, tmp_mode);
+	else
+		return NULL;
+}
+
 /* adjusted_mode has been preset to be the panel's fixed mode */
 void
 intel_pch_panel_fitting(struct intel_crtc *intel_crtc,
@@ -308,21 +361,43 @@ void intel_gmch_panel_fitting(struct intel_crtc *intel_crtc,
 		pfit_control |= ((intel_crtc->pipe << PFIT_PIPE_SHIFT) |
 				 PFIT_FILTER_FUZZY);
 
+	/* Make sure pre-965 set dither correctly for 18bpp panels. */
+	if (INTEL_INFO(dev)->gen < 4 && pipe_config->pipe_bpp == 18)
+		pfit_control |= PANEL_8TO6_DITHER_ENABLE;
+
 out:
 	if ((pfit_control & PFIT_ENABLE) == 0) {
 		pfit_control = 0;
 		pfit_pgm_ratios = 0;
 	}
 
-	/* Make sure pre-965 set dither correctly for 18bpp panels. */
-	if (INTEL_INFO(dev)->gen < 4 && pipe_config->pipe_bpp == 18)
-		pfit_control |= PANEL_8TO6_DITHER_ENABLE;
-
 	pipe_config->gmch_pfit.control = pfit_control;
 	pipe_config->gmch_pfit.pgm_ratios = pfit_pgm_ratios;
 	pipe_config->gmch_pfit.lvds_border_bits = border;
 }
 
+enum drm_connector_status
+intel_panel_detect(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	/* Assume that the BIOS does not lie through the OpRegion... */
+	if (!i915.panel_ignore_lid && dev_priv->opregion.lid_state) {
+		return ioread32(dev_priv->opregion.lid_state) & 0x1 ?
+			connector_status_connected :
+			connector_status_disconnected;
+	}
+
+	switch (i915.panel_ignore_lid) {
+	case -2:
+		return connector_status_connected;
+	case -1:
+		return connector_status_disconnected;
+	default:
+		return connector_status_unknown;
+	}
+}
+
 static u32 intel_panel_compute_brightness(struct intel_connector *connector,
 					  u32 val)
 {
@@ -795,40 +870,18 @@ void intel_panel_enable_backlight(struct intel_connector *connector)
 	spin_unlock_irqrestore(&dev_priv->backlight_lock, flags);
 }
 
-enum drm_connector_status
-intel_panel_detect(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-
-	/* Assume that the BIOS does not lie through the OpRegion... */
-	if (!i915.panel_ignore_lid && dev_priv->opregion.lid_state) {
-		return ioread32(dev_priv->opregion.lid_state) & 0x1 ?
-			connector_status_connected :
-			connector_status_disconnected;
-	}
-
-	switch (i915.panel_ignore_lid) {
-	case -2:
-		return connector_status_connected;
-	case -1:
-		return connector_status_disconnected;
-	default:
-		return connector_status_unknown;
-	}
-}
-
 #if IS_ENABLED(CONFIG_BACKLIGHT_CLASS_DEVICE)
 static int intel_backlight_device_update_status(struct backlight_device *bd)
 {
 	struct intel_connector *connector = bl_get_data(bd);
 	struct drm_device *dev = connector->base.dev;
 
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
 	DRM_DEBUG_KMS("updating intel_backlight, brightness=%d/%d\n",
 		      bd->props.brightness, bd->props.max_brightness);
 	intel_panel_set_backlight(connector, bd->props.brightness,
 				  bd->props.max_brightness);
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock(&dev->mode_config.connection_mutex);
 	return 0;
 }
 
@@ -840,9 +893,9 @@ static int intel_backlight_device_get_brightness(struct backlight_device *bd)
 	int ret;
 
 	intel_runtime_pm_get(dev_priv);
-	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
 	ret = intel_panel_get_backlight(connector);
-	mutex_unlock(&dev->mode_config.mutex);
+	drm_modeset_unlock(&dev->mode_config.connection_mutex);
 	intel_runtime_pm_put(dev_priv);
 
 	return ret;
@@ -1077,7 +1130,7 @@ int intel_panel_setup_backlight(struct drm_connector *connector)
 
 	if (ret) {
 		DRM_DEBUG_KMS("failed to setup backlight for connector %s\n",
-			      drm_get_connector_name(connector));
+			      connector->name);
 		return ret;
 	}
 
@@ -1103,59 +1156,6 @@ void intel_panel_destroy_backlight(struct drm_connector *connector)
 	intel_backlight_device_unregister(intel_connector);
 }
 
-/**
- * intel_find_panel_downclock - find the reduced downclock for LVDS in EDID
- * @dev: drm device
- * @fixed_mode : panel native mode
- * @connector: LVDS/eDP connector
- *
- * Return downclock_avail
- * Find the reduced downclock for LVDS/eDP in EDID.
- */
-struct drm_display_mode *
-intel_find_panel_downclock(struct drm_device *dev,
-			struct drm_display_mode *fixed_mode,
-			struct drm_connector *connector)
-{
-	struct drm_display_mode *scan, *tmp_mode;
-	int temp_downclock;
-
-	temp_downclock = fixed_mode->clock;
-	tmp_mode = NULL;
-
-	list_for_each_entry(scan, &connector->probed_modes, head) {
-		/*
-		 * If one mode has the same resolution with the fixed_panel
-		 * mode while they have the different refresh rate, it means
-		 * that the reduced downclock is found. In such
-		 * case we can set the different FPx0/1 to dynamically select
-		 * between low and high frequency.
-		 */
-		if (scan->hdisplay == fixed_mode->hdisplay &&
-		    scan->hsync_start == fixed_mode->hsync_start &&
-		    scan->hsync_end == fixed_mode->hsync_end &&
-		    scan->htotal == fixed_mode->htotal &&
-		    scan->vdisplay == fixed_mode->vdisplay &&
-		    scan->vsync_start == fixed_mode->vsync_start &&
-		    scan->vsync_end == fixed_mode->vsync_end &&
-		    scan->vtotal == fixed_mode->vtotal) {
-			if (scan->clock < temp_downclock) {
-				/*
-				 * The downclock is already found. But we
-				 * expect to find the lower downclock.
-				 */
-				temp_downclock = scan->clock;
-				tmp_mode = scan;
-			}
-		}
-	}
-
-	if (temp_downclock < fixed_mode->clock)
-		return drm_mode_duplicate(dev, tmp_mode);
-	else
-		return NULL;
-}
-
 /* Set up chip specific backlight functions */
 void intel_panel_init_backlight_funcs(struct drm_device *dev)
 {
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index d93dcf683e8c..d1e53abec1b5 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -487,7 +487,7 @@ void intel_update_fbc(struct drm_device *dev)
 	 *   - new fb is too large to fit in compressed buffer
 	 *   - going to an unsupported config (interlace, pixel multiply, etc.)
 	 */
-	list_for_each_entry(tmp_crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, tmp_crtc) {
 		if (intel_crtc_active(tmp_crtc) &&
 		    to_intel_crtc(tmp_crtc)->primary_enabled) {
 			if (crtc) {
@@ -1010,7 +1010,7 @@ static struct drm_crtc *single_enabled_crtc(struct drm_device *dev)
 {
 	struct drm_crtc *crtc, *enabled = NULL;
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+	for_each_crtc(dev, crtc) {
 		if (intel_crtc_active(crtc)) {
 			if (enabled)
 				return NULL;
@@ -1831,6 +1831,40 @@ static unsigned int ilk_display_fifo_size(const struct drm_device *dev)
 		return 512;
 }
 
+static unsigned int ilk_plane_wm_reg_max(const struct drm_device *dev,
+					 int level, bool is_sprite)
+{
+	if (INTEL_INFO(dev)->gen >= 8)
+		/* BDW primary/sprite plane watermarks */
+		return level == 0 ? 255 : 2047;
+	else if (INTEL_INFO(dev)->gen >= 7)
+		/* IVB/HSW primary/sprite plane watermarks */
+		return level == 0 ? 127 : 1023;
+	else if (!is_sprite)
+		/* ILK/SNB primary plane watermarks */
+		return level == 0 ? 127 : 511;
+	else
+		/* ILK/SNB sprite plane watermarks */
+		return level == 0 ? 63 : 255;
+}
+
+static unsigned int ilk_cursor_wm_reg_max(const struct drm_device *dev,
+					  int level)
+{
+	if (INTEL_INFO(dev)->gen >= 7)
+		return level == 0 ? 63 : 255;
+	else
+		return level == 0 ? 31 : 63;
+}
+
+static unsigned int ilk_fbc_wm_reg_max(const struct drm_device *dev)
+{
+	if (INTEL_INFO(dev)->gen >= 8)
+		return 31;
+	else
+		return 15;
+}
+
 /* Calculate the maximum primary/sprite plane watermark */
 static unsigned int ilk_plane_wm_max(const struct drm_device *dev,
 				     int level,
@@ -1839,7 +1873,6 @@ static unsigned int ilk_plane_wm_max(const struct drm_device *dev,
 				     bool is_sprite)
 {
 	unsigned int fifo_size = ilk_display_fifo_size(dev);
-	unsigned int max;
 
 	/* if sprites aren't enabled, sprites get nothing */
 	if (is_sprite && !config->sprites_enabled)
@@ -1870,19 +1903,7 @@ static unsigned int ilk_plane_wm_max(const struct drm_device *dev,
 	}
 
 	/* clamp to max that the registers can hold */
-	if (INTEL_INFO(dev)->gen >= 8)
-		max = level == 0 ? 255 : 2047;
-	else if (INTEL_INFO(dev)->gen >= 7)
-		/* IVB/HSW primary/sprite plane watermarks */
-		max = level == 0 ? 127 : 1023;
-	else if (!is_sprite)
-		/* ILK/SNB primary plane watermarks */
-		max = level == 0 ? 127 : 511;
-	else
-		/* ILK/SNB sprite plane watermarks */
-		max = level == 0 ? 63 : 255;
-
-	return min(fifo_size, max);
+	return min(fifo_size, ilk_plane_wm_reg_max(dev, level, is_sprite));
 }
 
 /* Calculate the maximum cursor plane watermark */
@@ -1895,20 +1916,7 @@ static unsigned int ilk_cursor_wm_max(const struct drm_device *dev,
 		return 64;
 
 	/* otherwise just report max that registers can hold */
-	if (INTEL_INFO(dev)->gen >= 7)
-		return level == 0 ? 63 : 255;
-	else
-		return level == 0 ? 31 : 63;
-}
-
-/* Calculate the maximum FBC watermark */
-static unsigned int ilk_fbc_wm_max(const struct drm_device *dev)
-{
-	/* max that registers can hold */
-	if (INTEL_INFO(dev)->gen >= 8)
-		return 31;
-	else
-		return 15;
+	return ilk_cursor_wm_reg_max(dev, level);
 }
 
 static void ilk_compute_wm_maximums(const struct drm_device *dev,
@@ -1920,7 +1928,17 @@ static void ilk_compute_wm_maximums(const struct drm_device *dev,
 	max->pri = ilk_plane_wm_max(dev, level, config, ddb_partitioning, false);
 	max->spr = ilk_plane_wm_max(dev, level, config, ddb_partitioning, true);
 	max->cur = ilk_cursor_wm_max(dev, level, config);
-	max->fbc = ilk_fbc_wm_max(dev);
+	max->fbc = ilk_fbc_wm_reg_max(dev);
+}
+
+static void ilk_compute_wm_reg_maximums(struct drm_device *dev,
+					int level,
+					struct ilk_wm_maximums *max)
+{
+	max->pri = ilk_plane_wm_reg_max(dev, level, false);
+	max->spr = ilk_plane_wm_reg_max(dev, level, true);
+	max->cur = ilk_cursor_wm_reg_max(dev, level);
+	max->fbc = ilk_fbc_wm_reg_max(dev);
 }
 
 static bool ilk_validate_wm_level(int level,
@@ -2059,7 +2077,7 @@ static void intel_fixup_cur_wm_latency(struct drm_device *dev, uint16_t wm[5])
 		wm[3] *= 2;
 }
 
-static int ilk_wm_max_level(const struct drm_device *dev)
+int ilk_wm_max_level(const struct drm_device *dev)
 {
 	/* how many WM levels are we expecting */
 	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
@@ -2155,38 +2173,52 @@ static void ilk_setup_wm_latency(struct drm_device *dev)
 }
 
 static void ilk_compute_wm_parameters(struct drm_crtc *crtc,
-				      struct ilk_pipe_wm_parameters *p,
-				      struct intel_wm_config *config)
+				      struct ilk_pipe_wm_parameters *p)
 {
 	struct drm_device *dev = crtc->dev;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	enum pipe pipe = intel_crtc->pipe;
 	struct drm_plane *plane;
 
-	p->active = intel_crtc_active(crtc);
-	if (p->active) {
-		p->pipe_htotal = intel_crtc->config.adjusted_mode.crtc_htotal;
-		p->pixel_rate = ilk_pipe_pixel_rate(dev, crtc);
-		p->pri.bytes_per_pixel = crtc->primary->fb->bits_per_pixel / 8;
-		p->cur.bytes_per_pixel = 4;
-		p->pri.horiz_pixels = intel_crtc->config.pipe_src_w;
-		p->cur.horiz_pixels = intel_crtc->cursor_width;
-		/* TODO: for now, assume primary and cursor planes are always enabled. */
-		p->pri.enabled = true;
-		p->cur.enabled = true;
-	}
+	if (!intel_crtc_active(crtc))
+		return;
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
-		config->num_pipes_active += intel_crtc_active(crtc);
+	p->active = true;
+	p->pipe_htotal = intel_crtc->config.adjusted_mode.crtc_htotal;
+	p->pixel_rate = ilk_pipe_pixel_rate(dev, crtc);
+	p->pri.bytes_per_pixel = crtc->primary->fb->bits_per_pixel / 8;
+	p->cur.bytes_per_pixel = 4;
+	p->pri.horiz_pixels = intel_crtc->config.pipe_src_w;
+	p->cur.horiz_pixels = intel_crtc->cursor_width;
+	/* TODO: for now, assume primary and cursor planes are always enabled. */
+	p->pri.enabled = true;
+	p->cur.enabled = true;
 
 	drm_for_each_legacy_plane(plane, &dev->mode_config.plane_list) {
 		struct intel_plane *intel_plane = to_intel_plane(plane);
 
-		if (intel_plane->pipe == pipe)
+		if (intel_plane->pipe == pipe) {
 			p->spr = intel_plane->wm;
+			break;
+		}
+	}
+}
+
+static void ilk_compute_wm_config(struct drm_device *dev,
+				  struct intel_wm_config *config)
+{
+	struct intel_crtc *intel_crtc;
 
-		config->sprites_enabled |= intel_plane->wm.enabled;
-		config->sprites_scaled |= intel_plane->wm.scaled;
+	/* Compute the currently _active_ config */
+	for_each_intel_crtc(dev, intel_crtc) {
+		const struct intel_pipe_wm *wm = &intel_crtc->wm.active;
+
+		if (!wm->pipe_enabled)
+			continue;
+
+		config->sprites_enabled |= wm->sprites_enabled;
+		config->sprites_scaled |= wm->sprites_scaled;
+		config->num_pipes_active++;
 	}
 }
 
@@ -2206,8 +2238,9 @@ static bool intel_compute_pipe_wm(struct drm_crtc *crtc,
 	};
 	struct ilk_wm_maximums max;
 
-	/* LP0 watermarks always use 1/2 DDB partitioning */
-	ilk_compute_wm_maximums(dev, 0, &config, INTEL_DDB_PART_1_2, &max);
+	pipe_wm->pipe_enabled = params->active;
+	pipe_wm->sprites_enabled = params->spr.enabled;
+	pipe_wm->sprites_scaled = params->spr.scaled;
 
 	/* ILK/SNB: LP2+ watermarks only w/o sprites */
 	if (INTEL_INFO(dev)->gen <= 6 && params->spr.enabled)
@@ -2217,15 +2250,37 @@ static bool intel_compute_pipe_wm(struct drm_crtc *crtc,
 	if (params->spr.scaled)
 		max_level = 0;
 
-	for (level = 0; level <= max_level; level++)
-		ilk_compute_wm_level(dev_priv, level, params,
-				     &pipe_wm->wm[level]);
+	ilk_compute_wm_level(dev_priv, 0, params, &pipe_wm->wm[0]);
 
 	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 		pipe_wm->linetime = hsw_compute_linetime_wm(dev, crtc);
 
+	/* LP0 watermarks always use 1/2 DDB partitioning */
+	ilk_compute_wm_maximums(dev, 0, &config, INTEL_DDB_PART_1_2, &max);
+
 	/* At least LP0 must be valid */
-	return ilk_validate_wm_level(0, &max, &pipe_wm->wm[0]);
+	if (!ilk_validate_wm_level(0, &max, &pipe_wm->wm[0]))
+		return false;
+
+	ilk_compute_wm_reg_maximums(dev, 1, &max);
+
+	for (level = 1; level <= max_level; level++) {
+		struct intel_wm_level wm = {};
+
+		ilk_compute_wm_level(dev_priv, level, params, &wm);
+
+		/*
+		 * Disable any watermark level that exceeds the
+		 * register maximums since such watermarks are
+		 * always invalid.
+		 */
+		if (!ilk_validate_wm_level(level, &max, &wm))
+			break;
+
+		pipe_wm->wm[level] = wm;
+	}
+
+	return true;
 }
 
 /*
@@ -2237,20 +2292,28 @@ static void ilk_merge_wm_level(struct drm_device *dev,
 {
 	const struct intel_crtc *intel_crtc;
 
-	list_for_each_entry(intel_crtc, &dev->mode_config.crtc_list, base.head) {
-		const struct intel_wm_level *wm =
-			&intel_crtc->wm.active.wm[level];
+	ret_wm->enable = true;
+
+	for_each_intel_crtc(dev, intel_crtc) {
+		const struct intel_pipe_wm *active = &intel_crtc->wm.active;
+		const struct intel_wm_level *wm = &active->wm[level];
+
+		if (!active->pipe_enabled)
+			continue;
 
+		/*
+		 * The watermark values may have been used in the past,
+		 * so we must maintain them in the registers for some
+		 * time even if the level is now disabled.
+		 */
 		if (!wm->enable)
-			return;
+			ret_wm->enable = false;
 
 		ret_wm->pri_val = max(ret_wm->pri_val, wm->pri_val);
 		ret_wm->spr_val = max(ret_wm->spr_val, wm->spr_val);
 		ret_wm->cur_val = max(ret_wm->cur_val, wm->cur_val);
 		ret_wm->fbc_val = max(ret_wm->fbc_val, wm->fbc_val);
 	}
-
-	ret_wm->enable = true;
 }
 
 /*
@@ -2262,6 +2325,7 @@ static void ilk_wm_merge(struct drm_device *dev,
 			 struct intel_pipe_wm *merged)
 {
 	int level, max_level = ilk_wm_max_level(dev);
+	int last_enabled_level = max_level;
 
 	/* ILK/SNB/IVB: LP1+ watermarks only w/ single pipe */
 	if ((INTEL_INFO(dev)->gen <= 6 || IS_IVYBRIDGE(dev)) &&
@@ -2277,15 +2341,19 @@ static void ilk_wm_merge(struct drm_device *dev,
 
 		ilk_merge_wm_level(dev, level, wm);
 
-		if (!ilk_validate_wm_level(level, max, wm))
-			break;
+		if (level > last_enabled_level)
+			wm->enable = false;
+		else if (!ilk_validate_wm_level(level, max, wm))
+			/* make sure all following levels get disabled */
+			last_enabled_level = level - 1;
 
 		/*
 		 * The spec says it is preferred to disable
 		 * FBC WMs instead of disabling a WM level.
 		 */
 		if (wm->fbc_val > max->fbc) {
-			merged->fbc_wm_enabled = false;
+			if (wm->enable)
+				merged->fbc_wm_enabled = false;
 			wm->fbc_val = 0;
 		}
 	}
@@ -2340,14 +2408,19 @@ static void ilk_compute_wm_results(struct drm_device *dev,
 		level = ilk_wm_lp_to_level(wm_lp, merged);
 
 		r = &merged->wm[level];
-		if (!r->enable)
-			break;
 
-		results->wm_lp[wm_lp - 1] = WM3_LP_EN |
+		/*
+		 * Maintain the watermark values even if the level is
+		 * disabled. Doing otherwise could cause underruns.
+		 */
+		results->wm_lp[wm_lp - 1] =
 			(ilk_wm_lp_latency(dev, level) << WM1_LP_LATENCY_SHIFT) |
 			(r->pri_val << WM1_LP_SR_SHIFT) |
 			r->cur_val;
 
+		if (r->enable)
+			results->wm_lp[wm_lp - 1] |= WM1_LP_SR_EN;
+
 		if (INTEL_INFO(dev)->gen >= 8)
 			results->wm_lp[wm_lp - 1] |=
 				r->fbc_val << WM1_LP_FBC_SHIFT_BDW;
@@ -2355,6 +2428,10 @@ static void ilk_compute_wm_results(struct drm_device *dev,
 			results->wm_lp[wm_lp - 1] |=
 				r->fbc_val << WM1_LP_FBC_SHIFT;
 
+		/*
+		 * Always set WM1S_LP_EN when spr_val != 0, even if the
+		 * level is disabled. Doing otherwise could cause underruns.
+		 */
 		if (INTEL_INFO(dev)->gen <= 6 && r->spr_val) {
 			WARN_ON(wm_lp != 1);
 			results->wm_lp_spr[wm_lp - 1] = WM1S_LP_EN | r->spr_val;
@@ -2363,7 +2440,7 @@ static void ilk_compute_wm_results(struct drm_device *dev,
 	}
 
 	/* LP0 register values */
-	list_for_each_entry(intel_crtc, &dev->mode_config.crtc_list, base.head) {
+	for_each_intel_crtc(dev, intel_crtc) {
 		enum pipe pipe = intel_crtc->pipe;
 		const struct intel_wm_level *r =
 			&intel_crtc->wm.active.wm[0];
@@ -2598,7 +2675,7 @@ static void ilk_update_wm(struct drm_crtc *crtc)
 	struct intel_pipe_wm lp_wm_1_2 = {}, lp_wm_5_6 = {}, *best_lp_wm;
 	struct intel_wm_config config = {};
 
-	ilk_compute_wm_parameters(crtc, &params, &config);
+	ilk_compute_wm_parameters(crtc, &params);
 
 	intel_compute_pipe_wm(crtc, &params, &pipe_wm);
 
@@ -2607,6 +2684,8 @@ static void ilk_update_wm(struct drm_crtc *crtc)
 
 	intel_crtc->wm.active = pipe_wm;
 
+	ilk_compute_wm_config(dev, &config);
+
 	ilk_compute_wm_maximums(dev, 1, &config, INTEL_DDB_PART_1_2, &max);
 	ilk_wm_merge(dev, &config, &max, &lp_wm_1_2);
 
@@ -2673,7 +2752,9 @@ static void ilk_pipe_wm_get_hw_state(struct drm_crtc *crtc)
 	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 		hw->wm_linetime[pipe] = I915_READ(PIPE_WM_LINETIME(pipe));
 
-	if (intel_crtc_active(crtc)) {
+	active->pipe_enabled = intel_crtc_active(crtc);
+
+	if (active->pipe_enabled) {
 		u32 tmp = hw->wm_pipe[pipe];
 
 		/*
@@ -2706,7 +2787,7 @@ void ilk_wm_get_hw_state(struct drm_device *dev)
 	struct ilk_wm_values *hw = &dev_priv->wm.hw;
 	struct drm_crtc *crtc;
 
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
+	for_each_crtc(dev, crtc)
 		ilk_pipe_wm_get_hw_state(crtc);
 
 	hw->wm_lp[0] = I915_READ(WM1_LP_ILK);
@@ -2714,8 +2795,10 @@ void ilk_wm_get_hw_state(struct drm_device *dev)
 	hw->wm_lp[2] = I915_READ(WM3_LP_ILK);
 
 	hw->wm_lp_spr[0] = I915_READ(WM1S_LP_ILK);
-	hw->wm_lp_spr[1] = I915_READ(WM2S_LP_IVB);
-	hw->wm_lp_spr[2] = I915_READ(WM3S_LP_IVB);
+	if (INTEL_INFO(dev)->gen >= 7) {
+		hw->wm_lp_spr[1] = I915_READ(WM2S_LP_IVB);
+		hw->wm_lp_spr[2] = I915_READ(WM3S_LP_IVB);
+	}
 
 	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 		hw->partitioning = (I915_READ(WM_MISC) & WM_MISC_DATA_PARTITION_5_6) ?
@@ -3071,6 +3154,9 @@ static u32 gen6_rps_pm_mask(struct drm_i915_private *dev_priv, u8 val)
 	if (INTEL_INFO(dev_priv->dev)->gen <= 7 && !IS_HASWELL(dev_priv->dev))
 		mask |= GEN6_PM_RP_UP_EI_EXPIRED;
 
+	if (IS_GEN8(dev_priv->dev))
+		mask |= GEN8_PMINTR_REDIRECT_TO_NON_DISP;
+
 	return ~mask;
 }
 
@@ -3091,7 +3177,7 @@ void gen6_set_rps(struct drm_device *dev, u8 val)
 	if (val != dev_priv->rps.cur_freq) {
 		gen6_set_rps_thresholds(dev_priv, val);
 
-		if (IS_HASWELL(dev))
+		if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 			I915_WRITE(GEN6_RPNSWREQ,
 				   HSW_FREQUENCY(val));
 		else
@@ -3134,16 +3220,7 @@ static void vlv_set_rps_idle(struct drm_i915_private *dev_priv)
 	/* Mask turbo interrupt so that they will not come in between */
 	I915_WRITE(GEN6_PMINTRMSK, 0xffffffff);
 
-	/* Bring up the Gfx clock */
-	I915_WRITE(VLV_GTLC_SURVIVABILITY_REG,
-		I915_READ(VLV_GTLC_SURVIVABILITY_REG) |
-				VLV_GFX_CLK_FORCE_ON_BIT);
-
-	if (wait_for(((VLV_GFX_CLK_STATUS_BIT &
-		I915_READ(VLV_GTLC_SURVIVABILITY_REG)) != 0), 5)) {
-			DRM_ERROR("GFX_CLK_ON request timed out\n");
-		return;
-	}
+	vlv_force_gfx_clock(dev_priv, true);
 
 	dev_priv->rps.cur_freq = dev_priv->rps.min_freq_softlimit;
 
@@ -3154,10 +3231,7 @@ static void vlv_set_rps_idle(struct drm_i915_private *dev_priv)
 				& GENFREQSTATUS) == 0, 5))
 		DRM_ERROR("timed out waiting for Punit\n");
 
-	/* Release the Gfx clock */
-	I915_WRITE(VLV_GTLC_SURVIVABILITY_REG,
-		I915_READ(VLV_GTLC_SURVIVABILITY_REG) &
-				~VLV_GFX_CLK_FORCE_ON_BIT);
+	vlv_force_gfx_clock(dev_priv, false);
 
 	I915_WRITE(GEN6_PMINTRMSK,
 		   gen6_rps_pm_mask(dev_priv, dev_priv->rps.cur_freq));
@@ -3215,6 +3289,26 @@ void valleyview_set_rps(struct drm_device *dev, u8 val)
 	trace_intel_gpu_freq_change(vlv_gpu_freq(dev_priv, val));
 }
 
+static void gen8_disable_rps_interrupts(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	I915_WRITE(GEN6_PMINTRMSK, ~GEN8_PMINTR_REDIRECT_TO_NON_DISP);
+	I915_WRITE(GEN8_GT_IER(2), I915_READ(GEN8_GT_IER(2)) &
+				   ~dev_priv->pm_rps_events);
+	/* Complete PM interrupt masking here doesn't race with the rps work
+	 * item again unmasking PM interrupts because that is using a different
+	 * register (GEN8_GT_IMR(2)) to mask PM interrupts. The only risk is in
+	 * leaving stale bits in GEN8_GT_IIR(2) and GEN8_GT_IMR(2) which
+	 * gen8_enable_rps will clean up. */
+
+	spin_lock_irq(&dev_priv->irq_lock);
+	dev_priv->rps.pm_iir = 0;
+	spin_unlock_irq(&dev_priv->irq_lock);
+
+	I915_WRITE(GEN8_GT_IIR(2), dev_priv->pm_rps_events);
+}
+
 static void gen6_disable_rps_interrupts(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -3241,7 +3335,10 @@ static void gen6_disable_rps(struct drm_device *dev)
 	I915_WRITE(GEN6_RC_CONTROL, 0);
 	I915_WRITE(GEN6_RPNSWREQ, 1 << 31);
 
-	gen6_disable_rps_interrupts(dev);
+	if (IS_BROADWELL(dev))
+		gen8_disable_rps_interrupts(dev);
+	else
+		gen6_disable_rps_interrupts(dev);
 }
 
 static void valleyview_disable_rps(struct drm_device *dev)
@@ -3255,21 +3352,44 @@ static void valleyview_disable_rps(struct drm_device *dev)
 
 static void intel_print_rc6_info(struct drm_device *dev, u32 mode)
 {
+	if (IS_VALLEYVIEW(dev)) {
+		if (mode & (GEN7_RC_CTL_TO_MODE | GEN6_RC_CTL_EI_MODE(1)))
+			mode = GEN6_RC_CTL_RC6_ENABLE;
+		else
+			mode = 0;
+	}
 	DRM_INFO("Enabling RC6 states: RC6 %s, RC6p %s, RC6pp %s\n",
 		 (mode & GEN6_RC_CTL_RC6_ENABLE) ? "on" : "off",
 		 (mode & GEN6_RC_CTL_RC6p_ENABLE) ? "on" : "off",
 		 (mode & GEN6_RC_CTL_RC6pp_ENABLE) ? "on" : "off");
 }
 
-int intel_enable_rc6(const struct drm_device *dev)
+static int sanitize_rc6_option(const struct drm_device *dev, int enable_rc6)
 {
 	/* No RC6 before Ironlake */
 	if (INTEL_INFO(dev)->gen < 5)
 		return 0;
 
+	/* RC6 is only on Ironlake mobile not on desktop */
+	if (INTEL_INFO(dev)->gen == 5 && !IS_IRONLAKE_M(dev))
+		return 0;
+
 	/* Respect the kernel parameter if it is set */
-	if (i915.enable_rc6 >= 0)
-		return i915.enable_rc6;
+	if (enable_rc6 >= 0) {
+		int mask;
+
+		if (INTEL_INFO(dev)->gen == 6 || IS_IVYBRIDGE(dev))
+			mask = INTEL_RC6_ENABLE | INTEL_RC6p_ENABLE |
+			       INTEL_RC6pp_ENABLE;
+		else
+			mask = INTEL_RC6_ENABLE;
+
+		if ((enable_rc6 & mask) != enable_rc6)
+			DRM_INFO("Adjusting RC6 mask to %d (requested %d, valid %d)\n",
+				 enable_rc6 & mask, enable_rc6, mask);
+
+		return enable_rc6 & mask;
+	}
 
 	/* Disable RC6 on Ironlake */
 	if (INTEL_INFO(dev)->gen == 5)
@@ -3281,6 +3401,22 @@ int intel_enable_rc6(const struct drm_device *dev)
 	return INTEL_RC6_ENABLE;
 }
 
+int intel_enable_rc6(const struct drm_device *dev)
+{
+	return i915.enable_rc6;
+}
+
+static void gen8_enable_rps_interrupts(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	spin_lock_irq(&dev_priv->irq_lock);
+	WARN_ON(dev_priv->rps.pm_iir);
+	bdw_enable_pm_irq(dev_priv, dev_priv->pm_rps_events);
+	I915_WRITE(GEN8_GT_IIR(2), dev_priv->pm_rps_events);
+	spin_unlock_irq(&dev_priv->irq_lock);
+}
+
 static void gen6_enable_rps_interrupts(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -3292,10 +3428,31 @@ static void gen6_enable_rps_interrupts(struct drm_device *dev)
 	spin_unlock_irq(&dev_priv->irq_lock);
 }
 
+static void parse_rp_state_cap(struct drm_i915_private *dev_priv, u32 rp_state_cap)
+{
+	/* All of these values are in units of 50MHz */
+	dev_priv->rps.cur_freq		= 0;
+	/* static values from HW: RP0 < RPe < RP1 < RPn (min_freq) */
+	dev_priv->rps.rp1_freq		= (rp_state_cap >>  8) & 0xff;
+	dev_priv->rps.rp0_freq		= (rp_state_cap >>  0) & 0xff;
+	dev_priv->rps.min_freq		= (rp_state_cap >> 16) & 0xff;
+	/* XXX: only BYT has a special efficient freq */
+	dev_priv->rps.efficient_freq	= dev_priv->rps.rp1_freq;
+	/* hw_max = RP0 until we check for overclocking */
+	dev_priv->rps.max_freq		= dev_priv->rps.rp0_freq;
+
+	/* Preserve min/max settings in case of re-init */
+	if (dev_priv->rps.max_freq_softlimit == 0)
+		dev_priv->rps.max_freq_softlimit = dev_priv->rps.max_freq;
+
+	if (dev_priv->rps.min_freq_softlimit == 0)
+		dev_priv->rps.min_freq_softlimit = dev_priv->rps.min_freq;
+}
+
 static void gen8_enable_rps(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	uint32_t rc6_mask = 0, rp_state_cap;
 	int unused;
 
@@ -3310,6 +3467,7 @@ static void gen8_enable_rps(struct drm_device *dev)
 	I915_WRITE(GEN6_RC_CONTROL, 0);
 
 	rp_state_cap = I915_READ(GEN6_RP_STATE_CAP);
+	parse_rp_state_cap(dev_priv, rp_state_cap);
 
 	/* 2b: Program RC6 thresholds.*/
 	I915_WRITE(GEN6_RC6_WAKE_RATE_LIMIT, 40 << 16);
@@ -3329,8 +3487,10 @@ static void gen8_enable_rps(struct drm_device *dev)
 				    rc6_mask);
 
 	/* 4 Program defaults and thresholds for RPS*/
-	I915_WRITE(GEN6_RPNSWREQ, HSW_FREQUENCY(10)); /* Request 500 MHz */
-	I915_WRITE(GEN6_RC_VIDEO_FREQ, HSW_FREQUENCY(12)); /* Request 600 MHz */
+	I915_WRITE(GEN6_RPNSWREQ,
+		   HSW_FREQUENCY(dev_priv->rps.rp1_freq));
+	I915_WRITE(GEN6_RC_VIDEO_FREQ,
+		   HSW_FREQUENCY(dev_priv->rps.rp1_freq));
 	/* NB: Docs say 1s, and 1000000 - which aren't equivalent */
 	I915_WRITE(GEN6_RP_DOWN_TIMEOUT, 100000000 / 128); /* 1 second timeout */
 
@@ -3346,11 +3506,15 @@ static void gen8_enable_rps(struct drm_device *dev)
 
 	I915_WRITE(GEN6_RP_IDLE_HYSTERSIS, 10);
 
+	/* WaDisablePwrmtrEvent:chv (pre-production hw) */
+	I915_WRITE(0xA80C, I915_READ(0xA80C) & 0x00ffffff);
+	I915_WRITE(0xA810, I915_READ(0xA810) & 0xffffff00);
+
 	/* 5: Enable RPS */
 	I915_WRITE(GEN6_RP_CONTROL,
 		   GEN6_RP_MEDIA_TURBO |
 		   GEN6_RP_MEDIA_HW_NORMAL_MODE |
-		   GEN6_RP_MEDIA_IS_GFX |
+		   GEN6_RP_MEDIA_IS_GFX | /* WaSetMaskForGfxBusyness:chv (pre-production hw ?) */
 		   GEN6_RP_ENABLE |
 		   GEN6_RP_UP_BUSY_AVG |
 		   GEN6_RP_DOWN_IDLE_AVG);
@@ -3359,7 +3523,7 @@ static void gen8_enable_rps(struct drm_device *dev)
 
 	gen6_set_rps(dev, (I915_READ(GEN6_GT_PERF_STATUS) & 0xff00) >> 8);
 
-	gen6_enable_rps_interrupts(dev);
+	gen8_enable_rps_interrupts(dev);
 
 	gen6_gt_force_wake_put(dev_priv, FORCEWAKE_ALL);
 }
@@ -3367,7 +3531,7 @@ static void gen8_enable_rps(struct drm_device *dev)
 static void gen6_enable_rps(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	u32 rp_state_cap;
 	u32 gt_perf_status;
 	u32 rc6vids, pcu_mbox = 0, rc6_mask = 0;
@@ -3396,23 +3560,7 @@ static void gen6_enable_rps(struct drm_device *dev)
 	rp_state_cap = I915_READ(GEN6_RP_STATE_CAP);
 	gt_perf_status = I915_READ(GEN6_GT_PERF_STATUS);
 
-	/* All of these values are in units of 50MHz */
-	dev_priv->rps.cur_freq		= 0;
-	/* static values from HW: RP0 < RPe < RP1 < RPn (min_freq) */
-	dev_priv->rps.rp1_freq		= (rp_state_cap >>  8) & 0xff;
-	dev_priv->rps.rp0_freq		= (rp_state_cap >>  0) & 0xff;
-	dev_priv->rps.min_freq		= (rp_state_cap >> 16) & 0xff;
-	/* XXX: only BYT has a special efficient freq */
-	dev_priv->rps.efficient_freq	= dev_priv->rps.rp1_freq;
-	/* hw_max = RP0 until we check for overclocking */
-	dev_priv->rps.max_freq		= dev_priv->rps.rp0_freq;
-
-	/* Preserve min/max settings in case of re-init */
-	if (dev_priv->rps.max_freq_softlimit == 0)
-		dev_priv->rps.max_freq_softlimit = dev_priv->rps.max_freq;
-
-	if (dev_priv->rps.min_freq_softlimit == 0)
-		dev_priv->rps.min_freq_softlimit = dev_priv->rps.min_freq;
+	parse_rp_state_cap(dev_priv, rp_state_cap);
 
 	/* disable the counters and set deterministic thresholds */
 	I915_WRITE(GEN6_RC_CONTROL, 0);
@@ -3494,7 +3642,7 @@ static void gen6_enable_rps(struct drm_device *dev)
 	gen6_gt_force_wake_put(dev_priv, FORCEWAKE_ALL);
 }
 
-void gen6_update_ring_freq(struct drm_device *dev)
+static void __gen6_update_ring_freq(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int min_freq = 15;
@@ -3564,6 +3712,18 @@ void gen6_update_ring_freq(struct drm_device *dev)
 	}
 }
 
+void gen6_update_ring_freq(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	if (INTEL_INFO(dev)->gen < 6 || IS_VALLEYVIEW(dev))
+		return;
+
+	mutex_lock(&dev_priv->rps.hw_lock);
+	__gen6_update_ring_freq(dev);
+	mutex_unlock(&dev_priv->rps.hw_lock);
+}
+
 int valleyview_rps_max_freq(struct drm_i915_private *dev_priv)
 {
 	u32 val, rp0;
@@ -3658,10 +3818,49 @@ static void valleyview_cleanup_pctx(struct drm_device *dev)
 	dev_priv->vlv_pctx = NULL;
 }
 
+static void valleyview_init_gt_powersave(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	valleyview_setup_pctx(dev);
+
+	mutex_lock(&dev_priv->rps.hw_lock);
+
+	dev_priv->rps.max_freq = valleyview_rps_max_freq(dev_priv);
+	dev_priv->rps.rp0_freq = dev_priv->rps.max_freq;
+	DRM_DEBUG_DRIVER("max GPU freq: %d MHz (%u)\n",
+			 vlv_gpu_freq(dev_priv, dev_priv->rps.max_freq),
+			 dev_priv->rps.max_freq);
+
+	dev_priv->rps.efficient_freq = valleyview_rps_rpe_freq(dev_priv);
+	DRM_DEBUG_DRIVER("RPe GPU freq: %d MHz (%u)\n",
+			 vlv_gpu_freq(dev_priv, dev_priv->rps.efficient_freq),
+			 dev_priv->rps.efficient_freq);
+
+	dev_priv->rps.min_freq = valleyview_rps_min_freq(dev_priv);
+	DRM_DEBUG_DRIVER("min GPU freq: %d MHz (%u)\n",
+			 vlv_gpu_freq(dev_priv, dev_priv->rps.min_freq),
+			 dev_priv->rps.min_freq);
+
+	/* Preserve min/max settings in case of re-init */
+	if (dev_priv->rps.max_freq_softlimit == 0)
+		dev_priv->rps.max_freq_softlimit = dev_priv->rps.max_freq;
+
+	if (dev_priv->rps.min_freq_softlimit == 0)
+		dev_priv->rps.min_freq_softlimit = dev_priv->rps.min_freq;
+
+	mutex_unlock(&dev_priv->rps.hw_lock);
+}
+
+static void valleyview_cleanup_gt_powersave(struct drm_device *dev)
+{
+	valleyview_cleanup_pctx(dev);
+}
+
 static void valleyview_enable_rps(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	u32 gtfifodbg, val, rc6_mode = 0;
 	int i;
 
@@ -3724,29 +3923,6 @@ static void valleyview_enable_rps(struct drm_device *dev)
 			 vlv_gpu_freq(dev_priv, dev_priv->rps.cur_freq),
 			 dev_priv->rps.cur_freq);
 
-	dev_priv->rps.max_freq = valleyview_rps_max_freq(dev_priv);
-	dev_priv->rps.rp0_freq  = dev_priv->rps.max_freq;
-	DRM_DEBUG_DRIVER("max GPU freq: %d MHz (%u)\n",
-			 vlv_gpu_freq(dev_priv, dev_priv->rps.max_freq),
-			 dev_priv->rps.max_freq);
-
-	dev_priv->rps.efficient_freq = valleyview_rps_rpe_freq(dev_priv);
-	DRM_DEBUG_DRIVER("RPe GPU freq: %d MHz (%u)\n",
-			 vlv_gpu_freq(dev_priv, dev_priv->rps.efficient_freq),
-			 dev_priv->rps.efficient_freq);
-
-	dev_priv->rps.min_freq = valleyview_rps_min_freq(dev_priv);
-	DRM_DEBUG_DRIVER("min GPU freq: %d MHz (%u)\n",
-			 vlv_gpu_freq(dev_priv, dev_priv->rps.min_freq),
-			 dev_priv->rps.min_freq);
-
-	/* Preserve min/max settings in case of re-init */
-	if (dev_priv->rps.max_freq_softlimit == 0)
-		dev_priv->rps.max_freq_softlimit = dev_priv->rps.max_freq;
-
-	if (dev_priv->rps.min_freq_softlimit == 0)
-		dev_priv->rps.min_freq_softlimit = dev_priv->rps.min_freq;
-
 	DRM_DEBUG_DRIVER("setting GPU freq to %d MHz (%u)\n",
 			 vlv_gpu_freq(dev_priv, dev_priv->rps.efficient_freq),
 			 dev_priv->rps.efficient_freq);
@@ -3815,7 +3991,7 @@ static int ironlake_setup_rc6(struct drm_device *dev)
 static void ironlake_enable_rc6(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
 	bool was_interruptible;
 	int ret;
 
@@ -3873,7 +4049,7 @@ static void ironlake_enable_rc6(struct drm_device *dev)
 	I915_WRITE(PWRCTXA, i915_gem_obj_ggtt_offset(dev_priv->ips.pwrctx) | PWRCTX_EN);
 	I915_WRITE(RSTDBYCTL, I915_READ(RSTDBYCTL) & ~RCX_SW_EXIT);
 
-	intel_print_rc6_info(dev, INTEL_RC6_ENABLE);
+	intel_print_rc6_info(dev, GEN6_RC_CTL_RC6_ENABLE);
 }
 
 static unsigned long intel_pxfreq(u32 vidfreq)
@@ -4327,7 +4503,7 @@ EXPORT_SYMBOL_GPL(i915_gpu_lower);
 bool i915_gpu_busy(void)
 {
 	struct drm_i915_private *dev_priv;
-	struct intel_ring_buffer *ring;
+	struct intel_engine_cs *ring;
 	bool ret = false;
 	int i;
 
@@ -4487,14 +4663,16 @@ static void intel_init_emon(struct drm_device *dev)
 
 void intel_init_gt_powersave(struct drm_device *dev)
 {
+	i915.enable_rc6 = sanitize_rc6_option(dev, i915.enable_rc6);
+
 	if (IS_VALLEYVIEW(dev))
-		valleyview_setup_pctx(dev);
+		valleyview_init_gt_powersave(dev);
 }
 
 void intel_cleanup_gt_powersave(struct drm_device *dev)
 {
 	if (IS_VALLEYVIEW(dev))
-		valleyview_cleanup_pctx(dev);
+		valleyview_cleanup_gt_powersave(dev);
 }
 
 void intel_disable_gt_powersave(struct drm_device *dev)
@@ -4507,8 +4685,10 @@ void intel_disable_gt_powersave(struct drm_device *dev)
 	if (IS_IRONLAKE_M(dev)) {
 		ironlake_disable_drps(dev);
 		ironlake_disable_rc6(dev);
-	} else if (INTEL_INFO(dev)->gen >= 6) {
-		cancel_delayed_work_sync(&dev_priv->rps.delayed_resume_work);
+	} else if (IS_GEN6(dev) || IS_GEN7(dev) || IS_BROADWELL(dev)) {
+		if (cancel_delayed_work_sync(&dev_priv->rps.delayed_resume_work))
+			intel_runtime_pm_put(dev_priv);
+
 		cancel_work_sync(&dev_priv->rps.work);
 		mutex_lock(&dev_priv->rps.hw_lock);
 		if (IS_VALLEYVIEW(dev))
@@ -4533,13 +4713,15 @@ static void intel_gen6_powersave_work(struct work_struct *work)
 		valleyview_enable_rps(dev);
 	} else if (IS_BROADWELL(dev)) {
 		gen8_enable_rps(dev);
-		gen6_update_ring_freq(dev);
+		__gen6_update_ring_freq(dev);
 	} else {
 		gen6_enable_rps(dev);
-		gen6_update_ring_freq(dev);
+		__gen6_update_ring_freq(dev);
 	}
 	dev_priv->rps.enabled = true;
 	mutex_unlock(&dev_priv->rps.hw_lock);
+
+	intel_runtime_pm_put(dev_priv);
 }
 
 void intel_enable_gt_powersave(struct drm_device *dev)
@@ -4547,20 +4729,38 @@ void intel_enable_gt_powersave(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	if (IS_IRONLAKE_M(dev)) {
+		mutex_lock(&dev->struct_mutex);
 		ironlake_enable_drps(dev);
 		ironlake_enable_rc6(dev);
 		intel_init_emon(dev);
-	} else if (IS_GEN6(dev) || IS_GEN7(dev)) {
+		mutex_unlock(&dev->struct_mutex);
+	} else if (IS_GEN6(dev) || IS_GEN7(dev) || IS_BROADWELL(dev)) {
 		/*
 		 * PCU communication is slow and this doesn't need to be
 		 * done at any specific time, so do this out of our fast path
 		 * to make resume and init faster.
+		 *
+		 * We depend on the HW RC6 power context save/restore
+		 * mechanism when entering D3 through runtime PM suspend. So
+		 * disable RPM until RPS/RC6 is properly setup. We can only
+		 * get here via the driver load/system resume/runtime resume
+		 * paths, so the _noresume version is enough (and in case of
+		 * runtime resume it's necessary).
 		 */
-		schedule_delayed_work(&dev_priv->rps.delayed_resume_work,
-				      round_jiffies_up_relative(HZ));
+		if (schedule_delayed_work(&dev_priv->rps.delayed_resume_work,
+					   round_jiffies_up_relative(HZ)))
+			intel_runtime_pm_get_noresume(dev_priv);
 	}
 }
 
+void intel_reset_gt_powersave(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	dev_priv->rps.enabled = false;
+	intel_enable_gt_powersave(dev);
+}
+
 static void ibx_init_clock_gating(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -4666,6 +4866,9 @@ static void ironlake_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(CACHE_MODE_0,
 		   _MASKED_BIT_ENABLE(CM0_PIPELINED_RENDER_FLUSH_DISABLE));
 
+	/* WaDisable_RenderCache_OperationalFlush:ilk */
+	I915_WRITE(CACHE_MODE_0, _MASKED_BIT_DISABLE(RC_OP_FLUSH_ENABLE));
+
 	g4x_disable_trickle_feed(dev);
 
 	ibx_init_clock_gating(dev);
@@ -4741,6 +4944,9 @@ static void gen6_init_clock_gating(struct drm_device *dev)
 		I915_WRITE(GEN6_GT_MODE,
 			   _MASKED_BIT_ENABLE(GEN6_TD_FOUR_ROW_DISPATCH_DISABLE));
 
+	/* WaDisable_RenderCache_OperationalFlush:snb */
+	I915_WRITE(CACHE_MODE_0, _MASKED_BIT_DISABLE(RC_OP_FLUSH_ENABLE));
+
 	/*
 	 * BSpec recoomends 8x4 when MSAA is used,
 	 * however in practice 16x4 seems fastest.
@@ -4909,6 +5115,10 @@ static void gen8_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(GEN7_HALF_SLICE_CHICKEN1,
 		   _MASKED_BIT_ENABLE(GEN7_SINGLE_SUBSCAN_DISPATCH_ENABLE));
 
+	/* WaDisableDopClockGating:bdw May not be needed for production */
+	I915_WRITE(GEN7_ROW_CHICKEN2,
+		   _MASKED_BIT_ENABLE(DOP_CLOCK_GATING_DISABLE));
+
 	/* WaSwitchSolVfFArbitrationPriority:bdw */
 	I915_WRITE(GAM_ECOCHK, I915_READ(GAM_ECOCHK) | HSW_ECOCHK_ARB_PRIO_SOL);
 
@@ -4980,6 +5190,9 @@ static void haswell_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(GEN7_FF_THREAD_MODE,
 		   I915_READ(GEN7_FF_THREAD_MODE) & ~GEN7_FF_VS_REF_CNT_FFME);
 
+	/* WaDisable_RenderCache_OperationalFlush:hsw */
+	I915_WRITE(CACHE_MODE_0_GEN7, _MASKED_BIT_DISABLE(RC_OP_FLUSH_ENABLE));
+
 	/* enable HiZ Raw Stall Optimization */
 	I915_WRITE(CACHE_MODE_0_GEN7,
 		   _MASKED_BIT_DISABLE(HIZ_RAW_STALL_OPT_DISABLE));
@@ -5032,6 +5245,9 @@ static void ivybridge_init_clock_gating(struct drm_device *dev)
 		I915_WRITE(GEN7_HALF_SLICE_CHICKEN1,
 			   _MASKED_BIT_ENABLE(GEN7_PSD_SINGLE_PORT_DISPATCH_ENABLE));
 
+	/* WaDisable_RenderCache_OperationalFlush:ivb */
+	I915_WRITE(CACHE_MODE_0_GEN7, _MASKED_BIT_DISABLE(RC_OP_FLUSH_ENABLE));
+
 	/* Apply the WaDisableRHWOOptimizationForRenderHang:ivb workaround. */
 	I915_WRITE(GEN7_COMMON_SLICE_CHICKEN1,
 		   GEN7_CSC1_RHWO_OPT_DISABLE_IN_RCC);
@@ -5126,6 +5342,10 @@ static void valleyview_init_clock_gating(struct drm_device *dev)
 	}
 	DRM_DEBUG_DRIVER("DDR speed: %d MHz", dev_priv->mem_freq);
 
+	dev_priv->vlv_cdclk_freq = valleyview_cur_cdclk(dev_priv);
+	DRM_DEBUG_DRIVER("Current CD clock rate: %d MHz",
+			 dev_priv->vlv_cdclk_freq);
+
 	I915_WRITE(DSPCLK_GATE_D, VRHUNIT_CLOCK_GATE_DISABLE);
 
 	/* WaDisableEarlyCull:vlv */
@@ -5143,6 +5363,9 @@ static void valleyview_init_clock_gating(struct drm_device *dev)
 		   _MASKED_BIT_ENABLE(GEN7_MAX_PS_THREAD_DEP |
 				      GEN7_PSD_SINGLE_PORT_DISPATCH_ENABLE));
 
+	/* WaDisable_RenderCache_OperationalFlush:vlv */
+	I915_WRITE(CACHE_MODE_0_GEN7, _MASKED_BIT_DISABLE(RC_OP_FLUSH_ENABLE));
+
 	/* WaForceL3Serialization:vlv */
 	I915_WRITE(GEN7_L3SQCREG4, I915_READ(GEN7_L3SQCREG4) &
 		   ~L3SQ_URB_READ_CAM_MATCH_DISABLE);
@@ -5165,8 +5388,11 @@ static void valleyview_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(GEN6_UCGCTL2,
 		   GEN6_RCZUNIT_CLOCK_GATE_DISABLE);
 
-	/* WaDisableL3Bank2xClockGate:vlv */
-	I915_WRITE(GEN7_UCGCTL4, GEN7_L3BANK2X_CLOCK_GATE_DISABLE);
+	/* WaDisableL3Bank2xClockGate:vlv
+	 * Disabling L3 clock gating- MMIO 940c[25] = 1
+	 * Set bit 25, to disable L3_BANK_2x_CLK_GATING */
+	I915_WRITE(GEN7_UCGCTL4,
+		   I915_READ(GEN7_UCGCTL4) | GEN7_L3BANK2X_CLOCK_GATE_DISABLE);
 
 	I915_WRITE(MI_ARB_VLV, MI_ARB_DISPLAY_TRICKLE_FEED_DISABLE);
 
@@ -5191,6 +5417,59 @@ static void valleyview_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(VLV_GUNIT_CLOCK_GATE, GCFG_DIS);
 }
 
+static void cherryview_init_clock_gating(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	I915_WRITE(DSPCLK_GATE_D, VRHUNIT_CLOCK_GATE_DISABLE);
+
+	I915_WRITE(MI_ARB_VLV, MI_ARB_DISPLAY_TRICKLE_FEED_DISABLE);
+
+	/* WaDisablePartialInstShootdown:chv */
+	I915_WRITE(GEN8_ROW_CHICKEN,
+		   _MASKED_BIT_ENABLE(PARTIAL_INSTRUCTION_SHOOTDOWN_DISABLE));
+
+	/* WaDisableThreadStallDopClockGating:chv */
+	I915_WRITE(GEN8_ROW_CHICKEN,
+		   _MASKED_BIT_ENABLE(STALL_DOP_GATING_DISABLE));
+
+	/* WaVSRefCountFullforceMissDisable:chv */
+	/* WaDSRefCountFullforceMissDisable:chv */
+	I915_WRITE(GEN7_FF_THREAD_MODE,
+		   I915_READ(GEN7_FF_THREAD_MODE) &
+		   ~(GEN8_FF_DS_REF_CNT_FFME | GEN7_FF_VS_REF_CNT_FFME));
+
+	/* WaDisableSemaphoreAndSyncFlipWait:chv */
+	I915_WRITE(GEN6_RC_SLEEP_PSMI_CONTROL,
+		   _MASKED_BIT_ENABLE(GEN8_RC_SEMA_IDLE_MSG_DISABLE));
+
+	/* WaDisableCSUnitClockGating:chv */
+	I915_WRITE(GEN6_UCGCTL1, I915_READ(GEN6_UCGCTL1) |
+		   GEN6_CSUNIT_CLOCK_GATE_DISABLE);
+
+	/* WaDisableSDEUnitClockGating:chv */
+	I915_WRITE(GEN8_UCGCTL6, I915_READ(GEN8_UCGCTL6) |
+		   GEN8_SDEUNIT_CLOCK_GATE_DISABLE);
+
+	/* WaDisableSamplerPowerBypass:chv (pre-production hw) */
+	I915_WRITE(HALF_SLICE_CHICKEN3,
+		   _MASKED_BIT_ENABLE(GEN8_SAMPLER_POWER_BYPASS_DIS));
+
+	/* WaDisableGunitClockGating:chv (pre-production hw) */
+	I915_WRITE(VLV_GUNIT_CLOCK_GATE, I915_READ(VLV_GUNIT_CLOCK_GATE) |
+		   GINT_DIS);
+
+	/* WaDisableFfDopClockGating:chv (pre-production hw) */
+	I915_WRITE(GEN6_RC_SLEEP_PSMI_CONTROL,
+		   _MASKED_BIT_ENABLE(GEN8_FF_DOP_CLOCK_GATE_DISABLE));
+
+	/* WaDisableDopClockGating:chv (pre-production hw) */
+	I915_WRITE(GEN7_ROW_CHICKEN2,
+		   _MASKED_BIT_ENABLE(DOP_CLOCK_GATING_DISABLE));
+	I915_WRITE(GEN6_UCGCTL1, I915_READ(GEN6_UCGCTL1) |
+		   GEN6_EU_TCUNIT_CLOCK_GATE_DISABLE);
+}
+
 static void g4x_init_clock_gating(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -5212,6 +5491,9 @@ static void g4x_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(CACHE_MODE_0,
 		   _MASKED_BIT_ENABLE(CM0_PIPELINED_RENDER_FLUSH_DISABLE));
 
+	/* WaDisable_RenderCache_OperationalFlush:g4x */
+	I915_WRITE(CACHE_MODE_0, _MASKED_BIT_DISABLE(RC_OP_FLUSH_ENABLE));
+
 	g4x_disable_trickle_feed(dev);
 }
 
@@ -5226,6 +5508,9 @@ static void crestline_init_clock_gating(struct drm_device *dev)
 	I915_WRITE16(DEUC, 0);
 	I915_WRITE(MI_ARB_STATE,
 		   _MASKED_BIT_ENABLE(MI_ARB_DISPLAY_TRICKLE_FEED_DISABLE));
+
+	/* WaDisable_RenderCache_OperationalFlush:gen4 */
+	I915_WRITE(CACHE_MODE_0, _MASKED_BIT_DISABLE(RC_OP_FLUSH_ENABLE));
 }
 
 static void broadwater_init_clock_gating(struct drm_device *dev)
@@ -5240,6 +5525,9 @@ static void broadwater_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(RENCLK_GATE_D2, 0);
 	I915_WRITE(MI_ARB_STATE,
 		   _MASKED_BIT_ENABLE(MI_ARB_DISPLAY_TRICKLE_FEED_DISABLE));
+
+	/* WaDisable_RenderCache_OperationalFlush:gen4 */
+	I915_WRITE(CACHE_MODE_0, _MASKED_BIT_DISABLE(RC_OP_FLUSH_ENABLE));
 }
 
 static void gen3_init_clock_gating(struct drm_device *dev)
@@ -5256,6 +5544,12 @@ static void gen3_init_clock_gating(struct drm_device *dev)
 
 	/* IIR "flip pending" means done if this bit is set */
 	I915_WRITE(ECOSKPD, _MASKED_BIT_DISABLE(ECO_FLIP_DONE));
+
+	/* interrupts should cause a wake up from C3 */
+	I915_WRITE(INSTPM, _MASKED_BIT_ENABLE(INSTPM_AGPBUSY_INT_EN));
+
+	/* On GEN3 we really need to make sure the ARB C3 LP bit is set */
+	I915_WRITE(MI_ARB_STATE, _MASKED_BIT_ENABLE(MI_ARB_C3_LP_WRITE_ENABLE));
 }
 
 static void i85x_init_clock_gating(struct drm_device *dev)
@@ -5263,6 +5557,10 @@ static void i85x_init_clock_gating(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	I915_WRITE(RENCLK_GATE_D1, SV_CLOCK_GATE_DISABLE);
+
+	/* interrupts should cause a wake up from C3 */
+	I915_WRITE(MI_STATE, _MASKED_BIT_ENABLE(MI_AGPBUSY_INT_EN) |
+		   _MASKED_BIT_DISABLE(MI_AGPBUSY_830_MODE));
 }
 
 static void i830_init_clock_gating(struct drm_device *dev)
@@ -5314,10 +5612,25 @@ bool intel_display_power_enabled_sw(struct drm_i915_private *dev_priv,
 				    enum intel_display_power_domain domain)
 {
 	struct i915_power_domains *power_domains;
+	struct i915_power_well *power_well;
+	bool is_enabled;
+	int i;
+
+	if (dev_priv->pm.suspended)
+		return false;
 
 	power_domains = &dev_priv->power_domains;
+	is_enabled = true;
+	for_each_power_well_rev(i, power_well, BIT(domain), power_domains) {
+		if (power_well->always_on)
+			continue;
 
-	return power_domains->domain_use_count[domain];
+		if (!power_well->count) {
+			is_enabled = false;
+			break;
+		}
+	}
+	return is_enabled;
 }
 
 bool intel_display_power_enabled(struct drm_i915_private *dev_priv,
@@ -5392,33 +5705,6 @@ static void hsw_power_well_post_enable(struct drm_i915_private *dev_priv)
 	}
 }
 
-static void reset_vblank_counter(struct drm_device *dev, enum pipe pipe)
-{
-	assert_spin_locked(&dev->vbl_lock);
-
-	dev->vblank[pipe].last = 0;
-}
-
-static void hsw_power_well_post_disable(struct drm_i915_private *dev_priv)
-{
-	struct drm_device *dev = dev_priv->dev;
-	enum pipe pipe;
-	unsigned long irqflags;
-
-	/*
-	 * After this, the registers on the pipes that are part of the power
-	 * well will become zero, so we have to adjust our counters according to
-	 * that.
-	 *
-	 * FIXME: Should we do this in general in drm_vblank_post_modeset?
-	 */
-	spin_lock_irqsave(&dev->vbl_lock, irqflags);
-	for_each_pipe(pipe)
-		if (pipe != PIPE_A)
-			reset_vblank_counter(dev, pipe);
-	spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
-}
-
 static void hsw_set_power_well(struct drm_i915_private *dev_priv,
 			       struct i915_power_well *power_well, bool enable)
 {
@@ -5447,8 +5733,6 @@ static void hsw_set_power_well(struct drm_i915_private *dev_priv,
 			I915_WRITE(HSW_PWR_WELL_DRIVER, 0);
 			POSTING_READ(HSW_PWR_WELL_DRIVER);
 			DRM_DEBUG_KMS("Requesting to disable the power well\n");
-
-			hsw_power_well_post_disable(dev_priv);
 		}
 	}
 }
@@ -5489,13 +5773,34 @@ static bool i9xx_always_on_power_well_enabled(struct drm_i915_private *dev_priv,
 	return true;
 }
 
-static void vlv_set_power_well(struct drm_i915_private *dev_priv,
-			       struct i915_power_well *power_well, bool enable)
+void __vlv_set_power_well(struct drm_i915_private *dev_priv,
+			  enum punit_power_well power_well_id, bool enable)
 {
-	enum punit_power_well power_well_id = power_well->data;
+	struct drm_device *dev = dev_priv->dev;
 	u32 mask;
 	u32 state;
 	u32 ctrl;
+	enum pipe pipe;
+
+	if (power_well_id == PUNIT_POWER_WELL_DPIO_CMN_BC) {
+		if (enable) {
+			/*
+			 * Enable the CRI clock source so we can get at the
+			 * display and the reference clock for VGA
+			 * hotplug / manual detection.
+			 */
+			I915_WRITE(DPLL(PIPE_B), I915_READ(DPLL(PIPE_B)) |
+				   DPLL_REFA_CLK_ENABLE_VLV |
+				   DPLL_INTEGRATED_CRI_CLK_VLV);
+			udelay(1); /* >10ns for cmnreset, >0ns for sidereset */
+		} else {
+			for_each_pipe(pipe)
+				assert_pll_disabled(dev_priv, pipe);
+			/* Assert common reset */
+			I915_WRITE(DPIO_CTL, I915_READ(DPIO_CTL) &
+				   ~DPIO_CMNRST);
+		}
+	}
 
 	mask = PUNIT_PWRGT_MASK(power_well_id);
 	state = enable ? PUNIT_PWRGT_PWR_ON(power_well_id) :
@@ -5523,6 +5828,28 @@ static void vlv_set_power_well(struct drm_i915_private *dev_priv,
 
 out:
 	mutex_unlock(&dev_priv->rps.hw_lock);
+
+	/*
+	 * From VLV2A0_DP_eDP_DPIO_driver_vbios_notes_10.docx -
+	 *  6.	De-assert cmn_reset/side_reset. Same as VLV X0.
+	 *   a.	GUnit 0x2110 bit[0] set to 1 (def 0)
+	 *   b.	The other bits such as sfr settings / modesel may all
+	 *	be set to 0.
+	 *
+	 * This should only be done on init and resume from S3 with
+	 * both PLLs disabled, or we risk losing DPIO and PLL
+	 * synchronization.
+	 */
+	if (power_well_id == PUNIT_POWER_WELL_DPIO_CMN_BC && enable)
+		I915_WRITE(DPIO_CTL, I915_READ(DPIO_CTL) | DPIO_CMNRST);
+}
+
+static void vlv_set_power_well(struct drm_i915_private *dev_priv,
+			       struct i915_power_well *power_well, bool enable)
+{
+	enum punit_power_well power_well_id = power_well->data;
+
+	__vlv_set_power_well(dev_priv, power_well_id, enable);
 }
 
 static void vlv_power_well_sync_hw(struct drm_i915_private *dev_priv,
@@ -5591,11 +5918,13 @@ static void vlv_display_power_well_enable(struct drm_i915_private *dev_priv,
 	spin_unlock_irq(&dev_priv->irq_lock);
 
 	/*
-	 * During driver initialization we need to defer enabling hotplug
-	 * processing until fbdev is set up.
+	 * During driver initialization/resume we can avoid restoring the
+	 * part of the HW/SW state that will be inited anyway explicitly.
 	 */
-	if (dev_priv->enable_hotplug_processing)
-		intel_hpd_init(dev_priv->dev);
+	if (dev_priv->power_domains.initializing)
+		return;
+
+	intel_hpd_init(dev_priv->dev);
 
 	i915_redisable_vga_power_on(dev_priv->dev);
 }
@@ -5603,23 +5932,12 @@ static void vlv_display_power_well_enable(struct drm_i915_private *dev_priv,
 static void vlv_display_power_well_disable(struct drm_i915_private *dev_priv,
 					   struct i915_power_well *power_well)
 {
-	struct drm_device *dev = dev_priv->dev;
-	enum pipe pipe;
-
 	WARN_ON_ONCE(power_well->data != PUNIT_POWER_WELL_DISP2D);
 
 	spin_lock_irq(&dev_priv->irq_lock);
-	for_each_pipe(pipe)
-		__intel_set_cpu_fifo_underrun_reporting(dev, pipe, false);
-
 	valleyview_disable_display_irqs(dev_priv);
 	spin_unlock_irq(&dev_priv->irq_lock);
 
-	spin_lock_irq(&dev->vbl_lock);
-	for_each_pipe(pipe)
-		reset_vblank_counter(dev, pipe);
-	spin_unlock_irq(&dev->vbl_lock);
-
 	vlv_set_power_well(dev_priv, power_well, false);
 }
 
@@ -5867,12 +6185,6 @@ static struct i915_power_well vlv_power_wells[] = {
 		.ops = &vlv_display_power_well_ops,
 	},
 	{
-		.name = "dpio-common",
-		.domains = VLV_DPIO_CMN_BC_POWER_DOMAINS,
-		.data = PUNIT_POWER_WELL_DPIO_CMN_BC,
-		.ops = &vlv_dpio_power_well_ops,
-	},
-	{
 		.name = "dpio-tx-b-01",
 		.domains = VLV_DPIO_TX_B_LANES_01_POWER_DOMAINS |
 			   VLV_DPIO_TX_B_LANES_23_POWER_DOMAINS |
@@ -5908,6 +6220,12 @@ static struct i915_power_well vlv_power_wells[] = {
 		.ops = &vlv_dpio_power_well_ops,
 		.data = PUNIT_POWER_WELL_DPIO_TX_C_LANES_23,
 	},
+	{
+		.name = "dpio-common",
+		.domains = VLV_DPIO_CMN_BC_POWER_DOMAINS,
+		.data = PUNIT_POWER_WELL_DPIO_CMN_BC,
+		.ops = &vlv_dpio_power_well_ops,
+	},
 };
 
 #define set_power_wells(power_domains, __power_wells) ({		\
@@ -5959,9 +6277,13 @@ static void intel_power_domains_resume(struct drm_i915_private *dev_priv)
 
 void intel_power_domains_init_hw(struct drm_i915_private *dev_priv)
 {
+	struct i915_power_domains *power_domains = &dev_priv->power_domains;
+
+	power_domains->initializing = true;
 	/* For now, we need the power well to be always enabled. */
 	intel_display_set_init_power(dev_priv, true);
 	intel_power_domains_resume(dev_priv);
+	power_domains->initializing = false;
 }
 
 void intel_aux_display_runtime_get(struct drm_i915_private *dev_priv)
@@ -5986,6 +6308,18 @@ void intel_runtime_pm_get(struct drm_i915_private *dev_priv)
 	WARN(dev_priv->pm.suspended, "Device still suspended.\n");
 }
 
+void intel_runtime_pm_get_noresume(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
+	struct device *device = &dev->pdev->dev;
+
+	if (!HAS_RUNTIME_PM(dev))
+		return;
+
+	WARN(dev_priv->pm.suspended, "Getting nosync-ref while suspended.\n");
+	pm_runtime_get_noresume(device);
+}
+
 void intel_runtime_pm_put(struct drm_i915_private *dev_priv)
 {
 	struct drm_device *dev = dev_priv->dev;
@@ -6008,6 +6342,15 @@ void intel_init_runtime_pm(struct drm_i915_private *dev_priv)
 
 	pm_runtime_set_active(device);
 
+	/*
+	 * RPM depends on RC6 to save restore the GT HW context, so make RC6 a
+	 * requirement.
+	 */
+	if (!intel_enable_rc6(dev)) {
+		DRM_INFO("RC6 disabled, disabling runtime PM support\n");
+		return;
+	}
+
 	pm_runtime_set_autosuspend_delay(device, 10000); /* 10s */
 	pm_runtime_mark_last_busy(device);
 	pm_runtime_use_autosuspend(device);
@@ -6023,6 +6366,9 @@ void intel_fini_runtime_pm(struct drm_i915_private *dev_priv)
 	if (!HAS_RUNTIME_PM(dev))
 		return;
 
+	if (!intel_enable_rc6(dev))
+		return;
+
 	/* Make sure we're not suspended first. */
 	pm_runtime_get_sync(device);
 	pm_runtime_disable(device);
@@ -6087,6 +6433,10 @@ void intel_init_pm(struct drm_device *dev)
 			dev_priv->display.init_clock_gating = haswell_init_clock_gating;
 		else if (INTEL_INFO(dev)->gen == 8)
 			dev_priv->display.init_clock_gating = gen8_init_clock_gating;
+	} else if (IS_CHERRYVIEW(dev)) {
+		dev_priv->display.update_wm = valleyview_update_wm;
+		dev_priv->display.init_clock_gating =
+			cherryview_init_clock_gating;
 	} else if (IS_VALLEYVIEW(dev)) {
 		dev_priv->display.update_wm = valleyview_update_wm;
 		dev_priv->display.init_clock_gating =
diff --git a/drivers/gpu/drm/i915/intel_renderstate.h b/drivers/gpu/drm/i915/intel_renderstate.h
new file mode 100644
index 000000000000..a5e783a9928a
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_renderstate.h
@@ -0,0 +1,48 @@
+/*
+ * Copyright © 2014 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef _INTEL_RENDERSTATE_H
+#define _INTEL_RENDERSTATE_H
+
+#include <linux/types.h>
+
+struct intel_renderstate_rodata {
+	const u32 *reloc;
+	const u32 reloc_items;
+	const u32 *batch;
+	const u32 batch_items;
+};
+
+extern const struct intel_renderstate_rodata gen6_null_state;
+extern const struct intel_renderstate_rodata gen7_null_state;
+extern const struct intel_renderstate_rodata gen8_null_state;
+
+#define RO_RENDERSTATE(_g)						\
+	const struct intel_renderstate_rodata gen ## _g ## _null_state = { \
+		.reloc = gen ## _g ## _null_state_relocs,		\
+		.reloc_items = sizeof(gen ## _g ## _null_state_relocs)/4, \
+		.batch = gen ## _g ## _null_state_batch,		\
+		.batch_items = sizeof(gen ## _g ## _null_state_batch)/4, \
+	}
+
+#endif /* INTEL_RENDERSTATE_H */
diff --git a/drivers/gpu/drm/i915/intel_renderstate_gen6.c b/drivers/gpu/drm/i915/intel_renderstate_gen6.c
new file mode 100644
index 000000000000..740538ad0977
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_renderstate_gen6.c
@@ -0,0 +1,289 @@
+#include "intel_renderstate.h"
+
+static const u32 gen6_null_state_relocs[] = {
+	0x00000020,
+	0x00000024,
+	0x0000002c,
+	0x000001e0,
+	0x000001e4,
+};
+
+static const u32 gen6_null_state_batch[] = {
+	0x69040000,
+	0x790d0001,
+	0x00000000,
+	0x00000000,
+	0x78180000,
+	0x00000001,
+	0x61010008,
+	0x00000000,
+	0x00000001,	 /* reloc */
+	0x00000001,	 /* reloc */
+	0x00000000,
+	0x00000001,	 /* reloc */
+	0x00000000,
+	0x00000001,
+	0x00000000,
+	0x00000001,
+	0x61020000,
+	0x00000000,
+	0x78050001,
+	0x00000018,
+	0x00000000,
+	0x780d1002,
+	0x00000000,
+	0x00000000,
+	0x00000420,
+	0x78150003,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78100004,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78160003,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78110005,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78120002,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78170003,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x79050005,
+	0xe0040000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x79100000,
+	0x00000000,
+	0x79000002,
+	0xffffffff,
+	0x00000000,
+	0x00000000,
+	0x780e0002,
+	0x00000441,
+	0x00000401,
+	0x00000401,
+	0x78021002,
+	0x00000000,
+	0x00000000,
+	0x00000400,
+	0x78130012,
+	0x00400810,
+	0x00000000,
+	0x20000000,
+	0x04000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78140007,
+	0x00000280,
+	0x08080000,
+	0x00000000,
+	0x00060000,
+	0x4e080002,
+	0x00100400,
+	0x00000000,
+	0x00000000,
+	0x78090005,
+	0x02000000,
+	0x22220000,
+	0x02f60000,
+	0x11330000,
+	0x02850004,
+	0x11220000,
+	0x78011002,
+	0x00000000,
+	0x00000000,
+	0x00000200,
+	0x78080003,
+	0x00002000,
+	0x00000448,	 /* reloc */
+	0x00000448,	 /* reloc */
+	0x00000000,
+	0x05000000,	 /* cmds end */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000220,	 /* state start */
+	0x00000240,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x0060005a,
+	0x204077be,
+	0x000000c0,
+	0x008d0040,
+	0x0060005a,
+	0x206077be,
+	0x000000c0,
+	0x008d0080,
+	0x0060005a,
+	0x208077be,
+	0x000000d0,
+	0x008d0040,
+	0x0060005a,
+	0x20a077be,
+	0x000000d0,
+	0x008d0080,
+	0x00000201,
+	0x20080061,
+	0x00000000,
+	0x00000000,
+	0x00600001,
+	0x20200022,
+	0x008d0000,
+	0x00000000,
+	0x02800031,
+	0x21c01cc9,
+	0x00000020,
+	0x0a8a0001,
+	0x00600001,
+	0x204003be,
+	0x008d01c0,
+	0x00000000,
+	0x00600001,
+	0x206003be,
+	0x008d01e0,
+	0x00000000,
+	0x00600001,
+	0x208003be,
+	0x008d0200,
+	0x00000000,
+	0x00600001,
+	0x20a003be,
+	0x008d0220,
+	0x00000000,
+	0x00600001,
+	0x20c003be,
+	0x008d0240,
+	0x00000000,
+	0x00600001,
+	0x20e003be,
+	0x008d0260,
+	0x00000000,
+	0x00600001,
+	0x210003be,
+	0x008d0280,
+	0x00000000,
+	0x00600001,
+	0x212003be,
+	0x008d02a0,
+	0x00000000,
+	0x05800031,
+	0x24001cc8,
+	0x00000040,
+	0x90019000,
+	0x0000007e,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x0000007e,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x0000007e,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x0000007e,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x0000007e,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x0000007e,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x0000007e,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x0000007e,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x30000000,
+	0x00000124,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0xf99a130c,
+	0x799a130c,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x80000031,
+	0x00000003,
+	0x00000000,	 /* state end */
+};
+
+RO_RENDERSTATE(6);
diff --git a/drivers/gpu/drm/i915/intel_renderstate_gen7.c b/drivers/gpu/drm/i915/intel_renderstate_gen7.c
new file mode 100644
index 000000000000..6fa7ff2a1298
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_renderstate_gen7.c
@@ -0,0 +1,253 @@
+#include "intel_renderstate.h"
+
+static const u32 gen7_null_state_relocs[] = {
+	0x0000000c,
+	0x00000010,
+	0x00000018,
+	0x000001ec,
+};
+
+static const u32 gen7_null_state_batch[] = {
+	0x69040000,
+	0x61010008,
+	0x00000000,
+	0x00000001,	 /* reloc */
+	0x00000001,	 /* reloc */
+	0x00000000,
+	0x00000001,	 /* reloc */
+	0x00000000,
+	0x00000001,
+	0x00000000,
+	0x00000001,
+	0x790d0002,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78180000,
+	0x00000001,
+	0x79160000,
+	0x00000008,
+	0x78300000,
+	0x02010040,
+	0x78310000,
+	0x04000000,
+	0x78320000,
+	0x04000000,
+	0x78330000,
+	0x02000000,
+	0x78100004,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x781b0005,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x781c0002,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x781d0004,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78110005,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78120002,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78210000,
+	0x00000000,
+	0x78130005,
+	0x00000000,
+	0x20000000,
+	0x04000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78140001,
+	0x20000800,
+	0x00000000,
+	0x781e0001,
+	0x00000000,
+	0x00000000,
+	0x78050005,
+	0xe0040000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78040001,
+	0x00000000,
+	0x00000000,
+	0x78240000,
+	0x00000240,
+	0x78230000,
+	0x00000260,
+	0x782f0000,
+	0x00000280,
+	0x781f000c,
+	0x00400810,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78200006,
+	0x000002c0,
+	0x08080000,
+	0x00000000,
+	0x28000402,
+	0x00060000,
+	0x00000000,
+	0x00000000,
+	0x78090005,
+	0x02000000,
+	0x22220000,
+	0x02f60000,
+	0x11230000,
+	0x02f60004,
+	0x11230000,
+	0x78080003,
+	0x00006008,
+	0x00000340,	 /* reloc */
+	0xffffffff,
+	0x00000000,
+	0x782a0000,
+	0x00000360,
+	0x79000002,
+	0xffffffff,
+	0x00000000,
+	0x00000000,
+	0x7b000005,
+	0x0000000f,
+	0x00000003,
+	0x00000000,
+	0x00000001,
+	0x00000000,
+	0x00000000,
+	0x05000000,	 /* cmds end */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000031,	 /* state start */
+	0x00000003,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0xf99a130c,
+	0x799a130c,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000492,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x0080005a,
+	0x2e2077bd,
+	0x000000c0,
+	0x008d0040,
+	0x0080005a,
+	0x2e6077bd,
+	0x000000d0,
+	0x008d0040,
+	0x02800031,
+	0x21801fa9,
+	0x008d0e20,
+	0x08840001,
+	0x00800001,
+	0x2e2003bd,
+	0x008d0180,
+	0x00000000,
+	0x00800001,
+	0x2e6003bd,
+	0x008d01c0,
+	0x00000000,
+	0x00800001,
+	0x2ea003bd,
+	0x008d0200,
+	0x00000000,
+	0x00800001,
+	0x2ee003bd,
+	0x008d0240,
+	0x00000000,
+	0x05800031,
+	0x20001fa8,
+	0x008d0e20,
+	0x90031000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000380,
+	0x000003a0,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,	 /* state end */
+};
+
+RO_RENDERSTATE(7);
diff --git a/drivers/gpu/drm/i915/intel_renderstate_gen8.c b/drivers/gpu/drm/i915/intel_renderstate_gen8.c
new file mode 100644
index 000000000000..5c875615d42a
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_renderstate_gen8.c
@@ -0,0 +1,479 @@
+#include "intel_renderstate.h"
+
+static const u32 gen8_null_state_relocs[] = {
+	0x00000048,
+	0x00000050,
+	0x00000060,
+	0x000003ec,
+};
+
+static const u32 gen8_null_state_batch[] = {
+	0x69040000,
+	0x61020001,
+	0x00000000,
+	0x00000000,
+	0x79120000,
+	0x00000000,
+	0x79130000,
+	0x00000000,
+	0x79140000,
+	0x00000000,
+	0x79150000,
+	0x00000000,
+	0x79160000,
+	0x00000000,
+	0x6101000e,
+	0x00000001,
+	0x00000000,
+	0x00000001,
+	0x00000001,	 /* reloc */
+	0x00000000,
+	0x00000001,	 /* reloc */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000001,	 /* reloc */
+	0x00000000,
+	0xfffff001,
+	0x00001001,
+	0xfffff001,
+	0x00001001,
+	0x78230000,
+	0x000006e0,
+	0x78210000,
+	0x00000700,
+	0x78300000,
+	0x08010040,
+	0x78330000,
+	0x08000000,
+	0x78310000,
+	0x08000000,
+	0x78320000,
+	0x08000000,
+	0x78240000,
+	0x00000641,
+	0x780e0000,
+	0x00000601,
+	0x780d0000,
+	0x00000000,
+	0x78180000,
+	0x00000001,
+	0x78520003,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78190009,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x781b0007,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78270000,
+	0x00000000,
+	0x782c0000,
+	0x00000000,
+	0x781c0002,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78160009,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78110008,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78290000,
+	0x00000000,
+	0x782e0000,
+	0x00000000,
+	0x781a0009,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x781d0007,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78280000,
+	0x00000000,
+	0x782d0000,
+	0x00000000,
+	0x78260000,
+	0x00000000,
+	0x782b0000,
+	0x00000000,
+	0x78150009,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78100007,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x781e0003,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78120002,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x781f0002,
+	0x30400820,
+	0x00000000,
+	0x00000000,
+	0x78510009,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78500003,
+	0x00210000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78130002,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x782a0000,
+	0x00000480,
+	0x782f0000,
+	0x00000540,
+	0x78140000,
+	0x00000800,
+	0x78170009,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x7820000a,
+	0x00000580,
+	0x00000000,
+	0x08080000,
+	0x00000000,
+	0x00000000,
+	0x1f000002,
+	0x00060000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x784d0000,
+	0x40000000,
+	0x784f0000,
+	0x80000100,
+	0x780f0000,
+	0x00000740,
+	0x78050006,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78070003,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78060003,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x78040001,
+	0x00000000,
+	0x00000001,
+	0x79000002,
+	0xffffffff,
+	0x00000000,
+	0x00000000,
+	0x78080003,
+	0x00006000,
+	0x000005e0,	 /* reloc */
+	0x00000000,
+	0x00000000,
+	0x78090005,
+	0x02000000,
+	0x22220000,
+	0x02f60000,
+	0x11230000,
+	0x02850004,
+	0x11230000,
+	0x784b0000,
+	0x0000000f,
+	0x78490001,
+	0x00000000,
+	0x00000000,
+	0x7b000005,
+	0x00000000,
+	0x00000003,
+	0x00000000,
+	0x00000001,
+	0x00000000,
+	0x00000000,
+	0x05000000,	 /* cmds end */
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x000004c0,	 /* state start */
+	0x00000500,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000092,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x0060005a,
+	0x21403ae8,
+	0x3a0000c0,
+	0x008d0040,
+	0x0060005a,
+	0x21603ae8,
+	0x3a0000c0,
+	0x008d0080,
+	0x0060005a,
+	0x21803ae8,
+	0x3a0000d0,
+	0x008d0040,
+	0x0060005a,
+	0x21a03ae8,
+	0x3a0000d0,
+	0x008d0080,
+	0x02800031,
+	0x2e0022e8,
+	0x0e000140,
+	0x08840001,
+	0x05800031,
+	0x200022e0,
+	0x0e000e00,
+	0x90031000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x06200000,
+	0x00000002,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0xf99a130c,
+	0x799a130c,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x3f800000,
+	0x00000000,
+	0x3f800000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,	 /* state end */
+};
+
+RO_RENDERSTATE(8);
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 79fb4cc2137c..279488addf3f 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -33,26 +33,44 @@
 #include "i915_trace.h"
 #include "intel_drv.h"
 
-static inline int ring_space(struct intel_ring_buffer *ring)
+/* Early gen2 devices have a cacheline of just 32 bytes, using 64 is overkill,
+ * but keeps the logic simple. Indeed, the whole purpose of this macro is just
+ * to give some inclination as to some of the magic values used in the various
+ * workarounds!
+ */
+#define CACHELINE_BYTES 64
+
+static inline int __ring_space(int head, int tail, int size)
 {
-	int space = (ring->head & HEAD_ADDR) - (ring->tail + I915_RING_FREE_SPACE);
+	int space = head - (tail + I915_RING_FREE_SPACE);
 	if (space < 0)
-		space += ring->size;
+		space += size;
 	return space;
 }
 
-void __intel_ring_advance(struct intel_ring_buffer *ring)
+static inline int ring_space(struct intel_engine_cs *ring)
+{
+	struct intel_ringbuffer *ringbuf = ring->buffer;
+	return __ring_space(ringbuf->head & HEAD_ADDR, ringbuf->tail, ringbuf->size);
+}
+
+static bool intel_ring_stopped(struct intel_engine_cs *ring)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	return dev_priv->gpu_error.stop_rings & intel_ring_flag(ring);
+}
 
-	ring->tail &= ring->size - 1;
-	if (dev_priv->gpu_error.stop_rings & intel_ring_flag(ring))
+void __intel_ring_advance(struct intel_engine_cs *ring)
+{
+	struct intel_ringbuffer *ringbuf = ring->buffer;
+	ringbuf->tail &= ringbuf->size - 1;
+	if (intel_ring_stopped(ring))
 		return;
-	ring->write_tail(ring, ring->tail);
+	ring->write_tail(ring, ringbuf->tail);
 }
 
 static int
-gen2_render_ring_flush(struct intel_ring_buffer *ring,
+gen2_render_ring_flush(struct intel_engine_cs *ring,
 		       u32	invalidate_domains,
 		       u32	flush_domains)
 {
@@ -78,7 +96,7 @@ gen2_render_ring_flush(struct intel_ring_buffer *ring,
 }
 
 static int
-gen4_render_ring_flush(struct intel_ring_buffer *ring,
+gen4_render_ring_flush(struct intel_engine_cs *ring,
 		       u32	invalidate_domains,
 		       u32	flush_domains)
 {
@@ -173,9 +191,9 @@ gen4_render_ring_flush(struct intel_ring_buffer *ring,
  * really our business.  That leaves only stall at scoreboard.
  */
 static int
-intel_emit_post_sync_nonzero_flush(struct intel_ring_buffer *ring)
+intel_emit_post_sync_nonzero_flush(struct intel_engine_cs *ring)
 {
-	u32 scratch_addr = ring->scratch.gtt_offset + 128;
+	u32 scratch_addr = ring->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 
 
@@ -208,11 +226,11 @@ intel_emit_post_sync_nonzero_flush(struct intel_ring_buffer *ring)
 }
 
 static int
-gen6_render_ring_flush(struct intel_ring_buffer *ring,
+gen6_render_ring_flush(struct intel_engine_cs *ring,
                          u32 invalidate_domains, u32 flush_domains)
 {
 	u32 flags = 0;
-	u32 scratch_addr = ring->scratch.gtt_offset + 128;
+	u32 scratch_addr = ring->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 
 	/* Force SNB workarounds for PIPE_CONTROL flushes */
@@ -260,7 +278,7 @@ gen6_render_ring_flush(struct intel_ring_buffer *ring,
 }
 
 static int
-gen7_render_ring_cs_stall_wa(struct intel_ring_buffer *ring)
+gen7_render_ring_cs_stall_wa(struct intel_engine_cs *ring)
 {
 	int ret;
 
@@ -278,7 +296,7 @@ gen7_render_ring_cs_stall_wa(struct intel_ring_buffer *ring)
 	return 0;
 }
 
-static int gen7_ring_fbc_flush(struct intel_ring_buffer *ring, u32 value)
+static int gen7_ring_fbc_flush(struct intel_engine_cs *ring, u32 value)
 {
 	int ret;
 
@@ -302,11 +320,11 @@ static int gen7_ring_fbc_flush(struct intel_ring_buffer *ring, u32 value)
 }
 
 static int
-gen7_render_ring_flush(struct intel_ring_buffer *ring,
+gen7_render_ring_flush(struct intel_engine_cs *ring,
 		       u32 invalidate_domains, u32 flush_domains)
 {
 	u32 flags = 0;
-	u32 scratch_addr = ring->scratch.gtt_offset + 128;
+	u32 scratch_addr = ring->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 
 	/*
@@ -363,11 +381,11 @@ gen7_render_ring_flush(struct intel_ring_buffer *ring,
 }
 
 static int
-gen8_render_ring_flush(struct intel_ring_buffer *ring,
+gen8_render_ring_flush(struct intel_engine_cs *ring,
 		       u32 invalidate_domains, u32 flush_domains)
 {
 	u32 flags = 0;
-	u32 scratch_addr = ring->scratch.gtt_offset + 128;
+	u32 scratch_addr = ring->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 
 	flags |= PIPE_CONTROL_CS_STALL;
@@ -403,14 +421,14 @@ gen8_render_ring_flush(struct intel_ring_buffer *ring,
 
 }
 
-static void ring_write_tail(struct intel_ring_buffer *ring,
+static void ring_write_tail(struct intel_engine_cs *ring,
 			    u32 value)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 	I915_WRITE_TAIL(ring, value);
 }
 
-u64 intel_ring_get_active_head(struct intel_ring_buffer *ring)
+u64 intel_ring_get_active_head(struct intel_engine_cs *ring)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 	u64 acthd;
@@ -426,7 +444,7 @@ u64 intel_ring_get_active_head(struct intel_ring_buffer *ring)
 	return acthd;
 }
 
-static void ring_setup_phys_status_page(struct intel_ring_buffer *ring)
+static void ring_setup_phys_status_page(struct intel_engine_cs *ring)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 	u32 addr;
@@ -437,7 +455,7 @@ static void ring_setup_phys_status_page(struct intel_ring_buffer *ring)
 	I915_WRITE(HWS_PGA, addr);
 }
 
-static bool stop_ring(struct intel_ring_buffer *ring)
+static bool stop_ring(struct intel_engine_cs *ring)
 {
 	struct drm_i915_private *dev_priv = to_i915(ring->dev);
 
@@ -461,11 +479,12 @@ static bool stop_ring(struct intel_ring_buffer *ring)
 	return (I915_READ_HEAD(ring) & HEAD_ADDR) == 0;
 }
 
-static int init_ring_common(struct intel_ring_buffer *ring)
+static int init_ring_common(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct drm_i915_gem_object *obj = ring->obj;
+	struct intel_ringbuffer *ringbuf = ring->buffer;
+	struct drm_i915_gem_object *obj = ringbuf->obj;
 	int ret = 0;
 
 	gen6_gt_force_wake_get(dev_priv, FORCEWAKE_ALL);
@@ -504,7 +523,7 @@ static int init_ring_common(struct intel_ring_buffer *ring)
 	 * register values. */
 	I915_WRITE_START(ring, i915_gem_obj_ggtt_offset(obj));
 	I915_WRITE_CTL(ring,
-			((ring->size - PAGE_SIZE) & RING_NR_PAGES)
+			((ringbuf->size - PAGE_SIZE) & RING_NR_PAGES)
 			| RING_VALID);
 
 	/* If the head is still not zero, the ring is dead */
@@ -512,12 +531,11 @@ static int init_ring_common(struct intel_ring_buffer *ring)
 		     I915_READ_START(ring) == i915_gem_obj_ggtt_offset(obj) &&
 		     (I915_READ_HEAD(ring) & HEAD_ADDR) == 0, 50)) {
 		DRM_ERROR("%s initialization failed "
-				"ctl %08x head %08x tail %08x start %08x\n",
-				ring->name,
-				I915_READ_CTL(ring),
-				I915_READ_HEAD(ring),
-				I915_READ_TAIL(ring),
-				I915_READ_START(ring));
+			  "ctl %08x (valid? %d) head %08x tail %08x start %08x [expected %08lx]\n",
+			  ring->name,
+			  I915_READ_CTL(ring), I915_READ_CTL(ring) & RING_VALID,
+			  I915_READ_HEAD(ring), I915_READ_TAIL(ring),
+			  I915_READ_START(ring), (unsigned long)i915_gem_obj_ggtt_offset(obj));
 		ret = -EIO;
 		goto out;
 	}
@@ -525,10 +543,10 @@ static int init_ring_common(struct intel_ring_buffer *ring)
 	if (!drm_core_check_feature(ring->dev, DRIVER_MODESET))
 		i915_kernel_lost_context(ring->dev);
 	else {
-		ring->head = I915_READ_HEAD(ring);
-		ring->tail = I915_READ_TAIL(ring) & TAIL_ADDR;
-		ring->space = ring_space(ring);
-		ring->last_retired_head = -1;
+		ringbuf->head = I915_READ_HEAD(ring);
+		ringbuf->tail = I915_READ_TAIL(ring) & TAIL_ADDR;
+		ringbuf->space = ring_space(ring);
+		ringbuf->last_retired_head = -1;
 	}
 
 	memset(&ring->hangcheck, 0, sizeof(ring->hangcheck));
@@ -540,7 +558,7 @@ out:
 }
 
 static int
-init_pipe_control(struct intel_ring_buffer *ring)
+init_pipe_control(struct intel_engine_cs *ring)
 {
 	int ret;
 
@@ -581,7 +599,7 @@ err:
 	return ret;
 }
 
-static int init_render_ring(struct intel_ring_buffer *ring)
+static int init_render_ring(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -595,19 +613,21 @@ static int init_render_ring(struct intel_ring_buffer *ring)
 	 * to use MI_WAIT_FOR_EVENT within the CS. It should already be
 	 * programmed to '1' on all products.
 	 *
-	 * WaDisableAsyncFlipPerfMode:snb,ivb,hsw,vlv,bdw
+	 * WaDisableAsyncFlipPerfMode:snb,ivb,hsw,vlv,bdw,chv
 	 */
 	if (INTEL_INFO(dev)->gen >= 6)
 		I915_WRITE(MI_MODE, _MASKED_BIT_ENABLE(ASYNC_FLIP_PERF_DISABLE));
 
 	/* Required for the hardware to program scanline values for waiting */
+	/* WaEnableFlushTlbInvalidationMode:snb */
 	if (INTEL_INFO(dev)->gen == 6)
 		I915_WRITE(GFX_MODE,
-			   _MASKED_BIT_ENABLE(GFX_TLB_INVALIDATE_ALWAYS));
+			   _MASKED_BIT_ENABLE(GFX_TLB_INVALIDATE_EXPLICIT));
 
+	/* WaBCSVCSTlbInvalidationMode:ivb,vlv,hsw */
 	if (IS_GEN7(dev))
 		I915_WRITE(GFX_MODE_GEN7,
-			   _MASKED_BIT_DISABLE(GFX_TLB_INVALIDATE_ALWAYS) |
+			   _MASKED_BIT_ENABLE(GFX_TLB_INVALIDATE_EXPLICIT) |
 			   _MASKED_BIT_ENABLE(GFX_REPLAY_MODE));
 
 	if (INTEL_INFO(dev)->gen >= 5) {
@@ -624,13 +644,6 @@ static int init_render_ring(struct intel_ring_buffer *ring)
 		 */
 		I915_WRITE(CACHE_MODE_0,
 			   _MASKED_BIT_DISABLE(CM0_STC_EVICT_DISABLE_LRA_SNB));
-
-		/* This is not explicitly set for GEN6, so read the register.
-		 * see intel_ring_mi_set_context() for why we care.
-		 * TODO: consider explicitly setting the bit for GEN5
-		 */
-		ring->itlb_before_ctx_switch =
-			!!(I915_READ(GFX_MODE) & GFX_TLB_INVALIDATE_ALWAYS);
 	}
 
 	if (INTEL_INFO(dev)->gen >= 6)
@@ -642,7 +655,7 @@ static int init_render_ring(struct intel_ring_buffer *ring)
 	return ret;
 }
 
-static void render_ring_cleanup(struct intel_ring_buffer *ring)
+static void render_ring_cleanup(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 
@@ -658,20 +671,46 @@ static void render_ring_cleanup(struct intel_ring_buffer *ring)
 	ring->scratch.obj = NULL;
 }
 
-static void
-update_mboxes(struct intel_ring_buffer *ring,
-	      u32 mmio_offset)
+static int gen6_signal(struct intel_engine_cs *signaller,
+		       unsigned int num_dwords)
 {
-/* NB: In order to be able to do semaphore MBOX updates for varying number
- * of rings, it's easiest if we round up each individual update to a
- * multiple of 2 (since ring updates must always be a multiple of 2)
- * even though the actual update only requires 3 dwords.
- */
+	struct drm_device *dev = signaller->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_engine_cs *useless;
+	int i, ret;
+
+	/* NB: In order to be able to do semaphore MBOX updates for varying
+	 * number of rings, it's easiest if we round up each individual update
+	 * to a multiple of 2 (since ring updates must always be a multiple of
+	 * 2) even though the actual update only requires 3 dwords.
+	 */
 #define MBOX_UPDATE_DWORDS 4
-	intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
-	intel_ring_emit(ring, mmio_offset);
-	intel_ring_emit(ring, ring->outstanding_lazy_seqno);
-	intel_ring_emit(ring, MI_NOOP);
+	if (i915_semaphore_is_enabled(dev))
+		num_dwords += ((I915_NUM_RINGS-1) * MBOX_UPDATE_DWORDS);
+	else
+		return intel_ring_begin(signaller, num_dwords);
+
+	ret = intel_ring_begin(signaller, num_dwords);
+	if (ret)
+		return ret;
+#undef MBOX_UPDATE_DWORDS
+
+	for_each_ring(useless, dev_priv, i) {
+		u32 mbox_reg = signaller->semaphore.mbox.signal[i];
+		if (mbox_reg != GEN6_NOSYNC) {
+			intel_ring_emit(signaller, MI_LOAD_REGISTER_IMM(1));
+			intel_ring_emit(signaller, mbox_reg);
+			intel_ring_emit(signaller, signaller->outstanding_lazy_seqno);
+			intel_ring_emit(signaller, MI_NOOP);
+		} else {
+			intel_ring_emit(signaller, MI_NOOP);
+			intel_ring_emit(signaller, MI_NOOP);
+			intel_ring_emit(signaller, MI_NOOP);
+			intel_ring_emit(signaller, MI_NOOP);
+		}
+	}
+
+	return 0;
 }
 
 /**
@@ -684,29 +723,14 @@ update_mboxes(struct intel_ring_buffer *ring,
  * This acts like a signal in the canonical semaphore.
  */
 static int
-gen6_add_request(struct intel_ring_buffer *ring)
+gen6_add_request(struct intel_engine_cs *ring)
 {
-	struct drm_device *dev = ring->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *useless;
-	int i, ret, num_dwords = 4;
-
-	if (i915_semaphore_is_enabled(dev))
-		num_dwords += ((I915_NUM_RINGS-1) * MBOX_UPDATE_DWORDS);
-#undef MBOX_UPDATE_DWORDS
+	int ret;
 
-	ret = intel_ring_begin(ring, num_dwords);
+	ret = ring->semaphore.signal(ring, 4);
 	if (ret)
 		return ret;
 
-	if (i915_semaphore_is_enabled(dev)) {
-		for_each_ring(useless, dev_priv, i) {
-			u32 mbox_reg = ring->signal_mbox[i];
-			if (mbox_reg != GEN6_NOSYNC)
-				update_mboxes(ring, mbox_reg);
-		}
-	}
-
 	intel_ring_emit(ring, MI_STORE_DWORD_INDEX);
 	intel_ring_emit(ring, I915_GEM_HWS_INDEX << MI_STORE_DWORD_INDEX_SHIFT);
 	intel_ring_emit(ring, ring->outstanding_lazy_seqno);
@@ -731,14 +755,15 @@ static inline bool i915_gem_has_seqno_wrapped(struct drm_device *dev,
  * @seqno - seqno which the waiter will block on
  */
 static int
-gen6_ring_sync(struct intel_ring_buffer *waiter,
-	       struct intel_ring_buffer *signaller,
+gen6_ring_sync(struct intel_engine_cs *waiter,
+	       struct intel_engine_cs *signaller,
 	       u32 seqno)
 {
-	int ret;
 	u32 dw1 = MI_SEMAPHORE_MBOX |
 		  MI_SEMAPHORE_COMPARE |
 		  MI_SEMAPHORE_REGISTER;
+	u32 wait_mbox = signaller->semaphore.mbox.wait[waiter->id];
+	int ret;
 
 	/* Throughout all of the GEM code, seqno passed implies our current
 	 * seqno is >= the last seqno executed. However for hardware the
@@ -746,8 +771,7 @@ gen6_ring_sync(struct intel_ring_buffer *waiter,
 	 */
 	seqno -= 1;
 
-	WARN_ON(signaller->semaphore_register[waiter->id] ==
-		MI_SEMAPHORE_SYNC_INVALID);
+	WARN_ON(wait_mbox == MI_SEMAPHORE_SYNC_INVALID);
 
 	ret = intel_ring_begin(waiter, 4);
 	if (ret)
@@ -755,9 +779,7 @@ gen6_ring_sync(struct intel_ring_buffer *waiter,
 
 	/* If seqno wrap happened, omit the wait with no-ops */
 	if (likely(!i915_gem_has_seqno_wrapped(waiter->dev, seqno))) {
-		intel_ring_emit(waiter,
-				dw1 |
-				signaller->semaphore_register[waiter->id]);
+		intel_ring_emit(waiter, dw1 | wait_mbox);
 		intel_ring_emit(waiter, seqno);
 		intel_ring_emit(waiter, 0);
 		intel_ring_emit(waiter, MI_NOOP);
@@ -782,9 +804,9 @@ do {									\
 } while (0)
 
 static int
-pc_render_add_request(struct intel_ring_buffer *ring)
+pc_render_add_request(struct intel_engine_cs *ring)
 {
-	u32 scratch_addr = ring->scratch.gtt_offset + 128;
+	u32 scratch_addr = ring->scratch.gtt_offset + 2 * CACHELINE_BYTES;
 	int ret;
 
 	/* For Ironlake, MI_USER_INTERRUPT was deprecated and apparently
@@ -806,15 +828,15 @@ pc_render_add_request(struct intel_ring_buffer *ring)
 	intel_ring_emit(ring, ring->outstanding_lazy_seqno);
 	intel_ring_emit(ring, 0);
 	PIPE_CONTROL_FLUSH(ring, scratch_addr);
-	scratch_addr += 128; /* write to separate cachelines */
+	scratch_addr += 2 * CACHELINE_BYTES; /* write to separate cachelines */
 	PIPE_CONTROL_FLUSH(ring, scratch_addr);
-	scratch_addr += 128;
+	scratch_addr += 2 * CACHELINE_BYTES;
 	PIPE_CONTROL_FLUSH(ring, scratch_addr);
-	scratch_addr += 128;
+	scratch_addr += 2 * CACHELINE_BYTES;
 	PIPE_CONTROL_FLUSH(ring, scratch_addr);
-	scratch_addr += 128;
+	scratch_addr += 2 * CACHELINE_BYTES;
 	PIPE_CONTROL_FLUSH(ring, scratch_addr);
-	scratch_addr += 128;
+	scratch_addr += 2 * CACHELINE_BYTES;
 	PIPE_CONTROL_FLUSH(ring, scratch_addr);
 
 	intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(4) | PIPE_CONTROL_QW_WRITE |
@@ -830,7 +852,7 @@ pc_render_add_request(struct intel_ring_buffer *ring)
 }
 
 static u32
-gen6_ring_get_seqno(struct intel_ring_buffer *ring, bool lazy_coherency)
+gen6_ring_get_seqno(struct intel_engine_cs *ring, bool lazy_coherency)
 {
 	/* Workaround to force correct ordering between irq and seqno writes on
 	 * ivb (and maybe also on snb) by reading from a CS register (like
@@ -844,31 +866,31 @@ gen6_ring_get_seqno(struct intel_ring_buffer *ring, bool lazy_coherency)
 }
 
 static u32
-ring_get_seqno(struct intel_ring_buffer *ring, bool lazy_coherency)
+ring_get_seqno(struct intel_engine_cs *ring, bool lazy_coherency)
 {
 	return intel_read_status_page(ring, I915_GEM_HWS_INDEX);
 }
 
 static void
-ring_set_seqno(struct intel_ring_buffer *ring, u32 seqno)
+ring_set_seqno(struct intel_engine_cs *ring, u32 seqno)
 {
 	intel_write_status_page(ring, I915_GEM_HWS_INDEX, seqno);
 }
 
 static u32
-pc_render_get_seqno(struct intel_ring_buffer *ring, bool lazy_coherency)
+pc_render_get_seqno(struct intel_engine_cs *ring, bool lazy_coherency)
 {
 	return ring->scratch.cpu_page[0];
 }
 
 static void
-pc_render_set_seqno(struct intel_ring_buffer *ring, u32 seqno)
+pc_render_set_seqno(struct intel_engine_cs *ring, u32 seqno)
 {
 	ring->scratch.cpu_page[0] = seqno;
 }
 
 static bool
-gen5_ring_get_irq(struct intel_ring_buffer *ring)
+gen5_ring_get_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -886,7 +908,7 @@ gen5_ring_get_irq(struct intel_ring_buffer *ring)
 }
 
 static void
-gen5_ring_put_irq(struct intel_ring_buffer *ring)
+gen5_ring_put_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -899,7 +921,7 @@ gen5_ring_put_irq(struct intel_ring_buffer *ring)
 }
 
 static bool
-i9xx_ring_get_irq(struct intel_ring_buffer *ring)
+i9xx_ring_get_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -920,7 +942,7 @@ i9xx_ring_get_irq(struct intel_ring_buffer *ring)
 }
 
 static void
-i9xx_ring_put_irq(struct intel_ring_buffer *ring)
+i9xx_ring_put_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -936,7 +958,7 @@ i9xx_ring_put_irq(struct intel_ring_buffer *ring)
 }
 
 static bool
-i8xx_ring_get_irq(struct intel_ring_buffer *ring)
+i8xx_ring_get_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -957,7 +979,7 @@ i8xx_ring_get_irq(struct intel_ring_buffer *ring)
 }
 
 static void
-i8xx_ring_put_irq(struct intel_ring_buffer *ring)
+i8xx_ring_put_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -972,7 +994,7 @@ i8xx_ring_put_irq(struct intel_ring_buffer *ring)
 	spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
 }
 
-void intel_ring_setup_status_page(struct intel_ring_buffer *ring)
+void intel_ring_setup_status_page(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
@@ -989,6 +1011,11 @@ void intel_ring_setup_status_page(struct intel_ring_buffer *ring)
 		case BCS:
 			mmio = BLT_HWS_PGA_GEN7;
 			break;
+		/*
+		 * VCS2 actually doesn't exist on Gen7. Only shut up
+		 * gcc switch check warning
+		 */
+		case VCS2:
 		case VCS:
 			mmio = BSD_HWS_PGA_GEN7;
 			break;
@@ -1030,7 +1057,7 @@ void intel_ring_setup_status_page(struct intel_ring_buffer *ring)
 }
 
 static int
-bsd_ring_flush(struct intel_ring_buffer *ring,
+bsd_ring_flush(struct intel_engine_cs *ring,
 	       u32     invalidate_domains,
 	       u32     flush_domains)
 {
@@ -1047,7 +1074,7 @@ bsd_ring_flush(struct intel_ring_buffer *ring,
 }
 
 static int
-i9xx_add_request(struct intel_ring_buffer *ring)
+i9xx_add_request(struct intel_engine_cs *ring)
 {
 	int ret;
 
@@ -1065,7 +1092,7 @@ i9xx_add_request(struct intel_ring_buffer *ring)
 }
 
 static bool
-gen6_ring_get_irq(struct intel_ring_buffer *ring)
+gen6_ring_get_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1090,7 +1117,7 @@ gen6_ring_get_irq(struct intel_ring_buffer *ring)
 }
 
 static void
-gen6_ring_put_irq(struct intel_ring_buffer *ring)
+gen6_ring_put_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1108,7 +1135,7 @@ gen6_ring_put_irq(struct intel_ring_buffer *ring)
 }
 
 static bool
-hsw_vebox_get_irq(struct intel_ring_buffer *ring)
+hsw_vebox_get_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1128,7 +1155,7 @@ hsw_vebox_get_irq(struct intel_ring_buffer *ring)
 }
 
 static void
-hsw_vebox_put_irq(struct intel_ring_buffer *ring)
+hsw_vebox_put_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1146,7 +1173,7 @@ hsw_vebox_put_irq(struct intel_ring_buffer *ring)
 }
 
 static bool
-gen8_ring_get_irq(struct intel_ring_buffer *ring)
+gen8_ring_get_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1172,7 +1199,7 @@ gen8_ring_get_irq(struct intel_ring_buffer *ring)
 }
 
 static void
-gen8_ring_put_irq(struct intel_ring_buffer *ring)
+gen8_ring_put_irq(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1192,8 +1219,8 @@ gen8_ring_put_irq(struct intel_ring_buffer *ring)
 }
 
 static int
-i965_dispatch_execbuffer(struct intel_ring_buffer *ring,
-			 u32 offset, u32 length,
+i965_dispatch_execbuffer(struct intel_engine_cs *ring,
+			 u64 offset, u32 length,
 			 unsigned flags)
 {
 	int ret;
@@ -1215,8 +1242,8 @@ i965_dispatch_execbuffer(struct intel_ring_buffer *ring,
 /* Just userspace ABI convention to limit the wa batch bo to a resonable size */
 #define I830_BATCH_LIMIT (256*1024)
 static int
-i830_dispatch_execbuffer(struct intel_ring_buffer *ring,
-				u32 offset, u32 len,
+i830_dispatch_execbuffer(struct intel_engine_cs *ring,
+				u64 offset, u32 len,
 				unsigned flags)
 {
 	int ret;
@@ -1266,8 +1293,8 @@ i830_dispatch_execbuffer(struct intel_ring_buffer *ring,
 }
 
 static int
-i915_dispatch_execbuffer(struct intel_ring_buffer *ring,
-			 u32 offset, u32 len,
+i915_dispatch_execbuffer(struct intel_engine_cs *ring,
+			 u64 offset, u32 len,
 			 unsigned flags)
 {
 	int ret;
@@ -1283,7 +1310,7 @@ i915_dispatch_execbuffer(struct intel_ring_buffer *ring,
 	return 0;
 }
 
-static void cleanup_status_page(struct intel_ring_buffer *ring)
+static void cleanup_status_page(struct intel_engine_cs *ring)
 {
 	struct drm_i915_gem_object *obj;
 
@@ -1297,50 +1324,44 @@ static void cleanup_status_page(struct intel_ring_buffer *ring)
 	ring->status_page.obj = NULL;
 }
 
-static int init_status_page(struct intel_ring_buffer *ring)
+static int init_status_page(struct intel_engine_cs *ring)
 {
-	struct drm_device *dev = ring->dev;
 	struct drm_i915_gem_object *obj;
-	int ret;
 
-	obj = i915_gem_alloc_object(dev, 4096);
-	if (obj == NULL) {
-		DRM_ERROR("Failed to allocate status page\n");
-		ret = -ENOMEM;
-		goto err;
-	}
+	if ((obj = ring->status_page.obj) == NULL) {
+		int ret;
 
-	ret = i915_gem_object_set_cache_level(obj, I915_CACHE_LLC);
-	if (ret)
-		goto err_unref;
+		obj = i915_gem_alloc_object(ring->dev, 4096);
+		if (obj == NULL) {
+			DRM_ERROR("Failed to allocate status page\n");
+			return -ENOMEM;
+		}
 
-	ret = i915_gem_obj_ggtt_pin(obj, 4096, 0);
-	if (ret)
-		goto err_unref;
+		ret = i915_gem_object_set_cache_level(obj, I915_CACHE_LLC);
+		if (ret)
+			goto err_unref;
+
+		ret = i915_gem_obj_ggtt_pin(obj, 4096, 0);
+		if (ret) {
+err_unref:
+			drm_gem_object_unreference(&obj->base);
+			return ret;
+		}
+
+		ring->status_page.obj = obj;
+	}
 
 	ring->status_page.gfx_addr = i915_gem_obj_ggtt_offset(obj);
 	ring->status_page.page_addr = kmap(sg_page(obj->pages->sgl));
-	if (ring->status_page.page_addr == NULL) {
-		ret = -ENOMEM;
-		goto err_unpin;
-	}
-	ring->status_page.obj = obj;
 	memset(ring->status_page.page_addr, 0, PAGE_SIZE);
 
 	DRM_DEBUG_DRIVER("%s hws offset: 0x%08x\n",
 			ring->name, ring->status_page.gfx_addr);
 
 	return 0;
-
-err_unpin:
-	i915_gem_object_ggtt_unpin(obj);
-err_unref:
-	drm_gem_object_unreference(&obj->base);
-err:
-	return ret;
 }
 
-static int init_phys_status_page(struct intel_ring_buffer *ring)
+static int init_phys_status_page(struct intel_engine_cs *ring)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 
@@ -1357,44 +1378,24 @@ static int init_phys_status_page(struct intel_ring_buffer *ring)
 	return 0;
 }
 
-static int intel_init_ring_buffer(struct drm_device *dev,
-				  struct intel_ring_buffer *ring)
+static int allocate_ring_buffer(struct intel_engine_cs *ring)
 {
+	struct drm_device *dev = ring->dev;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_ringbuffer *ringbuf = ring->buffer;
 	struct drm_i915_gem_object *obj;
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret;
 
-	ring->dev = dev;
-	INIT_LIST_HEAD(&ring->active_list);
-	INIT_LIST_HEAD(&ring->request_list);
-	ring->size = 32 * PAGE_SIZE;
-	memset(ring->sync_seqno, 0, sizeof(ring->sync_seqno));
-
-	init_waitqueue_head(&ring->irq_queue);
-
-	if (I915_NEED_GFX_HWS(dev)) {
-		ret = init_status_page(ring);
-		if (ret)
-			return ret;
-	} else {
-		BUG_ON(ring->id != RCS);
-		ret = init_phys_status_page(ring);
-		if (ret)
-			return ret;
-	}
+	if (intel_ring_initialized(ring))
+		return 0;
 
 	obj = NULL;
 	if (!HAS_LLC(dev))
-		obj = i915_gem_object_create_stolen(dev, ring->size);
+		obj = i915_gem_object_create_stolen(dev, ringbuf->size);
 	if (obj == NULL)
-		obj = i915_gem_alloc_object(dev, ring->size);
-	if (obj == NULL) {
-		DRM_ERROR("Failed to allocate ringbuffer\n");
-		ret = -ENOMEM;
-		goto err_hws;
-	}
-
-	ring->obj = obj;
+		obj = i915_gem_alloc_object(dev, ringbuf->size);
+	if (obj == NULL)
+		return -ENOMEM;
 
 	ret = i915_gem_obj_ggtt_pin(obj, PAGE_SIZE, PIN_MAPPABLE);
 	if (ret)
@@ -1404,65 +1405,102 @@ static int intel_init_ring_buffer(struct drm_device *dev,
 	if (ret)
 		goto err_unpin;
 
-	ring->virtual_start =
+	ringbuf->virtual_start =
 		ioremap_wc(dev_priv->gtt.mappable_base + i915_gem_obj_ggtt_offset(obj),
-			   ring->size);
-	if (ring->virtual_start == NULL) {
-		DRM_ERROR("Failed to map ringbuffer.\n");
+				ringbuf->size);
+	if (ringbuf->virtual_start == NULL) {
 		ret = -EINVAL;
 		goto err_unpin;
 	}
 
-	ret = ring->init(ring);
-	if (ret)
-		goto err_unmap;
+	ringbuf->obj = obj;
+	return 0;
+
+err_unpin:
+	i915_gem_object_ggtt_unpin(obj);
+err_unref:
+	drm_gem_object_unreference(&obj->base);
+	return ret;
+}
+
+static int intel_init_ring_buffer(struct drm_device *dev,
+				  struct intel_engine_cs *ring)
+{
+	struct intel_ringbuffer *ringbuf = ring->buffer;
+	int ret;
+
+	if (ringbuf == NULL) {
+		ringbuf = kzalloc(sizeof(*ringbuf), GFP_KERNEL);
+		if (!ringbuf)
+			return -ENOMEM;
+		ring->buffer = ringbuf;
+	}
+
+	ring->dev = dev;
+	INIT_LIST_HEAD(&ring->active_list);
+	INIT_LIST_HEAD(&ring->request_list);
+	ringbuf->size = 32 * PAGE_SIZE;
+	memset(ring->semaphore.sync_seqno, 0, sizeof(ring->semaphore.sync_seqno));
+
+	init_waitqueue_head(&ring->irq_queue);
+
+	if (I915_NEED_GFX_HWS(dev)) {
+		ret = init_status_page(ring);
+		if (ret)
+			goto error;
+	} else {
+		BUG_ON(ring->id != RCS);
+		ret = init_phys_status_page(ring);
+		if (ret)
+			goto error;
+	}
+
+	ret = allocate_ring_buffer(ring);
+	if (ret) {
+		DRM_ERROR("Failed to allocate ringbuffer %s: %d\n", ring->name, ret);
+		goto error;
+	}
 
 	/* Workaround an erratum on the i830 which causes a hang if
 	 * the TAIL pointer points to within the last 2 cachelines
 	 * of the buffer.
 	 */
-	ring->effective_size = ring->size;
-	if (IS_I830(ring->dev) || IS_845G(ring->dev))
-		ring->effective_size -= 128;
+	ringbuf->effective_size = ringbuf->size;
+	if (IS_I830(dev) || IS_845G(dev))
+		ringbuf->effective_size -= 2 * CACHELINE_BYTES;
 
-	i915_cmd_parser_init_ring(ring);
+	ret = i915_cmd_parser_init_ring(ring);
+	if (ret)
+		goto error;
+
+	ret = ring->init(ring);
+	if (ret)
+		goto error;
 
 	return 0;
 
-err_unmap:
-	iounmap(ring->virtual_start);
-err_unpin:
-	i915_gem_object_ggtt_unpin(obj);
-err_unref:
-	drm_gem_object_unreference(&obj->base);
-	ring->obj = NULL;
-err_hws:
-	cleanup_status_page(ring);
+error:
+	kfree(ringbuf);
+	ring->buffer = NULL;
 	return ret;
 }
 
-void intel_cleanup_ring_buffer(struct intel_ring_buffer *ring)
+void intel_cleanup_ring_buffer(struct intel_engine_cs *ring)
 {
-	struct drm_i915_private *dev_priv;
-	int ret;
+	struct drm_i915_private *dev_priv = to_i915(ring->dev);
+	struct intel_ringbuffer *ringbuf = ring->buffer;
 
-	if (ring->obj == NULL)
+	if (!intel_ring_initialized(ring))
 		return;
 
-	/* Disable the ring buffer. The ring must be idle at this point */
-	dev_priv = ring->dev->dev_private;
-	ret = intel_ring_idle(ring);
-	if (ret && !i915_reset_in_progress(&dev_priv->gpu_error))
-		DRM_ERROR("failed to quiesce %s whilst cleaning up: %d\n",
-			  ring->name, ret);
-
-	I915_WRITE_CTL(ring, 0);
+	intel_stop_ring_buffer(ring);
+	WARN_ON(!IS_GEN2(ring->dev) && (I915_READ_MODE(ring) & MODE_IDLE) == 0);
 
-	iounmap(ring->virtual_start);
+	iounmap(ringbuf->virtual_start);
 
-	i915_gem_object_ggtt_unpin(ring->obj);
-	drm_gem_object_unreference(&ring->obj->base);
-	ring->obj = NULL;
+	i915_gem_object_ggtt_unpin(ringbuf->obj);
+	drm_gem_object_unreference(&ringbuf->obj->base);
+	ringbuf->obj = NULL;
 	ring->preallocated_lazy_request = NULL;
 	ring->outstanding_lazy_seqno = 0;
 
@@ -1470,44 +1508,34 @@ void intel_cleanup_ring_buffer(struct intel_ring_buffer *ring)
 		ring->cleanup(ring);
 
 	cleanup_status_page(ring);
+
+	i915_cmd_parser_fini_ring(ring);
+
+	kfree(ringbuf);
+	ring->buffer = NULL;
 }
 
-static int intel_ring_wait_request(struct intel_ring_buffer *ring, int n)
+static int intel_ring_wait_request(struct intel_engine_cs *ring, int n)
 {
+	struct intel_ringbuffer *ringbuf = ring->buffer;
 	struct drm_i915_gem_request *request;
-	u32 seqno = 0, tail;
+	u32 seqno = 0;
 	int ret;
 
-	if (ring->last_retired_head != -1) {
-		ring->head = ring->last_retired_head;
-		ring->last_retired_head = -1;
+	if (ringbuf->last_retired_head != -1) {
+		ringbuf->head = ringbuf->last_retired_head;
+		ringbuf->last_retired_head = -1;
 
-		ring->space = ring_space(ring);
-		if (ring->space >= n)
+		ringbuf->space = ring_space(ring);
+		if (ringbuf->space >= n)
 			return 0;
 	}
 
 	list_for_each_entry(request, &ring->request_list, list) {
-		int space;
-
-		if (request->tail == -1)
-			continue;
-
-		space = request->tail - (ring->tail + I915_RING_FREE_SPACE);
-		if (space < 0)
-			space += ring->size;
-		if (space >= n) {
+		if (__ring_space(request->tail, ringbuf->tail, ringbuf->size) >= n) {
 			seqno = request->seqno;
-			tail = request->tail;
 			break;
 		}
-
-		/* Consume this request in case we need more space than
-		 * is available and so need to prevent a race between
-		 * updating last_retired_head and direct reads of
-		 * I915_RING_HEAD. It also provides a nice sanity check.
-		 */
-		request->tail = -1;
 	}
 
 	if (seqno == 0)
@@ -1517,18 +1545,19 @@ static int intel_ring_wait_request(struct intel_ring_buffer *ring, int n)
 	if (ret)
 		return ret;
 
-	ring->head = tail;
-	ring->space = ring_space(ring);
-	if (WARN_ON(ring->space < n))
-		return -ENOSPC;
+	i915_gem_retire_requests_ring(ring);
+	ringbuf->head = ringbuf->last_retired_head;
+	ringbuf->last_retired_head = -1;
 
+	ringbuf->space = ring_space(ring);
 	return 0;
 }
 
-static int ring_wait_for_space(struct intel_ring_buffer *ring, int n)
+static int ring_wait_for_space(struct intel_engine_cs *ring, int n)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_ringbuffer *ringbuf = ring->buffer;
 	unsigned long end;
 	int ret;
 
@@ -1539,7 +1568,6 @@ static int ring_wait_for_space(struct intel_ring_buffer *ring, int n)
 	/* force the tail write in case we have been skipping them */
 	__intel_ring_advance(ring);
 
-	trace_i915_ring_wait_begin(ring);
 	/* With GEM the hangcheck timer should kick us out of the loop,
 	 * leaving it early runs the risk of corrupting GEM state (due
 	 * to running on almost untested codepaths). But on resume
@@ -1547,12 +1575,13 @@ static int ring_wait_for_space(struct intel_ring_buffer *ring, int n)
 	 * case by choosing an insanely large timeout. */
 	end = jiffies + 60 * HZ;
 
+	trace_i915_ring_wait_begin(ring);
 	do {
-		ring->head = I915_READ_HEAD(ring);
-		ring->space = ring_space(ring);
-		if (ring->space >= n) {
-			trace_i915_ring_wait_end(ring);
-			return 0;
+		ringbuf->head = I915_READ_HEAD(ring);
+		ringbuf->space = ring_space(ring);
+		if (ringbuf->space >= n) {
+			ret = 0;
+			break;
 		}
 
 		if (!drm_core_check_feature(dev, DRIVER_MODESET) &&
@@ -1564,38 +1593,49 @@ static int ring_wait_for_space(struct intel_ring_buffer *ring, int n)
 
 		msleep(1);
 
+		if (dev_priv->mm.interruptible && signal_pending(current)) {
+			ret = -ERESTARTSYS;
+			break;
+		}
+
 		ret = i915_gem_check_wedge(&dev_priv->gpu_error,
 					   dev_priv->mm.interruptible);
 		if (ret)
-			return ret;
-	} while (!time_after(jiffies, end));
+			break;
+
+		if (time_after(jiffies, end)) {
+			ret = -EBUSY;
+			break;
+		}
+	} while (1);
 	trace_i915_ring_wait_end(ring);
-	return -EBUSY;
+	return ret;
 }
 
-static int intel_wrap_ring_buffer(struct intel_ring_buffer *ring)
+static int intel_wrap_ring_buffer(struct intel_engine_cs *ring)
 {
 	uint32_t __iomem *virt;
-	int rem = ring->size - ring->tail;
+	struct intel_ringbuffer *ringbuf = ring->buffer;
+	int rem = ringbuf->size - ringbuf->tail;
 
-	if (ring->space < rem) {
+	if (ringbuf->space < rem) {
 		int ret = ring_wait_for_space(ring, rem);
 		if (ret)
 			return ret;
 	}
 
-	virt = ring->virtual_start + ring->tail;
+	virt = ringbuf->virtual_start + ringbuf->tail;
 	rem /= 4;
 	while (rem--)
 		iowrite32(MI_NOOP, virt++);
 
-	ring->tail = 0;
-	ring->space = ring_space(ring);
+	ringbuf->tail = 0;
+	ringbuf->space = ring_space(ring);
 
 	return 0;
 }
 
-int intel_ring_idle(struct intel_ring_buffer *ring)
+int intel_ring_idle(struct intel_engine_cs *ring)
 {
 	u32 seqno;
 	int ret;
@@ -1619,7 +1659,7 @@ int intel_ring_idle(struct intel_ring_buffer *ring)
 }
 
 static int
-intel_ring_alloc_seqno(struct intel_ring_buffer *ring)
+intel_ring_alloc_seqno(struct intel_engine_cs *ring)
 {
 	if (ring->outstanding_lazy_seqno)
 		return 0;
@@ -1637,18 +1677,19 @@ intel_ring_alloc_seqno(struct intel_ring_buffer *ring)
 	return i915_gem_get_seqno(ring->dev, &ring->outstanding_lazy_seqno);
 }
 
-static int __intel_ring_prepare(struct intel_ring_buffer *ring,
+static int __intel_ring_prepare(struct intel_engine_cs *ring,
 				int bytes)
 {
+	struct intel_ringbuffer *ringbuf = ring->buffer;
 	int ret;
 
-	if (unlikely(ring->tail + bytes > ring->effective_size)) {
+	if (unlikely(ringbuf->tail + bytes > ringbuf->effective_size)) {
 		ret = intel_wrap_ring_buffer(ring);
 		if (unlikely(ret))
 			return ret;
 	}
 
-	if (unlikely(ring->space < bytes)) {
+	if (unlikely(ringbuf->space < bytes)) {
 		ret = ring_wait_for_space(ring, bytes);
 		if (unlikely(ret))
 			return ret;
@@ -1657,7 +1698,7 @@ static int __intel_ring_prepare(struct intel_ring_buffer *ring,
 	return 0;
 }
 
-int intel_ring_begin(struct intel_ring_buffer *ring,
+int intel_ring_begin(struct intel_engine_cs *ring,
 		     int num_dwords)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
@@ -1677,19 +1718,20 @@ int intel_ring_begin(struct intel_ring_buffer *ring,
 	if (ret)
 		return ret;
 
-	ring->space -= num_dwords * sizeof(uint32_t);
+	ring->buffer->space -= num_dwords * sizeof(uint32_t);
 	return 0;
 }
 
 /* Align the ring tail to a cacheline boundary */
-int intel_ring_cacheline_align(struct intel_ring_buffer *ring)
+int intel_ring_cacheline_align(struct intel_engine_cs *ring)
 {
-	int num_dwords = (64 - (ring->tail & 63)) / sizeof(uint32_t);
+	int num_dwords = (ring->buffer->tail & (CACHELINE_BYTES - 1)) / sizeof(uint32_t);
 	int ret;
 
 	if (num_dwords == 0)
 		return 0;
 
+	num_dwords = CACHELINE_BYTES / sizeof(uint32_t) - num_dwords;
 	ret = intel_ring_begin(ring, num_dwords);
 	if (ret)
 		return ret;
@@ -1702,7 +1744,7 @@ int intel_ring_cacheline_align(struct intel_ring_buffer *ring)
 	return 0;
 }
 
-void intel_ring_init_seqno(struct intel_ring_buffer *ring, u32 seqno)
+void intel_ring_init_seqno(struct intel_engine_cs *ring, u32 seqno)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 
@@ -1719,7 +1761,7 @@ void intel_ring_init_seqno(struct intel_ring_buffer *ring, u32 seqno)
 	ring->hangcheck.seqno = seqno;
 }
 
-static void gen6_bsd_ring_write_tail(struct intel_ring_buffer *ring,
+static void gen6_bsd_ring_write_tail(struct intel_engine_cs *ring,
 				     u32 value)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
@@ -1752,7 +1794,7 @@ static void gen6_bsd_ring_write_tail(struct intel_ring_buffer *ring,
 		   _MASKED_BIT_DISABLE(GEN6_BSD_SLEEP_MSG_DISABLE));
 }
 
-static int gen6_bsd_ring_flush(struct intel_ring_buffer *ring,
+static int gen6_bsd_ring_flush(struct intel_engine_cs *ring,
 			       u32 invalidate, u32 flush)
 {
 	uint32_t cmd;
@@ -1788,8 +1830,8 @@ static int gen6_bsd_ring_flush(struct intel_ring_buffer *ring,
 }
 
 static int
-gen8_ring_dispatch_execbuffer(struct intel_ring_buffer *ring,
-			      u32 offset, u32 len,
+gen8_ring_dispatch_execbuffer(struct intel_engine_cs *ring,
+			      u64 offset, u32 len,
 			      unsigned flags)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
@@ -1803,8 +1845,8 @@ gen8_ring_dispatch_execbuffer(struct intel_ring_buffer *ring,
 
 	/* FIXME(BDW): Address space and security selectors. */
 	intel_ring_emit(ring, MI_BATCH_BUFFER_START_GEN8 | (ppgtt<<8));
-	intel_ring_emit(ring, offset);
-	intel_ring_emit(ring, 0);
+	intel_ring_emit(ring, lower_32_bits(offset));
+	intel_ring_emit(ring, upper_32_bits(offset));
 	intel_ring_emit(ring, MI_NOOP);
 	intel_ring_advance(ring);
 
@@ -1812,8 +1854,8 @@ gen8_ring_dispatch_execbuffer(struct intel_ring_buffer *ring,
 }
 
 static int
-hsw_ring_dispatch_execbuffer(struct intel_ring_buffer *ring,
-			      u32 offset, u32 len,
+hsw_ring_dispatch_execbuffer(struct intel_engine_cs *ring,
+			      u64 offset, u32 len,
 			      unsigned flags)
 {
 	int ret;
@@ -1833,8 +1875,8 @@ hsw_ring_dispatch_execbuffer(struct intel_ring_buffer *ring,
 }
 
 static int
-gen6_ring_dispatch_execbuffer(struct intel_ring_buffer *ring,
-			      u32 offset, u32 len,
+gen6_ring_dispatch_execbuffer(struct intel_engine_cs *ring,
+			      u64 offset, u32 len,
 			      unsigned flags)
 {
 	int ret;
@@ -1855,7 +1897,7 @@ gen6_ring_dispatch_execbuffer(struct intel_ring_buffer *ring,
 
 /* Blitter support (SandyBridge+) */
 
-static int gen6_ring_flush(struct intel_ring_buffer *ring,
+static int gen6_ring_flush(struct intel_engine_cs *ring,
 			   u32 invalidate, u32 flush)
 {
 	struct drm_device *dev = ring->dev;
@@ -1898,7 +1940,7 @@ static int gen6_ring_flush(struct intel_ring_buffer *ring,
 int intel_init_render_ring_buffer(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
 
 	ring->name = "render ring";
 	ring->id = RCS;
@@ -1920,15 +1962,24 @@ int intel_init_render_ring_buffer(struct drm_device *dev)
 		ring->irq_enable_mask = GT_RENDER_USER_INTERRUPT;
 		ring->get_seqno = gen6_ring_get_seqno;
 		ring->set_seqno = ring_set_seqno;
-		ring->sync_to = gen6_ring_sync;
-		ring->semaphore_register[RCS] = MI_SEMAPHORE_SYNC_INVALID;
-		ring->semaphore_register[VCS] = MI_SEMAPHORE_SYNC_RV;
-		ring->semaphore_register[BCS] = MI_SEMAPHORE_SYNC_RB;
-		ring->semaphore_register[VECS] = MI_SEMAPHORE_SYNC_RVE;
-		ring->signal_mbox[RCS] = GEN6_NOSYNC;
-		ring->signal_mbox[VCS] = GEN6_VRSYNC;
-		ring->signal_mbox[BCS] = GEN6_BRSYNC;
-		ring->signal_mbox[VECS] = GEN6_VERSYNC;
+		ring->semaphore.sync_to = gen6_ring_sync;
+		ring->semaphore.signal = gen6_signal;
+		/*
+		 * The current semaphore is only applied on pre-gen8 platform.
+		 * And there is no VCS2 ring on the pre-gen8 platform. So the
+		 * semaphore between RCS and VCS2 is initialized as INVALID.
+		 * Gen8 will initialize the sema between VCS2 and RCS later.
+		 */
+		ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_INVALID;
+		ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_RV;
+		ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_RB;
+		ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_RVE;
+		ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
+		ring->semaphore.mbox.signal[RCS] = GEN6_NOSYNC;
+		ring->semaphore.mbox.signal[VCS] = GEN6_VRSYNC;
+		ring->semaphore.mbox.signal[BCS] = GEN6_BRSYNC;
+		ring->semaphore.mbox.signal[VECS] = GEN6_VERSYNC;
+		ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
 	} else if (IS_GEN5(dev)) {
 		ring->add_request = pc_render_add_request;
 		ring->flush = gen4_render_ring_flush;
@@ -1999,16 +2050,25 @@ int intel_init_render_ring_buffer(struct drm_device *dev)
 int intel_render_ring_init_dri(struct drm_device *dev, u64 start, u32 size)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[RCS];
+	struct intel_ringbuffer *ringbuf = ring->buffer;
 	int ret;
 
+	if (ringbuf == NULL) {
+		ringbuf = kzalloc(sizeof(*ringbuf), GFP_KERNEL);
+		if (!ringbuf)
+			return -ENOMEM;
+		ring->buffer = ringbuf;
+	}
+
 	ring->name = "render ring";
 	ring->id = RCS;
 	ring->mmio_base = RENDER_RING_BASE;
 
 	if (INTEL_INFO(dev)->gen >= 6) {
 		/* non-kms not supported on gen6+ */
-		return -ENODEV;
+		ret = -ENODEV;
+		goto err_ringbuf;
 	}
 
 	/* Note: gem is not supported on gen5/ilk without kms (the corresponding
@@ -2043,31 +2103,39 @@ int intel_render_ring_init_dri(struct drm_device *dev, u64 start, u32 size)
 	INIT_LIST_HEAD(&ring->active_list);
 	INIT_LIST_HEAD(&ring->request_list);
 
-	ring->size = size;
-	ring->effective_size = ring->size;
+	ringbuf->size = size;
+	ringbuf->effective_size = ringbuf->size;
 	if (IS_I830(ring->dev) || IS_845G(ring->dev))
-		ring->effective_size -= 128;
+		ringbuf->effective_size -= 2 * CACHELINE_BYTES;
 
-	ring->virtual_start = ioremap_wc(start, size);
-	if (ring->virtual_start == NULL) {
+	ringbuf->virtual_start = ioremap_wc(start, size);
+	if (ringbuf->virtual_start == NULL) {
 		DRM_ERROR("can not ioremap virtual address for"
 			  " ring buffer\n");
-		return -ENOMEM;
+		ret = -ENOMEM;
+		goto err_ringbuf;
 	}
 
 	if (!I915_NEED_GFX_HWS(dev)) {
 		ret = init_phys_status_page(ring);
 		if (ret)
-			return ret;
+			goto err_vstart;
 	}
 
 	return 0;
+
+err_vstart:
+	iounmap(ringbuf->virtual_start);
+err_ringbuf:
+	kfree(ringbuf);
+	ring->buffer = NULL;
+	return ret;
 }
 
 int intel_init_bsd_ring_buffer(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[VCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[VCS];
 
 	ring->name = "bsd ring";
 	ring->id = VCS;
@@ -2096,15 +2164,24 @@ int intel_init_bsd_ring_buffer(struct drm_device *dev)
 			ring->dispatch_execbuffer =
 				gen6_ring_dispatch_execbuffer;
 		}
-		ring->sync_to = gen6_ring_sync;
-		ring->semaphore_register[RCS] = MI_SEMAPHORE_SYNC_VR;
-		ring->semaphore_register[VCS] = MI_SEMAPHORE_SYNC_INVALID;
-		ring->semaphore_register[BCS] = MI_SEMAPHORE_SYNC_VB;
-		ring->semaphore_register[VECS] = MI_SEMAPHORE_SYNC_VVE;
-		ring->signal_mbox[RCS] = GEN6_RVSYNC;
-		ring->signal_mbox[VCS] = GEN6_NOSYNC;
-		ring->signal_mbox[BCS] = GEN6_BVSYNC;
-		ring->signal_mbox[VECS] = GEN6_VEVSYNC;
+		ring->semaphore.sync_to = gen6_ring_sync;
+		ring->semaphore.signal = gen6_signal;
+		/*
+		 * The current semaphore is only applied on pre-gen8 platform.
+		 * And there is no VCS2 ring on the pre-gen8 platform. So the
+		 * semaphore between VCS and VCS2 is initialized as INVALID.
+		 * Gen8 will initialize the sema between VCS2 and VCS later.
+		 */
+		ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_VR;
+		ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_INVALID;
+		ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_VB;
+		ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_VVE;
+		ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
+		ring->semaphore.mbox.signal[RCS] = GEN6_RVSYNC;
+		ring->semaphore.mbox.signal[VCS] = GEN6_NOSYNC;
+		ring->semaphore.mbox.signal[BCS] = GEN6_BVSYNC;
+		ring->semaphore.mbox.signal[VECS] = GEN6_VEVSYNC;
+		ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
 	} else {
 		ring->mmio_base = BSD_RING_BASE;
 		ring->flush = bsd_ring_flush;
@@ -2127,10 +2204,63 @@ int intel_init_bsd_ring_buffer(struct drm_device *dev)
 	return intel_init_ring_buffer(dev, ring);
 }
 
+/**
+ * Initialize the second BSD ring for Broadwell GT3.
+ * It is noted that this only exists on Broadwell GT3.
+ */
+int intel_init_bsd2_ring_buffer(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_engine_cs *ring = &dev_priv->ring[VCS2];
+
+	if ((INTEL_INFO(dev)->gen != 8)) {
+		DRM_ERROR("No dual-BSD ring on non-BDW machine\n");
+		return -EINVAL;
+	}
+
+	ring->name = "bds2_ring";
+	ring->id = VCS2;
+
+	ring->write_tail = ring_write_tail;
+	ring->mmio_base = GEN8_BSD2_RING_BASE;
+	ring->flush = gen6_bsd_ring_flush;
+	ring->add_request = gen6_add_request;
+	ring->get_seqno = gen6_ring_get_seqno;
+	ring->set_seqno = ring_set_seqno;
+	ring->irq_enable_mask =
+			GT_RENDER_USER_INTERRUPT << GEN8_VCS2_IRQ_SHIFT;
+	ring->irq_get = gen8_ring_get_irq;
+	ring->irq_put = gen8_ring_put_irq;
+	ring->dispatch_execbuffer =
+			gen8_ring_dispatch_execbuffer;
+	ring->semaphore.sync_to = gen6_ring_sync;
+	ring->semaphore.signal = gen6_signal;
+	/*
+	 * The current semaphore is only applied on the pre-gen8. And there
+	 * is no bsd2 ring on the pre-gen8. So now the semaphore_register
+	 * between VCS2 and other ring is initialized as invalid.
+	 * Gen8 will initialize the sema between VCS2 and other ring later.
+	 */
+	ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_INVALID;
+	ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_INVALID;
+	ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_INVALID;
+	ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_INVALID;
+	ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
+	ring->semaphore.mbox.signal[RCS] = GEN6_NOSYNC;
+	ring->semaphore.mbox.signal[VCS] = GEN6_NOSYNC;
+	ring->semaphore.mbox.signal[BCS] = GEN6_NOSYNC;
+	ring->semaphore.mbox.signal[VECS] = GEN6_NOSYNC;
+	ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
+
+	ring->init = init_ring_common;
+
+	return intel_init_ring_buffer(dev, ring);
+}
+
 int intel_init_blt_ring_buffer(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[BCS];
+	struct intel_engine_cs *ring = &dev_priv->ring[BCS];
 
 	ring->name = "blitter ring";
 	ring->id = BCS;
@@ -2153,15 +2283,24 @@ int intel_init_blt_ring_buffer(struct drm_device *dev)
 		ring->irq_put = gen6_ring_put_irq;
 		ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer;
 	}
-	ring->sync_to = gen6_ring_sync;
-	ring->semaphore_register[RCS] = MI_SEMAPHORE_SYNC_BR;
-	ring->semaphore_register[VCS] = MI_SEMAPHORE_SYNC_BV;
-	ring->semaphore_register[BCS] = MI_SEMAPHORE_SYNC_INVALID;
-	ring->semaphore_register[VECS] = MI_SEMAPHORE_SYNC_BVE;
-	ring->signal_mbox[RCS] = GEN6_RBSYNC;
-	ring->signal_mbox[VCS] = GEN6_VBSYNC;
-	ring->signal_mbox[BCS] = GEN6_NOSYNC;
-	ring->signal_mbox[VECS] = GEN6_VEBSYNC;
+	ring->semaphore.sync_to = gen6_ring_sync;
+	ring->semaphore.signal = gen6_signal;
+	/*
+	 * The current semaphore is only applied on pre-gen8 platform. And
+	 * there is no VCS2 ring on the pre-gen8 platform. So the semaphore
+	 * between BCS and VCS2 is initialized as INVALID.
+	 * Gen8 will initialize the sema between BCS and VCS2 later.
+	 */
+	ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_BR;
+	ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_BV;
+	ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_INVALID;
+	ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_BVE;
+	ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
+	ring->semaphore.mbox.signal[RCS] = GEN6_RBSYNC;
+	ring->semaphore.mbox.signal[VCS] = GEN6_VBSYNC;
+	ring->semaphore.mbox.signal[BCS] = GEN6_NOSYNC;
+	ring->semaphore.mbox.signal[VECS] = GEN6_VEBSYNC;
+	ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
 	ring->init = init_ring_common;
 
 	return intel_init_ring_buffer(dev, ring);
@@ -2170,7 +2309,7 @@ int intel_init_blt_ring_buffer(struct drm_device *dev)
 int intel_init_vebox_ring_buffer(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring = &dev_priv->ring[VECS];
+	struct intel_engine_cs *ring = &dev_priv->ring[VECS];
 
 	ring->name = "video enhancement ring";
 	ring->id = VECS;
@@ -2194,22 +2333,25 @@ int intel_init_vebox_ring_buffer(struct drm_device *dev)
 		ring->irq_put = hsw_vebox_put_irq;
 		ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer;
 	}
-	ring->sync_to = gen6_ring_sync;
-	ring->semaphore_register[RCS] = MI_SEMAPHORE_SYNC_VER;
-	ring->semaphore_register[VCS] = MI_SEMAPHORE_SYNC_VEV;
-	ring->semaphore_register[BCS] = MI_SEMAPHORE_SYNC_VEB;
-	ring->semaphore_register[VECS] = MI_SEMAPHORE_SYNC_INVALID;
-	ring->signal_mbox[RCS] = GEN6_RVESYNC;
-	ring->signal_mbox[VCS] = GEN6_VVESYNC;
-	ring->signal_mbox[BCS] = GEN6_BVESYNC;
-	ring->signal_mbox[VECS] = GEN6_NOSYNC;
+	ring->semaphore.sync_to = gen6_ring_sync;
+	ring->semaphore.signal = gen6_signal;
+	ring->semaphore.mbox.wait[RCS] = MI_SEMAPHORE_SYNC_VER;
+	ring->semaphore.mbox.wait[VCS] = MI_SEMAPHORE_SYNC_VEV;
+	ring->semaphore.mbox.wait[BCS] = MI_SEMAPHORE_SYNC_VEB;
+	ring->semaphore.mbox.wait[VECS] = MI_SEMAPHORE_SYNC_INVALID;
+	ring->semaphore.mbox.wait[VCS2] = MI_SEMAPHORE_SYNC_INVALID;
+	ring->semaphore.mbox.signal[RCS] = GEN6_RVESYNC;
+	ring->semaphore.mbox.signal[VCS] = GEN6_VVESYNC;
+	ring->semaphore.mbox.signal[BCS] = GEN6_BVESYNC;
+	ring->semaphore.mbox.signal[VECS] = GEN6_NOSYNC;
+	ring->semaphore.mbox.signal[VCS2] = GEN6_NOSYNC;
 	ring->init = init_ring_common;
 
 	return intel_init_ring_buffer(dev, ring);
 }
 
 int
-intel_ring_flush_all_caches(struct intel_ring_buffer *ring)
+intel_ring_flush_all_caches(struct intel_engine_cs *ring)
 {
 	int ret;
 
@@ -2227,7 +2369,7 @@ intel_ring_flush_all_caches(struct intel_ring_buffer *ring)
 }
 
 int
-intel_ring_invalidate_all_caches(struct intel_ring_buffer *ring)
+intel_ring_invalidate_all_caches(struct intel_engine_cs *ring)
 {
 	uint32_t flush_domains;
 	int ret;
@@ -2245,3 +2387,19 @@ intel_ring_invalidate_all_caches(struct intel_ring_buffer *ring)
 	ring->gpu_caches_dirty = false;
 	return 0;
 }
+
+void
+intel_stop_ring_buffer(struct intel_engine_cs *ring)
+{
+	int ret;
+
+	if (!intel_ring_initialized(ring))
+		return;
+
+	ret = intel_ring_idle(ring);
+	if (ret && !i915_reset_in_progress(&to_i915(ring->dev)->gpu_error))
+		DRM_ERROR("failed to quiesce %s whilst cleaning up: %d\n",
+			  ring->name, ret);
+
+	stop_ring(ring);
+}
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h
index 2b91c4b4d34b..910c83cf7d44 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.h
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
@@ -1,6 +1,10 @@
 #ifndef _INTEL_RINGBUFFER_H_
 #define _INTEL_RINGBUFFER_H_
 
+#include <linux/hashtable.h>
+
+#define I915_CMD_HASH_ORDER 9
+
 /*
  * Gen2 BSpec "1. Programming Environment" / 1.4.4.6 "Ring Buffer Use"
  * Gen3 BSpec "vol1c Memory Interface Functions" / 2.3.4.5 "Ring Buffer Use"
@@ -54,76 +58,93 @@ struct intel_ring_hangcheck {
 	bool deadlock;
 };
 
-struct  intel_ring_buffer {
+struct intel_ringbuffer {
+	struct drm_i915_gem_object *obj;
+	void __iomem *virtual_start;
+
+	u32 head;
+	u32 tail;
+	int space;
+	int size;
+	int effective_size;
+
+	/** We track the position of the requests in the ring buffer, and
+	 * when each is retired we increment last_retired_head as the GPU
+	 * must have finished processing the request and so we know we
+	 * can advance the ringbuffer up to that position.
+	 *
+	 * last_retired_head is set to -1 after the value is consumed so
+	 * we can detect new retirements.
+	 */
+	u32 last_retired_head;
+};
+
+struct  intel_engine_cs {
 	const char	*name;
 	enum intel_ring_id {
 		RCS = 0x0,
 		VCS,
 		BCS,
 		VECS,
+		VCS2
 	} id;
-#define I915_NUM_RINGS 4
+#define I915_NUM_RINGS 5
+#define LAST_USER_RING (VECS + 1)
 	u32		mmio_base;
-	void		__iomem *virtual_start;
 	struct		drm_device *dev;
-	struct		drm_i915_gem_object *obj;
+	struct intel_ringbuffer *buffer;
 
-	u32		head;
-	u32		tail;
-	int		space;
-	int		size;
-	int		effective_size;
 	struct intel_hw_status_page status_page;
 
-	/** We track the position of the requests in the ring buffer, and
-	 * when each is retired we increment last_retired_head as the GPU
-	 * must have finished processing the request and so we know we
-	 * can advance the ringbuffer up to that position.
-	 *
-	 * last_retired_head is set to -1 after the value is consumed so
-	 * we can detect new retirements.
-	 */
-	u32		last_retired_head;
-
 	unsigned irq_refcount; /* protected by dev_priv->irq_lock */
 	u32		irq_enable_mask;	/* bitmask to enable ring interrupt */
 	u32		trace_irq_seqno;
-	u32		sync_seqno[I915_NUM_RINGS-1];
-	bool __must_check (*irq_get)(struct intel_ring_buffer *ring);
-	void		(*irq_put)(struct intel_ring_buffer *ring);
+	bool __must_check (*irq_get)(struct intel_engine_cs *ring);
+	void		(*irq_put)(struct intel_engine_cs *ring);
 
-	int		(*init)(struct intel_ring_buffer *ring);
+	int		(*init)(struct intel_engine_cs *ring);
 
-	void		(*write_tail)(struct intel_ring_buffer *ring,
+	void		(*write_tail)(struct intel_engine_cs *ring,
 				      u32 value);
-	int __must_check (*flush)(struct intel_ring_buffer *ring,
+	int __must_check (*flush)(struct intel_engine_cs *ring,
 				  u32	invalidate_domains,
 				  u32	flush_domains);
-	int		(*add_request)(struct intel_ring_buffer *ring);
+	int		(*add_request)(struct intel_engine_cs *ring);
 	/* Some chipsets are not quite as coherent as advertised and need
 	 * an expensive kick to force a true read of the up-to-date seqno.
 	 * However, the up-to-date seqno is not always required and the last
 	 * seen value is good enough. Note that the seqno will always be
 	 * monotonic, even if not coherent.
 	 */
-	u32		(*get_seqno)(struct intel_ring_buffer *ring,
+	u32		(*get_seqno)(struct intel_engine_cs *ring,
 				     bool lazy_coherency);
-	void		(*set_seqno)(struct intel_ring_buffer *ring,
+	void		(*set_seqno)(struct intel_engine_cs *ring,
 				     u32 seqno);
-	int		(*dispatch_execbuffer)(struct intel_ring_buffer *ring,
-					       u32 offset, u32 length,
+	int		(*dispatch_execbuffer)(struct intel_engine_cs *ring,
+					       u64 offset, u32 length,
 					       unsigned flags);
 #define I915_DISPATCH_SECURE 0x1
 #define I915_DISPATCH_PINNED 0x2
-	void		(*cleanup)(struct intel_ring_buffer *ring);
-	int		(*sync_to)(struct intel_ring_buffer *ring,
-				   struct intel_ring_buffer *to,
-				   u32 seqno);
+	void		(*cleanup)(struct intel_engine_cs *ring);
 
-	/* our mbox written by others */
-	u32		semaphore_register[I915_NUM_RINGS];
-	/* mboxes this ring signals to */
-	u32		signal_mbox[I915_NUM_RINGS];
+	struct {
+		u32	sync_seqno[I915_NUM_RINGS-1];
+
+		struct {
+			/* our mbox written by others */
+			u32		wait[I915_NUM_RINGS];
+			/* mboxes this ring signals to */
+			u32		signal[I915_NUM_RINGS];
+		} mbox;
+
+		/* AKA wait() */
+		int	(*sync_to)(struct intel_engine_cs *ring,
+				   struct intel_engine_cs *to,
+				   u32 seqno);
+		int	(*signal)(struct intel_engine_cs *signaller,
+				  /* num_dwords needed by caller */
+				  unsigned int num_dwords);
+	} semaphore;
 
 	/**
 	 * List of objects currently involved in rendering from the
@@ -153,12 +174,8 @@ struct  intel_ring_buffer {
 
 	wait_queue_head_t irq_queue;
 
-	/**
-	 * Do an explicit TLB flush before MI_SET_CONTEXT
-	 */
-	bool itlb_before_ctx_switch;
-	struct i915_hw_context *default_context;
-	struct i915_hw_context *last_context;
+	struct intel_context *default_context;
+	struct intel_context *last_context;
 
 	struct intel_ring_hangcheck hangcheck;
 
@@ -168,12 +185,13 @@ struct  intel_ring_buffer {
 		volatile u32 *cpu_page;
 	} scratch;
 
+	bool needs_cmd_parser;
+
 	/*
-	 * Tables of commands the command parser needs to know about
+	 * Table of commands the command parser needs to know about
 	 * for this ring.
 	 */
-	const struct drm_i915_cmd_table *cmd_tables;
-	int cmd_table_count;
+	DECLARE_HASHTABLE(cmd_hash, I915_CMD_HASH_ORDER);
 
 	/*
 	 * Table of registers allowed in commands that read/write registers.
@@ -202,20 +220,20 @@ struct  intel_ring_buffer {
 };
 
 static inline bool
-intel_ring_initialized(struct intel_ring_buffer *ring)
+intel_ring_initialized(struct intel_engine_cs *ring)
 {
-	return ring->obj != NULL;
+	return ring->buffer && ring->buffer->obj;
 }
 
 static inline unsigned
-intel_ring_flag(struct intel_ring_buffer *ring)
+intel_ring_flag(struct intel_engine_cs *ring)
 {
 	return 1 << ring->id;
 }
 
 static inline u32
-intel_ring_sync_index(struct intel_ring_buffer *ring,
-		      struct intel_ring_buffer *other)
+intel_ring_sync_index(struct intel_engine_cs *ring,
+		      struct intel_engine_cs *other)
 {
 	int idx;
 
@@ -233,7 +251,7 @@ intel_ring_sync_index(struct intel_ring_buffer *ring,
 }
 
 static inline u32
-intel_read_status_page(struct intel_ring_buffer *ring,
+intel_read_status_page(struct intel_engine_cs *ring,
 		       int reg)
 {
 	/* Ensure that the compiler doesn't optimize away the load. */
@@ -242,7 +260,7 @@ intel_read_status_page(struct intel_ring_buffer *ring,
 }
 
 static inline void
-intel_write_status_page(struct intel_ring_buffer *ring,
+intel_write_status_page(struct intel_engine_cs *ring,
 			int reg, u32 value)
 {
 	ring->status_page.page_addr[reg] = value;
@@ -267,47 +285,51 @@ intel_write_status_page(struct intel_ring_buffer *ring,
 #define I915_GEM_HWS_SCRATCH_INDEX	0x30
 #define I915_GEM_HWS_SCRATCH_ADDR (I915_GEM_HWS_SCRATCH_INDEX << MI_STORE_DWORD_INDEX_SHIFT)
 
-void intel_cleanup_ring_buffer(struct intel_ring_buffer *ring);
+void intel_stop_ring_buffer(struct intel_engine_cs *ring);
+void intel_cleanup_ring_buffer(struct intel_engine_cs *ring);
 
-int __must_check intel_ring_begin(struct intel_ring_buffer *ring, int n);
-int __must_check intel_ring_cacheline_align(struct intel_ring_buffer *ring);
-static inline void intel_ring_emit(struct intel_ring_buffer *ring,
+int __must_check intel_ring_begin(struct intel_engine_cs *ring, int n);
+int __must_check intel_ring_cacheline_align(struct intel_engine_cs *ring);
+static inline void intel_ring_emit(struct intel_engine_cs *ring,
 				   u32 data)
 {
-	iowrite32(data, ring->virtual_start + ring->tail);
-	ring->tail += 4;
+	struct intel_ringbuffer *ringbuf = ring->buffer;
+	iowrite32(data, ringbuf->virtual_start + ringbuf->tail);
+	ringbuf->tail += 4;
 }
-static inline void intel_ring_advance(struct intel_ring_buffer *ring)
+static inline void intel_ring_advance(struct intel_engine_cs *ring)
 {
-	ring->tail &= ring->size - 1;
+	struct intel_ringbuffer *ringbuf = ring->buffer;
+	ringbuf->tail &= ringbuf->size - 1;
 }
-void __intel_ring_advance(struct intel_ring_buffer *ring);
+void __intel_ring_advance(struct intel_engine_cs *ring);
 
-int __must_check intel_ring_idle(struct intel_ring_buffer *ring);
-void intel_ring_init_seqno(struct intel_ring_buffer *ring, u32 seqno);
-int intel_ring_flush_all_caches(struct intel_ring_buffer *ring);
-int intel_ring_invalidate_all_caches(struct intel_ring_buffer *ring);
+int __must_check intel_ring_idle(struct intel_engine_cs *ring);
+void intel_ring_init_seqno(struct intel_engine_cs *ring, u32 seqno);
+int intel_ring_flush_all_caches(struct intel_engine_cs *ring);
+int intel_ring_invalidate_all_caches(struct intel_engine_cs *ring);
 
 int intel_init_render_ring_buffer(struct drm_device *dev);
 int intel_init_bsd_ring_buffer(struct drm_device *dev);
+int intel_init_bsd2_ring_buffer(struct drm_device *dev);
 int intel_init_blt_ring_buffer(struct drm_device *dev);
 int intel_init_vebox_ring_buffer(struct drm_device *dev);
 
-u64 intel_ring_get_active_head(struct intel_ring_buffer *ring);
-void intel_ring_setup_status_page(struct intel_ring_buffer *ring);
+u64 intel_ring_get_active_head(struct intel_engine_cs *ring);
+void intel_ring_setup_status_page(struct intel_engine_cs *ring);
 
-static inline u32 intel_ring_get_tail(struct intel_ring_buffer *ring)
+static inline u32 intel_ring_get_tail(struct intel_engine_cs *ring)
 {
-	return ring->tail;
+	return ring->buffer->tail;
 }
 
-static inline u32 intel_ring_get_seqno(struct intel_ring_buffer *ring)
+static inline u32 intel_ring_get_seqno(struct intel_engine_cs *ring)
 {
 	BUG_ON(ring->outstanding_lazy_seqno == 0);
 	return ring->outstanding_lazy_seqno;
 }
 
-static inline void i915_trace_irq_get(struct intel_ring_buffer *ring, u32 seqno)
+static inline void i915_trace_irq_get(struct intel_engine_cs *ring, u32 seqno)
 {
 	if (ring->trace_irq_seqno == 0 && ring->irq_get(ring))
 		ring->trace_irq_seqno = seqno;
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index 46be00d66df3..6a4d5bc17697 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -1153,20 +1153,21 @@ static bool intel_sdvo_compute_config(struct intel_encoder *encoder,
 	pipe_config->pixel_multiplier =
 		intel_sdvo_get_pixel_multiplier(adjusted_mode);
 
+	pipe_config->has_hdmi_sink = intel_sdvo->has_hdmi_monitor;
+
 	if (intel_sdvo->color_range_auto) {
 		/* See CEA-861-E - 5.1 Default Encoding Parameters */
 		/* FIXME: This bit is only valid when using TMDS encoding and 8
 		 * bit per color mode. */
-		if (intel_sdvo->has_hdmi_monitor &&
+		if (pipe_config->has_hdmi_sink &&
 		    drm_match_cea_mode(adjusted_mode) > 1)
-			intel_sdvo->color_range = HDMI_COLOR_RANGE_16_235;
-		else
-			intel_sdvo->color_range = 0;
+			pipe_config->limited_color_range = true;
+	} else {
+		if (pipe_config->has_hdmi_sink &&
+		    intel_sdvo->color_range == HDMI_COLOR_RANGE_16_235)
+			pipe_config->limited_color_range = true;
 	}
 
-	if (intel_sdvo->color_range)
-		pipe_config->limited_color_range = true;
-
 	/* Clock computation needs to happen after pixel multiplier. */
 	if (intel_sdvo->is_tv)
 		i9xx_adjust_sdvo_tv_clock(pipe_config);
@@ -1174,7 +1175,7 @@ static bool intel_sdvo_compute_config(struct intel_encoder *encoder,
 	return true;
 }
 
-static void intel_sdvo_mode_set(struct intel_encoder *intel_encoder)
+static void intel_sdvo_pre_enable(struct intel_encoder *intel_encoder)
 {
 	struct drm_device *dev = intel_encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1223,7 +1224,7 @@ static void intel_sdvo_mode_set(struct intel_encoder *intel_encoder)
 	if (!intel_sdvo_set_target_input(intel_sdvo))
 		return;
 
-	if (intel_sdvo->has_hdmi_monitor) {
+	if (crtc->config.has_hdmi_sink) {
 		intel_sdvo_set_encode(intel_sdvo, SDVO_ENCODE_HDMI);
 		intel_sdvo_set_colorimetry(intel_sdvo,
 					   SDVO_COLORIMETRY_RGB256);
@@ -1258,8 +1259,8 @@ static void intel_sdvo_mode_set(struct intel_encoder *intel_encoder)
 		/* The real mode polarity is set by the SDVO commands, using
 		 * struct intel_sdvo_dtd. */
 		sdvox = SDVO_VSYNC_ACTIVE_HIGH | SDVO_HSYNC_ACTIVE_HIGH;
-		if (!HAS_PCH_SPLIT(dev) && intel_sdvo->is_hdmi)
-			sdvox |= intel_sdvo->color_range;
+		if (!HAS_PCH_SPLIT(dev) && crtc->config.limited_color_range)
+			sdvox |= HDMI_COLOR_RANGE_16_235;
 		if (INTEL_INFO(dev)->gen < 5)
 			sdvox |= SDVO_BORDER_ENABLE;
 	} else {
@@ -1349,6 +1350,8 @@ static void intel_sdvo_get_config(struct intel_encoder *encoder,
 	u8 val;
 	bool ret;
 
+	sdvox = I915_READ(intel_sdvo->sdvo_reg);
+
 	ret = intel_sdvo_get_input_timing(intel_sdvo, &dtd);
 	if (!ret) {
 		/* Some sdvo encoders are not spec compliant and don't
@@ -1377,7 +1380,6 @@ static void intel_sdvo_get_config(struct intel_encoder *encoder,
 	 * other platfroms.
 	 */
 	if (IS_I915G(dev) || IS_I915GM(dev)) {
-		sdvox = I915_READ(intel_sdvo->sdvo_reg);
 		pipe_config->pixel_multiplier =
 			((sdvox & SDVO_PORT_MULTIPLY_MASK)
 			 >> SDVO_PORT_MULTIPLY_SHIFT) + 1;
@@ -1406,6 +1408,15 @@ static void intel_sdvo_get_config(struct intel_encoder *encoder,
 		}
 	}
 
+	if (sdvox & HDMI_COLOR_RANGE_16_235)
+		pipe_config->limited_color_range = true;
+
+	if (intel_sdvo_get_value(intel_sdvo, SDVO_CMD_GET_ENCODE,
+				 &val, 1)) {
+		if (val == SDVO_ENCODE_HDMI)
+			pipe_config->has_hdmi_sink = true;
+	}
+
 	WARN(encoder_pixel_multiplier != pipe_config->pixel_multiplier,
 	     "SDVO pixel multiplier mismatch, port: %i, encoder: %i\n",
 	     pipe_config->pixel_multiplier, encoder_pixel_multiplier);
@@ -1732,7 +1743,7 @@ intel_sdvo_detect(struct drm_connector *connector, bool force)
 	enum drm_connector_status ret;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
-		      connector->base.id, drm_get_connector_name(connector));
+		      connector->base.id, connector->name);
 
 	if (!intel_sdvo_get_value(intel_sdvo,
 				  SDVO_CMD_GET_ATTACHED_DISPLAYS,
@@ -1794,7 +1805,7 @@ static void intel_sdvo_get_ddc_modes(struct drm_connector *connector)
 	struct edid *edid;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
-		      connector->base.id, drm_get_connector_name(connector));
+		      connector->base.id, connector->name);
 
 	/* set the bus switch and get the modes */
 	edid = intel_sdvo_get_edid(connector);
@@ -1892,7 +1903,7 @@ static void intel_sdvo_get_tv_modes(struct drm_connector *connector)
 	int i;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
-		      connector->base.id, drm_get_connector_name(connector));
+		      connector->base.id, connector->name);
 
 	/* Read the list of supported input resolutions for the selected TV
 	 * format.
@@ -1929,7 +1940,7 @@ static void intel_sdvo_get_lvds_modes(struct drm_connector *connector)
 	struct drm_display_mode *newmode;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
-		      connector->base.id, drm_get_connector_name(connector));
+		      connector->base.id, connector->name);
 
 	/*
 	 * Fetch modes from VBT. For SDVO prefer the VBT mode since some
@@ -2999,7 +3010,7 @@ bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg, bool is_sdvob)
 
 	intel_encoder->compute_config = intel_sdvo_compute_config;
 	intel_encoder->disable = intel_disable_sdvo;
-	intel_encoder->mode_set = intel_sdvo_mode_set;
+	intel_encoder->pre_enable = intel_sdvo_pre_enable;
 	intel_encoder->enable = intel_enable_sdvo;
 	intel_encoder->get_hw_state = intel_sdvo_get_hw_state;
 	intel_encoder->get_config = intel_sdvo_get_config;
diff --git a/drivers/gpu/drm/i915/intel_sideband.c b/drivers/gpu/drm/i915/intel_sideband.c
index 0954f132726e..01d841ea3140 100644
--- a/drivers/gpu/drm/i915/intel_sideband.c
+++ b/drivers/gpu/drm/i915/intel_sideband.c
@@ -29,12 +29,21 @@
  * IOSF sideband, see VLV2_SidebandMsg_HAS.docx and
  * VLV_VLV2_PUNIT_HAS_0.8.docx
  */
+
+/* Standard MMIO read, non-posted */
+#define SB_MRD_NP	0x00
+/* Standard MMIO write, non-posted */
+#define SB_MWR_NP	0x01
+/* Private register read, double-word addressing, non-posted */
+#define SB_CRRDDA_NP	0x06
+/* Private register write, double-word addressing, non-posted */
+#define SB_CRWRDA_NP	0x07
+
 static int vlv_sideband_rw(struct drm_i915_private *dev_priv, u32 devfn,
 			   u32 port, u32 opcode, u32 addr, u32 *val)
 {
 	u32 cmd, be = 0xf, bar = 0;
-	bool is_read = (opcode == PUNIT_OPCODE_REG_READ ||
-			opcode == DPIO_OPCODE_REG_READ);
+	bool is_read = (opcode == SB_MRD_NP || opcode == SB_CRRDDA_NP);
 
 	cmd = (devfn << IOSF_DEVFN_SHIFT) | (opcode << IOSF_OPCODE_SHIFT) |
 		(port << IOSF_PORT_SHIFT) | (be << IOSF_BYTE_ENABLES_SHIFT) |
@@ -74,7 +83,7 @@ u32 vlv_punit_read(struct drm_i915_private *dev_priv, u8 addr)
 
 	mutex_lock(&dev_priv->dpio_lock);
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_PUNIT,
-			PUNIT_OPCODE_REG_READ, addr, &val);
+			SB_CRRDDA_NP, addr, &val);
 	mutex_unlock(&dev_priv->dpio_lock);
 
 	return val;
@@ -86,7 +95,7 @@ void vlv_punit_write(struct drm_i915_private *dev_priv, u8 addr, u32 val)
 
 	mutex_lock(&dev_priv->dpio_lock);
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_PUNIT,
-			PUNIT_OPCODE_REG_WRITE, addr, &val);
+			SB_CRWRDA_NP, addr, &val);
 	mutex_unlock(&dev_priv->dpio_lock);
 }
 
@@ -95,7 +104,7 @@ u32 vlv_bunit_read(struct drm_i915_private *dev_priv, u32 reg)
 	u32 val = 0;
 
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_BUNIT,
-			PUNIT_OPCODE_REG_READ, reg, &val);
+			SB_CRRDDA_NP, reg, &val);
 
 	return val;
 }
@@ -103,7 +112,7 @@ u32 vlv_bunit_read(struct drm_i915_private *dev_priv, u32 reg)
 void vlv_bunit_write(struct drm_i915_private *dev_priv, u32 reg, u32 val)
 {
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_BUNIT,
-			PUNIT_OPCODE_REG_WRITE, reg, &val);
+			SB_CRWRDA_NP, reg, &val);
 }
 
 u32 vlv_nc_read(struct drm_i915_private *dev_priv, u8 addr)
@@ -114,7 +123,7 @@ u32 vlv_nc_read(struct drm_i915_private *dev_priv, u8 addr)
 
 	mutex_lock(&dev_priv->dpio_lock);
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_NC,
-			PUNIT_OPCODE_REG_READ, addr, &val);
+			SB_CRRDDA_NP, addr, &val);
 	mutex_unlock(&dev_priv->dpio_lock);
 
 	return val;
@@ -124,56 +133,56 @@ u32 vlv_gpio_nc_read(struct drm_i915_private *dev_priv, u32 reg)
 {
 	u32 val = 0;
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_GPIO_NC,
-			PUNIT_OPCODE_REG_READ, reg, &val);
+			SB_CRRDDA_NP, reg, &val);
 	return val;
 }
 
 void vlv_gpio_nc_write(struct drm_i915_private *dev_priv, u32 reg, u32 val)
 {
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_GPIO_NC,
-			PUNIT_OPCODE_REG_WRITE, reg, &val);
+			SB_CRWRDA_NP, reg, &val);
 }
 
 u32 vlv_cck_read(struct drm_i915_private *dev_priv, u32 reg)
 {
 	u32 val = 0;
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_CCK,
-			PUNIT_OPCODE_REG_READ, reg, &val);
+			SB_CRRDDA_NP, reg, &val);
 	return val;
 }
 
 void vlv_cck_write(struct drm_i915_private *dev_priv, u32 reg, u32 val)
 {
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_CCK,
-			PUNIT_OPCODE_REG_WRITE, reg, &val);
+			SB_CRWRDA_NP, reg, &val);
 }
 
 u32 vlv_ccu_read(struct drm_i915_private *dev_priv, u32 reg)
 {
 	u32 val = 0;
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_CCU,
-			PUNIT_OPCODE_REG_READ, reg, &val);
+			SB_CRRDDA_NP, reg, &val);
 	return val;
 }
 
 void vlv_ccu_write(struct drm_i915_private *dev_priv, u32 reg, u32 val)
 {
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_CCU,
-			PUNIT_OPCODE_REG_WRITE, reg, &val);
+			SB_CRWRDA_NP, reg, &val);
 }
 
 u32 vlv_gps_core_read(struct drm_i915_private *dev_priv, u32 reg)
 {
 	u32 val = 0;
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_GPS_CORE,
-			PUNIT_OPCODE_REG_READ, reg, &val);
+			SB_CRRDDA_NP, reg, &val);
 	return val;
 }
 
 void vlv_gps_core_write(struct drm_i915_private *dev_priv, u32 reg, u32 val)
 {
 	vlv_sideband_rw(dev_priv, PCI_DEVFN(2, 0), IOSF_PORT_GPS_CORE,
-			PUNIT_OPCODE_REG_WRITE, reg, &val);
+			SB_CRWRDA_NP, reg, &val);
 }
 
 u32 vlv_dpio_read(struct drm_i915_private *dev_priv, enum pipe pipe, int reg)
@@ -181,14 +190,22 @@ u32 vlv_dpio_read(struct drm_i915_private *dev_priv, enum pipe pipe, int reg)
 	u32 val = 0;
 
 	vlv_sideband_rw(dev_priv, DPIO_DEVFN, DPIO_PHY_IOSF_PORT(DPIO_PHY(pipe)),
-			DPIO_OPCODE_REG_READ, reg, &val);
+			SB_MRD_NP, reg, &val);
+
+	/*
+	 * FIXME: There might be some registers where all 1's is a valid value,
+	 * so ideally we should check the register offset instead...
+	 */
+	WARN(val == 0xffffffff, "DPIO read pipe %c reg 0x%x == 0x%x\n",
+	     pipe_name(pipe), reg, val);
+
 	return val;
 }
 
 void vlv_dpio_write(struct drm_i915_private *dev_priv, enum pipe pipe, int reg, u32 val)
 {
 	vlv_sideband_rw(dev_priv, DPIO_DEVFN, DPIO_PHY_IOSF_PORT(DPIO_PHY(pipe)),
-			DPIO_OPCODE_REG_WRITE, reg, &val);
+			SB_MWR_NP, reg, &val);
 }
 
 /* SBI access */
@@ -253,13 +270,13 @@ void intel_sbi_write(struct drm_i915_private *dev_priv, u16 reg, u32 value,
 u32 vlv_flisdsi_read(struct drm_i915_private *dev_priv, u32 reg)
 {
 	u32 val = 0;
-	vlv_sideband_rw(dev_priv, DPIO_DEVFN, IOSF_PORT_FLISDSI,
-					DPIO_OPCODE_REG_READ, reg, &val);
+	vlv_sideband_rw(dev_priv, DPIO_DEVFN, IOSF_PORT_FLISDSI, SB_CRRDDA_NP,
+			reg, &val);
 	return val;
 }
 
 void vlv_flisdsi_write(struct drm_i915_private *dev_priv, u32 reg, u32 val)
 {
-	vlv_sideband_rw(dev_priv, DPIO_DEVFN, IOSF_PORT_FLISDSI,
-					DPIO_OPCODE_REG_WRITE, reg, &val);
+	vlv_sideband_rw(dev_priv, DPIO_DEVFN, IOSF_PORT_FLISDSI, SB_CRWRDA_NP,
+			reg, &val);
 }
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index 336ae6c602f2..1b66ddcdfb33 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -37,6 +37,106 @@
 #include <drm/i915_drm.h>
 #include "i915_drv.h"
 
+static int usecs_to_scanlines(const struct drm_display_mode *mode, int usecs)
+{
+	/* paranoia */
+	if (!mode->crtc_htotal)
+		return 1;
+
+	return DIV_ROUND_UP(usecs * mode->crtc_clock, 1000 * mode->crtc_htotal);
+}
+
+static bool intel_pipe_update_start(struct intel_crtc *crtc, uint32_t *start_vbl_count)
+{
+	struct drm_device *dev = crtc->base.dev;
+	const struct drm_display_mode *mode = &crtc->config.adjusted_mode;
+	enum pipe pipe = crtc->pipe;
+	long timeout = msecs_to_jiffies_timeout(1);
+	int scanline, min, max, vblank_start;
+	DEFINE_WAIT(wait);
+
+	WARN_ON(!drm_modeset_is_locked(&crtc->base.mutex));
+
+	vblank_start = mode->crtc_vblank_start;
+	if (mode->flags & DRM_MODE_FLAG_INTERLACE)
+		vblank_start = DIV_ROUND_UP(vblank_start, 2);
+
+	/* FIXME needs to be calibrated sensibly */
+	min = vblank_start - usecs_to_scanlines(mode, 100);
+	max = vblank_start - 1;
+
+	if (min <= 0 || max <= 0)
+		return false;
+
+	if (WARN_ON(drm_vblank_get(dev, pipe)))
+		return false;
+
+	local_irq_disable();
+
+	trace_i915_pipe_update_start(crtc, min, max);
+
+	for (;;) {
+		/*
+		 * prepare_to_wait() has a memory barrier, which guarantees
+		 * other CPUs can see the task state update by the time we
+		 * read the scanline.
+		 */
+		prepare_to_wait(&crtc->vbl_wait, &wait, TASK_UNINTERRUPTIBLE);
+
+		scanline = intel_get_crtc_scanline(crtc);
+		if (scanline < min || scanline > max)
+			break;
+
+		if (timeout <= 0) {
+			DRM_ERROR("Potential atomic update failure on pipe %c\n",
+				  pipe_name(crtc->pipe));
+			break;
+		}
+
+		local_irq_enable();
+
+		timeout = schedule_timeout(timeout);
+
+		local_irq_disable();
+	}
+
+	finish_wait(&crtc->vbl_wait, &wait);
+
+	drm_vblank_put(dev, pipe);
+
+	*start_vbl_count = dev->driver->get_vblank_counter(dev, pipe);
+
+	trace_i915_pipe_update_vblank_evaded(crtc, min, max, *start_vbl_count);
+
+	return true;
+}
+
+static void intel_pipe_update_end(struct intel_crtc *crtc, u32 start_vbl_count)
+{
+	struct drm_device *dev = crtc->base.dev;
+	enum pipe pipe = crtc->pipe;
+	u32 end_vbl_count = dev->driver->get_vblank_counter(dev, pipe);
+
+	trace_i915_pipe_update_end(crtc, end_vbl_count);
+
+	local_irq_enable();
+
+	if (start_vbl_count != end_vbl_count)
+		DRM_ERROR("Atomic update failure on pipe %c (start=%u end=%u)\n",
+			  pipe_name(pipe), start_vbl_count, end_vbl_count);
+}
+
+static void intel_update_primary_plane(struct intel_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	int reg = DSPCNTR(crtc->plane);
+
+	if (crtc->primary_enabled)
+		I915_WRITE(reg, I915_READ(reg) | DISPLAY_PLANE_ENABLE);
+	else
+		I915_WRITE(reg, I915_READ(reg) & ~DISPLAY_PLANE_ENABLE);
+}
+
 static void
 vlv_update_plane(struct drm_plane *dplane, struct drm_crtc *crtc,
 		 struct drm_framebuffer *fb,
@@ -48,11 +148,14 @@ vlv_update_plane(struct drm_plane *dplane, struct drm_crtc *crtc,
 	struct drm_device *dev = dplane->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_plane *intel_plane = to_intel_plane(dplane);
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_plane->pipe;
 	int plane = intel_plane->plane;
 	u32 sprctl;
 	unsigned long sprsurf_offset, linear_offset;
 	int pixel_size = drm_format_plane_cpp(fb->pixel_format, 0);
+	u32 start_vbl_count;
+	bool atomic_update;
 
 	sprctl = I915_READ(SPCNTR(pipe, plane));
 
@@ -131,6 +234,10 @@ vlv_update_plane(struct drm_plane *dplane, struct drm_crtc *crtc,
 							fb->pitches[0]);
 	linear_offset -= sprsurf_offset;
 
+	atomic_update = intel_pipe_update_start(intel_crtc, &start_vbl_count);
+
+	intel_update_primary_plane(intel_crtc);
+
 	I915_WRITE(SPSTRIDE(pipe, plane), fb->pitches[0]);
 	I915_WRITE(SPPOS(pipe, plane), (crtc_y << 16) | crtc_x);
 
@@ -143,7 +250,11 @@ vlv_update_plane(struct drm_plane *dplane, struct drm_crtc *crtc,
 	I915_WRITE(SPCNTR(pipe, plane), sprctl);
 	I915_WRITE(SPSURF(pipe, plane), i915_gem_obj_ggtt_offset(obj) +
 		   sprsurf_offset);
-	POSTING_READ(SPSURF(pipe, plane));
+
+	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
+
+	if (atomic_update)
+		intel_pipe_update_end(intel_crtc, start_vbl_count);
 }
 
 static void
@@ -152,14 +263,25 @@ vlv_disable_plane(struct drm_plane *dplane, struct drm_crtc *crtc)
 	struct drm_device *dev = dplane->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_plane *intel_plane = to_intel_plane(dplane);
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_plane->pipe;
 	int plane = intel_plane->plane;
+	u32 start_vbl_count;
+	bool atomic_update;
+
+	atomic_update = intel_pipe_update_start(intel_crtc, &start_vbl_count);
+
+	intel_update_primary_plane(intel_crtc);
 
 	I915_WRITE(SPCNTR(pipe, plane), I915_READ(SPCNTR(pipe, plane)) &
 		   ~SP_ENABLE);
 	/* Activate double buffered register update */
 	I915_WRITE(SPSURF(pipe, plane), 0);
-	POSTING_READ(SPSURF(pipe, plane));
+
+	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
+
+	if (atomic_update)
+		intel_pipe_update_end(intel_crtc, start_vbl_count);
 
 	intel_update_sprite_watermarks(dplane, crtc, 0, 0, false, false);
 }
@@ -226,10 +348,13 @@ ivb_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	struct drm_device *dev = plane->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_plane *intel_plane = to_intel_plane(plane);
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_plane->pipe;
 	u32 sprctl, sprscale = 0;
 	unsigned long sprsurf_offset, linear_offset;
 	int pixel_size = drm_format_plane_cpp(fb->pixel_format, 0);
+	u32 start_vbl_count;
+	bool atomic_update;
 
 	sprctl = I915_READ(SPRCTL(pipe));
 
@@ -299,6 +424,10 @@ ivb_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 					       pixel_size, fb->pitches[0]);
 	linear_offset -= sprsurf_offset;
 
+	atomic_update = intel_pipe_update_start(intel_crtc, &start_vbl_count);
+
+	intel_update_primary_plane(intel_crtc);
+
 	I915_WRITE(SPRSTRIDE(pipe), fb->pitches[0]);
 	I915_WRITE(SPRPOS(pipe), (crtc_y << 16) | crtc_x);
 
@@ -317,7 +446,11 @@ ivb_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	I915_WRITE(SPRCTL(pipe), sprctl);
 	I915_WRITE(SPRSURF(pipe),
 		   i915_gem_obj_ggtt_offset(obj) + sprsurf_offset);
-	POSTING_READ(SPRSURF(pipe));
+
+	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
+
+	if (atomic_update)
+		intel_pipe_update_end(intel_crtc, start_vbl_count);
 }
 
 static void
@@ -326,7 +459,14 @@ ivb_disable_plane(struct drm_plane *plane, struct drm_crtc *crtc)
 	struct drm_device *dev = plane->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_plane *intel_plane = to_intel_plane(plane);
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_plane->pipe;
+	u32 start_vbl_count;
+	bool atomic_update;
+
+	atomic_update = intel_pipe_update_start(intel_crtc, &start_vbl_count);
+
+	intel_update_primary_plane(intel_crtc);
 
 	I915_WRITE(SPRCTL(pipe), I915_READ(SPRCTL(pipe)) & ~SPRITE_ENABLE);
 	/* Can't leave the scaler enabled... */
@@ -334,7 +474,11 @@ ivb_disable_plane(struct drm_plane *plane, struct drm_crtc *crtc)
 		I915_WRITE(SPRSCALE(pipe), 0);
 	/* Activate double buffered register update */
 	I915_WRITE(SPRSURF(pipe), 0);
-	POSTING_READ(SPRSURF(pipe));
+
+	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
+
+	if (atomic_update)
+		intel_pipe_update_end(intel_crtc, start_vbl_count);
 
 	/*
 	 * Avoid underruns when disabling the sprite.
@@ -410,10 +554,13 @@ ilk_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	struct drm_device *dev = plane->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_plane *intel_plane = to_intel_plane(plane);
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_plane->pipe;
 	unsigned long dvssurf_offset, linear_offset;
 	u32 dvscntr, dvsscale;
 	int pixel_size = drm_format_plane_cpp(fb->pixel_format, 0);
+	u32 start_vbl_count;
+	bool atomic_update;
 
 	dvscntr = I915_READ(DVSCNTR(pipe));
 
@@ -478,6 +625,10 @@ ilk_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 					       pixel_size, fb->pitches[0]);
 	linear_offset -= dvssurf_offset;
 
+	atomic_update = intel_pipe_update_start(intel_crtc, &start_vbl_count);
+
+	intel_update_primary_plane(intel_crtc);
+
 	I915_WRITE(DVSSTRIDE(pipe), fb->pitches[0]);
 	I915_WRITE(DVSPOS(pipe), (crtc_y << 16) | crtc_x);
 
@@ -491,7 +642,11 @@ ilk_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	I915_WRITE(DVSCNTR(pipe), dvscntr);
 	I915_WRITE(DVSSURF(pipe),
 		   i915_gem_obj_ggtt_offset(obj) + dvssurf_offset);
-	POSTING_READ(DVSSURF(pipe));
+
+	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
+
+	if (atomic_update)
+		intel_pipe_update_end(intel_crtc, start_vbl_count);
 }
 
 static void
@@ -500,14 +655,25 @@ ilk_disable_plane(struct drm_plane *plane, struct drm_crtc *crtc)
 	struct drm_device *dev = plane->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_plane *intel_plane = to_intel_plane(plane);
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_plane->pipe;
+	u32 start_vbl_count;
+	bool atomic_update;
+
+	atomic_update = intel_pipe_update_start(intel_crtc, &start_vbl_count);
+
+	intel_update_primary_plane(intel_crtc);
 
 	I915_WRITE(DVSCNTR(pipe), I915_READ(DVSCNTR(pipe)) & ~DVS_ENABLE);
 	/* Disable the scaler */
 	I915_WRITE(DVSSCALE(pipe), 0);
 	/* Flush double buffered register updates */
 	I915_WRITE(DVSSURF(pipe), 0);
-	POSTING_READ(DVSSURF(pipe));
+
+	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
+
+	if (atomic_update)
+		intel_pipe_update_end(intel_crtc, start_vbl_count);
 
 	/*
 	 * Avoid underruns when disabling the sprite.
@@ -519,20 +685,10 @@ ilk_disable_plane(struct drm_plane *plane, struct drm_crtc *crtc)
 }
 
 static void
-intel_enable_primary(struct drm_crtc *crtc)
+intel_post_enable_primary(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	int reg = DSPCNTR(intel_crtc->plane);
-
-	if (intel_crtc->primary_enabled)
-		return;
-
-	intel_crtc->primary_enabled = true;
-
-	I915_WRITE(reg, I915_READ(reg) | DISPLAY_PLANE_ENABLE);
-	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
 
 	/*
 	 * FIXME IPS should be fine as long as one plane is
@@ -540,10 +696,7 @@ intel_enable_primary(struct drm_crtc *crtc)
 	 * when going from primary only to sprite only and vice
 	 * versa.
 	 */
-	if (intel_crtc->config.ips_enabled) {
-		intel_wait_for_vblank(dev, intel_crtc->pipe);
-		hsw_enable_ips(intel_crtc);
-	}
+	hsw_enable_ips(intel_crtc);
 
 	mutex_lock(&dev->struct_mutex);
 	intel_update_fbc(dev);
@@ -551,17 +704,11 @@ intel_enable_primary(struct drm_crtc *crtc)
 }
 
 static void
-intel_disable_primary(struct drm_crtc *crtc)
+intel_pre_disable_primary(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	int reg = DSPCNTR(intel_crtc->plane);
-
-	if (!intel_crtc->primary_enabled)
-		return;
-
-	intel_crtc->primary_enabled = false;
 
 	mutex_lock(&dev->struct_mutex);
 	if (dev_priv->fbc.plane == intel_crtc->plane)
@@ -575,9 +722,6 @@ intel_disable_primary(struct drm_crtc *crtc)
 	 * versa.
 	 */
 	hsw_disable_ips(intel_crtc);
-
-	I915_WRITE(reg, I915_READ(reg) & ~DISPLAY_PLANE_ENABLE);
-	intel_flush_primary_plane(dev_priv, intel_crtc->plane);
 }
 
 static int
@@ -671,7 +815,7 @@ intel_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	struct drm_i915_gem_object *obj = intel_fb->obj;
 	struct drm_i915_gem_object *old_obj = intel_plane->obj;
 	int ret;
-	bool disable_primary = false;
+	bool primary_enabled;
 	bool visible;
 	int hscale, vscale;
 	int max_scale, min_scale;
@@ -842,8 +986,8 @@ intel_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	 * If the sprite is completely covering the primary plane,
 	 * we can disable the primary and save power.
 	 */
-	disable_primary = drm_rect_equals(&dst, &clip) && !colorkey_enabled(intel_plane);
-	WARN_ON(disable_primary && !visible && intel_crtc->active);
+	primary_enabled = !drm_rect_equals(&dst, &clip) || colorkey_enabled(intel_plane);
+	WARN_ON(!primary_enabled && !visible && intel_crtc->active);
 
 	mutex_lock(&dev->struct_mutex);
 
@@ -870,12 +1014,15 @@ intel_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	intel_plane->obj = obj;
 
 	if (intel_crtc->active) {
-		/*
-		 * Be sure to re-enable the primary before the sprite is no longer
-		 * covering it fully.
-		 */
-		if (!disable_primary)
-			intel_enable_primary(crtc);
+		bool primary_was_enabled = intel_crtc->primary_enabled;
+
+		intel_crtc->primary_enabled = primary_enabled;
+
+		if (primary_was_enabled != primary_enabled)
+			intel_crtc_wait_for_pending_flips(crtc);
+
+		if (primary_was_enabled && !primary_enabled)
+			intel_pre_disable_primary(crtc);
 
 		if (visible)
 			intel_plane->update_plane(plane, crtc, fb, obj,
@@ -884,8 +1031,8 @@ intel_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 		else
 			intel_plane->disable_plane(plane, crtc);
 
-		if (disable_primary)
-			intel_disable_primary(crtc);
+		if (!primary_was_enabled && primary_enabled)
+			intel_post_enable_primary(crtc);
 	}
 
 	/* Unpin old obj after new one is active to avoid ugliness */
@@ -923,8 +1070,14 @@ intel_disable_plane(struct drm_plane *plane)
 	intel_crtc = to_intel_crtc(plane->crtc);
 
 	if (intel_crtc->active) {
-		intel_enable_primary(plane->crtc);
+		bool primary_was_enabled = intel_crtc->primary_enabled;
+
+		intel_crtc->primary_enabled = true;
+
 		intel_plane->disable_plane(plane, plane->crtc);
+
+		if (!primary_was_enabled && intel_crtc->primary_enabled)
+			intel_post_enable_primary(plane->crtc);
 	}
 
 	if (intel_plane->obj) {
diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c
index bafe92e317d5..67c6c9a2eb1c 100644
--- a/drivers/gpu/drm/i915/intel_tv.c
+++ b/drivers/gpu/drm/i915/intel_tv.c
@@ -934,7 +934,86 @@ intel_tv_compute_config(struct intel_encoder *encoder,
 	return true;
 }
 
-static void intel_tv_mode_set(struct intel_encoder *encoder)
+static void
+set_tv_mode_timings(struct drm_i915_private *dev_priv,
+		    const struct tv_mode *tv_mode,
+		    bool burst_ena)
+{
+	u32 hctl1, hctl2, hctl3;
+	u32 vctl1, vctl2, vctl3, vctl4, vctl5, vctl6, vctl7;
+
+	hctl1 = (tv_mode->hsync_end << TV_HSYNC_END_SHIFT) |
+		(tv_mode->htotal << TV_HTOTAL_SHIFT);
+
+	hctl2 = (tv_mode->hburst_start << 16) |
+		(tv_mode->hburst_len << TV_HBURST_LEN_SHIFT);
+
+	if (burst_ena)
+		hctl2 |= TV_BURST_ENA;
+
+	hctl3 = (tv_mode->hblank_start << TV_HBLANK_START_SHIFT) |
+		(tv_mode->hblank_end << TV_HBLANK_END_SHIFT);
+
+	vctl1 = (tv_mode->nbr_end << TV_NBR_END_SHIFT) |
+		(tv_mode->vi_end_f1 << TV_VI_END_F1_SHIFT) |
+		(tv_mode->vi_end_f2 << TV_VI_END_F2_SHIFT);
+
+	vctl2 = (tv_mode->vsync_len << TV_VSYNC_LEN_SHIFT) |
+		(tv_mode->vsync_start_f1 << TV_VSYNC_START_F1_SHIFT) |
+		(tv_mode->vsync_start_f2 << TV_VSYNC_START_F2_SHIFT);
+
+	vctl3 = (tv_mode->veq_len << TV_VEQ_LEN_SHIFT) |
+		(tv_mode->veq_start_f1 << TV_VEQ_START_F1_SHIFT) |
+		(tv_mode->veq_start_f2 << TV_VEQ_START_F2_SHIFT);
+
+	if (tv_mode->veq_ena)
+		vctl3 |= TV_EQUAL_ENA;
+
+	vctl4 = (tv_mode->vburst_start_f1 << TV_VBURST_START_F1_SHIFT) |
+		(tv_mode->vburst_end_f1 << TV_VBURST_END_F1_SHIFT);
+
+	vctl5 = (tv_mode->vburst_start_f2 << TV_VBURST_START_F2_SHIFT) |
+		(tv_mode->vburst_end_f2 << TV_VBURST_END_F2_SHIFT);
+
+	vctl6 = (tv_mode->vburst_start_f3 << TV_VBURST_START_F3_SHIFT) |
+		(tv_mode->vburst_end_f3 << TV_VBURST_END_F3_SHIFT);
+
+	vctl7 = (tv_mode->vburst_start_f4 << TV_VBURST_START_F4_SHIFT) |
+		(tv_mode->vburst_end_f4 << TV_VBURST_END_F4_SHIFT);
+
+	I915_WRITE(TV_H_CTL_1, hctl1);
+	I915_WRITE(TV_H_CTL_2, hctl2);
+	I915_WRITE(TV_H_CTL_3, hctl3);
+	I915_WRITE(TV_V_CTL_1, vctl1);
+	I915_WRITE(TV_V_CTL_2, vctl2);
+	I915_WRITE(TV_V_CTL_3, vctl3);
+	I915_WRITE(TV_V_CTL_4, vctl4);
+	I915_WRITE(TV_V_CTL_5, vctl5);
+	I915_WRITE(TV_V_CTL_6, vctl6);
+	I915_WRITE(TV_V_CTL_7, vctl7);
+}
+
+static void set_color_conversion(struct drm_i915_private *dev_priv,
+				 const struct color_conversion *color_conversion)
+{
+	if (!color_conversion)
+		return;
+
+	I915_WRITE(TV_CSC_Y, (color_conversion->ry << 16) |
+		   color_conversion->gy);
+	I915_WRITE(TV_CSC_Y2, (color_conversion->by << 16) |
+		   color_conversion->ay);
+	I915_WRITE(TV_CSC_U, (color_conversion->ru << 16) |
+		   color_conversion->gu);
+	I915_WRITE(TV_CSC_U2, (color_conversion->bu << 16) |
+		   color_conversion->au);
+	I915_WRITE(TV_CSC_V, (color_conversion->rv << 16) |
+		   color_conversion->gv);
+	I915_WRITE(TV_CSC_V2, (color_conversion->bv << 16) |
+		   color_conversion->av);
+}
+
+static void intel_tv_pre_enable(struct intel_encoder *encoder)
 {
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -942,14 +1021,13 @@ static void intel_tv_mode_set(struct intel_encoder *encoder)
 	struct intel_tv *intel_tv = enc_to_tv(encoder);
 	const struct tv_mode *tv_mode = intel_tv_mode_find(intel_tv);
 	u32 tv_ctl;
-	u32 hctl1, hctl2, hctl3;
-	u32 vctl1, vctl2, vctl3, vctl4, vctl5, vctl6, vctl7;
 	u32 scctl1, scctl2, scctl3;
 	int i, j;
 	const struct video_levels *video_levels;
 	const struct color_conversion *color_conversion;
 	bool burst_ena;
-	int pipe = intel_crtc->pipe;
+	int xpos = 0x0, ypos = 0x0;
+	unsigned int xsize, ysize;
 
 	if (!tv_mode)
 		return;	/* can't happen (mode_prepare prevents this) */
@@ -982,44 +1060,6 @@ static void intel_tv_mode_set(struct intel_encoder *encoder)
 		burst_ena = tv_mode->burst_ena;
 		break;
 	}
-	hctl1 = (tv_mode->hsync_end << TV_HSYNC_END_SHIFT) |
-		(tv_mode->htotal << TV_HTOTAL_SHIFT);
-
-	hctl2 = (tv_mode->hburst_start << 16) |
-		(tv_mode->hburst_len << TV_HBURST_LEN_SHIFT);
-
-	if (burst_ena)
-		hctl2 |= TV_BURST_ENA;
-
-	hctl3 = (tv_mode->hblank_start << TV_HBLANK_START_SHIFT) |
-		(tv_mode->hblank_end << TV_HBLANK_END_SHIFT);
-
-	vctl1 = (tv_mode->nbr_end << TV_NBR_END_SHIFT) |
-		(tv_mode->vi_end_f1 << TV_VI_END_F1_SHIFT) |
-		(tv_mode->vi_end_f2 << TV_VI_END_F2_SHIFT);
-
-	vctl2 = (tv_mode->vsync_len << TV_VSYNC_LEN_SHIFT) |
-		(tv_mode->vsync_start_f1 << TV_VSYNC_START_F1_SHIFT) |
-		(tv_mode->vsync_start_f2 << TV_VSYNC_START_F2_SHIFT);
-
-	vctl3 = (tv_mode->veq_len << TV_VEQ_LEN_SHIFT) |
-		(tv_mode->veq_start_f1 << TV_VEQ_START_F1_SHIFT) |
-		(tv_mode->veq_start_f2 << TV_VEQ_START_F2_SHIFT);
-
-	if (tv_mode->veq_ena)
-		vctl3 |= TV_EQUAL_ENA;
-
-	vctl4 = (tv_mode->vburst_start_f1 << TV_VBURST_START_F1_SHIFT) |
-		(tv_mode->vburst_end_f1 << TV_VBURST_END_F1_SHIFT);
-
-	vctl5 = (tv_mode->vburst_start_f2 << TV_VBURST_START_F2_SHIFT) |
-		(tv_mode->vburst_end_f2 << TV_VBURST_END_F2_SHIFT);
-
-	vctl6 = (tv_mode->vburst_start_f3 << TV_VBURST_START_F3_SHIFT) |
-		(tv_mode->vburst_end_f3 << TV_VBURST_END_F3_SHIFT);
-
-	vctl7 = (tv_mode->vburst_start_f4 << TV_VBURST_START_F4_SHIFT) |
-		(tv_mode->vburst_end_f4 << TV_VBURST_END_F4_SHIFT);
 
 	if (intel_crtc->pipe == 1)
 		tv_ctl |= TV_ENC_PIPEB_SELECT;
@@ -1051,37 +1091,16 @@ static void intel_tv_mode_set(struct intel_encoder *encoder)
 		tv_mode->dda3_inc << TV_SCDDA3_INC_SHIFT;
 
 	/* Enable two fixes for the chips that need them. */
-	if (dev->pdev->device < 0x2772)
+	if (IS_I915GM(dev))
 		tv_ctl |= TV_ENC_C0_FIX | TV_ENC_SDP_FIX;
 
-	I915_WRITE(TV_H_CTL_1, hctl1);
-	I915_WRITE(TV_H_CTL_2, hctl2);
-	I915_WRITE(TV_H_CTL_3, hctl3);
-	I915_WRITE(TV_V_CTL_1, vctl1);
-	I915_WRITE(TV_V_CTL_2, vctl2);
-	I915_WRITE(TV_V_CTL_3, vctl3);
-	I915_WRITE(TV_V_CTL_4, vctl4);
-	I915_WRITE(TV_V_CTL_5, vctl5);
-	I915_WRITE(TV_V_CTL_6, vctl6);
-	I915_WRITE(TV_V_CTL_7, vctl7);
+	set_tv_mode_timings(dev_priv, tv_mode, burst_ena);
+
 	I915_WRITE(TV_SC_CTL_1, scctl1);
 	I915_WRITE(TV_SC_CTL_2, scctl2);
 	I915_WRITE(TV_SC_CTL_3, scctl3);
 
-	if (color_conversion) {
-		I915_WRITE(TV_CSC_Y, (color_conversion->ry << 16) |
-			   color_conversion->gy);
-		I915_WRITE(TV_CSC_Y2, (color_conversion->by << 16) |
-			   color_conversion->ay);
-		I915_WRITE(TV_CSC_U, (color_conversion->ru << 16) |
-			   color_conversion->gu);
-		I915_WRITE(TV_CSC_U2, (color_conversion->bu << 16) |
-			   color_conversion->au);
-		I915_WRITE(TV_CSC_V, (color_conversion->rv << 16) |
-			   color_conversion->gv);
-		I915_WRITE(TV_CSC_V2, (color_conversion->bv << 16) |
-			   color_conversion->av);
-	}
+	set_color_conversion(dev_priv, color_conversion);
 
 	if (INTEL_INFO(dev)->gen >= 4)
 		I915_WRITE(TV_CLR_KNOBS, 0x00404000);
@@ -1092,46 +1111,25 @@ static void intel_tv_mode_set(struct intel_encoder *encoder)
 		I915_WRITE(TV_CLR_LEVEL,
 			   ((video_levels->black << TV_BLACK_LEVEL_SHIFT) |
 			    (video_levels->blank << TV_BLANK_LEVEL_SHIFT)));
-	{
-		int pipeconf_reg = PIPECONF(pipe);
-		int dspcntr_reg = DSPCNTR(intel_crtc->plane);
-		int pipeconf = I915_READ(pipeconf_reg);
-		int dspcntr = I915_READ(dspcntr_reg);
-		int xpos = 0x0, ypos = 0x0;
-		unsigned int xsize, ysize;
-		/* Pipe must be off here */
-		I915_WRITE(dspcntr_reg, dspcntr & ~DISPLAY_PLANE_ENABLE);
-		intel_flush_primary_plane(dev_priv, intel_crtc->plane);
-
-		/* Wait for vblank for the disable to take effect */
-		if (IS_GEN2(dev))
-			intel_wait_for_vblank(dev, intel_crtc->pipe);
-
-		I915_WRITE(pipeconf_reg, pipeconf & ~PIPECONF_ENABLE);
-		/* Wait for vblank for the disable to take effect. */
-		intel_wait_for_pipe_off(dev, intel_crtc->pipe);
-
-		/* Filter ctl must be set before TV_WIN_SIZE */
-		I915_WRITE(TV_FILTER_CTL_1, TV_AUTO_SCALE);
-		xsize = tv_mode->hblank_start - tv_mode->hblank_end;
-		if (tv_mode->progressive)
-			ysize = tv_mode->nbr_end + 1;
-		else
-			ysize = 2*tv_mode->nbr_end + 1;
-
-		xpos += intel_tv->margin[TV_MARGIN_LEFT];
-		ypos += intel_tv->margin[TV_MARGIN_TOP];
-		xsize -= (intel_tv->margin[TV_MARGIN_LEFT] +
-			  intel_tv->margin[TV_MARGIN_RIGHT]);
-		ysize -= (intel_tv->margin[TV_MARGIN_TOP] +
-			  intel_tv->margin[TV_MARGIN_BOTTOM]);
-		I915_WRITE(TV_WIN_POS, (xpos<<16)|ypos);
-		I915_WRITE(TV_WIN_SIZE, (xsize<<16)|ysize);
-
-		I915_WRITE(pipeconf_reg, pipeconf);
-		I915_WRITE(dspcntr_reg, dspcntr);
-		intel_flush_primary_plane(dev_priv, intel_crtc->plane);
-	}
+
+	assert_pipe_disabled(dev_priv, intel_crtc->pipe);
+
+	/* Filter ctl must be set before TV_WIN_SIZE */
+	I915_WRITE(TV_FILTER_CTL_1, TV_AUTO_SCALE);
+	xsize = tv_mode->hblank_start - tv_mode->hblank_end;
+	if (tv_mode->progressive)
+		ysize = tv_mode->nbr_end + 1;
+	else
+		ysize = 2*tv_mode->nbr_end + 1;
+
+	xpos += intel_tv->margin[TV_MARGIN_LEFT];
+	ypos += intel_tv->margin[TV_MARGIN_TOP];
+	xsize -= (intel_tv->margin[TV_MARGIN_LEFT] +
+		  intel_tv->margin[TV_MARGIN_RIGHT]);
+	ysize -= (intel_tv->margin[TV_MARGIN_TOP] +
+		  intel_tv->margin[TV_MARGIN_BOTTOM]);
+	I915_WRITE(TV_WIN_POS, (xpos<<16)|ypos);
+	I915_WRITE(TV_WIN_SIZE, (xsize<<16)|ysize);
 
 	j = 0;
 	for (i = 0; i < 60; i++)
@@ -1316,17 +1314,18 @@ intel_tv_detect(struct drm_connector *connector, bool force)
 	int type;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s] force=%d\n",
-		      connector->base.id, drm_get_connector_name(connector),
+		      connector->base.id, connector->name,
 		      force);
 
 	mode = reported_modes[0];
 
 	if (force) {
 		struct intel_load_detect_pipe tmp;
+		struct drm_modeset_acquire_ctx ctx;
 
-		if (intel_get_load_detect_pipe(connector, &mode, &tmp)) {
+		if (intel_get_load_detect_pipe(connector, &mode, &tmp, &ctx)) {
 			type = intel_tv_detect_type(intel_tv, connector);
-			intel_release_load_detect_pipe(connector, &tmp);
+			intel_release_load_detect_pipe(connector, &tmp, &ctx);
 		} else
 			return connector_status_unknown;
 	} else
@@ -1634,7 +1633,7 @@ intel_tv_init(struct drm_device *dev)
 
 	intel_encoder->compute_config = intel_tv_compute_config;
 	intel_encoder->get_config = intel_tv_get_config;
-	intel_encoder->mode_set = intel_tv_mode_set;
+	intel_encoder->pre_enable = intel_tv_pre_enable;
 	intel_encoder->enable = intel_enable_tv;
 	intel_encoder->disable = intel_disable_tv;
 	intel_encoder->get_hw_state = intel_tv_get_hw_state;
diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index d0c75779d3f6..79cba593df0d 100644
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -255,8 +255,7 @@ static void __vlv_force_wake_put(struct drm_i915_private *dev_priv,
 
 }
 
-void vlv_force_wake_get(struct drm_i915_private *dev_priv,
-						int fw_engine)
+static void vlv_force_wake_get(struct drm_i915_private *dev_priv, int fw_engine)
 {
 	unsigned long irqflags;
 
@@ -275,8 +274,7 @@ void vlv_force_wake_get(struct drm_i915_private *dev_priv,
 	spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
 }
 
-void vlv_force_wake_put(struct drm_i915_private *dev_priv,
-						int fw_engine)
+static void vlv_force_wake_put(struct drm_i915_private *dev_priv, int fw_engine)
 {
 	unsigned long irqflags;
 
@@ -374,7 +372,7 @@ void intel_uncore_early_sanitize(struct drm_device *dev)
 	if (HAS_FPGA_DBG_UNCLAIMED(dev))
 		__raw_i915_write32(dev_priv, FPGA_DBG, FPGA_DBG_RM_NOCLAIM);
 
-	if (IS_HASWELL(dev) &&
+	if ((IS_HASWELL(dev) || IS_BROADWELL(dev)) &&
 	    (__raw_i915_read32(dev_priv, HSW_EDRAM_PRESENT) == 1)) {
 		/* The docs do not explain exactly how the calculation can be
 		 * made. It is somewhat guessable, but for now, it's always
@@ -395,26 +393,8 @@ void intel_uncore_early_sanitize(struct drm_device *dev)
 
 void intel_uncore_sanitize(struct drm_device *dev)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 reg_val;
-
 	/* BIOS often leaves RC6 enabled, but disable it for hw init */
 	intel_disable_gt_powersave(dev);
-
-	/* Turn off power gate, require especially for the BIOS less system */
-	if (IS_VALLEYVIEW(dev)) {
-
-		mutex_lock(&dev_priv->rps.hw_lock);
-		reg_val = vlv_punit_read(dev_priv, PUNIT_REG_PWRGT_STATUS);
-
-		if (reg_val & (PUNIT_PWRGT_PWR_GATE(PUNIT_POWER_WELL_RENDER) |
-			       PUNIT_PWRGT_PWR_GATE(PUNIT_POWER_WELL_MEDIA) |
-			       PUNIT_PWRGT_PWR_GATE(PUNIT_POWER_WELL_DISP2D)))
-			vlv_punit_write(dev_priv, PUNIT_REG_PWRGT_CTRL, 0x0);
-
-		mutex_unlock(&dev_priv->rps.hw_lock);
-
-	}
 }
 
 /*
@@ -488,6 +468,17 @@ void assert_force_wake_inactive(struct drm_i915_private *dev_priv)
 #define NEEDS_FORCE_WAKE(dev_priv, reg) \
 	 ((reg) < 0x40000 && (reg) != FORCEWAKE)
 
+#define FORCEWAKE_VLV_RENDER_RANGE_OFFSET(reg) \
+	(((reg) >= 0x2000 && (reg) < 0x4000) ||\
+	((reg) >= 0x5000 && (reg) < 0x8000) ||\
+	((reg) >= 0xB000 && (reg) < 0x12000) ||\
+	((reg) >= 0x2E000 && (reg) < 0x30000))
+
+#define FORCEWAKE_VLV_MEDIA_RANGE_OFFSET(reg)\
+	(((reg) >= 0x12000 && (reg) < 0x14000) ||\
+	((reg) >= 0x22000 && (reg) < 0x24000) ||\
+	((reg) >= 0x30000 && (reg) < 0x40000))
+
 static void
 ilk_dummy_write(struct drm_i915_private *dev_priv)
 {
@@ -854,12 +845,15 @@ void intel_uncore_fini(struct drm_device *dev)
 	intel_uncore_forcewake_reset(dev, false);
 }
 
+#define GEN_RANGE(l, h) GENMASK(h, l)
+
 static const struct register_whitelist {
 	uint64_t offset;
 	uint32_t size;
-	uint32_t gen_bitmask; /* support gens, 0x10 for 4, 0x30 for 4 and 5, etc. */
+	/* supported gens, 0x10 for 4, 0x30 for 4 and 5, etc. */
+	uint32_t gen_bitmask;
 } whitelist[] = {
-	{ RING_TIMESTAMP(RENDER_RING_BASE), 8, 0x1F0 },
+	{ RING_TIMESTAMP(RENDER_RING_BASE), 8, GEN_RANGE(4, 8) },
 };
 
 int i915_reg_read_ioctl(struct drm_device *dev,
@@ -911,7 +905,7 @@ int i915_get_reset_stats_ioctl(struct drm_device *dev,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_reset_stats *args = data;
 	struct i915_ctx_hang_stats *hs;
-	struct i915_hw_context *ctx;
+	struct intel_context *ctx;
 	int ret;
 
 	if (args->flags || args->pad)
@@ -955,6 +949,9 @@ static int i965_do_reset(struct drm_device *dev)
 {
 	int ret;
 
+	/* FIXME: i965g/gm need a display save/restore for gpu reset. */
+	return -ENODEV;
+
 	/*
 	 * Set the domains we want to reset (GRDOM/bits 2 and 3) as
 	 * well as the reset bit (GR/bit 0).  Setting the GR bit
@@ -966,7 +963,6 @@ static int i965_do_reset(struct drm_device *dev)
 	if (ret)
 		return ret;
 
-	/* We can't reset render&media without also resetting display ... */
 	pci_write_config_byte(dev->pdev, I965_GDRST,
 			      GRDOM_MEDIA | GRDOM_RESET_ENABLE);
 
@@ -979,26 +975,58 @@ static int i965_do_reset(struct drm_device *dev)
 	return 0;
 }
 
+static int g4x_do_reset(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int ret;
+
+	pci_write_config_byte(dev->pdev, I965_GDRST,
+			      GRDOM_RENDER | GRDOM_RESET_ENABLE);
+	ret =  wait_for(i965_reset_complete(dev), 500);
+	if (ret)
+		return ret;
+
+	/* WaVcpClkGateDisableForMediaReset:ctg,elk */
+	I915_WRITE(VDECCLK_GATE_D, I915_READ(VDECCLK_GATE_D) | VCP_UNIT_CLOCK_GATE_DISABLE);
+	POSTING_READ(VDECCLK_GATE_D);
+
+	pci_write_config_byte(dev->pdev, I965_GDRST,
+			      GRDOM_MEDIA | GRDOM_RESET_ENABLE);
+	ret =  wait_for(i965_reset_complete(dev), 500);
+	if (ret)
+		return ret;
+
+	/* WaVcpClkGateDisableForMediaReset:ctg,elk */
+	I915_WRITE(VDECCLK_GATE_D, I915_READ(VDECCLK_GATE_D) & ~VCP_UNIT_CLOCK_GATE_DISABLE);
+	POSTING_READ(VDECCLK_GATE_D);
+
+	pci_write_config_byte(dev->pdev, I965_GDRST, 0);
+
+	return 0;
+}
+
 static int ironlake_do_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 gdrst;
 	int ret;
 
-	gdrst = I915_READ(MCHBAR_MIRROR_BASE + ILK_GDSR);
-	gdrst &= ~GRDOM_MASK;
 	I915_WRITE(MCHBAR_MIRROR_BASE + ILK_GDSR,
-		   gdrst | GRDOM_RENDER | GRDOM_RESET_ENABLE);
-	ret = wait_for(I915_READ(MCHBAR_MIRROR_BASE + ILK_GDSR) & 0x1, 500);
+		   ILK_GRDOM_RENDER | ILK_GRDOM_RESET_ENABLE);
+	ret = wait_for((I915_READ(MCHBAR_MIRROR_BASE + ILK_GDSR) &
+			ILK_GRDOM_RESET_ENABLE) == 0, 500);
 	if (ret)
 		return ret;
 
-	/* We can't reset render&media without also resetting display ... */
-	gdrst = I915_READ(MCHBAR_MIRROR_BASE + ILK_GDSR);
-	gdrst &= ~GRDOM_MASK;
 	I915_WRITE(MCHBAR_MIRROR_BASE + ILK_GDSR,
-		   gdrst | GRDOM_MEDIA | GRDOM_RESET_ENABLE);
-	return wait_for(I915_READ(MCHBAR_MIRROR_BASE + ILK_GDSR) & 0x1, 500);
+		   ILK_GRDOM_MEDIA | ILK_GRDOM_RESET_ENABLE);
+	ret = wait_for((I915_READ(MCHBAR_MIRROR_BASE + ILK_GDSR) &
+			ILK_GRDOM_RESET_ENABLE) == 0, 500);
+	if (ret)
+		return ret;
+
+	I915_WRITE(MCHBAR_MIRROR_BASE + ILK_GDSR, 0);
+
+	return 0;
 }
 
 static int gen6_do_reset(struct drm_device *dev)
@@ -1029,7 +1057,11 @@ int intel_gpu_reset(struct drm_device *dev)
 	case 7:
 	case 6: return gen6_do_reset(dev);
 	case 5: return ironlake_do_reset(dev);
-	case 4: return i965_do_reset(dev);
+	case 4:
+		if (IS_G4X(dev))
+			return g4x_do_reset(dev);
+		else
+			return i965_do_reset(dev);
 	default: return -ENODEV;
 	}
 }
diff --git a/drivers/gpu/drm/mga/mga_ioc32.c b/drivers/gpu/drm/mga/mga_ioc32.c
index 86b4bb804852..729bfd56b55f 100644
--- a/drivers/gpu/drm/mga/mga_ioc32.c
+++ b/drivers/gpu/drm/mga/mga_ioc32.c
@@ -214,7 +214,7 @@ long mga_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 	if (nr < DRM_COMMAND_BASE)
 		return drm_compat_ioctl(filp, cmd, arg);
 
-	if (nr < DRM_COMMAND_BASE + DRM_ARRAY_SIZE(mga_compat_ioctls))
+	if (nr < DRM_COMMAND_BASE + ARRAY_SIZE(mga_compat_ioctls))
 		fn = mga_compat_ioctls[nr - DRM_COMMAND_BASE];
 
 	if (fn != NULL)
diff --git a/drivers/gpu/drm/mga/mga_state.c b/drivers/gpu/drm/mga/mga_state.c
index 314685b7f41f..792f924496fc 100644
--- a/drivers/gpu/drm/mga/mga_state.c
+++ b/drivers/gpu/drm/mga/mga_state.c
@@ -1020,7 +1020,7 @@ static int mga_getparam(struct drm_device *dev, void *data, struct drm_file *fil
 
 	switch (param->param) {
 	case MGA_PARAM_IRQ_NR:
-		value = drm_dev_to_irq(dev);
+		value = dev->pdev->irq;
 		break;
 	case MGA_PARAM_CARD_TYPE:
 		value = dev_priv->chipset;
@@ -1099,4 +1099,4 @@ const struct drm_ioctl_desc mga_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(MGA_DMA_BOOTSTRAP, mga_dma_bootstrap, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
 };
 
-int mga_max_ioctl = DRM_ARRAY_SIZE(mga_ioctls);
+int mga_max_ioctl = ARRAY_SIZE(mga_ioctls);
diff --git a/drivers/gpu/drm/mgag200/mgag200_main.c b/drivers/gpu/drm/mgag200/mgag200_main.c
index 26868e5c55b0..f6b283b8375e 100644
--- a/drivers/gpu/drm/mgag200/mgag200_main.c
+++ b/drivers/gpu/drm/mgag200/mgag200_main.c
@@ -322,17 +322,13 @@ static void mgag200_bo_unref(struct mgag200_bo **bo)
 
 	tbo = &((*bo)->bo);
 	ttm_bo_unref(&tbo);
-	if (tbo == NULL)
-		*bo = NULL;
-
+	*bo = NULL;
 }
 
 void mgag200_gem_free_object(struct drm_gem_object *obj)
 {
 	struct mgag200_bo *mgag200_bo = gem_to_mga_bo(obj);
 
-	if (!mgag200_bo)
-		return;
 	mgag200_bo_unref(&mgag200_bo);
 }
 
diff --git a/drivers/gpu/drm/msm/Kconfig b/drivers/gpu/drm/msm/Kconfig
index b6984971ce0c..f12388967856 100644
--- a/drivers/gpu/drm/msm/Kconfig
+++ b/drivers/gpu/drm/msm/Kconfig
@@ -3,7 +3,7 @@ config DRM_MSM
 	tristate "MSM DRM"
 	depends on DRM
 	depends on MSM_IOMMU
-	depends on ARCH_MSM8960 || (ARM && COMPILE_TEST)
+	depends on ARCH_QCOM || (ARM && COMPILE_TEST)
 	select DRM_KMS_HELPER
 	select SHMEM
 	select TMPFS
diff --git a/drivers/gpu/drm/msm/Makefile b/drivers/gpu/drm/msm/Makefile
index 5e1e6b0cd8ac..93ca49c8df44 100644
--- a/drivers/gpu/drm/msm/Makefile
+++ b/drivers/gpu/drm/msm/Makefile
@@ -34,6 +34,8 @@ msm-y := \
 	msm_gem_submit.o \
 	msm_gpu.o \
 	msm_iommu.o \
+	msm_perf.o \
+	msm_rd.o \
 	msm_ringbuffer.o
 
 msm-$(CONFIG_DRM_MSM_FBDEV) += msm_fbdev.o
diff --git a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c
index f20fbde5dc49..942e09d898a8 100644
--- a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c
@@ -207,11 +207,11 @@ static int a3xx_hw_init(struct msm_gpu *gpu)
 	/* Turn on performance counters: */
 	gpu_write(gpu, REG_A3XX_RBBM_PERFCTR_CTL, 0x01);
 
-	/* Set SP perfcounter 7 to count SP_FS_FULL_ALU_INSTRUCTIONS
-	 * we will use this to augment our hang detection:
-	 */
-	gpu_write(gpu, REG_A3XX_SP_PERFCOUNTER7_SELECT,
-			SP_FS_FULL_ALU_INSTRUCTIONS);
+	/* Enable the perfcntrs that we use.. */
+	for (i = 0; i < gpu->num_perfcntrs; i++) {
+		const struct msm_gpu_perfcntr *perfcntr = &gpu->perfcntrs[i];
+		gpu_write(gpu, perfcntr->select_reg, perfcntr->select_val);
+	}
 
 	gpu_write(gpu, REG_A3XX_RBBM_INT_0_MASK, A3XX_INT0_MASK);
 
@@ -465,6 +465,13 @@ static const struct adreno_gpu_funcs funcs = {
 	},
 };
 
+static const struct msm_gpu_perfcntr perfcntrs[] = {
+	{ REG_A3XX_SP_PERFCOUNTER6_SELECT, REG_A3XX_RBBM_PERFCTR_SP_6_LO,
+			SP_ALU_ACTIVE_CYCLES, "ALUACTIVE" },
+	{ REG_A3XX_SP_PERFCOUNTER7_SELECT, REG_A3XX_RBBM_PERFCTR_SP_7_LO,
+			SP_FS_FULL_ALU_INSTRUCTIONS, "ALUFULL" },
+};
+
 struct msm_gpu *a3xx_gpu_init(struct drm_device *dev)
 {
 	struct a3xx_gpu *a3xx_gpu = NULL;
@@ -504,6 +511,9 @@ struct msm_gpu *a3xx_gpu_init(struct drm_device *dev)
 	DBG("fast_rate=%u, slow_rate=%u, bus_freq=%u",
 			gpu->fast_rate, gpu->slow_rate, gpu->bus_freq);
 
+	gpu->perfcntrs = perfcntrs;
+	gpu->num_perfcntrs = ARRAY_SIZE(perfcntrs);
+
 	ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, config->rev);
 	if (ret)
 		goto fail;
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_connector.c b/drivers/gpu/drm/msm/hdmi/hdmi_connector.c
index 7dedfdd12075..e56a6196867c 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi_connector.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_connector.c
@@ -247,36 +247,49 @@ void hdmi_connector_irq(struct drm_connector *connector)
 	}
 }
 
+static enum drm_connector_status detect_reg(struct hdmi *hdmi)
+{
+	uint32_t hpd_int_status = hdmi_read(hdmi, REG_HDMI_HPD_INT_STATUS);
+	return (hpd_int_status & HDMI_HPD_INT_STATUS_CABLE_DETECTED) ?
+			connector_status_connected : connector_status_disconnected;
+}
+
+static enum drm_connector_status detect_gpio(struct hdmi *hdmi)
+{
+	const struct hdmi_platform_config *config = hdmi->config;
+	return gpio_get_value(config->hpd_gpio) ?
+			connector_status_connected :
+			connector_status_disconnected;
+}
+
 static enum drm_connector_status hdmi_connector_detect(
 		struct drm_connector *connector, bool force)
 {
 	struct hdmi_connector *hdmi_connector = to_hdmi_connector(connector);
 	struct hdmi *hdmi = hdmi_connector->hdmi;
-	const struct hdmi_platform_config *config = hdmi->config;
-	uint32_t hpd_int_status;
+	enum drm_connector_status stat_gpio, stat_reg;
 	int retry = 20;
 
-	hpd_int_status = hdmi_read(hdmi, REG_HDMI_HPD_INT_STATUS);
+	do {
+		stat_gpio = detect_gpio(hdmi);
+		stat_reg  = detect_reg(hdmi);
 
-	/* sense seems to in some cases be momentarily de-asserted, don't
-	 * let that trick us into thinking the monitor is gone:
-	 */
-	while (retry-- && !(hpd_int_status & HDMI_HPD_INT_STATUS_CABLE_DETECTED)) {
-		/* hdmi debounce logic seems to get stuck sometimes,
-		 * read directly the gpio to get a second opinion:
-		 */
-		if (gpio_get_value(config->hpd_gpio)) {
-			DBG("gpio tells us we are connected!");
-			hpd_int_status |= HDMI_HPD_INT_STATUS_CABLE_DETECTED;
+		if (stat_gpio == stat_reg)
 			break;
-		}
+
 		mdelay(10);
-		hpd_int_status = hdmi_read(hdmi, REG_HDMI_HPD_INT_STATUS);
-		DBG("status=%08x", hpd_int_status);
+	} while (--retry);
+
+	/* the status we get from reading gpio seems to be more reliable,
+	 * so trust that one the most if we didn't manage to get hdmi and
+	 * gpio status to agree:
+	 */
+	if (stat_gpio != stat_reg) {
+		DBG("HDMI_HPD_INT_STATUS tells us: %d", stat_reg);
+		DBG("hpd gpio tells us: %d", stat_gpio);
 	}
 
-	return (hpd_int_status & HDMI_HPD_INT_STATUS_CABLE_DETECTED) ?
-			connector_status_connected : connector_status_disconnected;
+	return stat_gpio;
 }
 
 static void hdmi_connector_destroy(struct drm_connector *connector)
@@ -389,7 +402,8 @@ struct drm_connector *hdmi_connector_init(struct hdmi *hdmi)
 			DRM_MODE_CONNECTOR_HDMIA);
 	drm_connector_helper_add(connector, &hdmi_connector_helper_funcs);
 
-	connector->polled = DRM_CONNECTOR_POLL_HPD;
+	connector->polled = DRM_CONNECTOR_POLL_CONNECT |
+			DRM_CONNECTOR_POLL_DISCONNECT;
 
 	connector->interlace_allowed = 1;
 	connector->doublescan_allowed = 0;
diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c
index ef9957dbac94..74cebb51e8c2 100644
--- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c
+++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c
@@ -217,8 +217,6 @@ static void mdp4_crtc_destroy(struct drm_crtc *crtc)
 {
 	struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc);
 
-	mdp4_crtc->plane->funcs->destroy(mdp4_crtc->plane);
-
 	drm_crtc_cleanup(crtc);
 	drm_flip_work_cleanup(&mdp4_crtc->unref_fb_work);
 	drm_flip_work_cleanup(&mdp4_crtc->unref_cursor_work);
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c
index 6ea10bdb6e8f..ebe2e60f3ab1 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c
@@ -195,8 +195,6 @@ static void mdp5_crtc_destroy(struct drm_crtc *crtc)
 {
 	struct mdp5_crtc *mdp5_crtc = to_mdp5_crtc(crtc);
 
-	mdp5_crtc->plane->funcs->destroy(mdp5_crtc->plane);
-
 	drm_crtc_cleanup(crtc);
 	drm_flip_work_cleanup(&mdp5_crtc->unref_fb_work);
 
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
index ee8446c1b5f6..42caf7fcb0b9 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
@@ -280,12 +280,22 @@ struct msm_kms *mdp5_kms_init(struct drm_device *dev)
 		goto fail;
 	}
 
-	ret = get_clk(pdev, &mdp5_kms->axi_clk, "bus_clk") ||
-			get_clk(pdev, &mdp5_kms->ahb_clk, "iface_clk") ||
-			get_clk(pdev, &mdp5_kms->src_clk, "core_clk_src") ||
-			get_clk(pdev, &mdp5_kms->core_clk, "core_clk") ||
-			get_clk(pdev, &mdp5_kms->lut_clk, "lut_clk") ||
-			get_clk(pdev, &mdp5_kms->vsync_clk, "vsync_clk");
+	ret = get_clk(pdev, &mdp5_kms->axi_clk, "bus_clk");
+	if (ret)
+		goto fail;
+	ret = get_clk(pdev, &mdp5_kms->ahb_clk, "iface_clk");
+	if (ret)
+		goto fail;
+	ret = get_clk(pdev, &mdp5_kms->src_clk, "core_clk_src");
+	if (ret)
+		goto fail;
+	ret = get_clk(pdev, &mdp5_kms->core_clk, "core_clk");
+	if (ret)
+		goto fail;
+	ret = get_clk(pdev, &mdp5_kms->lut_clk, "lut_clk");
+	if (ret)
+		goto fail;
+	ret = get_clk(pdev, &mdp5_kms->vsync_clk, "vsync_clk");
 	if (ret)
 		goto fail;
 
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
index 47f7bbb9c15a..f3daec4412ad 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
@@ -85,8 +85,11 @@ static int mdp5_plane_disable(struct drm_plane *plane)
 static void mdp5_plane_destroy(struct drm_plane *plane)
 {
 	struct mdp5_plane *mdp5_plane = to_mdp5_plane(plane);
+	struct msm_drm_private *priv = plane->dev->dev_private;
+
+	if (priv->kms)
+		mdp5_plane_disable(plane);
 
-	mdp5_plane_disable(plane);
 	drm_plane_cleanup(plane);
 
 	kfree(mdp5_plane);
diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c
index f9de156b9e65..0d2562fb681e 100644
--- a/drivers/gpu/drm/msm/msm_drv.c
+++ b/drivers/gpu/drm/msm/msm_drv.c
@@ -220,7 +220,7 @@ static int msm_load(struct drm_device *dev, unsigned long flags)
 		 * is bogus, but non-null if allocation succeeded:
 		 */
 		p = dma_alloc_attrs(dev->dev, size,
-				&priv->vram.paddr, 0, &attrs);
+				&priv->vram.paddr, GFP_KERNEL, &attrs);
 		if (!p) {
 			dev_err(dev->dev, "failed to allocate VRAM\n");
 			priv->vram.paddr = 0;
@@ -288,7 +288,7 @@ static int msm_load(struct drm_device *dev, unsigned long flags)
 	}
 
 	pm_runtime_get_sync(dev->dev);
-	ret = drm_irq_install(dev);
+	ret = drm_irq_install(dev, platform_get_irq(dev->platformdev, 0));
 	pm_runtime_put_sync(dev->dev);
 	if (ret < 0) {
 		dev_err(dev->dev, "failed to install IRQ handler\n");
@@ -299,6 +299,10 @@ static int msm_load(struct drm_device *dev, unsigned long flags)
 	priv->fbdev = msm_fbdev_init(dev);
 #endif
 
+	ret = msm_debugfs_late_init(dev);
+	if (ret)
+		goto fail;
+
 	drm_kms_helper_poll_init(dev);
 
 	return 0;
@@ -382,11 +386,8 @@ static void msm_preclose(struct drm_device *dev, struct drm_file *file)
 static void msm_lastclose(struct drm_device *dev)
 {
 	struct msm_drm_private *priv = dev->dev_private;
-	if (priv->fbdev) {
-		drm_modeset_lock_all(dev);
-		drm_fb_helper_restore_fbdev_mode(priv->fbdev);
-		drm_modeset_unlock_all(dev);
-	}
+	if (priv->fbdev)
+		drm_fb_helper_restore_fbdev_mode_unlocked(priv->fbdev);
 }
 
 static irqreturn_t msm_irq(int irq, void *arg)
@@ -531,6 +532,41 @@ static struct drm_info_list msm_debugfs_list[] = {
 		{ "fb", show_locked, 0, msm_fb_show },
 };
 
+static int late_init_minor(struct drm_minor *minor)
+{
+	int ret;
+
+	if (!minor)
+		return 0;
+
+	ret = msm_rd_debugfs_init(minor);
+	if (ret) {
+		dev_err(minor->dev->dev, "could not install rd debugfs\n");
+		return ret;
+	}
+
+	ret = msm_perf_debugfs_init(minor);
+	if (ret) {
+		dev_err(minor->dev->dev, "could not install perf debugfs\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+int msm_debugfs_late_init(struct drm_device *dev)
+{
+	int ret;
+	ret = late_init_minor(dev->primary);
+	if (ret)
+		return ret;
+	ret = late_init_minor(dev->render);
+	if (ret)
+		return ret;
+	ret = late_init_minor(dev->control);
+	return ret;
+}
+
 static int msm_debugfs_init(struct drm_minor *minor)
 {
 	struct drm_device *dev = minor->dev;
@@ -545,13 +581,17 @@ static int msm_debugfs_init(struct drm_minor *minor)
 		return ret;
 	}
 
-	return ret;
+	return 0;
 }
 
 static void msm_debugfs_cleanup(struct drm_minor *minor)
 {
 	drm_debugfs_remove_files(msm_debugfs_list,
 			ARRAY_SIZE(msm_debugfs_list), minor);
+	if (!minor->dev->dev_private)
+		return;
+	msm_rd_debugfs_cleanup(minor);
+	msm_perf_debugfs_cleanup(minor);
 }
 #endif
 
diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h
index 9d10ee0b5aac..8a2c5fd0893e 100644
--- a/drivers/gpu/drm/msm/msm_drv.h
+++ b/drivers/gpu/drm/msm/msm_drv.h
@@ -33,7 +33,7 @@
 #include <asm/sizes.h>
 
 
-#if defined(CONFIG_COMPILE_TEST) && !defined(CONFIG_ARCH_MSM)
+#if defined(CONFIG_COMPILE_TEST) && !defined(CONFIG_ARCH_QCOM)
 /* stubs we need for compile-test: */
 static inline struct device *msm_iommu_get_ctx(const char *ctx_name)
 {
@@ -55,6 +55,9 @@ static inline struct device *msm_iommu_get_ctx(const char *ctx_name)
 struct msm_kms;
 struct msm_gpu;
 struct msm_mmu;
+struct msm_rd_state;
+struct msm_perf_state;
+struct msm_gem_submit;
 
 #define NUM_DOMAINS 2    /* one for KMS, then one per gpu core (?) */
 
@@ -82,6 +85,9 @@ struct msm_drm_private {
 	uint32_t next_fence, completed_fence;
 	wait_queue_head_t fence_event;
 
+	struct msm_rd_state *rd;
+	struct msm_perf_state *perf;
+
 	/* list of GEM objects: */
 	struct list_head inactive_list;
 
@@ -204,6 +210,15 @@ void __exit hdmi_unregister(void);
 void msm_gem_describe(struct drm_gem_object *obj, struct seq_file *m);
 void msm_gem_describe_objects(struct list_head *list, struct seq_file *m);
 void msm_framebuffer_describe(struct drm_framebuffer *fb, struct seq_file *m);
+int msm_debugfs_late_init(struct drm_device *dev);
+int msm_rd_debugfs_init(struct drm_minor *minor);
+void msm_rd_debugfs_cleanup(struct drm_minor *minor);
+void msm_rd_dump_submit(struct msm_gem_submit *submit);
+int msm_perf_debugfs_init(struct drm_minor *minor);
+void msm_perf_debugfs_cleanup(struct drm_minor *minor);
+#else
+static inline int msm_debugfs_late_init(struct drm_device *dev) { return 0; }
+static inline void msm_rd_dump_submit(struct msm_gem_submit *submit) {}
 #endif
 
 void __iomem *msm_ioremap(struct platform_device *pdev, const char *name,
diff --git a/drivers/gpu/drm/msm/msm_gem.h b/drivers/gpu/drm/msm/msm_gem.h
index 3246bb46c4f2..bfb052688f8e 100644
--- a/drivers/gpu/drm/msm/msm_gem.h
+++ b/drivers/gpu/drm/msm/msm_gem.h
@@ -90,6 +90,7 @@ struct msm_gem_submit {
 		uint32_t type;
 		uint32_t size;  /* in dwords */
 		uint32_t iova;
+		uint32_t idx;   /* cmdstream buffer idx in bos[] */
 	} cmd[MAX_CMDS];
 	struct {
 		uint32_t flags;
diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c
index 1f1f4cffdaed..cd0554f68316 100644
--- a/drivers/gpu/drm/msm/msm_gem_submit.c
+++ b/drivers/gpu/drm/msm/msm_gem_submit.c
@@ -402,6 +402,7 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 		submit->cmd[i].type = submit_cmd.type;
 		submit->cmd[i].size = submit_cmd.size / 4;
 		submit->cmd[i].iova = iova + submit_cmd.submit_offset;
+		submit->cmd[i].idx  = submit_cmd.submit_idx;
 
 		if (submit->valid)
 			continue;
diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c
index 3e667ca1f2b9..c6322197db8c 100644
--- a/drivers/gpu/drm/msm/msm_gpu.c
+++ b/drivers/gpu/drm/msm/msm_gpu.c
@@ -320,6 +320,101 @@ static void hangcheck_handler(unsigned long data)
 }
 
 /*
+ * Performance Counters:
+ */
+
+/* called under perf_lock */
+static int update_hw_cntrs(struct msm_gpu *gpu, uint32_t ncntrs, uint32_t *cntrs)
+{
+	uint32_t current_cntrs[ARRAY_SIZE(gpu->last_cntrs)];
+	int i, n = min(ncntrs, gpu->num_perfcntrs);
+
+	/* read current values: */
+	for (i = 0; i < gpu->num_perfcntrs; i++)
+		current_cntrs[i] = gpu_read(gpu, gpu->perfcntrs[i].sample_reg);
+
+	/* update cntrs: */
+	for (i = 0; i < n; i++)
+		cntrs[i] = current_cntrs[i] - gpu->last_cntrs[i];
+
+	/* save current values: */
+	for (i = 0; i < gpu->num_perfcntrs; i++)
+		gpu->last_cntrs[i] = current_cntrs[i];
+
+	return n;
+}
+
+static void update_sw_cntrs(struct msm_gpu *gpu)
+{
+	ktime_t time;
+	uint32_t elapsed;
+	unsigned long flags;
+
+	spin_lock_irqsave(&gpu->perf_lock, flags);
+	if (!gpu->perfcntr_active)
+		goto out;
+
+	time = ktime_get();
+	elapsed = ktime_to_us(ktime_sub(time, gpu->last_sample.time));
+
+	gpu->totaltime += elapsed;
+	if (gpu->last_sample.active)
+		gpu->activetime += elapsed;
+
+	gpu->last_sample.active = msm_gpu_active(gpu);
+	gpu->last_sample.time = time;
+
+out:
+	spin_unlock_irqrestore(&gpu->perf_lock, flags);
+}
+
+void msm_gpu_perfcntr_start(struct msm_gpu *gpu)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&gpu->perf_lock, flags);
+	/* we could dynamically enable/disable perfcntr registers too.. */
+	gpu->last_sample.active = msm_gpu_active(gpu);
+	gpu->last_sample.time = ktime_get();
+	gpu->activetime = gpu->totaltime = 0;
+	gpu->perfcntr_active = true;
+	update_hw_cntrs(gpu, 0, NULL);
+	spin_unlock_irqrestore(&gpu->perf_lock, flags);
+}
+
+void msm_gpu_perfcntr_stop(struct msm_gpu *gpu)
+{
+	gpu->perfcntr_active = false;
+}
+
+/* returns -errno or # of cntrs sampled */
+int msm_gpu_perfcntr_sample(struct msm_gpu *gpu, uint32_t *activetime,
+		uint32_t *totaltime, uint32_t ncntrs, uint32_t *cntrs)
+{
+	unsigned long flags;
+	int ret;
+
+	spin_lock_irqsave(&gpu->perf_lock, flags);
+
+	if (!gpu->perfcntr_active) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	*activetime = gpu->activetime;
+	*totaltime = gpu->totaltime;
+
+	gpu->activetime = gpu->totaltime = 0;
+
+	ret = update_hw_cntrs(gpu, ncntrs, cntrs);
+
+out:
+	spin_unlock_irqrestore(&gpu->perf_lock, flags);
+
+	return ret;
+}
+
+/*
  * Cmdstream submission/retirement:
  */
 
@@ -361,6 +456,7 @@ void msm_gpu_retire(struct msm_gpu *gpu)
 {
 	struct msm_drm_private *priv = gpu->dev->dev_private;
 	queue_work(priv->wq, &gpu->retire_work);
+	update_sw_cntrs(gpu);
 }
 
 /* add bo's to gpu's ring, and kick gpu: */
@@ -377,6 +473,12 @@ int msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
 
 	inactive_cancel(gpu);
 
+	msm_rd_dump_submit(submit);
+
+	gpu->submitted_fence = submit->fence;
+
+	update_sw_cntrs(gpu);
+
 	ret = gpu->funcs->submit(gpu, submit, ctx);
 	priv->lastctx = ctx;
 
@@ -429,6 +531,9 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev,
 	struct iommu_domain *iommu;
 	int i, ret;
 
+	if (WARN_ON(gpu->num_perfcntrs > ARRAY_SIZE(gpu->last_cntrs)))
+		gpu->num_perfcntrs = ARRAY_SIZE(gpu->last_cntrs);
+
 	gpu->dev = drm;
 	gpu->funcs = funcs;
 	gpu->name = name;
@@ -444,6 +549,8 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev,
 	setup_timer(&gpu->hangcheck_timer, hangcheck_handler,
 			(unsigned long)gpu);
 
+	spin_lock_init(&gpu->perf_lock);
+
 	BUG_ON(ARRAY_SIZE(clk_names) != ARRAY_SIZE(gpu->grp_clks));
 
 	/* Map registers: */
diff --git a/drivers/gpu/drm/msm/msm_gpu.h b/drivers/gpu/drm/msm/msm_gpu.h
index fad27008922f..9b579b792840 100644
--- a/drivers/gpu/drm/msm/msm_gpu.h
+++ b/drivers/gpu/drm/msm/msm_gpu.h
@@ -25,6 +25,7 @@
 #include "msm_ringbuffer.h"
 
 struct msm_gem_submit;
+struct msm_gpu_perfcntr;
 
 /* So far, with hardware that I've seen to date, we can have:
  *  + zero, one, or two z180 2d cores
@@ -64,6 +65,18 @@ struct msm_gpu {
 	struct drm_device *dev;
 	const struct msm_gpu_funcs *funcs;
 
+	/* performance counters (hw & sw): */
+	spinlock_t perf_lock;
+	bool perfcntr_active;
+	struct {
+		bool active;
+		ktime_t time;
+	} last_sample;
+	uint32_t totaltime, activetime;    /* sw counters */
+	uint32_t last_cntrs[5];            /* hw counters */
+	const struct msm_gpu_perfcntr *perfcntrs;
+	uint32_t num_perfcntrs;
+
 	struct msm_ringbuffer *rb;
 	uint32_t rb_iova;
 
@@ -113,6 +126,19 @@ static inline bool msm_gpu_active(struct msm_gpu *gpu)
 	return gpu->submitted_fence > gpu->funcs->last_fence(gpu);
 }
 
+/* Perf-Counters:
+ * The select_reg and select_val are just there for the benefit of the child
+ * class that actually enables the perf counter..  but msm_gpu base class
+ * will handle sampling/displaying the counters.
+ */
+
+struct msm_gpu_perfcntr {
+	uint32_t select_reg;
+	uint32_t sample_reg;
+	uint32_t select_val;
+	const char *name;
+};
+
 static inline void gpu_write(struct msm_gpu *gpu, u32 reg, u32 data)
 {
 	msm_writel(data, gpu->mmio + (reg << 2));
@@ -126,6 +152,11 @@ static inline u32 gpu_read(struct msm_gpu *gpu, u32 reg)
 int msm_gpu_pm_suspend(struct msm_gpu *gpu);
 int msm_gpu_pm_resume(struct msm_gpu *gpu);
 
+void msm_gpu_perfcntr_start(struct msm_gpu *gpu);
+void msm_gpu_perfcntr_stop(struct msm_gpu *gpu);
+int msm_gpu_perfcntr_sample(struct msm_gpu *gpu, uint32_t *activetime,
+		uint32_t *totaltime, uint32_t ncntrs, uint32_t *cntrs);
+
 void msm_gpu_retire(struct msm_gpu *gpu);
 int msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
 		struct msm_file_private *ctx);
diff --git a/drivers/gpu/drm/msm/msm_perf.c b/drivers/gpu/drm/msm/msm_perf.c
new file mode 100644
index 000000000000..830857c47c86
--- /dev/null
+++ b/drivers/gpu/drm/msm/msm_perf.c
@@ -0,0 +1,275 @@
+/*
+ * Copyright (C) 2013 Red Hat
+ * Author: Rob Clark <robdclark@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/* For profiling, userspace can:
+ *
+ *   tail -f /sys/kernel/debug/dri/<minor>/gpu
+ *
+ * This will enable performance counters/profiling to track the busy time
+ * and any gpu specific performance counters that are supported.
+ */
+
+#ifdef CONFIG_DEBUG_FS
+
+#include <linux/debugfs.h>
+
+#include "msm_drv.h"
+#include "msm_gpu.h"
+
+struct msm_perf_state {
+	struct drm_device *dev;
+
+	bool open;
+	int cnt;
+	struct mutex read_lock;
+
+	char buf[256];
+	int buftot, bufpos;
+
+	unsigned long next_jiffies;
+
+	struct dentry *ent;
+	struct drm_info_node *node;
+};
+
+#define SAMPLE_TIME (HZ/4)
+
+/* wait for next sample time: */
+static int wait_sample(struct msm_perf_state *perf)
+{
+	unsigned long start_jiffies = jiffies;
+
+	if (time_after(perf->next_jiffies, start_jiffies)) {
+		unsigned long remaining_jiffies =
+			perf->next_jiffies - start_jiffies;
+		int ret = schedule_timeout_interruptible(remaining_jiffies);
+		if (ret > 0) {
+			/* interrupted */
+			return -ERESTARTSYS;
+		}
+	}
+	perf->next_jiffies += SAMPLE_TIME;
+	return 0;
+}
+
+static int refill_buf(struct msm_perf_state *perf)
+{
+	struct msm_drm_private *priv = perf->dev->dev_private;
+	struct msm_gpu *gpu = priv->gpu;
+	char *ptr = perf->buf;
+	int rem = sizeof(perf->buf);
+	int i, n;
+
+	if ((perf->cnt++ % 32) == 0) {
+		/* Header line: */
+		n = snprintf(ptr, rem, "%%BUSY");
+		ptr += n;
+		rem -= n;
+
+		for (i = 0; i < gpu->num_perfcntrs; i++) {
+			const struct msm_gpu_perfcntr *perfcntr = &gpu->perfcntrs[i];
+			n = snprintf(ptr, rem, "\t%s", perfcntr->name);
+			ptr += n;
+			rem -= n;
+		}
+	} else {
+		/* Sample line: */
+		uint32_t activetime = 0, totaltime = 0;
+		uint32_t cntrs[5];
+		uint32_t val;
+		int ret;
+
+		/* sleep until next sample time: */
+		ret = wait_sample(perf);
+		if (ret)
+			return ret;
+
+		ret = msm_gpu_perfcntr_sample(gpu, &activetime, &totaltime,
+				ARRAY_SIZE(cntrs), cntrs);
+		if (ret < 0)
+			return ret;
+
+		val = totaltime ? 1000 * activetime / totaltime : 0;
+		n = snprintf(ptr, rem, "%3d.%d%%", val / 10, val % 10);
+		ptr += n;
+		rem -= n;
+
+		for (i = 0; i < ret; i++) {
+			/* cycle counters (I think).. convert to MHz.. */
+			val = cntrs[i] / 10000;
+			n = snprintf(ptr, rem, "\t%5d.%02d",
+					val / 100, val % 100);
+			ptr += n;
+			rem -= n;
+		}
+	}
+
+	n = snprintf(ptr, rem, "\n");
+	ptr += n;
+	rem -= n;
+
+	perf->bufpos = 0;
+	perf->buftot = ptr - perf->buf;
+
+	return 0;
+}
+
+static ssize_t perf_read(struct file *file, char __user *buf,
+		size_t sz, loff_t *ppos)
+{
+	struct msm_perf_state *perf = file->private_data;
+	int n = 0, ret;
+
+	mutex_lock(&perf->read_lock);
+
+	if (perf->bufpos >= perf->buftot) {
+		ret = refill_buf(perf);
+		if (ret)
+			goto out;
+	}
+
+	n = min((int)sz, perf->buftot - perf->bufpos);
+	ret = copy_to_user(buf, &perf->buf[perf->bufpos], n);
+	if (ret)
+		goto out;
+
+	perf->bufpos += n;
+	*ppos += n;
+
+out:
+	mutex_unlock(&perf->read_lock);
+	if (ret)
+		return ret;
+	return n;
+}
+
+static int perf_open(struct inode *inode, struct file *file)
+{
+	struct msm_perf_state *perf = inode->i_private;
+	struct drm_device *dev = perf->dev;
+	struct msm_drm_private *priv = dev->dev_private;
+	struct msm_gpu *gpu = priv->gpu;
+	int ret = 0;
+
+	mutex_lock(&dev->struct_mutex);
+
+	if (perf->open || !gpu) {
+		ret = -EBUSY;
+		goto out;
+	}
+
+	file->private_data = perf;
+	perf->open = true;
+	perf->cnt = 0;
+	perf->buftot = 0;
+	perf->bufpos = 0;
+	msm_gpu_perfcntr_start(gpu);
+	perf->next_jiffies = jiffies + SAMPLE_TIME;
+
+out:
+	mutex_unlock(&dev->struct_mutex);
+	return ret;
+}
+
+static int perf_release(struct inode *inode, struct file *file)
+{
+	struct msm_perf_state *perf = inode->i_private;
+	struct msm_drm_private *priv = perf->dev->dev_private;
+	msm_gpu_perfcntr_stop(priv->gpu);
+	perf->open = false;
+	return 0;
+}
+
+
+static const struct file_operations perf_debugfs_fops = {
+	.owner = THIS_MODULE,
+	.open = perf_open,
+	.read = perf_read,
+	.llseek = no_llseek,
+	.release = perf_release,
+};
+
+int msm_perf_debugfs_init(struct drm_minor *minor)
+{
+	struct msm_drm_private *priv = minor->dev->dev_private;
+	struct msm_perf_state *perf;
+
+	/* only create on first minor: */
+	if (priv->perf)
+		return 0;
+
+	perf = kzalloc(sizeof(*perf), GFP_KERNEL);
+	if (!perf)
+		return -ENOMEM;
+
+	perf->dev = minor->dev;
+
+	mutex_init(&perf->read_lock);
+	priv->perf = perf;
+
+	perf->node = kzalloc(sizeof(*perf->node), GFP_KERNEL);
+	if (!perf->node)
+		goto fail;
+
+	perf->ent = debugfs_create_file("perf", S_IFREG | S_IRUGO,
+			minor->debugfs_root, perf, &perf_debugfs_fops);
+	if (!perf->ent) {
+		DRM_ERROR("Cannot create /sys/kernel/debug/dri/%s/perf\n",
+				minor->debugfs_root->d_name.name);
+		goto fail;
+	}
+
+	perf->node->minor = minor;
+	perf->node->dent  = perf->ent;
+	perf->node->info_ent = NULL;
+
+	mutex_lock(&minor->debugfs_lock);
+	list_add(&perf->node->list, &minor->debugfs_list);
+	mutex_unlock(&minor->debugfs_lock);
+
+	return 0;
+
+fail:
+	msm_perf_debugfs_cleanup(minor);
+	return -1;
+}
+
+void msm_perf_debugfs_cleanup(struct drm_minor *minor)
+{
+	struct msm_drm_private *priv = minor->dev->dev_private;
+	struct msm_perf_state *perf = priv->perf;
+
+	if (!perf)
+		return;
+
+	priv->perf = NULL;
+
+	debugfs_remove(perf->ent);
+
+	if (perf->node) {
+		mutex_lock(&minor->debugfs_lock);
+		list_del(&perf->node->list);
+		mutex_unlock(&minor->debugfs_lock);
+		kfree(perf->node);
+	}
+
+	mutex_destroy(&perf->read_lock);
+
+	kfree(perf);
+}
+
+#endif
diff --git a/drivers/gpu/drm/msm/msm_rd.c b/drivers/gpu/drm/msm/msm_rd.c
new file mode 100644
index 000000000000..9a78c48817c6
--- /dev/null
+++ b/drivers/gpu/drm/msm/msm_rd.c
@@ -0,0 +1,337 @@
+/*
+ * Copyright (C) 2013 Red Hat
+ * Author: Rob Clark <robdclark@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/* For debugging crashes, userspace can:
+ *
+ *   tail -f /sys/kernel/debug/dri/<minor>/rd > logfile.rd
+ *
+ * To log the cmdstream in a format that is understood by freedreno/cffdump
+ * utility.  By comparing the last successfully completed fence #, to the
+ * cmdstream for the next fence, you can narrow down which process and submit
+ * caused the gpu crash/lockup.
+ *
+ * This bypasses drm_debugfs_create_files() mainly because we need to use
+ * our own fops for a bit more control.  In particular, we don't want to
+ * do anything if userspace doesn't have the debugfs file open.
+ */
+
+#ifdef CONFIG_DEBUG_FS
+
+#include <linux/kfifo.h>
+#include <linux/debugfs.h>
+#include <linux/circ_buf.h>
+#include <linux/wait.h>
+
+#include "msm_drv.h"
+#include "msm_gpu.h"
+#include "msm_gem.h"
+
+enum rd_sect_type {
+	RD_NONE,
+	RD_TEST,       /* ascii text */
+	RD_CMD,        /* ascii text */
+	RD_GPUADDR,    /* u32 gpuaddr, u32 size */
+	RD_CONTEXT,    /* raw dump */
+	RD_CMDSTREAM,  /* raw dump */
+	RD_CMDSTREAM_ADDR, /* gpu addr of cmdstream */
+	RD_PARAM,      /* u32 param_type, u32 param_val, u32 bitlen */
+	RD_FLUSH,      /* empty, clear previous params */
+	RD_PROGRAM,    /* shader program, raw dump */
+	RD_VERT_SHADER,
+	RD_FRAG_SHADER,
+	RD_BUFFER_CONTENTS,
+	RD_GPU_ID,
+};
+
+#define BUF_SZ 512  /* should be power of 2 */
+
+/* space used: */
+#define circ_count(circ) \
+	(CIRC_CNT((circ)->head, (circ)->tail, BUF_SZ))
+#define circ_count_to_end(circ) \
+	(CIRC_CNT_TO_END((circ)->head, (circ)->tail, BUF_SZ))
+/* space available: */
+#define circ_space(circ) \
+	(CIRC_SPACE((circ)->head, (circ)->tail, BUF_SZ))
+#define circ_space_to_end(circ) \
+	(CIRC_SPACE_TO_END((circ)->head, (circ)->tail, BUF_SZ))
+
+struct msm_rd_state {
+	struct drm_device *dev;
+
+	bool open;
+
+	struct dentry *ent;
+	struct drm_info_node *node;
+
+	/* current submit to read out: */
+	struct msm_gem_submit *submit;
+
+	/* fifo access is synchronized on the producer side by
+	 * struct_mutex held by submit code (otherwise we could
+	 * end up w/ cmds logged in different order than they
+	 * were executed).  And read_lock synchronizes the reads
+	 */
+	struct mutex read_lock;
+
+	wait_queue_head_t fifo_event;
+	struct circ_buf fifo;
+
+	char buf[BUF_SZ];
+};
+
+static void rd_write(struct msm_rd_state *rd, const void *buf, int sz)
+{
+	struct circ_buf *fifo = &rd->fifo;
+	const char *ptr = buf;
+
+	while (sz > 0) {
+		char *fptr = &fifo->buf[fifo->head];
+		int n;
+
+		wait_event(rd->fifo_event, circ_space(&rd->fifo) > 0);
+
+		n = min(sz, circ_space_to_end(&rd->fifo));
+		memcpy(fptr, ptr, n);
+
+		fifo->head = (fifo->head + n) & (BUF_SZ - 1);
+		sz  -= n;
+		ptr += n;
+
+		wake_up_all(&rd->fifo_event);
+	}
+}
+
+static void rd_write_section(struct msm_rd_state *rd,
+		enum rd_sect_type type, const void *buf, int sz)
+{
+	rd_write(rd, &type, 4);
+	rd_write(rd, &sz, 4);
+	rd_write(rd, buf, sz);
+}
+
+static ssize_t rd_read(struct file *file, char __user *buf,
+		size_t sz, loff_t *ppos)
+{
+	struct msm_rd_state *rd = file->private_data;
+	struct circ_buf *fifo = &rd->fifo;
+	const char *fptr = &fifo->buf[fifo->tail];
+	int n = 0, ret = 0;
+
+	mutex_lock(&rd->read_lock);
+
+	ret = wait_event_interruptible(rd->fifo_event,
+			circ_count(&rd->fifo) > 0);
+	if (ret)
+		goto out;
+
+	n = min_t(int, sz, circ_count_to_end(&rd->fifo));
+	ret = copy_to_user(buf, fptr, n);
+	if (ret)
+		goto out;
+
+	fifo->tail = (fifo->tail + n) & (BUF_SZ - 1);
+	*ppos += n;
+
+	wake_up_all(&rd->fifo_event);
+
+out:
+	mutex_unlock(&rd->read_lock);
+	if (ret)
+		return ret;
+	return n;
+}
+
+static int rd_open(struct inode *inode, struct file *file)
+{
+	struct msm_rd_state *rd = inode->i_private;
+	struct drm_device *dev = rd->dev;
+	struct msm_drm_private *priv = dev->dev_private;
+	struct msm_gpu *gpu = priv->gpu;
+	uint64_t val;
+	uint32_t gpu_id;
+	int ret = 0;
+
+	mutex_lock(&dev->struct_mutex);
+
+	if (rd->open || !gpu) {
+		ret = -EBUSY;
+		goto out;
+	}
+
+	file->private_data = rd;
+	rd->open = true;
+
+	/* the parsing tools need to know gpu-id to know which
+	 * register database to load.
+	 */
+	gpu->funcs->get_param(gpu, MSM_PARAM_GPU_ID, &val);
+	gpu_id = val;
+
+	rd_write_section(rd, RD_GPU_ID, &gpu_id, sizeof(gpu_id));
+
+out:
+	mutex_unlock(&dev->struct_mutex);
+	return ret;
+}
+
+static int rd_release(struct inode *inode, struct file *file)
+{
+	struct msm_rd_state *rd = inode->i_private;
+	rd->open = false;
+	return 0;
+}
+
+
+static const struct file_operations rd_debugfs_fops = {
+	.owner = THIS_MODULE,
+	.open = rd_open,
+	.read = rd_read,
+	.llseek = no_llseek,
+	.release = rd_release,
+};
+
+int msm_rd_debugfs_init(struct drm_minor *minor)
+{
+	struct msm_drm_private *priv = minor->dev->dev_private;
+	struct msm_rd_state *rd;
+
+	/* only create on first minor: */
+	if (priv->rd)
+		return 0;
+
+	rd = kzalloc(sizeof(*rd), GFP_KERNEL);
+	if (!rd)
+		return -ENOMEM;
+
+	rd->dev = minor->dev;
+	rd->fifo.buf = rd->buf;
+
+	mutex_init(&rd->read_lock);
+	priv->rd = rd;
+
+	init_waitqueue_head(&rd->fifo_event);
+
+	rd->node = kzalloc(sizeof(*rd->node), GFP_KERNEL);
+	if (!rd->node)
+		goto fail;
+
+	rd->ent = debugfs_create_file("rd", S_IFREG | S_IRUGO,
+			minor->debugfs_root, rd, &rd_debugfs_fops);
+	if (!rd->ent) {
+		DRM_ERROR("Cannot create /sys/kernel/debug/dri/%s/rd\n",
+				minor->debugfs_root->d_name.name);
+		goto fail;
+	}
+
+	rd->node->minor = minor;
+	rd->node->dent  = rd->ent;
+	rd->node->info_ent = NULL;
+
+	mutex_lock(&minor->debugfs_lock);
+	list_add(&rd->node->list, &minor->debugfs_list);
+	mutex_unlock(&minor->debugfs_lock);
+
+	return 0;
+
+fail:
+	msm_rd_debugfs_cleanup(minor);
+	return -1;
+}
+
+void msm_rd_debugfs_cleanup(struct drm_minor *minor)
+{
+	struct msm_drm_private *priv = minor->dev->dev_private;
+	struct msm_rd_state *rd = priv->rd;
+
+	if (!rd)
+		return;
+
+	priv->rd = NULL;
+
+	debugfs_remove(rd->ent);
+
+	if (rd->node) {
+		mutex_lock(&minor->debugfs_lock);
+		list_del(&rd->node->list);
+		mutex_unlock(&minor->debugfs_lock);
+		kfree(rd->node);
+	}
+
+	mutex_destroy(&rd->read_lock);
+
+	kfree(rd);
+}
+
+/* called under struct_mutex */
+void msm_rd_dump_submit(struct msm_gem_submit *submit)
+{
+	struct drm_device *dev = submit->dev;
+	struct msm_drm_private *priv = dev->dev_private;
+	struct msm_rd_state *rd = priv->rd;
+	char msg[128];
+	int i, n;
+
+	if (!rd->open)
+		return;
+
+	/* writing into fifo is serialized by caller, and
+	 * rd->read_lock is used to serialize the reads
+	 */
+	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
+
+	n = snprintf(msg, sizeof(msg), "%.*s/%d: fence=%u",
+			TASK_COMM_LEN, current->comm, task_pid_nr(current),
+			submit->fence);
+
+	rd_write_section(rd, RD_CMD, msg, ALIGN(n, 4));
+
+	/* could be nice to have an option (module-param?) to snapshot
+	 * all the bo's associated with the submit.  Handy to see vtx
+	 * buffers, etc.  For now just the cmdstream bo's is enough.
+	 */
+
+	for (i = 0; i < submit->nr_cmds; i++) {
+		uint32_t idx  = submit->cmd[i].idx;
+		uint32_t iova = submit->cmd[i].iova;
+		uint32_t szd  = submit->cmd[i].size; /* in dwords */
+		struct msm_gem_object *obj = submit->bos[idx].obj;
+		const char *buf = msm_gem_vaddr_locked(&obj->base);
+
+		buf += iova - submit->bos[idx].iova;
+
+		rd_write_section(rd, RD_GPUADDR,
+				(uint32_t[2]){ iova, szd * 4 }, 8);
+		rd_write_section(rd, RD_BUFFER_CONTENTS,
+				buf, szd * 4);
+
+		switch (submit->cmd[i].type) {
+		case MSM_SUBMIT_CMD_IB_TARGET_BUF:
+			/* ignore IB-targets, we've logged the buffer, the
+			 * parser tool will follow the IB based on the logged
+			 * buffer/gpuaddr, so nothing more to do.
+			 */
+			break;
+		case MSM_SUBMIT_CMD_CTX_RESTORE_BUF:
+		case MSM_SUBMIT_CMD_BUF:
+			rd_write_section(rd, RD_CMDSTREAM_ADDR,
+					(uint32_t[2]){ iova, szd }, 8);
+			break;
+		}
+	}
+}
+#endif
diff --git a/drivers/gpu/drm/nouveau/Makefile b/drivers/gpu/drm/nouveau/Makefile
index b7d216264775..2b6156d0e4b5 100644
--- a/drivers/gpu/drm/nouveau/Makefile
+++ b/drivers/gpu/drm/nouveau/Makefile
@@ -102,6 +102,7 @@ nouveau-y += core/subdev/fb/nvaa.o
 nouveau-y += core/subdev/fb/nvaf.o
 nouveau-y += core/subdev/fb/nvc0.o
 nouveau-y += core/subdev/fb/nve0.o
+nouveau-y += core/subdev/fb/gk20a.o
 nouveau-y += core/subdev/fb/gm107.o
 nouveau-y += core/subdev/fb/ramnv04.o
 nouveau-y += core/subdev/fb/ramnv10.o
@@ -117,25 +118,32 @@ nouveau-y += core/subdev/fb/ramnva3.o
 nouveau-y += core/subdev/fb/ramnvaa.o
 nouveau-y += core/subdev/fb/ramnvc0.o
 nouveau-y += core/subdev/fb/ramnve0.o
+nouveau-y += core/subdev/fb/ramgk20a.o
 nouveau-y += core/subdev/fb/ramgm107.o
 nouveau-y += core/subdev/fb/sddr3.o
 nouveau-y += core/subdev/fb/gddr5.o
 nouveau-y += core/subdev/gpio/base.o
 nouveau-y += core/subdev/gpio/nv10.o
 nouveau-y += core/subdev/gpio/nv50.o
+nouveau-y += core/subdev/gpio/nv92.o
 nouveau-y += core/subdev/gpio/nvd0.o
 nouveau-y += core/subdev/gpio/nve0.o
 nouveau-y += core/subdev/i2c/base.o
 nouveau-y += core/subdev/i2c/anx9805.o
 nouveau-y += core/subdev/i2c/aux.o
 nouveau-y += core/subdev/i2c/bit.o
+nouveau-y += core/subdev/i2c/pad.o
+nouveau-y += core/subdev/i2c/padnv04.o
+nouveau-y += core/subdev/i2c/padnv94.o
 nouveau-y += core/subdev/i2c/nv04.o
 nouveau-y += core/subdev/i2c/nv4e.o
 nouveau-y += core/subdev/i2c/nv50.o
 nouveau-y += core/subdev/i2c/nv94.o
 nouveau-y += core/subdev/i2c/nvd0.o
+nouveau-y += core/subdev/i2c/nve0.o
 nouveau-y += core/subdev/ibus/nvc0.o
 nouveau-y += core/subdev/ibus/nve0.o
+nouveau-y += core/subdev/ibus/gk20a.o
 nouveau-y += core/subdev/instmem/base.o
 nouveau-y += core/subdev/instmem/nv04.o
 nouveau-y += core/subdev/instmem/nv40.o
@@ -214,6 +222,9 @@ nouveau-y += core/engine/device/nvc0.o
 nouveau-y += core/engine/device/nve0.o
 nouveau-y += core/engine/device/gm100.o
 nouveau-y += core/engine/disp/base.o
+nouveau-y += core/engine/disp/conn.o
+nouveau-y += core/engine/disp/outp.o
+nouveau-y += core/engine/disp/outpdp.o
 nouveau-y += core/engine/disp/nv04.o
 nouveau-y += core/engine/disp/nv50.o
 nouveau-y += core/engine/disp/nv84.o
@@ -245,6 +256,7 @@ nouveau-y += core/engine/fifo/nv50.o
 nouveau-y += core/engine/fifo/nv84.o
 nouveau-y += core/engine/fifo/nvc0.o
 nouveau-y += core/engine/fifo/nve0.o
+nouveau-y += core/engine/fifo/gk20a.o
 nouveau-y += core/engine/fifo/nv108.o
 nouveau-y += core/engine/graph/ctxnv40.o
 nouveau-y += core/engine/graph/ctxnv50.o
@@ -255,6 +267,7 @@ nouveau-y += core/engine/graph/ctxnvc8.o
 nouveau-y += core/engine/graph/ctxnvd7.o
 nouveau-y += core/engine/graph/ctxnvd9.o
 nouveau-y += core/engine/graph/ctxnve4.o
+nouveau-y += core/engine/graph/ctxgk20a.o
 nouveau-y += core/engine/graph/ctxnvf0.o
 nouveau-y += core/engine/graph/ctxnv108.o
 nouveau-y += core/engine/graph/ctxgm107.o
@@ -275,6 +288,7 @@ nouveau-y += core/engine/graph/nvc8.o
 nouveau-y += core/engine/graph/nvd7.o
 nouveau-y += core/engine/graph/nvd9.o
 nouveau-y += core/engine/graph/nve4.o
+nouveau-y += core/engine/graph/gk20a.o
 nouveau-y += core/engine/graph/nvf0.o
 nouveau-y += core/engine/graph/nv108.o
 nouveau-y += core/engine/graph/gm107.o
diff --git a/drivers/gpu/drm/nouveau/core/core/event.c b/drivers/gpu/drm/nouveau/core/core/event.c
index 3f3c76581a9e..ae81d3b5d8b7 100644
--- a/drivers/gpu/drm/nouveau/core/core/event.c
+++ b/drivers/gpu/drm/nouveau/core/core/event.c
@@ -28,14 +28,20 @@ nouveau_event_put(struct nouveau_eventh *handler)
 {
 	struct nouveau_event *event = handler->event;
 	unsigned long flags;
-	if (__test_and_clear_bit(NVKM_EVENT_ENABLE, &handler->flags)) {
-		spin_lock_irqsave(&event->refs_lock, flags);
-		if (!--event->index[handler->index].refs) {
+	u32 m, t;
+
+	if (!__test_and_clear_bit(NVKM_EVENT_ENABLE, &handler->flags))
+		return;
+
+	spin_lock_irqsave(&event->refs_lock, flags);
+	for (m = handler->types; t = __ffs(m), m; m &= ~(1 << t)) {
+		if (!--event->refs[handler->index * event->types_nr + t]) {
 			if (event->disable)
-				event->disable(event, handler->index);
+				event->disable(event, 1 << t, handler->index);
 		}
-		spin_unlock_irqrestore(&event->refs_lock, flags);
+
 	}
+	spin_unlock_irqrestore(&event->refs_lock, flags);
 }
 
 void
@@ -43,14 +49,20 @@ nouveau_event_get(struct nouveau_eventh *handler)
 {
 	struct nouveau_event *event = handler->event;
 	unsigned long flags;
-	if (!__test_and_set_bit(NVKM_EVENT_ENABLE, &handler->flags)) {
-		spin_lock_irqsave(&event->refs_lock, flags);
-		if (!event->index[handler->index].refs++) {
+	u32 m, t;
+
+	if (__test_and_set_bit(NVKM_EVENT_ENABLE, &handler->flags))
+		return;
+
+	spin_lock_irqsave(&event->refs_lock, flags);
+	for (m = handler->types; t = __ffs(m), m; m &= ~(1 << t)) {
+		if (!event->refs[handler->index * event->types_nr + t]++) {
 			if (event->enable)
-				event->enable(event, handler->index);
+				event->enable(event, 1 << t, handler->index);
 		}
-		spin_unlock_irqrestore(&event->refs_lock, flags);
+
 	}
+	spin_unlock_irqrestore(&event->refs_lock, flags);
 }
 
 static void
@@ -65,38 +77,47 @@ nouveau_event_fini(struct nouveau_eventh *handler)
 }
 
 static int
-nouveau_event_init(struct nouveau_event *event, int index,
-		   int (*func)(void *, int), void *priv,
+nouveau_event_init(struct nouveau_event *event, u32 types, int index,
+		   int (*func)(void *, u32, int), void *priv,
 		   struct nouveau_eventh *handler)
 {
 	unsigned long flags;
 
+	if (types & ~((1 << event->types_nr) - 1))
+		return -EINVAL;
 	if (index >= event->index_nr)
 		return -EINVAL;
 
 	handler->event = event;
 	handler->flags = 0;
+	handler->types = types;
 	handler->index = index;
 	handler->func = func;
 	handler->priv = priv;
 
 	spin_lock_irqsave(&event->list_lock, flags);
-	list_add_tail(&handler->head, &event->index[index].list);
+	list_add_tail(&handler->head, &event->list[index]);
 	spin_unlock_irqrestore(&event->list_lock, flags);
 	return 0;
 }
 
 int
-nouveau_event_new(struct nouveau_event *event, int index,
-		  int (*func)(void *, int), void *priv,
+nouveau_event_new(struct nouveau_event *event, u32 types, int index,
+		  int (*func)(void *, u32, int), void *priv,
 		  struct nouveau_eventh **phandler)
 {
 	struct nouveau_eventh *handler;
 	int ret = -ENOMEM;
 
+	if (event->check) {
+		ret = event->check(event, types, index);
+		if (ret)
+			return ret;
+	}
+
 	handler = *phandler = kmalloc(sizeof(*handler), GFP_KERNEL);
 	if (handler) {
-		ret = nouveau_event_init(event, index, func, priv, handler);
+		ret = nouveau_event_init(event, types, index, func, priv, handler);
 		if (ret)
 			kfree(handler);
 	}
@@ -116,7 +137,7 @@ nouveau_event_ref(struct nouveau_eventh *handler, struct nouveau_eventh **ref)
 }
 
 void
-nouveau_event_trigger(struct nouveau_event *event, int index)
+nouveau_event_trigger(struct nouveau_event *event, u32 types, int index)
 {
 	struct nouveau_eventh *handler;
 	unsigned long flags;
@@ -125,10 +146,15 @@ nouveau_event_trigger(struct nouveau_event *event, int index)
 		return;
 
 	spin_lock_irqsave(&event->list_lock, flags);
-	list_for_each_entry(handler, &event->index[index].list, head) {
-		if (test_bit(NVKM_EVENT_ENABLE, &handler->flags) &&
-		    handler->func(handler->priv, index) == NVKM_EVENT_DROP)
-			nouveau_event_put(handler);
+	list_for_each_entry(handler, &event->list[index], head) {
+		if (!test_bit(NVKM_EVENT_ENABLE, &handler->flags))
+			continue;
+		if (!(handler->types & types))
+			continue;
+		if (handler->func(handler->priv, handler->types & types, index)
+				!= NVKM_EVENT_DROP)
+			continue;
+		nouveau_event_put(handler);
 	}
 	spin_unlock_irqrestore(&event->list_lock, flags);
 }
@@ -144,20 +170,27 @@ nouveau_event_destroy(struct nouveau_event **pevent)
 }
 
 int
-nouveau_event_create(int index_nr, struct nouveau_event **pevent)
+nouveau_event_create(int types_nr, int index_nr, struct nouveau_event **pevent)
 {
 	struct nouveau_event *event;
 	int i;
 
-	event = *pevent = kzalloc(sizeof(*event) + index_nr *
-				  sizeof(event->index[0]), GFP_KERNEL);
+	event = *pevent = kzalloc(sizeof(*event) + (index_nr * types_nr) *
+				  sizeof(event->refs[0]), GFP_KERNEL);
 	if (!event)
 		return -ENOMEM;
 
+	event->list = kmalloc(sizeof(*event->list) * index_nr, GFP_KERNEL);
+	if (!event->list) {
+		kfree(event);
+		return -ENOMEM;
+	}
+
 	spin_lock_init(&event->list_lock);
 	spin_lock_init(&event->refs_lock);
 	for (i = 0; i < index_nr; i++)
-		INIT_LIST_HEAD(&event->index[i].list);
+		INIT_LIST_HEAD(&event->list[i]);
+	event->types_nr = types_nr;
 	event->index_nr = index_nr;
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/core/core/object.c b/drivers/gpu/drm/nouveau/core/core/object.c
index 7f48e288215f..124538555904 100644
--- a/drivers/gpu/drm/nouveau/core/core/object.c
+++ b/drivers/gpu/drm/nouveau/core/core/object.c
@@ -156,7 +156,7 @@ nouveau_object_ctor(struct nouveau_object *parent,
 	}
 
 	if (ret == 0) {
-		nv_debug(object, "created\n");
+		nv_trace(object, "created\n");
 		atomic_set(&object->refcount, 1);
 	}
 
@@ -166,7 +166,7 @@ nouveau_object_ctor(struct nouveau_object *parent,
 static void
 nouveau_object_dtor(struct nouveau_object *object)
 {
-	nv_debug(object, "destroying\n");
+	nv_trace(object, "destroying\n");
 	nv_ofuncs(object)->dtor(object);
 }
 
@@ -337,7 +337,7 @@ nouveau_object_inc(struct nouveau_object *object)
 		goto fail_self;
 	}
 
-	nv_debug(object, "initialised\n");
+	nv_trace(object, "initialised\n");
 	return 0;
 
 fail_self:
@@ -375,7 +375,7 @@ nouveau_object_decf(struct nouveau_object *object)
 	if (object->parent)
 		nouveau_object_dec(object->parent, false);
 
-	nv_debug(object, "stopped\n");
+	nv_trace(object, "stopped\n");
 	return 0;
 }
 
@@ -411,7 +411,7 @@ nouveau_object_decs(struct nouveau_object *object)
 		}
 	}
 
-	nv_debug(object, "suspended\n");
+	nv_trace(object, "suspended\n");
 	return 0;
 
 fail_parent:
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/gm100.c b/drivers/gpu/drm/nouveau/core/engine/device/gm100.c
index d258c21c4a22..a520029e25d9 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/gm100.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/gm100.c
@@ -60,8 +60,8 @@ gm100_identify(struct nouveau_device *device)
 	case 0x117:
 		device->cname = "GM107";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nve0_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nve0_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nvd0_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nve0_clock_oclass;
 #if 0
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nv04.c b/drivers/gpu/drm/nouveau/core/engine/device/nv04.c
index 0a51ff4e9e00..40b29d0214cb 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nv04.c
@@ -47,7 +47,7 @@ nv04_identify(struct nouveau_device *device)
 	case 0x04:
 		device->cname = "NV04";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv04_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -65,7 +65,7 @@ nv04_identify(struct nouveau_device *device)
 	case 0x05:
 		device->cname = "NV05";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv05_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nv10.c b/drivers/gpu/drm/nouveau/core/engine/device/nv10.c
index e008de8b51b0..5f7c25ff523d 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nv10.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nv10.c
@@ -48,8 +48,8 @@ nv10_identify(struct nouveau_device *device)
 	case 0x10:
 		device->cname = "NV10";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv10_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -65,8 +65,8 @@ nv10_identify(struct nouveau_device *device)
 	case 0x15:
 		device->cname = "NV15";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv10_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -84,8 +84,8 @@ nv10_identify(struct nouveau_device *device)
 	case 0x16:
 		device->cname = "NV16";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv10_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -103,8 +103,8 @@ nv10_identify(struct nouveau_device *device)
 	case 0x1a:
 		device->cname = "nForce";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -122,8 +122,8 @@ nv10_identify(struct nouveau_device *device)
 	case 0x11:
 		device->cname = "NV11";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv10_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -141,8 +141,8 @@ nv10_identify(struct nouveau_device *device)
 	case 0x17:
 		device->cname = "NV17";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv10_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -160,8 +160,8 @@ nv10_identify(struct nouveau_device *device)
 	case 0x1f:
 		device->cname = "nForce2";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -179,8 +179,8 @@ nv10_identify(struct nouveau_device *device)
 	case 0x18:
 		device->cname = "NV18";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv10_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nv20.c b/drivers/gpu/drm/nouveau/core/engine/device/nv20.c
index 7b629a3aed05..75fed11bba0a 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nv20.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nv20.c
@@ -49,8 +49,8 @@ nv20_identify(struct nouveau_device *device)
 	case 0x20:
 		device->cname = "NV20";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -68,8 +68,8 @@ nv20_identify(struct nouveau_device *device)
 	case 0x25:
 		device->cname = "NV25";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -87,8 +87,8 @@ nv20_identify(struct nouveau_device *device)
 	case 0x28:
 		device->cname = "NV28";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -106,8 +106,8 @@ nv20_identify(struct nouveau_device *device)
 	case 0x2a:
 		device->cname = "NV2A";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nv30.c b/drivers/gpu/drm/nouveau/core/engine/device/nv30.c
index 7dfddd5a1908..36919d7db7cc 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nv30.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nv30.c
@@ -49,8 +49,8 @@ nv30_identify(struct nouveau_device *device)
 	case 0x30:
 		device->cname = "NV30";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -68,8 +68,8 @@ nv30_identify(struct nouveau_device *device)
 	case 0x35:
 		device->cname = "NV35";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -87,8 +87,8 @@ nv30_identify(struct nouveau_device *device)
 	case 0x31:
 		device->cname = "NV31";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -107,8 +107,8 @@ nv30_identify(struct nouveau_device *device)
 	case 0x36:
 		device->cname = "NV36";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv20_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
@@ -127,8 +127,8 @@ nv30_identify(struct nouveau_device *device)
 	case 0x34:
 		device->cname = "NV34";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv04_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv10_devinit_oclass;
 		device->oclass[NVDEV_SUBDEV_MC     ] =  nv04_mc_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nv40.c b/drivers/gpu/drm/nouveau/core/engine/device/nv40.c
index 7c1ce6cf4f1f..1130a62be2c7 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nv40.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nv40.c
@@ -53,8 +53,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x40:
 		device->cname = "NV40";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -76,8 +76,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x41:
 		device->cname = "NV41";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -99,8 +99,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x42:
 		device->cname = "NV42";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -122,8 +122,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x43:
 		device->cname = "NV43";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -145,8 +145,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x45:
 		device->cname = "NV45";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -168,8 +168,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x47:
 		device->cname = "G70";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -191,8 +191,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x49:
 		device->cname = "G71";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -214,8 +214,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x4b:
 		device->cname = "G73";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -237,8 +237,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x44:
 		device->cname = "NV44";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -260,8 +260,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x46:
 		device->cname = "G72";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -283,8 +283,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x4a:
 		device->cname = "NV44A";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -306,8 +306,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x4c:
 		device->cname = "C61";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -329,8 +329,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x4e:
 		device->cname = "C51";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv4e_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv4e_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -352,8 +352,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x63:
 		device->cname = "C73";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -375,8 +375,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x67:
 		device->cname = "C67";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
@@ -398,8 +398,8 @@ nv40_identify(struct nouveau_device *device)
 	case 0x68:
 		device->cname = "C68";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv10_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv04_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv10_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv04_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nv50.c b/drivers/gpu/drm/nouveau/core/engine/device/nv50.c
index 66499fa0f758..ef0b0bde1a91 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nv50.c
@@ -60,8 +60,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0x50:
 		device->cname = "G80";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv50_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv50_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv50_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv50_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv50_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -85,8 +85,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0x84:
 		device->cname = "G84";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv50_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv50_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv50_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -113,8 +113,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0x86:
 		device->cname = "G86";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv50_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv50_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv50_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -141,8 +141,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0x92:
 		device->cname = "G92";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv50_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv50_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -169,8 +169,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0x94:
 		device->cname = "G94";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -197,8 +197,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0x96:
 		device->cname = "G96";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -225,8 +225,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0x98:
 		device->cname = "G98";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -253,8 +253,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0xa0:
 		device->cname = "G200";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv50_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv50_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -281,8 +281,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0xaa:
 		device->cname = "MCP77/MCP78";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nvaa_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -309,8 +309,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0xac:
 		device->cname = "MCP79/MCP7A";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nvaa_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -337,8 +337,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0xa3:
 		device->cname = "GT215";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nva3_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -367,8 +367,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0xa5:
 		device->cname = "GT216";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nva3_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -396,8 +396,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0xa8:
 		device->cname = "GT218";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nva3_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -425,8 +425,8 @@ nv50_identify(struct nouveau_device *device)
 	case 0xaf:
 		device->cname = "MCP89";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nva3_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nvc0.c b/drivers/gpu/drm/nouveau/core/engine/device/nvc0.c
index 2075b3027052..f199957995fa 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nvc0.c
@@ -60,8 +60,8 @@ nvc0_identify(struct nouveau_device *device)
 	case 0xc0:
 		device->cname = "GF100";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -92,8 +92,8 @@ nvc0_identify(struct nouveau_device *device)
 	case 0xc4:
 		device->cname = "GF104";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -124,8 +124,8 @@ nvc0_identify(struct nouveau_device *device)
 	case 0xc3:
 		device->cname = "GF106";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -155,8 +155,8 @@ nvc0_identify(struct nouveau_device *device)
 	case 0xce:
 		device->cname = "GF114";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -187,8 +187,8 @@ nvc0_identify(struct nouveau_device *device)
 	case 0xcf:
 		device->cname = "GF116";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -219,8 +219,8 @@ nvc0_identify(struct nouveau_device *device)
 	case 0xc1:
 		device->cname = "GF108";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -250,8 +250,8 @@ nvc0_identify(struct nouveau_device *device)
 	case 0xc8:
 		device->cname = "GF110";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nv92_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nv94_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nva3_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -282,8 +282,8 @@ nvc0_identify(struct nouveau_device *device)
 	case 0xd9:
 		device->cname = "GF119";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nvd0_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nvd0_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nvd0_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -313,8 +313,8 @@ nvc0_identify(struct nouveau_device *device)
 	case 0xd7:
 		device->cname = "GF117";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nvd0_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nvd0_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nvd0_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nvc0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nve0.c b/drivers/gpu/drm/nouveau/core/engine/device/nve0.c
index 9784cbf8a9d2..2d1e97d4264f 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nve0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nve0.c
@@ -60,8 +60,8 @@ nve0_identify(struct nouveau_device *device)
 	case 0xe4:
 		device->cname = "GK104";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nve0_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nve0_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nve0_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nve0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -93,8 +93,8 @@ nve0_identify(struct nouveau_device *device)
 	case 0xe7:
 		device->cname = "GK107";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nve0_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nve0_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nve0_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nve0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -126,8 +126,8 @@ nve0_identify(struct nouveau_device *device)
 	case 0xe6:
 		device->cname = "GK106";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nve0_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nve0_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nve0_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nve0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -156,11 +156,61 @@ nve0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_ENGINE_PPP    ] = &nvc0_ppp_oclass;
 		device->oclass[NVDEV_ENGINE_PERFMON] = &nve0_perfmon_oclass;
 		break;
+	case 0xea:
+		device->cname = "GK20A";
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nvc3_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nvc0_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_TIMER  ] = &gk20a_timer_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  gk20a_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_IBUS   ] = &gk20a_ibus_oclass;
+		device->oclass[NVDEV_SUBDEV_INSTMEM] = nv50_instmem_oclass;
+		device->oclass[NVDEV_SUBDEV_VM     ] = &nvc0_vmmgr_oclass;
+		device->oclass[NVDEV_SUBDEV_BAR    ] = &nvc0_bar_oclass;
+		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nvd0_dmaeng_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  gk20a_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nvc0_software_oclass;
+		device->oclass[NVDEV_ENGINE_GR     ] =  gk20a_graph_oclass;
+		device->oclass[NVDEV_ENGINE_COPY2  ] = &nve0_copy2_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] = &nve0_perfmon_oclass;
+		break;
 	case 0xf0:
 		device->cname = "GK110";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nve0_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nve0_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nve0_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nve0_clock_oclass;
+		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
+		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
+		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nvc0_devinit_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nvc3_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  nvc0_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  nve0_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_LTCG   ] =  gf100_ltcg_oclass;
+		device->oclass[NVDEV_SUBDEV_IBUS   ] = &nve0_ibus_oclass;
+		device->oclass[NVDEV_SUBDEV_INSTMEM] =  nv50_instmem_oclass;
+		device->oclass[NVDEV_SUBDEV_VM     ] = &nvc0_vmmgr_oclass;
+		device->oclass[NVDEV_SUBDEV_BAR    ] = &nvc0_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_PWR    ] = &nvd0_pwr_oclass;
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
+		device->oclass[NVDEV_ENGINE_DMAOBJ ] = &nvd0_dmaeng_oclass;
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  nve0_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  nvc0_software_oclass;
+		device->oclass[NVDEV_ENGINE_GR     ] =  nvf0_graph_oclass;
+		device->oclass[NVDEV_ENGINE_DISP   ] =  nvf0_disp_oclass;
+		device->oclass[NVDEV_ENGINE_COPY0  ] = &nve0_copy0_oclass;
+		device->oclass[NVDEV_ENGINE_COPY1  ] = &nve0_copy1_oclass;
+		device->oclass[NVDEV_ENGINE_COPY2  ] = &nve0_copy2_oclass;
+		device->oclass[NVDEV_ENGINE_BSP    ] = &nve0_bsp_oclass;
+		device->oclass[NVDEV_ENGINE_VP     ] = &nve0_vp_oclass;
+		device->oclass[NVDEV_ENGINE_PPP    ] = &nvc0_ppp_oclass;
+		device->oclass[NVDEV_ENGINE_PERFMON] = &nvf0_perfmon_oclass;
+		break;
+	case 0xf1:
+		device->cname = "GK110B";
+		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nve0_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nvd0_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nve0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -184,18 +234,16 @@ nve0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_ENGINE_COPY0  ] = &nve0_copy0_oclass;
 		device->oclass[NVDEV_ENGINE_COPY1  ] = &nve0_copy1_oclass;
 		device->oclass[NVDEV_ENGINE_COPY2  ] = &nve0_copy2_oclass;
-#if 0
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nve0_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nve0_vp_oclass;
 		device->oclass[NVDEV_ENGINE_PPP    ] = &nvc0_ppp_oclass;
-#endif
 		device->oclass[NVDEV_ENGINE_PERFMON] = &nvf0_perfmon_oclass;
 		break;
 	case 0x108:
 		device->cname = "GK208";
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
-		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nve0_gpio_oclass;
-		device->oclass[NVDEV_SUBDEV_I2C    ] = &nvd0_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  nve0_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  nve0_i2c_oclass;
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nve0_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nvd0_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
@@ -219,11 +267,9 @@ nve0_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_ENGINE_COPY0  ] = &nve0_copy0_oclass;
 		device->oclass[NVDEV_ENGINE_COPY1  ] = &nve0_copy1_oclass;
 		device->oclass[NVDEV_ENGINE_COPY2  ] = &nve0_copy2_oclass;
-#if 0
 		device->oclass[NVDEV_ENGINE_BSP    ] = &nve0_bsp_oclass;
 		device->oclass[NVDEV_ENGINE_VP     ] = &nve0_vp_oclass;
 		device->oclass[NVDEV_ENGINE_PPP    ] = &nvc0_ppp_oclass;
-#endif
 		break;
 	default:
 		nv_fatal(device, "unknown Kepler chipset\n");
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/base.c b/drivers/gpu/drm/nouveau/core/engine/disp/base.c
index 7a5cae42834f..c41f656abe64 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/base.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/base.c
@@ -22,13 +22,87 @@
  * Authors: Ben Skeggs
  */
 
-#include <engine/disp.h>
+#include "priv.h"
+#include "outp.h"
+#include "conn.h"
+
+static int
+nouveau_disp_hpd_check(struct nouveau_event *event, u32 types, int index)
+{
+	struct nouveau_disp *disp = event->priv;
+	struct nvkm_output *outp;
+	list_for_each_entry(outp, &disp->outp, head) {
+		if (outp->conn->index == index) {
+			if (outp->conn->hpd.event)
+				return 0;
+			break;
+		}
+	}
+	return -ENOSYS;
+}
+
+int
+_nouveau_disp_fini(struct nouveau_object *object, bool suspend)
+{
+	struct nouveau_disp *disp = (void *)object;
+	struct nvkm_output *outp;
+	int ret;
+
+	list_for_each_entry(outp, &disp->outp, head) {
+		ret = nv_ofuncs(outp)->fini(nv_object(outp), suspend);
+		if (ret && suspend)
+			goto fail_outp;
+	}
+
+	return nouveau_engine_fini(&disp->base, suspend);
+
+fail_outp:
+	list_for_each_entry_continue_reverse(outp, &disp->outp, head) {
+		nv_ofuncs(outp)->init(nv_object(outp));
+	}
+
+	return ret;
+}
+
+int
+_nouveau_disp_init(struct nouveau_object *object)
+{
+	struct nouveau_disp *disp = (void *)object;
+	struct nvkm_output *outp;
+	int ret;
+
+	ret = nouveau_engine_init(&disp->base);
+	if (ret)
+		return ret;
+
+	list_for_each_entry(outp, &disp->outp, head) {
+		ret = nv_ofuncs(outp)->init(nv_object(outp));
+		if (ret)
+			goto fail_outp;
+	}
+
+	return ret;
+
+fail_outp:
+	list_for_each_entry_continue_reverse(outp, &disp->outp, head) {
+		nv_ofuncs(outp)->fini(nv_object(outp), false);
+	}
+
+	return ret;
+}
 
 void
 _nouveau_disp_dtor(struct nouveau_object *object)
 {
 	struct nouveau_disp *disp = (void *)object;
+	struct nvkm_output *outp, *outt;
+
 	nouveau_event_destroy(&disp->vblank);
+
+	list_for_each_entry_safe(outp, outt, &disp->outp, head) {
+		nouveau_object_ref(NULL, (struct nouveau_object **)&outp);
+	}
+
 	nouveau_engine_destroy(&disp->base);
 }
 
@@ -39,8 +113,15 @@ nouveau_disp_create_(struct nouveau_object *parent,
 		     const char *intname, const char *extname,
 		     int length, void **pobject)
 {
+	struct nouveau_disp_impl *impl = (void *)oclass;
+	struct nouveau_bios *bios = nouveau_bios(parent);
 	struct nouveau_disp *disp;
-	int ret;
+	struct nouveau_oclass **sclass;
+	struct nouveau_object *object;
+	struct dcb_output dcbE;
+	u8  hpd = 0, ver, hdr;
+	u32 data;
+	int ret, i;
 
 	ret = nouveau_engine_create_(parent, engine, oclass, true,
 				     intname, extname, length, pobject);
@@ -48,5 +129,42 @@ nouveau_disp_create_(struct nouveau_object *parent,
 	if (ret)
 		return ret;
 
-	return nouveau_event_create(heads, &disp->vblank);
+	INIT_LIST_HEAD(&disp->outp);
+
+	/* create output objects for each display path in the vbios */
+	i = -1;
+	while ((data = dcb_outp_parse(bios, ++i, &ver, &hdr, &dcbE))) {
+		if (dcbE.type == DCB_OUTPUT_UNUSED)
+			continue;
+		if (dcbE.type == DCB_OUTPUT_EOL)
+			break;
+		data = dcbE.location << 4 | dcbE.type;
+
+		oclass = nvkm_output_oclass;
+		sclass = impl->outp;
+		while (sclass && sclass[0]) {
+			if (sclass[0]->handle == data) {
+				oclass = sclass[0];
+				break;
+			}
+			sclass++;
+		}
+
+		nouveau_object_ctor(*pobject, *pobject, oclass,
+				    &dcbE, i, &object);
+		hpd = max(hpd, (u8)(dcbE.connector + 1));
+	}
+
+	ret = nouveau_event_create(3, hpd, &disp->hpd);
+	if (ret)
+		return ret;
+
+	disp->hpd->priv = disp;
+	disp->hpd->check = nouveau_disp_hpd_check;
+
+	ret = nouveau_event_create(1, heads, &disp->vblank);
+	if (ret)
+		return ret;
+
+	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/conn.c b/drivers/gpu/drm/nouveau/core/engine/disp/conn.c
new file mode 100644
index 000000000000..4ffbc70ecf5a
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/conn.c
@@ -0,0 +1,172 @@
+/*
+ * Copyright 2014 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/gpio.h>
+
+#include "conn.h"
+#include "outp.h"
+
+static void
+nvkm_connector_hpd_work(struct work_struct *w)
+{
+	struct nvkm_connector *conn = container_of(w, typeof(*conn), hpd.work);
+	struct nouveau_disp *disp = nouveau_disp(conn);
+	struct nouveau_gpio *gpio = nouveau_gpio(conn);
+	u32 send = NVKM_HPD_UNPLUG;
+	if (gpio->get(gpio, 0, DCB_GPIO_UNUSED, conn->hpd.event->index))
+		send = NVKM_HPD_PLUG;
+	nouveau_event_trigger(disp->hpd, send, conn->index);
+	nouveau_event_get(conn->hpd.event);
+}
+
+static int
+nvkm_connector_hpd(void *data, u32 type, int index)
+{
+	struct nvkm_connector *conn = data;
+	DBG("HPD: %d\n", type);
+	schedule_work(&conn->hpd.work);
+	return NVKM_EVENT_DROP;
+}
+
+int
+_nvkm_connector_fini(struct nouveau_object *object, bool suspend)
+{
+	struct nvkm_connector *conn = (void *)object;
+	if (conn->hpd.event)
+		nouveau_event_put(conn->hpd.event);
+	return nouveau_object_fini(&conn->base, suspend);
+}
+
+int
+_nvkm_connector_init(struct nouveau_object *object)
+{
+	struct nvkm_connector *conn = (void *)object;
+	int ret = nouveau_object_init(&conn->base);
+	if (ret == 0) {
+		if (conn->hpd.event)
+			nouveau_event_get(conn->hpd.event);
+	}
+	return ret;
+}
+
+void
+_nvkm_connector_dtor(struct nouveau_object *object)
+{
+	struct nvkm_connector *conn = (void *)object;
+	nouveau_event_ref(NULL, &conn->hpd.event);
+	nouveau_object_destroy(&conn->base);
+}
+
+int
+nvkm_connector_create_(struct nouveau_object *parent,
+		       struct nouveau_object *engine,
+		       struct nouveau_oclass *oclass,
+		       struct nvbios_connE *info, int index,
+		       int length, void **pobject)
+{
+	static const u8 hpd[] = { 0x07, 0x08, 0x51, 0x52, 0x5e, 0x5f, 0x60 };
+	struct nouveau_gpio *gpio = nouveau_gpio(parent);
+	struct nouveau_disp *disp = (void *)engine;
+	struct nvkm_connector *conn;
+	struct nvkm_output *outp;
+	struct dcb_gpio_func func;
+	int ret;
+
+	list_for_each_entry(outp, &disp->outp, head) {
+		if (outp->conn && outp->conn->index == index) {
+			atomic_inc(&nv_object(outp->conn)->refcount);
+			*pobject = outp->conn;
+			return 1;
+		}
+	}
+
+	ret = nouveau_object_create_(parent, engine, oclass, 0, length, pobject);
+	conn = *pobject;
+	if (ret)
+		return ret;
+
+	conn->info = *info;
+	conn->index = index;
+
+	DBG("type %02x loc %d hpd %02x dp %x di %x sr %x lcdid %x\n",
+	    info->type, info->location, info->hpd, info->dp,
+	    info->di, info->sr, info->lcdid);
+
+	if ((info->hpd = ffs(info->hpd))) {
+		if (--info->hpd >= ARRAY_SIZE(hpd)) {
+			ERR("hpd %02x unknown\n", info->hpd);
+			goto done;
+		}
+		info->hpd = hpd[info->hpd];
+
+		ret = gpio->find(gpio, 0, info->hpd, DCB_GPIO_UNUSED, &func);
+		if (ret) {
+			ERR("func %02x lookup failed, %d\n", info->hpd, ret);
+			goto done;
+		}
+
+		ret = nouveau_event_new(gpio->events, NVKM_GPIO_TOGGLED,
+					func.line, nvkm_connector_hpd,
+					conn, &conn->hpd.event);
+		if (ret) {
+			ERR("func %02x failed, %d\n", info->hpd, ret);
+		} else {
+			DBG("func %02x (HPD)\n", info->hpd);
+		}
+	}
+
+done:
+	INIT_WORK(&conn->hpd.work, nvkm_connector_hpd_work);
+	return 0;
+}
+
+int
+_nvkm_connector_ctor(struct nouveau_object *parent,
+		     struct nouveau_object *engine,
+		     struct nouveau_oclass *oclass, void *info, u32 index,
+		     struct nouveau_object **pobject)
+{
+	struct nvkm_connector *conn;
+	int ret;
+
+	ret = nvkm_connector_create(parent, engine, oclass, info, index, &conn);
+	*pobject = nv_object(conn);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+struct nouveau_oclass *
+nvkm_connector_oclass = &(struct nvkm_connector_impl) {
+	.base = {
+		.handle = 0,
+		.ofuncs = &(struct nouveau_ofuncs) {
+			.ctor = _nvkm_connector_ctor,
+			.dtor = _nvkm_connector_dtor,
+			.init = _nvkm_connector_init,
+			.fini = _nvkm_connector_fini,
+		},
+	},
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/conn.h b/drivers/gpu/drm/nouveau/core/engine/disp/conn.h
new file mode 100644
index 000000000000..035ebeacbb1c
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/conn.h
@@ -0,0 +1,59 @@
+#ifndef __NVKM_DISP_CONN_H__
+#define __NVKM_DISP_CONN_H__
+
+#include "priv.h"
+
+struct nvkm_connector {
+	struct nouveau_object base;
+	struct list_head head;
+
+	struct nvbios_connE info;
+	int index;
+
+	struct {
+		struct nouveau_eventh *event;
+		struct work_struct work;
+	} hpd;
+};
+
+#define nvkm_connector_create(p,e,c,b,i,d)                                     \
+	nvkm_connector_create_((p), (e), (c), (b), (i), sizeof(**d), (void **)d)
+#define nvkm_connector_destroy(d) ({                                           \
+	struct nvkm_connector *disp = (d);                                     \
+	_nvkm_connector_dtor(nv_object(disp));                                 \
+})
+#define nvkm_connector_init(d) ({                                              \
+	struct nvkm_connector *disp = (d);                                     \
+	_nvkm_connector_init(nv_object(disp));                                 \
+})
+#define nvkm_connector_fini(d,s) ({                                            \
+	struct nvkm_connector *disp = (d);                                     \
+	_nvkm_connector_fini(nv_object(disp), (s));                            \
+})
+
+int nvkm_connector_create_(struct nouveau_object *, struct nouveau_object *,
+			   struct nouveau_oclass *, struct nvbios_connE *,
+			   int, int, void **);
+
+int  _nvkm_connector_ctor(struct nouveau_object *, struct nouveau_object *,
+			  struct nouveau_oclass *, void *, u32,
+			  struct nouveau_object **);
+void _nvkm_connector_dtor(struct nouveau_object *);
+int  _nvkm_connector_init(struct nouveau_object *);
+int  _nvkm_connector_fini(struct nouveau_object *, bool);
+
+struct nvkm_connector_impl {
+	struct nouveau_oclass base;
+};
+
+#ifndef MSG
+#define MSG(l,f,a...) do {                                                     \
+	struct nvkm_connector *_conn = (void *)conn;                           \
+	nv_##l(nv_object(conn)->engine, "%02x:%02x%02x: "f, _conn->index,      \
+	       _conn->info.location, _conn->info.type, ##a);                   \
+} while(0)
+#define DBG(f,a...) MSG(debug, f, ##a)
+#define ERR(f,a...) MSG(error, f, ##a)
+#endif
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/dport.c b/drivers/gpu/drm/nouveau/core/engine/disp/dport.c
index 3ca2d25b7f5e..39562d48101d 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/dport.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/dport.c
@@ -30,42 +30,38 @@
 
 #include <engine/disp.h>
 
-#include "dport.h"
+#include <core/class.h>
 
-#define DBG(fmt, args...) nv_debug(dp->disp, "DP:%04x:%04x: " fmt,             \
-				   dp->outp->hasht, dp->outp->hashm, ##args)
-#define ERR(fmt, args...) nv_error(dp->disp, "DP:%04x:%04x: " fmt,             \
-				   dp->outp->hasht, dp->outp->hashm, ##args)
+#include "dport.h"
+#include "outpdp.h"
 
 /******************************************************************************
  * link training
  *****************************************************************************/
 struct dp_state {
-	const struct nouveau_dp_func *func;
-	struct nouveau_disp *disp;
-	struct dcb_output *outp;
-	struct nvbios_dpout info;
-	u8 version;
-	struct nouveau_i2c_port *aux;
-	int head;
-	u8  dpcd[4];
+	struct nvkm_output_dp *outp;
 	int link_nr;
 	u32 link_bw;
 	u8  stat[6];
 	u8  conf[4];
+	bool pc2;
+	u8  pc2stat;
+	u8  pc2conf[2];
 };
 
 static int
 dp_set_link_config(struct dp_state *dp)
 {
-	struct nouveau_disp *disp = dp->disp;
+	struct nvkm_output_dp_impl *impl = (void *)nv_oclass(dp->outp);
+	struct nvkm_output_dp *outp = dp->outp;
+	struct nouveau_disp *disp = nouveau_disp(outp);
 	struct nouveau_bios *bios = nouveau_bios(disp);
 	struct nvbios_init init = {
-		.subdev = nv_subdev(dp->disp),
+		.subdev = nv_subdev(disp),
 		.bios = bios,
 		.offset = 0x0000,
-		.outp = dp->outp,
-		.crtc = dp->head,
+		.outp = &outp->base.info,
+		.crtc = -1,
 		.execute = 1,
 	};
 	u32 lnkcmp;
@@ -75,8 +71,8 @@ dp_set_link_config(struct dp_state *dp)
 	DBG("%d lanes at %d KB/s\n", dp->link_nr, dp->link_bw);
 
 	/* set desired link configuration on the source */
-	if ((lnkcmp = dp->info.lnkcmp)) {
-		if (dp->version < 0x30) {
+	if ((lnkcmp = dp->outp->info.lnkcmp)) {
+		if (outp->version < 0x30) {
 			while ((dp->link_bw / 10) < nv_ro16(bios, lnkcmp))
 				lnkcmp += 4;
 			init.offset = nv_ro16(bios, lnkcmp + 2);
@@ -89,73 +85,112 @@ dp_set_link_config(struct dp_state *dp)
 		nvbios_exec(&init);
 	}
 
-	ret = dp->func->lnk_ctl(dp->disp, dp->outp, dp->head,
-				dp->link_nr, dp->link_bw / 27000,
-				dp->dpcd[DPCD_RC02] &
-					 DPCD_RC02_ENHANCED_FRAME_CAP);
+	ret = impl->lnk_ctl(outp, dp->link_nr, dp->link_bw / 27000,
+			    outp->dpcd[DPCD_RC02] &
+				       DPCD_RC02_ENHANCED_FRAME_CAP);
 	if (ret) {
-		ERR("lnk_ctl failed with %d\n", ret);
+		if (ret < 0)
+			ERR("lnk_ctl failed with %d\n", ret);
 		return ret;
 	}
 
+	impl->lnk_pwr(outp, dp->link_nr);
+
 	/* set desired link configuration on the sink */
 	sink[0] = dp->link_bw / 27000;
 	sink[1] = dp->link_nr;
-	if (dp->dpcd[DPCD_RC02] & DPCD_RC02_ENHANCED_FRAME_CAP)
+	if (outp->dpcd[DPCD_RC02] & DPCD_RC02_ENHANCED_FRAME_CAP)
 		sink[1] |= DPCD_LC01_ENHANCED_FRAME_EN;
 
-	return nv_wraux(dp->aux, DPCD_LC00, sink, 2);
+	return nv_wraux(outp->base.edid, DPCD_LC00_LINK_BW_SET, sink, 2);
 }
 
 static void
 dp_set_training_pattern(struct dp_state *dp, u8 pattern)
 {
+	struct nvkm_output_dp_impl *impl = (void *)nv_oclass(dp->outp);
+	struct nvkm_output_dp *outp = dp->outp;
 	u8 sink_tp;
 
 	DBG("training pattern %d\n", pattern);
-	dp->func->pattern(dp->disp, dp->outp, dp->head, pattern);
+	impl->pattern(outp, pattern);
 
-	nv_rdaux(dp->aux, DPCD_LC02, &sink_tp, 1);
+	nv_rdaux(outp->base.edid, DPCD_LC02, &sink_tp, 1);
 	sink_tp &= ~DPCD_LC02_TRAINING_PATTERN_SET;
 	sink_tp |= pattern;
-	nv_wraux(dp->aux, DPCD_LC02, &sink_tp, 1);
+	nv_wraux(outp->base.edid, DPCD_LC02, &sink_tp, 1);
 }
 
 static int
-dp_link_train_commit(struct dp_state *dp)
+dp_link_train_commit(struct dp_state *dp, bool pc)
 {
-	int i;
+	struct nvkm_output_dp_impl *impl = (void *)nv_oclass(dp->outp);
+	struct nvkm_output_dp *outp = dp->outp;
+	int ret, i;
 
 	for (i = 0; i < dp->link_nr; i++) {
 		u8 lane = (dp->stat[4 + (i >> 1)] >> ((i & 1) * 4)) & 0xf;
+		u8 lpc2 = (dp->pc2stat >> (i * 2)) & 0x3;
 		u8 lpre = (lane & 0x0c) >> 2;
 		u8 lvsw = (lane & 0x03) >> 0;
+		u8 hivs = 3 - lpre;
+		u8 hipe = 3;
+		u8 hipc = 3;
+
+		if (lpc2 >= hipc)
+			lpc2 = hipc | DPCD_LC0F_LANE0_MAX_POST_CURSOR2_REACHED;
+		if (lpre >= hipe) {
+			lpre = hipe | DPCD_LC03_MAX_SWING_REACHED; /* yes. */
+			lvsw = hivs = 3 - (lpre & 3);
+		} else
+		if (lvsw >= hivs) {
+			lvsw = hivs | DPCD_LC03_MAX_SWING_REACHED;
+		}
 
 		dp->conf[i] = (lpre << 3) | lvsw;
-		if (lvsw == 3)
-			dp->conf[i] |= DPCD_LC03_MAX_SWING_REACHED;
-		if (lpre == 3)
-			dp->conf[i] |= DPCD_LC03_MAX_PRE_EMPHASIS_REACHED;
+		dp->pc2conf[i >> 1] |= lpc2 << ((i & 1) * 4);
+
+		DBG("config lane %d %02x %02x\n", i, dp->conf[i], lpc2);
+		impl->drv_ctl(outp, i, lvsw & 3, lpre & 3, lpc2 & 3);
+	}
+
+	ret = nv_wraux(outp->base.edid, DPCD_LC03(0), dp->conf, 4);
+	if (ret)
+		return ret;
 
-		DBG("config lane %d %02x\n", i, dp->conf[i]);
-		dp->func->drv_ctl(dp->disp, dp->outp, dp->head, i, lvsw, lpre);
+	if (pc) {
+		ret = nv_wraux(outp->base.edid, DPCD_LC0F, dp->pc2conf, 2);
+		if (ret)
+			return ret;
 	}
 
-	return nv_wraux(dp->aux, DPCD_LC03(0), dp->conf, 4);
+	return 0;
 }
 
 static int
-dp_link_train_update(struct dp_state *dp, u32 delay)
+dp_link_train_update(struct dp_state *dp, bool pc, u32 delay)
 {
+	struct nvkm_output_dp *outp = dp->outp;
 	int ret;
 
-	udelay(delay);
+	if (outp->dpcd[DPCD_RC0E_AUX_RD_INTERVAL])
+		mdelay(outp->dpcd[DPCD_RC0E_AUX_RD_INTERVAL] * 4);
+	else
+		udelay(delay);
 
-	ret = nv_rdaux(dp->aux, DPCD_LS02, dp->stat, 6);
+	ret = nv_rdaux(outp->base.edid, DPCD_LS02, dp->stat, 6);
 	if (ret)
 		return ret;
 
-	DBG("status %6ph\n", dp->stat);
+	if (pc) {
+		ret = nv_rdaux(outp->base.edid, DPCD_LS0C, &dp->pc2stat, 1);
+		if (ret)
+			dp->pc2stat = 0x00;
+		DBG("status %6ph pc2 %02x\n", dp->stat, dp->pc2stat);
+	} else {
+		DBG("status %6ph\n", dp->stat);
+	}
+
 	return 0;
 }
 
@@ -169,8 +204,8 @@ dp_link_train_cr(struct dp_state *dp)
 	dp_set_training_pattern(dp, 1);
 
 	do {
-		if (dp_link_train_commit(dp) ||
-		    dp_link_train_update(dp, 100))
+		if (dp_link_train_commit(dp, false) ||
+		    dp_link_train_update(dp, false, 100))
 			break;
 
 		cr_done = true;
@@ -196,13 +231,17 @@ dp_link_train_cr(struct dp_state *dp)
 static int
 dp_link_train_eq(struct dp_state *dp)
 {
+	struct nvkm_output_dp *outp = dp->outp;
 	bool eq_done = false, cr_done = true;
 	int tries = 0, i;
 
-	dp_set_training_pattern(dp, 2);
+	if (outp->dpcd[2] & DPCD_RC02_TPS3_SUPPORTED)
+		dp_set_training_pattern(dp, 3);
+	else
+		dp_set_training_pattern(dp, 2);
 
 	do {
-		if (dp_link_train_update(dp, 400))
+		if (dp_link_train_update(dp, dp->pc2, 400))
 			break;
 
 		eq_done = !!(dp->stat[2] & DPCD_LS04_INTERLANE_ALIGN_DONE);
@@ -215,7 +254,7 @@ dp_link_train_eq(struct dp_state *dp)
 				eq_done = false;
 		}
 
-		if (dp_link_train_commit(dp))
+		if (dp_link_train_commit(dp, dp->pc2))
 			break;
 	} while (!eq_done && cr_done && ++tries <= 5);
 
@@ -225,121 +264,109 @@ dp_link_train_eq(struct dp_state *dp)
 static void
 dp_link_train_init(struct dp_state *dp, bool spread)
 {
+	struct nvkm_output_dp *outp = dp->outp;
+	struct nouveau_disp *disp = nouveau_disp(outp);
+	struct nouveau_bios *bios = nouveau_bios(disp);
 	struct nvbios_init init = {
-		.subdev = nv_subdev(dp->disp),
-		.bios = nouveau_bios(dp->disp),
-		.outp = dp->outp,
-		.crtc = dp->head,
+		.subdev = nv_subdev(disp),
+		.bios = bios,
+		.outp = &outp->base.info,
+		.crtc = -1,
 		.execute = 1,
 	};
 
 	/* set desired spread */
 	if (spread)
-		init.offset = dp->info.script[2];
+		init.offset = outp->info.script[2];
 	else
-		init.offset = dp->info.script[3];
+		init.offset = outp->info.script[3];
 	nvbios_exec(&init);
 
 	/* pre-train script */
-	init.offset = dp->info.script[0];
+	init.offset = outp->info.script[0];
 	nvbios_exec(&init);
 }
 
 static void
 dp_link_train_fini(struct dp_state *dp)
 {
+	struct nvkm_output_dp *outp = dp->outp;
+	struct nouveau_disp *disp = nouveau_disp(outp);
+	struct nouveau_bios *bios = nouveau_bios(disp);
 	struct nvbios_init init = {
-		.subdev = nv_subdev(dp->disp),
-		.bios = nouveau_bios(dp->disp),
-		.outp = dp->outp,
-		.crtc = dp->head,
+		.subdev = nv_subdev(disp),
+		.bios = bios,
+		.outp = &outp->base.info,
+		.crtc = -1,
 		.execute = 1,
 	};
 
 	/* post-train script */
-	init.offset = dp->info.script[1],
+	init.offset = outp->info.script[1],
 	nvbios_exec(&init);
 }
 
-int
-nouveau_dp_train(struct nouveau_disp *disp, const struct nouveau_dp_func *func,
-		 struct dcb_output *outp, int head, u32 datarate)
+static const struct dp_rates {
+	u32 rate;
+	u8  bw;
+	u8  nr;
+} nouveau_dp_rates[] = {
+	{ 2160000, 0x14, 4 },
+	{ 1080000, 0x0a, 4 },
+	{ 1080000, 0x14, 2 },
+	{  648000, 0x06, 4 },
+	{  540000, 0x0a, 2 },
+	{  540000, 0x14, 1 },
+	{  324000, 0x06, 2 },
+	{  270000, 0x0a, 1 },
+	{  162000, 0x06, 1 },
+	{}
+};
+
+void
+nouveau_dp_train(struct work_struct *w)
 {
-	struct nouveau_bios *bios = nouveau_bios(disp);
-	struct nouveau_i2c *i2c = nouveau_i2c(disp);
+	struct nvkm_output_dp *outp = container_of(w, typeof(*outp), lt.work);
+	struct nouveau_disp *disp = nouveau_disp(outp);
+	const struct dp_rates *cfg = nouveau_dp_rates;
 	struct dp_state _dp = {
-		.disp = disp,
-		.func = func,
 		.outp = outp,
-		.head = head,
 	}, *dp = &_dp;
-	const u32 bw_list[] = { 540000, 270000, 162000, 0 };
-	const u32 *link_bw = bw_list;
-	u8  hdr, cnt, len;
-	u32 data;
+	u32 datarate = 0;
 	int ret;
 
-	/* find the bios displayport data relevant to this output */
-	data = nvbios_dpout_match(bios, outp->hasht, outp->hashm, &dp->version,
-				 &hdr, &cnt, &len, &dp->info);
-	if (!data) {
-		ERR("bios data not found\n");
-		return -EINVAL;
-	}
-
-	/* acquire the aux channel and fetch some info about the display */
-	if (outp->location)
-		dp->aux = i2c->find_type(i2c, NV_I2C_TYPE_EXTAUX(outp->extdev));
-	else
-		dp->aux = i2c->find(i2c, NV_I2C_TYPE_DCBI2C(outp->i2c_index));
-	if (!dp->aux) {
-		ERR("no aux channel?!\n");
-		return -ENODEV;
-	}
-
-	ret = nv_rdaux(dp->aux, 0x00000, dp->dpcd, sizeof(dp->dpcd));
-	if (ret) {
-		/* it's possible the display has been unplugged before we
-		 * get here.  we still need to execute the full set of
-		 * vbios scripts, and program the OR at a high enough
-		 * frequency to satisfy the target mode.  failure to do
-		 * so results at best in an UPDATE hanging, and at worst
-		 * with PDISP running away to join the circus.
-		 */
-		dp->dpcd[1] = link_bw[0] / 27000;
-		dp->dpcd[2] = 4;
-		dp->dpcd[3] = 0x00;
-		ERR("failed to read DPCD\n");
-	}
-
 	/* bring capabilities within encoder limits */
-	if ((dp->dpcd[2] & 0x1f) > dp->outp->dpconf.link_nr) {
-		dp->dpcd[2] &= ~0x1f;
-		dp->dpcd[2] |= dp->outp->dpconf.link_nr;
+	if (nv_mclass(disp) < NVD0_DISP_CLASS)
+		outp->dpcd[2] &= ~DPCD_RC02_TPS3_SUPPORTED;
+	if ((outp->dpcd[2] & 0x1f) > outp->base.info.dpconf.link_nr) {
+		outp->dpcd[2] &= ~DPCD_RC02_MAX_LANE_COUNT;
+		outp->dpcd[2] |= outp->base.info.dpconf.link_nr;
+	}
+	if (outp->dpcd[1] > outp->base.info.dpconf.link_bw)
+		outp->dpcd[1] = outp->base.info.dpconf.link_bw;
+	dp->pc2 = outp->dpcd[2] & DPCD_RC02_TPS3_SUPPORTED;
+
+	/* restrict link config to the lowest required rate, if requested */
+	if (datarate) {
+		datarate = (datarate / 8) * 10; /* 8B/10B coding overhead */
+		while (cfg[1].rate >= datarate)
+			cfg++;
 	}
-	if (dp->dpcd[1] > dp->outp->dpconf.link_bw)
-		dp->dpcd[1] = dp->outp->dpconf.link_bw;
+	cfg--;
 
-	/* adjust required bandwidth for 8B/10B coding overhead */
-	datarate = (datarate / 8) * 10;
+	/* disable link interrupt handling during link training */
+	nouveau_event_put(outp->irq);
 
 	/* enable down-spreading and execute pre-train script from vbios */
-	dp_link_train_init(dp, dp->dpcd[3] & 0x01);
+	dp_link_train_init(dp, outp->dpcd[3] & 0x01);
 
-	/* start off at highest link rate supported by encoder and display */
-	while (*link_bw > (dp->dpcd[1] * 27000))
-		link_bw++;
-
-	while ((ret = -EIO) && link_bw[0]) {
-		/* find minimum required lane count at this link rate */
-		dp->link_nr = dp->dpcd[2] & DPCD_RC02_MAX_LANE_COUNT;
-		while ((dp->link_nr >> 1) * link_bw[0] > datarate)
-			dp->link_nr >>= 1;
-
-		/* drop link rate to minimum with this lane count */
-		while ((link_bw[1] * dp->link_nr) > datarate)
-			link_bw++;
-		dp->link_bw = link_bw[0];
+	while (ret = -EIO, (++cfg)->rate) {
+		/* select next configuration supported by encoder and sink */
+		while (cfg->nr > (outp->dpcd[2] & DPCD_RC02_MAX_LANE_COUNT) ||
+		       cfg->bw > (outp->dpcd[DPCD_RC01_MAX_LINK_RATE]))
+			cfg++;
+		dp->link_bw = cfg->bw * 27000;
+		dp->link_nr = cfg->nr;
 
 		/* program selected link configuration */
 		ret = dp_set_link_config(dp);
@@ -356,17 +383,18 @@ nouveau_dp_train(struct nouveau_disp *disp, const struct nouveau_dp_func *func,
 			 */
 			break;
 		}
-
-		/* retry at lower rate */
-		link_bw++;
 	}
 
-	/* finish link training */
+	/* finish link training and execute post-train script from vbios */
 	dp_set_training_pattern(dp, 0);
 	if (ret < 0)
 		ERR("link training failed\n");
 
-	/* execute post-train script from vbios */
 	dp_link_train_fini(dp);
-	return (ret < 0) ? false : true;
+
+	/* signal completion and enable link interrupt handling */
+	DBG("training complete\n");
+	atomic_set(&outp->lt.done, 1);
+	wake_up(&outp->lt.wait);
+	nouveau_event_get(outp->irq);
 }
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/dport.h b/drivers/gpu/drm/nouveau/core/engine/disp/dport.h
index 0e1bbd18ff6c..5628d2d5ec71 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/dport.h
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/dport.h
@@ -2,19 +2,18 @@
 #define __NVKM_DISP_DPORT_H__
 
 /* DPCD Receiver Capabilities */
-#define DPCD_RC00                                                       0x00000
-#define DPCD_RC00_DPCD_REV                                                 0xff
-#define DPCD_RC01                                                       0x00001
-#define DPCD_RC01_MAX_LINK_RATE                                            0xff
+#define DPCD_RC00_DPCD_REV                                              0x00000
+#define DPCD_RC01_MAX_LINK_RATE                                         0x00001
 #define DPCD_RC02                                                       0x00002
 #define DPCD_RC02_ENHANCED_FRAME_CAP                                       0x80
+#define DPCD_RC02_TPS3_SUPPORTED                                           0x40
 #define DPCD_RC02_MAX_LANE_COUNT                                           0x1f
 #define DPCD_RC03                                                       0x00003
 #define DPCD_RC03_MAX_DOWNSPREAD                                           0x01
+#define DPCD_RC0E_AUX_RD_INTERVAL                                       0x0000e
 
 /* DPCD Link Configuration */
-#define DPCD_LC00                                                       0x00100
-#define DPCD_LC00_LINK_BW_SET                                              0xff
+#define DPCD_LC00_LINK_BW_SET                                           0x00100
 #define DPCD_LC01                                                       0x00101
 #define DPCD_LC01_ENHANCED_FRAME_EN                                        0x80
 #define DPCD_LC01_LANE_COUNT_SET                                           0x1f
@@ -25,6 +24,16 @@
 #define DPCD_LC03_PRE_EMPHASIS_SET                                         0x18
 #define DPCD_LC03_MAX_SWING_REACHED                                        0x04
 #define DPCD_LC03_VOLTAGE_SWING_SET                                        0x03
+#define DPCD_LC0F                                                       0x0010f
+#define DPCD_LC0F_LANE1_MAX_POST_CURSOR2_REACHED                           0x40
+#define DPCD_LC0F_LANE1_POST_CURSOR2_SET                                   0x30
+#define DPCD_LC0F_LANE0_MAX_POST_CURSOR2_REACHED                           0x04
+#define DPCD_LC0F_LANE0_POST_CURSOR2_SET                                   0x03
+#define DPCD_LC10                                                       0x00110
+#define DPCD_LC10_LANE3_MAX_POST_CURSOR2_REACHED                           0x40
+#define DPCD_LC10_LANE3_POST_CURSOR2_SET                                   0x30
+#define DPCD_LC10_LANE2_MAX_POST_CURSOR2_REACHED                           0x04
+#define DPCD_LC10_LANE2_POST_CURSOR2_SET                                   0x03
 
 /* DPCD Link/Sink Status */
 #define DPCD_LS02                                                       0x00202
@@ -55,24 +64,12 @@
 #define DPCD_LS07_LANE3_VOLTAGE_SWING                                      0x30
 #define DPCD_LS07_LANE2_PRE_EMPHASIS                                       0x0c
 #define DPCD_LS07_LANE2_VOLTAGE_SWING                                      0x03
+#define DPCD_LS0C                                                       0x0020c
+#define DPCD_LS0C_LANE3_POST_CURSOR2                                       0xc0
+#define DPCD_LS0C_LANE2_POST_CURSOR2                                       0x30
+#define DPCD_LS0C_LANE1_POST_CURSOR2                                       0x0c
+#define DPCD_LS0C_LANE0_POST_CURSOR2                                       0x03
 
-struct nouveau_disp;
-struct dcb_output;
-
-struct nouveau_dp_func {
-	int (*pattern)(struct nouveau_disp *, struct dcb_output *,
-		       int head, int pattern);
-	int (*lnk_ctl)(struct nouveau_disp *, struct dcb_output *, int head,
-		       int link_nr, int link_bw, bool enh_frame);
-	int (*drv_ctl)(struct nouveau_disp *, struct dcb_output *, int head,
-		       int lane, int swing, int preem);
-};
-
-extern const struct nouveau_dp_func nv94_sor_dp_func;
-extern const struct nouveau_dp_func nvd0_sor_dp_func;
-extern const struct nouveau_dp_func nv50_pior_dp_func;
-
-int nouveau_dp_train(struct nouveau_disp *, const struct nouveau_dp_func *,
-		     struct dcb_output *, int, u32);
+void nouveau_dp_train(struct work_struct *);
 
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/gm107.c b/drivers/gpu/drm/nouveau/core/engine/disp/gm107.c
index cf6f59677b74..9fc7447fec90 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/gm107.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/gm107.c
@@ -81,7 +81,6 @@ gm107_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	priv->sor.power = nv50_sor_power;
 	priv->sor.hda_eld = nvd0_hda_eld;
 	priv->sor.hdmi = nvd0_hdmi_ctrl;
-	priv->sor.dp = &nvd0_sor_dp_func;
 	return 0;
 }
 
@@ -94,6 +93,7 @@ gm107_disp_oclass = &(struct nv50_disp_impl) {
 		.init = _nouveau_disp_init,
 		.fini = _nouveau_disp_fini,
 	},
+	.base.outp =  nvd0_disp_outp_sclass,
 	.mthd.core = &nve0_disp_mast_mthd_chan,
 	.mthd.base = &nvd0_disp_sync_mthd_chan,
 	.mthd.ovly = &nve0_disp_ovly_mthd_chan,
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nv04.c b/drivers/gpu/drm/nouveau/core/engine/disp/nv04.c
index 6c89af792889..a32666ed0c47 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nv04.c
@@ -51,6 +51,14 @@ nv04_disp_scanoutpos(struct nouveau_object *object, u32 mthd,
 	args->htotal  = nv_rd32(priv, 0x680824 + (head * 0x2000)) & 0xffff;
 	args->hblanke = args->htotal - 1;
 
+	/*
+	 * If output is vga instead of digital then vtotal/htotal is invalid
+	 * so we have to give up and trigger the timestamping fallback in the
+	 * drm core.
+	 */
+	if (!args->vtotal || !args->htotal)
+		return -ENOTSUPP;
+
 	args->time[0] = ktime_to_ns(ktime_get());
 	line = nv_rd32(priv, 0x600868 + (head * 0x2000));
 	args->time[1] = ktime_to_ns(ktime_get());
@@ -78,13 +86,13 @@ nv04_disp_sclass[] = {
  ******************************************************************************/
 
 static void
-nv04_disp_vblank_enable(struct nouveau_event *event, int head)
+nv04_disp_vblank_enable(struct nouveau_event *event, int type, int head)
 {
 	nv_wr32(event->priv, 0x600140 + (head * 0x2000) , 0x00000001);
 }
 
 static void
-nv04_disp_vblank_disable(struct nouveau_event *event, int head)
+nv04_disp_vblank_disable(struct nouveau_event *event, int type, int head)
 {
 	nv_wr32(event->priv, 0x600140 + (head * 0x2000) , 0x00000000);
 }
@@ -98,12 +106,12 @@ nv04_disp_intr(struct nouveau_subdev *subdev)
 	u32 pvideo;
 
 	if (crtc0 & 0x00000001) {
-		nouveau_event_trigger(priv->base.vblank, 0);
+		nouveau_event_trigger(priv->base.vblank, 1, 0);
 		nv_wr32(priv, 0x600100, 0x00000001);
 	}
 
 	if (crtc1 & 0x00000001) {
-		nouveau_event_trigger(priv->base.vblank, 1);
+		nouveau_event_trigger(priv->base.vblank, 1, 1);
 		nv_wr32(priv, 0x602100, 0x00000001);
 	}
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nv50.c b/drivers/gpu/drm/nouveau/core/engine/disp/nv50.c
index 9a0cab9c3adb..1e85f36c705f 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nv50.c
@@ -829,13 +829,13 @@ nv50_disp_base_scanoutpos(struct nouveau_object *object, u32 mthd,
 }
 
 static void
-nv50_disp_base_vblank_enable(struct nouveau_event *event, int head)
+nv50_disp_base_vblank_enable(struct nouveau_event *event, int type, int head)
 {
 	nv_mask(event->priv, 0x61002c, (4 << head), (4 << head));
 }
 
 static void
-nv50_disp_base_vblank_disable(struct nouveau_event *event, int head)
+nv50_disp_base_vblank_disable(struct nouveau_event *event, int type, int head)
 {
 	nv_mask(event->priv, 0x61002c, (4 << head), 0);
 }
@@ -1114,19 +1114,20 @@ nv50_disp_intr_error(struct nv50_disp_priv *priv, int chid)
 	nv_wr32(priv, 0x610080 + (chid * 0x08), 0x90000000);
 }
 
-static u16
-exec_lookup(struct nv50_disp_priv *priv, int head, int outp, u32 ctrl,
-	    struct dcb_output *dcb, u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
+static struct nvkm_output *
+exec_lookup(struct nv50_disp_priv *priv, int head, int or, u32 ctrl,
+	    u32 *data, u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
 	    struct nvbios_outp *info)
 {
 	struct nouveau_bios *bios = nouveau_bios(priv);
-	u16 mask, type, data;
+	struct nvkm_output *outp;
+	u16 mask, type;
 
-	if (outp < 4) {
+	if (or < 4) {
 		type = DCB_OUTPUT_ANALOG;
 		mask = 0;
 	} else
-	if (outp < 8) {
+	if (or < 8) {
 		switch (ctrl & 0x00000f00) {
 		case 0x00000000: type = DCB_OUTPUT_LVDS; mask = 1; break;
 		case 0x00000100: type = DCB_OUTPUT_TMDS; mask = 1; break;
@@ -1136,45 +1137,48 @@ exec_lookup(struct nv50_disp_priv *priv, int head, int outp, u32 ctrl,
 		case 0x00000900: type = DCB_OUTPUT_DP; mask = 2; break;
 		default:
 			nv_error(priv, "unknown SOR mc 0x%08x\n", ctrl);
-			return 0x0000;
+			return NULL;
 		}
-		outp -= 4;
+		or  -= 4;
 	} else {
-		outp = outp - 8;
+		or   = or - 8;
 		type = 0x0010;
 		mask = 0;
 		switch (ctrl & 0x00000f00) {
-		case 0x00000000: type |= priv->pior.type[outp]; break;
+		case 0x00000000: type |= priv->pior.type[or]; break;
 		default:
 			nv_error(priv, "unknown PIOR mc 0x%08x\n", ctrl);
-			return 0x0000;
+			return NULL;
 		}
 	}
 
 	mask  = 0x00c0 & (mask << 6);
-	mask |= 0x0001 << outp;
+	mask |= 0x0001 << or;
 	mask |= 0x0100 << head;
 
-	data = dcb_outp_match(bios, type, mask, ver, hdr, dcb);
-	if (!data)
-		return 0x0000;
-
-	/* off-chip encoders require matching the exact encoder type */
-	if (dcb->location != 0)
-		type |= dcb->extdev << 8;
+	list_for_each_entry(outp, &priv->base.outp, head) {
+		if ((outp->info.hasht & 0xff) == type &&
+		    (outp->info.hashm & mask) == mask) {
+			*data = nvbios_outp_match(bios, outp->info.hasht,
+							outp->info.hashm,
+						  ver, hdr, cnt, len, info);
+			if (!*data)
+				return NULL;
+			return outp;
+		}
+	}
 
-	return nvbios_outp_match(bios, type, mask, ver, hdr, cnt, len, info);
+	return NULL;
 }
 
-static bool
+static struct nvkm_output *
 exec_script(struct nv50_disp_priv *priv, int head, int id)
 {
 	struct nouveau_bios *bios = nouveau_bios(priv);
+	struct nvkm_output *outp;
 	struct nvbios_outp info;
-	struct dcb_output dcb;
 	u8  ver, hdr, cnt, len;
-	u16 data;
-	u32 ctrl = 0x00000000;
+	u32 data, ctrl = 0;
 	u32 reg;
 	int i;
 
@@ -1204,36 +1208,35 @@ exec_script(struct nv50_disp_priv *priv, int head, int id)
 	}
 
 	if (!(ctrl & (1 << head)))
-		return false;
+		return NULL;
 	i--;
 
-	data = exec_lookup(priv, head, i, ctrl, &dcb, &ver, &hdr, &cnt, &len, &info);
-	if (data) {
+	outp = exec_lookup(priv, head, i, ctrl, &data, &ver, &hdr, &cnt, &len, &info);
+	if (outp) {
 		struct nvbios_init init = {
 			.subdev = nv_subdev(priv),
 			.bios = bios,
 			.offset = info.script[id],
-			.outp = &dcb,
+			.outp = &outp->info,
 			.crtc = head,
 			.execute = 1,
 		};
 
-		return nvbios_exec(&init) == 0;
+		nvbios_exec(&init);
 	}
 
-	return false;
+	return outp;
 }
 
-static u32
-exec_clkcmp(struct nv50_disp_priv *priv, int head, int id, u32 pclk,
-	    struct dcb_output *outp)
+static struct nvkm_output *
+exec_clkcmp(struct nv50_disp_priv *priv, int head, int id, u32 pclk, u32 *conf)
 {
 	struct nouveau_bios *bios = nouveau_bios(priv);
+	struct nvkm_output *outp;
 	struct nvbios_outp info1;
 	struct nvbios_ocfg info2;
 	u8  ver, hdr, cnt, len;
-	u32 ctrl = 0x00000000;
-	u32 data, conf = ~0;
+	u32 data, ctrl = 0;
 	u32 reg;
 	int i;
 
@@ -1263,37 +1266,37 @@ exec_clkcmp(struct nv50_disp_priv *priv, int head, int id, u32 pclk,
 	}
 
 	if (!(ctrl & (1 << head)))
-		return conf;
+		return NULL;
 	i--;
 
-	data = exec_lookup(priv, head, i, ctrl, outp, &ver, &hdr, &cnt, &len, &info1);
+	outp = exec_lookup(priv, head, i, ctrl, &data, &ver, &hdr, &cnt, &len, &info1);
 	if (!data)
-		return conf;
+		return NULL;
 
-	if (outp->location == 0) {
-		switch (outp->type) {
+	if (outp->info.location == 0) {
+		switch (outp->info.type) {
 		case DCB_OUTPUT_TMDS:
-			conf = (ctrl & 0x00000f00) >> 8;
+			*conf = (ctrl & 0x00000f00) >> 8;
 			if (pclk >= 165000)
-				conf |= 0x0100;
+				*conf |= 0x0100;
 			break;
 		case DCB_OUTPUT_LVDS:
-			conf = priv->sor.lvdsconf;
+			*conf = priv->sor.lvdsconf;
 			break;
 		case DCB_OUTPUT_DP:
-			conf = (ctrl & 0x00000f00) >> 8;
+			*conf = (ctrl & 0x00000f00) >> 8;
 			break;
 		case DCB_OUTPUT_ANALOG:
 		default:
-			conf = 0x00ff;
+			*conf = 0x00ff;
 			break;
 		}
 	} else {
-		conf = (ctrl & 0x00000f00) >> 8;
+		*conf = (ctrl & 0x00000f00) >> 8;
 		pclk = pclk / 2;
 	}
 
-	data = nvbios_ocfg_match(bios, data, conf, &ver, &hdr, &cnt, &len, &info2);
+	data = nvbios_ocfg_match(bios, data, *conf, &ver, &hdr, &cnt, &len, &info2);
 	if (data && id < 0xff) {
 		data = nvbios_oclk_match(bios, info2.clkcmp[id], pclk);
 		if (data) {
@@ -1301,7 +1304,7 @@ exec_clkcmp(struct nv50_disp_priv *priv, int head, int id, u32 pclk,
 				.subdev = nv_subdev(priv),
 				.bios = bios,
 				.offset = data,
-				.outp = outp,
+				.outp = &outp->info,
 				.crtc = head,
 				.execute = 1,
 			};
@@ -1310,7 +1313,7 @@ exec_clkcmp(struct nv50_disp_priv *priv, int head, int id, u32 pclk,
 		}
 	}
 
-	return conf;
+	return outp;
 }
 
 static void
@@ -1322,7 +1325,35 @@ nv50_disp_intr_unk10_0(struct nv50_disp_priv *priv, int head)
 static void
 nv50_disp_intr_unk20_0(struct nv50_disp_priv *priv, int head)
 {
-	exec_script(priv, head, 2);
+	struct nvkm_output *outp = exec_script(priv, head, 2);
+
+	/* the binary driver does this outside of the supervisor handling
+	 * (after the third supervisor from a detach).  we (currently?)
+	 * allow both detach/attach to happen in the same set of
+	 * supervisor interrupts, so it would make sense to execute this
+	 * (full power down?) script after all the detach phases of the
+	 * supervisor handling.  like with training if needed from the
+	 * second supervisor, nvidia doesn't do this, so who knows if it's
+	 * entirely safe, but it does appear to work..
+	 *
+	 * without this script being run, on some configurations i've
+	 * seen, switching from DP to TMDS on a DP connector may result
+	 * in a blank screen (SOR_PWR off/on can restore it)
+	 */
+	if (outp && outp->info.type == DCB_OUTPUT_DP) {
+		struct nvkm_output_dp *outpdp = (void *)outp;
+		struct nvbios_init init = {
+			.subdev = nv_subdev(priv),
+			.bios = nouveau_bios(priv),
+			.outp = &outp->info,
+			.crtc = head,
+			.offset = outpdp->info.script[4],
+			.execute = 1,
+		};
+
+		nvbios_exec(&init);
+		atomic_set(&outpdp->lt.done, 0);
+	}
 }
 
 static void
@@ -1444,56 +1475,83 @@ nv50_disp_intr_unk20_2_dp(struct nv50_disp_priv *priv,
 static void
 nv50_disp_intr_unk20_2(struct nv50_disp_priv *priv, int head)
 {
-	struct dcb_output outp;
+	struct nvkm_output *outp;
 	u32 pclk = nv_rd32(priv, 0x610ad0 + (head * 0x540)) & 0x3fffff;
 	u32 hval, hreg = 0x614200 + (head * 0x800);
 	u32 oval, oreg;
-	u32 mask;
-	u32 conf = exec_clkcmp(priv, head, 0xff, pclk, &outp);
-	if (conf != ~0) {
-		if (outp.location == 0 && outp.type == DCB_OUTPUT_DP) {
-			u32 soff = (ffs(outp.or) - 1) * 0x08;
-			u32 ctrl = nv_rd32(priv, 0x610794 + soff);
-			u32 datarate;
-
-			switch ((ctrl & 0x000f0000) >> 16) {
-			case 6: datarate = pclk * 30 / 8; break;
-			case 5: datarate = pclk * 24 / 8; break;
-			case 2:
-			default:
-				datarate = pclk * 18 / 8;
-				break;
-			}
+	u32 mask, conf;
 
-			nouveau_dp_train(&priv->base, priv->sor.dp,
-					 &outp, head, datarate);
-		}
+	outp = exec_clkcmp(priv, head, 0xff, pclk, &conf);
+	if (!outp)
+		return;
+
+	/* we allow both encoder attach and detach operations to occur
+	 * within a single supervisor (ie. modeset) sequence.  the
+	 * encoder detach scripts quite often switch off power to the
+	 * lanes, which requires the link to be re-trained.
+	 *
+	 * this is not generally an issue as the sink "must" (heh)
+	 * signal an irq when it's lost sync so the driver can
+	 * re-train.
+	 *
+	 * however, on some boards, if one does not configure at least
+	 * the gpu side of the link *before* attaching, then various
+	 * things can go horribly wrong (PDISP disappearing from mmio,
+	 * third supervisor never happens, etc).
+	 *
+	 * the solution is simply to retrain here, if necessary.  last
+	 * i checked, the binary driver userspace does not appear to
+	 * trigger this situation (it forces an UPDATE between steps).
+	 */
+	if (outp->info.type == DCB_OUTPUT_DP) {
+		u32 soff = (ffs(outp->info.or) - 1) * 0x08;
+		u32 ctrl, datarate;
 
-		exec_clkcmp(priv, head, 0, pclk, &outp);
-
-		if (!outp.location && outp.type == DCB_OUTPUT_ANALOG) {
-			oreg = 0x614280 + (ffs(outp.or) - 1) * 0x800;
-			oval = 0x00000000;
-			hval = 0x00000000;
-			mask = 0xffffffff;
-		} else
-		if (!outp.location) {
-			if (outp.type == DCB_OUTPUT_DP)
-				nv50_disp_intr_unk20_2_dp(priv, &outp, pclk);
-			oreg = 0x614300 + (ffs(outp.or) - 1) * 0x800;
-			oval = (conf & 0x0100) ? 0x00000101 : 0x00000000;
-			hval = 0x00000000;
-			mask = 0x00000707;
+		if (outp->info.location == 0) {
+			ctrl = nv_rd32(priv, 0x610794 + soff);
+			soff = 1;
 		} else {
-			oreg = 0x614380 + (ffs(outp.or) - 1) * 0x800;
-			oval = 0x00000001;
-			hval = 0x00000001;
-			mask = 0x00000707;
+			ctrl = nv_rd32(priv, 0x610b80 + soff);
+			soff = 2;
 		}
 
-		nv_mask(priv, hreg, 0x0000000f, hval);
-		nv_mask(priv, oreg, mask, oval);
+		switch ((ctrl & 0x000f0000) >> 16) {
+		case 6: datarate = pclk * 30 / 8; break;
+		case 5: datarate = pclk * 24 / 8; break;
+		case 2:
+		default:
+			datarate = pclk * 18 / 8;
+			break;
+		}
+
+		if (nvkm_output_dp_train(outp, datarate / soff, true))
+			ERR("link not trained before attach\n");
 	}
+
+	exec_clkcmp(priv, head, 0, pclk, &conf);
+
+	if (!outp->info.location && outp->info.type == DCB_OUTPUT_ANALOG) {
+		oreg = 0x614280 + (ffs(outp->info.or) - 1) * 0x800;
+		oval = 0x00000000;
+		hval = 0x00000000;
+		mask = 0xffffffff;
+	} else
+	if (!outp->info.location) {
+		if (outp->info.type == DCB_OUTPUT_DP)
+			nv50_disp_intr_unk20_2_dp(priv, &outp->info, pclk);
+		oreg = 0x614300 + (ffs(outp->info.or) - 1) * 0x800;
+		oval = (conf & 0x0100) ? 0x00000101 : 0x00000000;
+		hval = 0x00000000;
+		mask = 0x00000707;
+	} else {
+		oreg = 0x614380 + (ffs(outp->info.or) - 1) * 0x800;
+		oval = 0x00000001;
+		hval = 0x00000001;
+		mask = 0x00000707;
+	}
+
+	nv_mask(priv, hreg, 0x0000000f, hval);
+	nv_mask(priv, oreg, mask, oval);
 }
 
 /* If programming a TMDS output on a SOR that can also be configured for
@@ -1521,30 +1579,16 @@ nv50_disp_intr_unk40_0_tmds(struct nv50_disp_priv *priv, struct dcb_output *outp
 static void
 nv50_disp_intr_unk40_0(struct nv50_disp_priv *priv, int head)
 {
-	struct dcb_output outp;
+	struct nvkm_output *outp;
 	u32 pclk = nv_rd32(priv, 0x610ad0 + (head * 0x540)) & 0x3fffff;
-	if (exec_clkcmp(priv, head, 1, pclk, &outp) != ~0) {
-		if (outp.location == 0 && outp.type == DCB_OUTPUT_TMDS)
-			nv50_disp_intr_unk40_0_tmds(priv, &outp);
-		else
-		if (outp.location == 1 && outp.type == DCB_OUTPUT_DP) {
-			u32 soff = (ffs(outp.or) - 1) * 0x08;
-			u32 ctrl = nv_rd32(priv, 0x610b84 + soff);
-			u32 datarate;
-
-			switch ((ctrl & 0x000f0000) >> 16) {
-			case 6: datarate = pclk * 30 / 8; break;
-			case 5: datarate = pclk * 24 / 8; break;
-			case 2:
-			default:
-				datarate = pclk * 18 / 8;
-				break;
-			}
+	u32 conf;
 
-			nouveau_dp_train(&priv->base, priv->pior.dp,
-					 &outp, head, datarate);
-		}
-	}
+	outp = exec_clkcmp(priv, head, 1, pclk, &conf);
+	if (!outp)
+		return;
+
+	if (outp->info.location == 0 && outp->info.type == DCB_OUTPUT_TMDS)
+		nv50_disp_intr_unk40_0_tmds(priv, &outp->info);
 }
 
 void
@@ -1610,13 +1654,13 @@ nv50_disp_intr(struct nouveau_subdev *subdev)
 	}
 
 	if (intr1 & 0x00000004) {
-		nouveau_event_trigger(priv->base.vblank, 0);
+		nouveau_event_trigger(priv->base.vblank, 1, 0);
 		nv_wr32(priv, 0x610024, 0x00000004);
 		intr1 &= ~0x00000004;
 	}
 
 	if (intr1 & 0x00000008) {
-		nouveau_event_trigger(priv->base.vblank, 1);
+		nouveau_event_trigger(priv->base.vblank, 1, 1);
 		nv_wr32(priv, 0x610024, 0x00000008);
 		intr1 &= ~0x00000008;
 	}
@@ -1656,11 +1700,16 @@ nv50_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	priv->dac.sense = nv50_dac_sense;
 	priv->sor.power = nv50_sor_power;
 	priv->pior.power = nv50_pior_power;
-	priv->pior.dp = &nv50_pior_dp_func;
 	return 0;
 }
 
 struct nouveau_oclass *
+nv50_disp_outp_sclass[] = {
+	&nv50_pior_dp_impl.base.base,
+	NULL
+};
+
+struct nouveau_oclass *
 nv50_disp_oclass = &(struct nv50_disp_impl) {
 	.base.base.handle = NV_ENGINE(DISP, 0x50),
 	.base.base.ofuncs = &(struct nouveau_ofuncs) {
@@ -1669,6 +1718,7 @@ nv50_disp_oclass = &(struct nv50_disp_impl) {
 		.init = _nouveau_disp_init,
 		.fini = _nouveau_disp_fini,
 	},
+	.base.outp =  nv50_disp_outp_sclass,
 	.mthd.core = &nv50_disp_mast_mthd_chan,
 	.mthd.base = &nv50_disp_sync_mthd_chan,
 	.mthd.ovly = &nv50_disp_ovly_mthd_chan,
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nv50.h b/drivers/gpu/drm/nouveau/core/engine/disp/nv50.h
index 48d59db47f0d..1a886472b6f5 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nv50.h
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nv50.h
@@ -11,6 +11,8 @@
 
 #include "dport.h"
 #include "priv.h"
+#include "outp.h"
+#include "outpdp.h"
 
 struct nv50_disp_impl {
 	struct nouveau_disp_impl base;
@@ -43,13 +45,11 @@ struct nv50_disp_priv {
 		int (*hda_eld)(struct nv50_disp_priv *, int sor, u8 *, u32);
 		int (*hdmi)(struct nv50_disp_priv *, int head, int sor, u32);
 		u32 lvdsconf;
-		const struct nouveau_dp_func *dp;
 	} sor;
 	struct {
 		int nr;
 		int (*power)(struct nv50_disp_priv *, int ext, u32 data);
 		u8 type[3];
-		const struct nouveau_dp_func *dp;
 	} pior;
 };
 
@@ -199,4 +199,14 @@ void nvd0_disp_intr(struct nouveau_subdev *);
 extern const struct nv50_disp_mthd_chan nve0_disp_mast_mthd_chan;
 extern const struct nv50_disp_mthd_chan nve0_disp_ovly_mthd_chan;
 
+extern struct nvkm_output_dp_impl nv50_pior_dp_impl;
+extern struct nouveau_oclass *nv50_disp_outp_sclass[];
+
+extern struct nvkm_output_dp_impl nv94_sor_dp_impl;
+int nv94_sor_dp_lnk_pwr(struct nvkm_output_dp *, int);
+extern struct nouveau_oclass *nv94_disp_outp_sclass[];
+
+extern struct nvkm_output_dp_impl nvd0_sor_dp_impl;
+extern struct nouveau_oclass *nvd0_disp_outp_sclass[];
+
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nv84.c b/drivers/gpu/drm/nouveau/core/engine/disp/nv84.c
index 98c5b19bc2b0..1cc62e434683 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nv84.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nv84.c
@@ -264,7 +264,6 @@ nv84_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	priv->sor.power = nv50_sor_power;
 	priv->sor.hdmi = nv84_hdmi_ctrl;
 	priv->pior.power = nv50_pior_power;
-	priv->pior.dp = &nv50_pior_dp_func;
 	return 0;
 }
 
@@ -277,6 +276,7 @@ nv84_disp_oclass = &(struct nv50_disp_impl) {
 		.init = _nouveau_disp_init,
 		.fini = _nouveau_disp_fini,
 	},
+	.base.outp =  nv50_disp_outp_sclass,
 	.mthd.core = &nv84_disp_mast_mthd_chan,
 	.mthd.base = &nv84_disp_sync_mthd_chan,
 	.mthd.ovly = &nv84_disp_ovly_mthd_chan,
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nv94.c b/drivers/gpu/drm/nouveau/core/engine/disp/nv94.c
index 6844061c7e04..4f718a9f5aef 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nv94.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nv94.c
@@ -77,6 +77,7 @@ nv94_disp_base_omthds[] = {
 	{ SOR_MTHD(NV50_DISP_SOR_PWR)         , nv50_sor_mthd },
 	{ SOR_MTHD(NV84_DISP_SOR_HDMI_PWR)    , nv50_sor_mthd },
 	{ SOR_MTHD(NV50_DISP_SOR_LVDS_SCRIPT) , nv50_sor_mthd },
+	{ SOR_MTHD(NV94_DISP_SOR_DP_PWR)      , nv50_sor_mthd },
 	{ DAC_MTHD(NV50_DISP_DAC_PWR)         , nv50_dac_mthd },
 	{ DAC_MTHD(NV50_DISP_DAC_LOAD)        , nv50_dac_mthd },
 	{ PIOR_MTHD(NV50_DISP_PIOR_PWR)       , nv50_pior_mthd },
@@ -122,13 +123,18 @@ nv94_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	priv->dac.sense = nv50_dac_sense;
 	priv->sor.power = nv50_sor_power;
 	priv->sor.hdmi = nv84_hdmi_ctrl;
-	priv->sor.dp = &nv94_sor_dp_func;
 	priv->pior.power = nv50_pior_power;
-	priv->pior.dp = &nv50_pior_dp_func;
 	return 0;
 }
 
 struct nouveau_oclass *
+nv94_disp_outp_sclass[] = {
+	&nv50_pior_dp_impl.base.base,
+	&nv94_sor_dp_impl.base.base,
+	NULL
+};
+
+struct nouveau_oclass *
 nv94_disp_oclass = &(struct nv50_disp_impl) {
 	.base.base.handle = NV_ENGINE(DISP, 0x88),
 	.base.base.ofuncs = &(struct nouveau_ofuncs) {
@@ -137,6 +143,7 @@ nv94_disp_oclass = &(struct nv50_disp_impl) {
 		.init = _nouveau_disp_init,
 		.fini = _nouveau_disp_fini,
 	},
+	.base.outp =  nv94_disp_outp_sclass,
 	.mthd.core = &nv94_disp_mast_mthd_chan,
 	.mthd.base = &nv84_disp_sync_mthd_chan,
 	.mthd.ovly = &nv84_disp_ovly_mthd_chan,
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nva0.c b/drivers/gpu/drm/nouveau/core/engine/disp/nva0.c
index 88c96241c02a..6237a9a36f70 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nva0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nva0.c
@@ -126,7 +126,6 @@ nva0_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	priv->sor.power = nv50_sor_power;
 	priv->sor.hdmi = nv84_hdmi_ctrl;
 	priv->pior.power = nv50_pior_power;
-	priv->pior.dp = &nv50_pior_dp_func;
 	return 0;
 }
 
@@ -139,6 +138,7 @@ nva0_disp_oclass = &(struct nv50_disp_impl) {
 		.init = _nouveau_disp_init,
 		.fini = _nouveau_disp_fini,
 	},
+	.base.outp =  nv50_disp_outp_sclass,
 	.mthd.core = &nv84_disp_mast_mthd_chan,
 	.mthd.base = &nv84_disp_sync_mthd_chan,
 	.mthd.ovly = &nva0_disp_ovly_mthd_chan,
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nva3.c b/drivers/gpu/drm/nouveau/core/engine/disp/nva3.c
index 46cb2ce0e82a..019124d4782b 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nva3.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nva3.c
@@ -50,6 +50,7 @@ nva3_disp_base_omthds[] = {
 	{ SOR_MTHD(NVA3_DISP_SOR_HDA_ELD)     , nv50_sor_mthd },
 	{ SOR_MTHD(NV84_DISP_SOR_HDMI_PWR)    , nv50_sor_mthd },
 	{ SOR_MTHD(NV50_DISP_SOR_LVDS_SCRIPT) , nv50_sor_mthd },
+	{ SOR_MTHD(NV94_DISP_SOR_DP_PWR)      , nv50_sor_mthd },
 	{ DAC_MTHD(NV50_DISP_DAC_PWR)         , nv50_dac_mthd },
 	{ DAC_MTHD(NV50_DISP_DAC_LOAD)        , nv50_dac_mthd },
 	{ PIOR_MTHD(NV50_DISP_PIOR_PWR)       , nv50_pior_mthd },
@@ -96,9 +97,7 @@ nva3_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	priv->sor.power = nv50_sor_power;
 	priv->sor.hda_eld = nva3_hda_eld;
 	priv->sor.hdmi = nva3_hdmi_ctrl;
-	priv->sor.dp = &nv94_sor_dp_func;
 	priv->pior.power = nv50_pior_power;
-	priv->pior.dp = &nv50_pior_dp_func;
 	return 0;
 }
 
@@ -111,6 +110,7 @@ nva3_disp_oclass = &(struct nv50_disp_impl) {
 		.init = _nouveau_disp_init,
 		.fini = _nouveau_disp_fini,
 	},
+	.base.outp =  nv94_disp_outp_sclass,
 	.mthd.core = &nv94_disp_mast_mthd_chan,
 	.mthd.base = &nv84_disp_sync_mthd_chan,
 	.mthd.ovly = &nv84_disp_ovly_mthd_chan,
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nvd0.c b/drivers/gpu/drm/nouveau/core/engine/disp/nvd0.c
index 876de9ac3793..48aa38a87e3f 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nvd0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nvd0.c
@@ -748,13 +748,13 @@ nvd0_disp_base_scanoutpos(struct nouveau_object *object, u32 mthd,
 }
 
 static void
-nvd0_disp_base_vblank_enable(struct nouveau_event *event, int head)
+nvd0_disp_base_vblank_enable(struct nouveau_event *event, int type, int head)
 {
 	nv_mask(event->priv, 0x6100c0 + (head * 0x800), 0x00000001, 0x00000001);
 }
 
 static void
-nvd0_disp_base_vblank_disable(struct nouveau_event *event, int head)
+nvd0_disp_base_vblank_disable(struct nouveau_event *event, int type, int head)
 {
 	nv_mask(event->priv, 0x6100c0 + (head * 0x800), 0x00000001, 0x00000000);
 }
@@ -887,6 +887,7 @@ nvd0_disp_base_omthds[] = {
 	{ SOR_MTHD(NVA3_DISP_SOR_HDA_ELD)     , nv50_sor_mthd },
 	{ SOR_MTHD(NV84_DISP_SOR_HDMI_PWR)    , nv50_sor_mthd },
 	{ SOR_MTHD(NV50_DISP_SOR_LVDS_SCRIPT) , nv50_sor_mthd },
+	{ SOR_MTHD(NV94_DISP_SOR_DP_PWR)      , nv50_sor_mthd },
 	{ DAC_MTHD(NV50_DISP_DAC_PWR)         , nv50_dac_mthd },
 	{ DAC_MTHD(NV50_DISP_DAC_LOAD)        , nv50_dac_mthd },
 	{ PIOR_MTHD(NV50_DISP_PIOR_PWR)       , nv50_pior_mthd },
@@ -915,19 +916,20 @@ nvd0_disp_sclass[] = {
  * Display engine implementation
  ******************************************************************************/
 
-static u16
-exec_lookup(struct nv50_disp_priv *priv, int head, int outp, u32 ctrl,
-	    struct dcb_output *dcb, u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
+static struct nvkm_output *
+exec_lookup(struct nv50_disp_priv *priv, int head, int or, u32 ctrl,
+	    u32 *data, u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
 	    struct nvbios_outp *info)
 {
 	struct nouveau_bios *bios = nouveau_bios(priv);
-	u16 mask, type, data;
+	struct nvkm_output *outp;
+	u16 mask, type;
 
-	if (outp < 4) {
+	if (or < 4) {
 		type = DCB_OUTPUT_ANALOG;
 		mask = 0;
 	} else {
-		outp -= 4;
+		or -= 4;
 		switch (ctrl & 0x00000f00) {
 		case 0x00000000: type = DCB_OUTPUT_LVDS; mask = 1; break;
 		case 0x00000100: type = DCB_OUTPUT_TMDS; mask = 1; break;
@@ -939,101 +941,106 @@ exec_lookup(struct nv50_disp_priv *priv, int head, int outp, u32 ctrl,
 			nv_error(priv, "unknown SOR mc 0x%08x\n", ctrl);
 			return 0x0000;
 		}
-		dcb->sorconf.link = mask;
 	}
 
 	mask  = 0x00c0 & (mask << 6);
-	mask |= 0x0001 << outp;
+	mask |= 0x0001 << or;
 	mask |= 0x0100 << head;
 
-	data = dcb_outp_match(bios, type, mask, ver, hdr, dcb);
-	if (!data)
-		return 0x0000;
+	list_for_each_entry(outp, &priv->base.outp, head) {
+		if ((outp->info.hasht & 0xff) == type &&
+		    (outp->info.hashm & mask) == mask) {
+			*data = nvbios_outp_match(bios, outp->info.hasht,
+							outp->info.hashm,
+						  ver, hdr, cnt, len, info);
+			if (!*data)
+				return NULL;
+			return outp;
+		}
+	}
 
-	return nvbios_outp_match(bios, type, mask, ver, hdr, cnt, len, info);
+	return NULL;
 }
 
-static bool
+static struct nvkm_output *
 exec_script(struct nv50_disp_priv *priv, int head, int id)
 {
 	struct nouveau_bios *bios = nouveau_bios(priv);
+	struct nvkm_output *outp;
 	struct nvbios_outp info;
-	struct dcb_output dcb;
 	u8  ver, hdr, cnt, len;
-	u32 ctrl = 0x00000000;
-	u16 data;
-	int outp;
+	u32 data, ctrl = 0;
+	int or;
 
-	for (outp = 0; !(ctrl & (1 << head)) && outp < 8; outp++) {
-		ctrl = nv_rd32(priv, 0x640180 + (outp * 0x20));
+	for (or = 0; !(ctrl & (1 << head)) && or < 8; or++) {
+		ctrl = nv_rd32(priv, 0x640180 + (or * 0x20));
 		if (ctrl & (1 << head))
 			break;
 	}
 
-	if (outp == 8)
-		return false;
+	if (or == 8)
+		return NULL;
 
-	data = exec_lookup(priv, head, outp, ctrl, &dcb, &ver, &hdr, &cnt, &len, &info);
-	if (data) {
+	outp = exec_lookup(priv, head, or, ctrl, &data, &ver, &hdr, &cnt, &len, &info);
+	if (outp) {
 		struct nvbios_init init = {
 			.subdev = nv_subdev(priv),
 			.bios = bios,
 			.offset = info.script[id],
-			.outp = &dcb,
+			.outp = &outp->info,
 			.crtc = head,
 			.execute = 1,
 		};
 
-		return nvbios_exec(&init) == 0;
+		nvbios_exec(&init);
 	}
 
-	return false;
+	return outp;
 }
 
-static u32
-exec_clkcmp(struct nv50_disp_priv *priv, int head, int id,
-	    u32 pclk, struct dcb_output *dcb)
+static struct nvkm_output *
+exec_clkcmp(struct nv50_disp_priv *priv, int head, int id, u32 pclk, u32 *conf)
 {
 	struct nouveau_bios *bios = nouveau_bios(priv);
+	struct nvkm_output *outp;
 	struct nvbios_outp info1;
 	struct nvbios_ocfg info2;
 	u8  ver, hdr, cnt, len;
-	u32 ctrl = 0x00000000;
-	u32 data, conf = ~0;
-	int outp;
+	u32 data, ctrl = 0;
+	int or;
 
-	for (outp = 0; !(ctrl & (1 << head)) && outp < 8; outp++) {
-		ctrl = nv_rd32(priv, 0x660180 + (outp * 0x20));
+	for (or = 0; !(ctrl & (1 << head)) && or < 8; or++) {
+		ctrl = nv_rd32(priv, 0x660180 + (or * 0x20));
 		if (ctrl & (1 << head))
 			break;
 	}
 
-	if (outp == 8)
-		return conf;
+	if (or == 8)
+		return NULL;
 
-	data = exec_lookup(priv, head, outp, ctrl, dcb, &ver, &hdr, &cnt, &len, &info1);
-	if (data == 0x0000)
-		return conf;
+	outp = exec_lookup(priv, head, or, ctrl, &data, &ver, &hdr, &cnt, &len, &info1);
+	if (!outp)
+		return NULL;
 
-	switch (dcb->type) {
+	switch (outp->info.type) {
 	case DCB_OUTPUT_TMDS:
-		conf = (ctrl & 0x00000f00) >> 8;
+		*conf = (ctrl & 0x00000f00) >> 8;
 		if (pclk >= 165000)
-			conf |= 0x0100;
+			*conf |= 0x0100;
 		break;
 	case DCB_OUTPUT_LVDS:
-		conf = priv->sor.lvdsconf;
+		*conf = priv->sor.lvdsconf;
 		break;
 	case DCB_OUTPUT_DP:
-		conf = (ctrl & 0x00000f00) >> 8;
+		*conf = (ctrl & 0x00000f00) >> 8;
 		break;
 	case DCB_OUTPUT_ANALOG:
 	default:
-		conf = 0x00ff;
+		*conf = 0x00ff;
 		break;
 	}
 
-	data = nvbios_ocfg_match(bios, data, conf, &ver, &hdr, &cnt, &len, &info2);
+	data = nvbios_ocfg_match(bios, data, *conf, &ver, &hdr, &cnt, &len, &info2);
 	if (data && id < 0xff) {
 		data = nvbios_oclk_match(bios, info2.clkcmp[id], pclk);
 		if (data) {
@@ -1041,7 +1048,7 @@ exec_clkcmp(struct nv50_disp_priv *priv, int head, int id,
 				.subdev = nv_subdev(priv),
 				.bios = bios,
 				.offset = data,
-				.outp = dcb,
+				.outp = &outp->info,
 				.crtc = head,
 				.execute = 1,
 			};
@@ -1050,7 +1057,7 @@ exec_clkcmp(struct nv50_disp_priv *priv, int head, int id,
 		}
 	}
 
-	return conf;
+	return outp;
 }
 
 static void
@@ -1062,7 +1069,23 @@ nvd0_disp_intr_unk1_0(struct nv50_disp_priv *priv, int head)
 static void
 nvd0_disp_intr_unk2_0(struct nv50_disp_priv *priv, int head)
 {
-	exec_script(priv, head, 2);
+	struct nvkm_output *outp = exec_script(priv, head, 2);
+
+	/* see note in nv50_disp_intr_unk20_0() */
+	if (outp && outp->info.type == DCB_OUTPUT_DP) {
+		struct nvkm_output_dp *outpdp = (void *)outp;
+		struct nvbios_init init = {
+			.subdev = nv_subdev(priv),
+			.bios = nouveau_bios(priv),
+			.outp = &outp->info,
+			.crtc = head,
+			.offset = outpdp->info.script[4],
+			.execute = 1,
+		};
+
+		nvbios_exec(&init);
+		atomic_set(&outpdp->lt.done, 0);
+	}
 }
 
 static void
@@ -1124,49 +1147,52 @@ nvd0_disp_intr_unk2_2_tu(struct nv50_disp_priv *priv, int head,
 static void
 nvd0_disp_intr_unk2_2(struct nv50_disp_priv *priv, int head)
 {
-	struct dcb_output outp;
+	struct nvkm_output *outp;
 	u32 pclk = nv_rd32(priv, 0x660450 + (head * 0x300)) / 1000;
-	u32 conf = exec_clkcmp(priv, head, 0xff, pclk, &outp);
-	if (conf != ~0) {
-		u32 addr, data;
-
-		if (outp.type == DCB_OUTPUT_DP) {
-			u32 sync = nv_rd32(priv, 0x660404 + (head * 0x300));
-			switch ((sync & 0x000003c0) >> 6) {
-			case 6: pclk = pclk * 30 / 8; break;
-			case 5: pclk = pclk * 24 / 8; break;
-			case 2:
-			default:
-				pclk = pclk * 18 / 8;
-				break;
-			}
-
-			nouveau_dp_train(&priv->base, priv->sor.dp,
-					 &outp, head, pclk);
+	u32 conf, addr, data;
+
+	outp = exec_clkcmp(priv, head, 0xff, pclk, &conf);
+	if (!outp)
+		return;
+
+	/* see note in nv50_disp_intr_unk20_2() */
+	if (outp->info.type == DCB_OUTPUT_DP) {
+		u32 sync = nv_rd32(priv, 0x660404 + (head * 0x300));
+		switch ((sync & 0x000003c0) >> 6) {
+		case 6: pclk = pclk * 30 / 8; break;
+		case 5: pclk = pclk * 24 / 8; break;
+		case 2:
+		default:
+			pclk = pclk * 18 / 8;
+			break;
 		}
 
-		exec_clkcmp(priv, head, 0, pclk, &outp);
+		if (nvkm_output_dp_train(outp, pclk, true))
+			ERR("link not trained before attach\n");
+	}
 
-		if (outp.type == DCB_OUTPUT_ANALOG) {
-			addr = 0x612280 + (ffs(outp.or) - 1) * 0x800;
-			data = 0x00000000;
-		} else {
-			if (outp.type == DCB_OUTPUT_DP)
-				nvd0_disp_intr_unk2_2_tu(priv, head, &outp);
-			addr = 0x612300 + (ffs(outp.or) - 1) * 0x800;
-			data = (conf & 0x0100) ? 0x00000101 : 0x00000000;
-		}
+	exec_clkcmp(priv, head, 0, pclk, &conf);
 
-		nv_mask(priv, addr, 0x00000707, data);
+	if (outp->info.type == DCB_OUTPUT_ANALOG) {
+		addr = 0x612280 + (ffs(outp->info.or) - 1) * 0x800;
+		data = 0x00000000;
+	} else {
+		if (outp->info.type == DCB_OUTPUT_DP)
+			nvd0_disp_intr_unk2_2_tu(priv, head, &outp->info);
+		addr = 0x612300 + (ffs(outp->info.or) - 1) * 0x800;
+		data = (conf & 0x0100) ? 0x00000101 : 0x00000000;
 	}
+
+	nv_mask(priv, addr, 0x00000707, data);
 }
 
 static void
 nvd0_disp_intr_unk4_0(struct nv50_disp_priv *priv, int head)
 {
-	struct dcb_output outp;
 	u32 pclk = nv_rd32(priv, 0x660450 + (head * 0x300)) / 1000;
-	exec_clkcmp(priv, head, 1, pclk, &outp);
+	u32 conf;
+
+	exec_clkcmp(priv, head, 1, pclk, &conf);
 }
 
 void
@@ -1240,7 +1266,7 @@ nvd0_disp_intr_error(struct nv50_disp_priv *priv, int chid)
 		 chid, (mthd & 0x0000ffc), data, mthd, unkn);
 
 	if (chid == 0) {
-		switch (mthd) {
+		switch (mthd & 0xffc) {
 		case 0x0080:
 			nv50_disp_mthd_chan(priv, NV_DBG_ERROR, chid - 0,
 					    impl->mthd.core);
@@ -1250,7 +1276,7 @@ nvd0_disp_intr_error(struct nv50_disp_priv *priv, int chid)
 		}
 	} else
 	if (chid <= 4) {
-		switch (mthd) {
+		switch (mthd & 0xffc) {
 		case 0x0080:
 			nv50_disp_mthd_chan(priv, NV_DBG_ERROR, chid - 1,
 					    impl->mthd.base);
@@ -1260,7 +1286,7 @@ nvd0_disp_intr_error(struct nv50_disp_priv *priv, int chid)
 		}
 	} else
 	if (chid <= 8) {
-		switch (mthd) {
+		switch (mthd & 0xffc) {
 		case 0x0080:
 			nv50_disp_mthd_chan(priv, NV_DBG_ERROR, chid - 5,
 					    impl->mthd.ovly);
@@ -1317,7 +1343,7 @@ nvd0_disp_intr(struct nouveau_subdev *subdev)
 		if (mask & intr) {
 			u32 stat = nv_rd32(priv, 0x6100bc + (i * 0x800));
 			if (stat & 0x00000001)
-				nouveau_event_trigger(priv->base.vblank, i);
+				nouveau_event_trigger(priv->base.vblank, 1, i);
 			nv_mask(priv, 0x6100bc + (i * 0x800), 0, 0);
 			nv_rd32(priv, 0x6100c0 + (i * 0x800));
 		}
@@ -1352,11 +1378,16 @@ nvd0_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	priv->sor.power = nv50_sor_power;
 	priv->sor.hda_eld = nvd0_hda_eld;
 	priv->sor.hdmi = nvd0_hdmi_ctrl;
-	priv->sor.dp = &nvd0_sor_dp_func;
 	return 0;
 }
 
 struct nouveau_oclass *
+nvd0_disp_outp_sclass[] = {
+	&nvd0_sor_dp_impl.base.base,
+	NULL
+};
+
+struct nouveau_oclass *
 nvd0_disp_oclass = &(struct nv50_disp_impl) {
 	.base.base.handle = NV_ENGINE(DISP, 0x90),
 	.base.base.ofuncs = &(struct nouveau_ofuncs) {
@@ -1365,6 +1396,7 @@ nvd0_disp_oclass = &(struct nv50_disp_impl) {
 		.init = _nouveau_disp_init,
 		.fini = _nouveau_disp_fini,
 	},
+	.base.outp =  nvd0_disp_outp_sclass,
 	.mthd.core = &nvd0_disp_mast_mthd_chan,
 	.mthd.base = &nvd0_disp_sync_mthd_chan,
 	.mthd.ovly = &nvd0_disp_ovly_mthd_chan,
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nve0.c b/drivers/gpu/drm/nouveau/core/engine/disp/nve0.c
index 44e0b8f34c1a..11328e3f5df1 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nve0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nve0.c
@@ -246,7 +246,6 @@ nve0_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	priv->sor.power = nv50_sor_power;
 	priv->sor.hda_eld = nvd0_hda_eld;
 	priv->sor.hdmi = nvd0_hdmi_ctrl;
-	priv->sor.dp = &nvd0_sor_dp_func;
 	return 0;
 }
 
@@ -259,6 +258,7 @@ nve0_disp_oclass = &(struct nv50_disp_impl) {
 		.init = _nouveau_disp_init,
 		.fini = _nouveau_disp_fini,
 	},
+	.base.outp =  nvd0_disp_outp_sclass,
 	.mthd.core = &nve0_disp_mast_mthd_chan,
 	.mthd.base = &nvd0_disp_sync_mthd_chan,
 	.mthd.ovly = &nve0_disp_ovly_mthd_chan,
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nvf0.c b/drivers/gpu/drm/nouveau/core/engine/disp/nvf0.c
index 482585d375fa..104388081d73 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nvf0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nvf0.c
@@ -81,7 +81,6 @@ nvf0_disp_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	priv->sor.power = nv50_sor_power;
 	priv->sor.hda_eld = nvd0_hda_eld;
 	priv->sor.hdmi = nvd0_hdmi_ctrl;
-	priv->sor.dp = &nvd0_sor_dp_func;
 	return 0;
 }
 
@@ -94,6 +93,7 @@ nvf0_disp_oclass = &(struct nv50_disp_impl) {
 		.init = _nouveau_disp_init,
 		.fini = _nouveau_disp_fini,
 	},
+	.base.outp =  nvd0_disp_outp_sclass,
 	.mthd.core = &nve0_disp_mast_mthd_chan,
 	.mthd.base = &nvd0_disp_sync_mthd_chan,
 	.mthd.ovly = &nve0_disp_ovly_mthd_chan,
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/outp.c b/drivers/gpu/drm/nouveau/core/engine/disp/outp.c
new file mode 100644
index 000000000000..ad9ba7ccec7f
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/outp.c
@@ -0,0 +1,137 @@
+/*
+ * Copyright 2014 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/i2c.h>
+#include <subdev/bios.h>
+#include <subdev/bios/conn.h>
+
+#include "outp.h"
+
+int
+_nvkm_output_fini(struct nouveau_object *object, bool suspend)
+{
+	struct nvkm_output *outp = (void *)object;
+	nv_ofuncs(outp->conn)->fini(nv_object(outp->conn), suspend);
+	return nouveau_object_fini(&outp->base, suspend);
+}
+
+int
+_nvkm_output_init(struct nouveau_object *object)
+{
+	struct nvkm_output *outp = (void *)object;
+	int ret = nouveau_object_init(&outp->base);
+	if (ret == 0)
+		nv_ofuncs(outp->conn)->init(nv_object(outp->conn));
+	return 0;
+}
+
+void
+_nvkm_output_dtor(struct nouveau_object *object)
+{
+	struct nvkm_output *outp = (void *)object;
+	list_del(&outp->head);
+	nouveau_object_ref(NULL, (void *)&outp->conn);
+	nouveau_object_destroy(&outp->base);
+}
+
+int
+nvkm_output_create_(struct nouveau_object *parent,
+		    struct nouveau_object *engine,
+		    struct nouveau_oclass *oclass,
+		    struct dcb_output *dcbE, int index,
+		    int length, void **pobject)
+{
+	struct nouveau_bios *bios = nouveau_bios(engine);
+	struct nouveau_i2c *i2c = nouveau_i2c(parent);
+	struct nouveau_disp *disp = (void *)engine;
+	struct nvbios_connE connE;
+	struct nvkm_output *outp;
+	u8  ver, hdr;
+	u32 data;
+	int ret;
+
+	ret = nouveau_object_create_(parent, engine, oclass, 0, length, pobject);
+	outp = *pobject;
+	if (ret)
+		return ret;
+
+	outp->info = *dcbE;
+	outp->index = index;
+
+	DBG("type %02x loc %d or %d link %d con %x edid %x bus %d head %x\n",
+	    dcbE->type, dcbE->location, dcbE->or, dcbE->type >= 2 ?
+	    dcbE->sorconf.link : 0, dcbE->connector, dcbE->i2c_index,
+	    dcbE->bus, dcbE->heads);
+
+	outp->port = i2c->find(i2c, outp->info.i2c_index);
+	outp->edid = outp->port;
+
+	data = nvbios_connEp(bios, outp->info.connector, &ver, &hdr, &connE);
+	if (!data) {
+		DBG("vbios connector data not found\n");
+		memset(&connE, 0x00, sizeof(connE));
+		connE.type = DCB_CONNECTOR_NONE;
+	}
+
+	ret = nouveau_object_ctor(parent, engine, nvkm_connector_oclass,
+				 &connE, outp->info.connector,
+				 (struct nouveau_object **)&outp->conn);
+	if (ret < 0) {
+		ERR("error %d creating connector, disabling\n", ret);
+		return ret;
+	}
+
+	list_add_tail(&outp->head, &disp->outp);
+	return 0;
+}
+
+int
+_nvkm_output_ctor(struct nouveau_object *parent,
+		  struct nouveau_object *engine,
+		  struct nouveau_oclass *oclass, void *dcbE, u32 index,
+		  struct nouveau_object **pobject)
+{
+	struct nvkm_output *outp;
+	int ret;
+
+	ret = nvkm_output_create(parent, engine, oclass, dcbE, index, &outp);
+	*pobject = nv_object(outp);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+struct nouveau_oclass *
+nvkm_output_oclass = &(struct nvkm_output_impl) {
+	.base = {
+		.handle = 0,
+		.ofuncs = &(struct nouveau_ofuncs) {
+			.ctor = _nvkm_output_ctor,
+			.dtor = _nvkm_output_dtor,
+			.init = _nvkm_output_init,
+			.fini = _nvkm_output_fini,
+		},
+	},
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/outp.h b/drivers/gpu/drm/nouveau/core/engine/disp/outp.h
new file mode 100644
index 000000000000..bc76fbf85710
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/outp.h
@@ -0,0 +1,59 @@
+#ifndef __NVKM_DISP_OUTP_H__
+#define __NVKM_DISP_OUTP_H__
+
+#include "priv.h"
+
+struct nvkm_output {
+	struct nouveau_object base;
+	struct list_head head;
+
+	struct dcb_output info;
+	int index;
+
+	struct nouveau_i2c_port *port;
+	struct nouveau_i2c_port *edid;
+
+	struct nvkm_connector *conn;
+};
+
+#define nvkm_output_create(p,e,c,b,i,d)                                        \
+	nvkm_output_create_((p), (e), (c), (b), (i), sizeof(**d), (void **)d)
+#define nvkm_output_destroy(d) ({                                              \
+	struct nvkm_output *_outp = (d);                                       \
+	_nvkm_output_dtor(nv_object(_outp));                                   \
+})
+#define nvkm_output_init(d) ({                                                 \
+	struct nvkm_output *_outp = (d);                                       \
+	_nvkm_output_init(nv_object(_outp));                                   \
+})
+#define nvkm_output_fini(d,s) ({                                               \
+	struct nvkm_output *_outp = (d);                                       \
+	_nvkm_output_fini(nv_object(_outp), (s));                              \
+})
+
+int nvkm_output_create_(struct nouveau_object *, struct nouveau_object *,
+			struct nouveau_oclass *, struct dcb_output *,
+			int, int, void **);
+
+int  _nvkm_output_ctor(struct nouveau_object *, struct nouveau_object *,
+		       struct nouveau_oclass *, void *, u32,
+		       struct nouveau_object **);
+void _nvkm_output_dtor(struct nouveau_object *);
+int  _nvkm_output_init(struct nouveau_object *);
+int  _nvkm_output_fini(struct nouveau_object *, bool);
+
+struct nvkm_output_impl {
+	struct nouveau_oclass base;
+};
+
+#ifndef MSG
+#define MSG(l,f,a...) do {                                                     \
+	struct nvkm_output *_outp = (void *)outp;                              \
+	nv_##l(nv_object(outp)->engine, "%02x:%04x:%04x: "f, _outp->index,     \
+	       _outp->info.hasht, _outp->info.hashm, ##a);                     \
+} while(0)
+#define DBG(f,a...) MSG(debug, f, ##a)
+#define ERR(f,a...) MSG(error, f, ##a)
+#endif
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/outpdp.c b/drivers/gpu/drm/nouveau/core/engine/disp/outpdp.c
new file mode 100644
index 000000000000..52c299c3d300
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/outpdp.c
@@ -0,0 +1,276 @@
+/*
+ * Copyright 2014 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <subdev/i2c.h>
+
+#include "outpdp.h"
+#include "conn.h"
+#include "dport.h"
+
+int
+nvkm_output_dp_train(struct nvkm_output *base, u32 datarate, bool wait)
+{
+	struct nvkm_output_dp *outp = (void *)base;
+	bool retrain = true;
+	u8 link[2], stat[3];
+	u32 rate;
+	int ret, i;
+
+	/* check that the link is trained at a high enough rate */
+	ret = nv_rdaux(outp->base.edid, DPCD_LC00_LINK_BW_SET, link, 2);
+	if (ret) {
+		DBG("failed to read link config, assuming no sink\n");
+		goto done;
+	}
+
+	rate = link[0] * 27000 * (link[1] & DPCD_LC01_LANE_COUNT_SET);
+	if (rate < ((datarate / 8) * 10)) {
+		DBG("link not trained at sufficient rate\n");
+		goto done;
+	}
+
+	/* check that link is still trained */
+	ret = nv_rdaux(outp->base.edid, DPCD_LS02, stat, 3);
+	if (ret) {
+		DBG("failed to read link status, assuming no sink\n");
+		goto done;
+	}
+
+	if (stat[2] & DPCD_LS04_INTERLANE_ALIGN_DONE) {
+		for (i = 0; i < (link[1] & DPCD_LC01_LANE_COUNT_SET); i++) {
+			u8 lane = (stat[i >> 1] >> ((i & 1) * 4)) & 0x0f;
+			if (!(lane & DPCD_LS02_LANE0_CR_DONE) ||
+			    !(lane & DPCD_LS02_LANE0_CHANNEL_EQ_DONE) ||
+			    !(lane & DPCD_LS02_LANE0_SYMBOL_LOCKED)) {
+				DBG("lane %d not equalised\n", lane);
+				goto done;
+			}
+		}
+		retrain = false;
+	} else {
+		DBG("no inter-lane alignment\n");
+	}
+
+done:
+	if (retrain || !atomic_read(&outp->lt.done)) {
+		/* no sink, but still need to configure source */
+		if (outp->dpcd[DPCD_RC00_DPCD_REV] == 0x00) {
+			outp->dpcd[DPCD_RC01_MAX_LINK_RATE] =
+				outp->base.info.dpconf.link_bw;
+			outp->dpcd[DPCD_RC02] =
+				outp->base.info.dpconf.link_nr;
+		}
+		atomic_set(&outp->lt.done, 0);
+		schedule_work(&outp->lt.work);
+	} else {
+		nouveau_event_get(outp->irq);
+	}
+
+	if (wait) {
+		if (!wait_event_timeout(outp->lt.wait,
+					atomic_read(&outp->lt.done),
+					msecs_to_jiffies(2000)))
+			ret = -ETIMEDOUT;
+	}
+
+	return ret;
+}
+
+static void
+nvkm_output_dp_enable(struct nvkm_output_dp *outp, bool present)
+{
+	struct nouveau_i2c_port *port = outp->base.edid;
+	if (present) {
+		if (!outp->present) {
+			nouveau_i2c(port)->acquire_pad(port, 0);
+			DBG("aux power -> always\n");
+			outp->present = true;
+		}
+		nvkm_output_dp_train(&outp->base, 0, true);
+	} else {
+		if (outp->present) {
+			nouveau_i2c(port)->release_pad(port);
+			DBG("aux power -> demand\n");
+			outp->present = false;
+		}
+		atomic_set(&outp->lt.done, 0);
+	}
+}
+
+static void
+nvkm_output_dp_detect(struct nvkm_output_dp *outp)
+{
+	struct nouveau_i2c_port *port = outp->base.edid;
+	int ret = nouveau_i2c(port)->acquire_pad(port, 0);
+	if (ret == 0) {
+		ret = nv_rdaux(outp->base.edid, DPCD_RC00_DPCD_REV,
+			       outp->dpcd, sizeof(outp->dpcd));
+		nvkm_output_dp_enable(outp, ret == 0);
+		nouveau_i2c(port)->release_pad(port);
+	}
+}
+
+static void
+nvkm_output_dp_service_work(struct work_struct *work)
+{
+	struct nvkm_output_dp *outp = container_of(work, typeof(*outp), work);
+	struct nouveau_disp *disp = nouveau_disp(outp);
+	int type = atomic_xchg(&outp->pending, 0);
+	u32 send = 0;
+
+	if (type & (NVKM_I2C_PLUG | NVKM_I2C_UNPLUG)) {
+		nvkm_output_dp_detect(outp);
+		if (type & NVKM_I2C_UNPLUG)
+			send |= NVKM_HPD_UNPLUG;
+		if (type & NVKM_I2C_PLUG)
+			send |= NVKM_HPD_PLUG;
+		nouveau_event_get(outp->base.conn->hpd.event);
+	}
+
+	if (type & NVKM_I2C_IRQ) {
+		nvkm_output_dp_train(&outp->base, 0, true);
+		send |= NVKM_HPD_IRQ;
+	}
+
+	nouveau_event_trigger(disp->hpd, send, outp->base.info.connector);
+}
+
+static int
+nvkm_output_dp_service(void *data, u32 type, int index)
+{
+	struct nvkm_output_dp *outp = data;
+	DBG("HPD: %d\n", type);
+	atomic_or(type, &outp->pending);
+	schedule_work(&outp->work);
+	return NVKM_EVENT_DROP;
+}
+
+int
+_nvkm_output_dp_fini(struct nouveau_object *object, bool suspend)
+{
+	struct nvkm_output_dp *outp = (void *)object;
+	nouveau_event_put(outp->irq);
+	nvkm_output_dp_enable(outp, false);
+	return nvkm_output_fini(&outp->base, suspend);
+}
+
+int
+_nvkm_output_dp_init(struct nouveau_object *object)
+{
+	struct nvkm_output_dp *outp = (void *)object;
+	nvkm_output_dp_detect(outp);
+	return nvkm_output_init(&outp->base);
+}
+
+void
+_nvkm_output_dp_dtor(struct nouveau_object *object)
+{
+	struct nvkm_output_dp *outp = (void *)object;
+	nouveau_event_ref(NULL, &outp->irq);
+	nvkm_output_destroy(&outp->base);
+}
+
+int
+nvkm_output_dp_create_(struct nouveau_object *parent,
+		       struct nouveau_object *engine,
+		       struct nouveau_oclass *oclass,
+		       struct dcb_output *info, int index,
+		       int length, void **pobject)
+{
+	struct nouveau_bios *bios = nouveau_bios(parent);
+	struct nouveau_i2c *i2c = nouveau_i2c(parent);
+	struct nvkm_output_dp *outp;
+	u8  hdr, cnt, len;
+	u32 data;
+	int ret;
+
+	ret = nvkm_output_create_(parent, engine, oclass, info, index,
+				  length, pobject);
+	outp = *pobject;
+	if (ret)
+		return ret;
+
+	nouveau_event_ref(NULL, &outp->base.conn->hpd.event);
+
+	/* access to the aux channel is not optional... */
+	if (!outp->base.edid) {
+		ERR("aux channel not found\n");
+		return -ENODEV;
+	}
+
+	/* nor is the bios data for this output... */
+	data = nvbios_dpout_match(bios, outp->base.info.hasht,
+				  outp->base.info.hashm, &outp->version,
+				  &hdr, &cnt, &len, &outp->info);
+	if (!data) {
+		ERR("no bios dp data\n");
+		return -ENODEV;
+	}
+
+	DBG("bios dp %02x %02x %02x %02x\n", outp->version, hdr, cnt, len);
+
+	/* link training */
+	INIT_WORK(&outp->lt.work, nouveau_dp_train);
+	init_waitqueue_head(&outp->lt.wait);
+	atomic_set(&outp->lt.done, 0);
+
+	/* link maintenance */
+	ret = nouveau_event_new(i2c->ntfy, NVKM_I2C_IRQ, outp->base.edid->index,
+				nvkm_output_dp_service, outp, &outp->irq);
+	if (ret) {
+		ERR("error monitoring aux irq event: %d\n", ret);
+		return ret;
+	}
+
+	INIT_WORK(&outp->work, nvkm_output_dp_service_work);
+
+	/* hotplug detect, replaces gpio-based mechanism with aux events */
+	ret = nouveau_event_new(i2c->ntfy, NVKM_I2C_PLUG | NVKM_I2C_UNPLUG,
+				outp->base.edid->index,
+				nvkm_output_dp_service, outp,
+			       &outp->base.conn->hpd.event);
+	if (ret) {
+		ERR("error monitoring aux hpd events: %d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+int
+_nvkm_output_dp_ctor(struct nouveau_object *parent,
+		     struct nouveau_object *engine,
+		     struct nouveau_oclass *oclass, void *info, u32 index,
+		     struct nouveau_object **pobject)
+{
+	struct nvkm_output_dp *outp;
+	int ret;
+
+	ret = nvkm_output_dp_create(parent, engine, oclass, info, index, &outp);
+	*pobject = nv_object(outp);
+	if (ret)
+		return ret;
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/outpdp.h b/drivers/gpu/drm/nouveau/core/engine/disp/outpdp.h
new file mode 100644
index 000000000000..ff33ba12cb67
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/outpdp.h
@@ -0,0 +1,65 @@
+#ifndef __NVKM_DISP_OUTP_DP_H__
+#define __NVKM_DISP_OUTP_DP_H__
+
+#include <subdev/bios.h>
+#include <subdev/bios/dp.h>
+
+#include "outp.h"
+
+struct nvkm_output_dp {
+	struct nvkm_output base;
+
+	struct nvbios_dpout info;
+	u8 version;
+
+	struct nouveau_eventh *irq;
+	struct nouveau_eventh *hpd;
+	struct work_struct work;
+	atomic_t pending;
+	bool present;
+	u8 dpcd[16];
+
+	struct {
+		struct work_struct work;
+		wait_queue_head_t wait;
+		atomic_t done;
+	} lt;
+};
+
+#define nvkm_output_dp_create(p,e,c,b,i,d)                                     \
+	nvkm_output_dp_create_((p), (e), (c), (b), (i), sizeof(**d), (void **)d)
+#define nvkm_output_dp_destroy(d) ({                                           \
+	struct nvkm_output_dp *_outp = (d);                                    \
+	_nvkm_output_dp_dtor(nv_object(_outp));                                \
+})
+#define nvkm_output_dp_init(d) ({                                              \
+	struct nvkm_output_dp *_outp = (d);                                    \
+	_nvkm_output_dp_init(nv_object(_outp));                                \
+})
+#define nvkm_output_dp_fini(d,s) ({                                            \
+	struct nvkm_output_dp *_outp = (d);                                    \
+	_nvkm_output_dp_fini(nv_object(_outp), (s));                           \
+})
+
+int nvkm_output_dp_create_(struct nouveau_object *, struct nouveau_object *,
+			   struct nouveau_oclass *, struct dcb_output *,
+			   int, int, void **);
+
+int  _nvkm_output_dp_ctor(struct nouveau_object *, struct nouveau_object *,
+			  struct nouveau_oclass *, void *, u32,
+			  struct nouveau_object **);
+void _nvkm_output_dp_dtor(struct nouveau_object *);
+int  _nvkm_output_dp_init(struct nouveau_object *);
+int  _nvkm_output_dp_fini(struct nouveau_object *, bool);
+
+struct nvkm_output_dp_impl {
+	struct nvkm_output_impl base;
+	int (*pattern)(struct nvkm_output_dp *, int);
+	int (*lnk_pwr)(struct nvkm_output_dp *, int nr);
+	int (*lnk_ctl)(struct nvkm_output_dp *, int nr, int bw, bool ef);
+	int (*drv_ctl)(struct nvkm_output_dp *, int ln, int vs, int pe, int pc);
+};
+
+int nvkm_output_dp_train(struct nvkm_output *, u32 rate, bool wait);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/piornv50.c b/drivers/gpu/drm/nouveau/core/engine/disp/piornv50.c
index 2c8ce351b52d..fe0f256f11bf 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/piornv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/piornv50.c
@@ -33,68 +33,107 @@
 #include "nv50.h"
 
 /******************************************************************************
- * DisplayPort
+ * TMDS
  *****************************************************************************/
-static struct nouveau_i2c_port *
-nv50_pior_dp_find(struct nouveau_disp *disp, struct dcb_output *outp)
+
+static int
+nv50_pior_tmds_ctor(struct nouveau_object *parent,
+		    struct nouveau_object *engine,
+		    struct nouveau_oclass *oclass, void *info, u32 index,
+		    struct nouveau_object **pobject)
 {
-	struct nouveau_i2c *i2c = nouveau_i2c(disp);
-	return i2c->find_type(i2c, NV_I2C_TYPE_EXTAUX(outp->extdev));
+	struct nouveau_i2c *i2c = nouveau_i2c(parent);
+	struct nvkm_output *outp;
+	int ret;
+
+	ret = nvkm_output_create(parent, engine, oclass, info, index, &outp);
+	*pobject = nv_object(outp);
+	if (ret)
+		return ret;
+
+	outp->edid = i2c->find_type(i2c, NV_I2C_TYPE_EXTDDC(outp->info.extdev));
+	return 0;
 }
 
+struct nvkm_output_impl
+nv50_pior_tmds_impl = {
+	.base.handle = DCB_OUTPUT_TMDS | 0x0100,
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv50_pior_tmds_ctor,
+		.dtor = _nvkm_output_dtor,
+		.init = _nvkm_output_init,
+		.fini = _nvkm_output_fini,
+	},
+};
+
+/******************************************************************************
+ * DisplayPort
+ *****************************************************************************/
+
 static int
-nv50_pior_dp_pattern(struct nouveau_disp *disp, struct dcb_output *outp,
-		     int head, int pattern)
+nv50_pior_dp_pattern(struct nvkm_output_dp *outp, int pattern)
 {
-	struct nouveau_i2c_port *port;
-	int ret = -EINVAL;
-
-	port = nv50_pior_dp_find(disp, outp);
-	if (port) {
-		if (port->func->pattern)
-			ret = port->func->pattern(port, pattern);
-		else
-			ret = 0;
-	}
-
-	return ret;
+	struct nouveau_i2c_port *port = outp->base.edid;
+	if (port && port->func->pattern)
+		return port->func->pattern(port, pattern);
+	return port ? 0 : -ENODEV;
 }
 
 static int
-nv50_pior_dp_lnk_ctl(struct nouveau_disp *disp, struct dcb_output *outp,
-		     int head, int lane_nr, int link_bw, bool enh)
+nv50_pior_dp_lnk_pwr(struct nvkm_output_dp *outp, int nr)
 {
-	struct nouveau_i2c_port *port;
-	int ret = -EINVAL;
+	return 0;
+}
 
-	port = nv50_pior_dp_find(disp, outp);
+static int
+nv50_pior_dp_lnk_ctl(struct nvkm_output_dp *outp, int nr, int bw, bool ef)
+{
+	struct nouveau_i2c_port *port = outp->base.edid;
 	if (port && port->func->lnk_ctl)
-		ret = port->func->lnk_ctl(port, lane_nr, link_bw, enh);
+		return port->func->lnk_ctl(port, nr, bw, ef);
+	return port ? 0 : -ENODEV;
+}
 
-	return ret;
+static int
+nv50_pior_dp_drv_ctl(struct nvkm_output_dp *outp, int ln, int vs, int pe, int pc)
+{
+	struct nouveau_i2c_port *port = outp->base.edid;
+	if (port && port->func->drv_ctl)
+		return port->func->drv_ctl(port, ln, vs, pe);
+	return port ? 0 : -ENODEV;
 }
 
 static int
-nv50_pior_dp_drv_ctl(struct nouveau_disp *disp, struct dcb_output *outp,
-		     int head, int lane, int vsw, int pre)
+nv50_pior_dp_ctor(struct nouveau_object *parent,
+		  struct nouveau_object *engine,
+		  struct nouveau_oclass *oclass, void *info, u32 index,
+		  struct nouveau_object **pobject)
 {
-	struct nouveau_i2c_port *port;
-	int ret = -EINVAL;
-
-	port = nv50_pior_dp_find(disp, outp);
-	if (port) {
-		if (port->func->drv_ctl)
-			ret = port->func->drv_ctl(port, lane, vsw, pre);
-		else
-			ret = 0;
-	}
+	struct nouveau_i2c *i2c = nouveau_i2c(parent);
+	struct nvkm_output_dp *outp;
+	int ret;
 
-	return ret;
+	ret = nvkm_output_dp_create(parent, engine, oclass, info, index, &outp);
+	*pobject = nv_object(outp);
+	if (ret)
+		return ret;
+
+	outp->base.edid = i2c->find_type(i2c, NV_I2C_TYPE_EXTAUX(
+					 outp->base.info.extdev));
+	return 0;
 }
 
-const struct nouveau_dp_func
-nv50_pior_dp_func = {
+struct nvkm_output_dp_impl
+nv50_pior_dp_impl = {
+	.base.base.handle = DCB_OUTPUT_DP | 0x0010,
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv50_pior_dp_ctor,
+		.dtor = _nvkm_output_dp_dtor,
+		.init = _nvkm_output_dp_init,
+		.fini = _nvkm_output_dp_fini,
+	},
 	.pattern = nv50_pior_dp_pattern,
+	.lnk_pwr = nv50_pior_dp_lnk_pwr,
 	.lnk_ctl = nv50_pior_dp_lnk_ctl,
 	.drv_ctl = nv50_pior_dp_drv_ctl,
 };
@@ -102,6 +141,7 @@ nv50_pior_dp_func = {
 /******************************************************************************
  * General PIOR handling
  *****************************************************************************/
+
 int
 nv50_pior_power(struct nv50_disp_priv *priv, int or, u32 data)
 {
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/priv.h b/drivers/gpu/drm/nouveau/core/engine/disp/priv.h
index cc3c7a4ca747..26e9a42569c7 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/priv.h
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/priv.h
@@ -1,10 +1,42 @@
 #ifndef __NVKM_DISP_PRIV_H__
 #define __NVKM_DISP_PRIV_H__
 
+#include <subdev/bios.h>
+#include <subdev/bios/dcb.h>
+#include <subdev/bios/conn.h>
+
 #include <engine/disp.h>
 
 struct nouveau_disp_impl {
 	struct nouveau_oclass base;
+	struct nouveau_oclass **outp;
+	struct nouveau_oclass **conn;
 };
 
+#define nouveau_disp_create(p,e,c,h,i,x,d)                                     \
+	nouveau_disp_create_((p), (e), (c), (h), (i), (x),                     \
+			     sizeof(**d), (void **)d)
+#define nouveau_disp_destroy(d) ({                                             \
+	struct nouveau_disp *disp = (d);                                       \
+	_nouveau_disp_dtor(nv_object(disp));                                   \
+})
+#define nouveau_disp_init(d) ({                                                \
+	struct nouveau_disp *disp = (d);                                       \
+	_nouveau_disp_init(nv_object(disp));                                   \
+})
+#define nouveau_disp_fini(d,s) ({                                              \
+	struct nouveau_disp *disp = (d);                                       \
+	_nouveau_disp_fini(nv_object(disp), (s));                              \
+})
+
+int  nouveau_disp_create_(struct nouveau_object *, struct nouveau_object *,
+			  struct nouveau_oclass *, int heads,
+			  const char *, const char *, int, void **);
+void _nouveau_disp_dtor(struct nouveau_object *);
+int  _nouveau_disp_init(struct nouveau_object *);
+int  _nouveau_disp_fini(struct nouveau_object *, bool);
+
+extern struct nouveau_oclass *nvkm_output_oclass;
+extern struct nouveau_oclass *nvkm_connector_oclass;
+
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/sornv50.c b/drivers/gpu/drm/nouveau/core/engine/disp/sornv50.c
index 526b75242899..e1832778e8b6 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/sornv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/sornv50.c
@@ -47,8 +47,12 @@ int
 nv50_sor_mthd(struct nouveau_object *object, u32 mthd, void *args, u32 size)
 {
 	struct nv50_disp_priv *priv = (void *)object->engine;
+	const u8  type = (mthd & NV50_DISP_SOR_MTHD_TYPE) >> 12;
 	const u8  head = (mthd & NV50_DISP_SOR_MTHD_HEAD) >> 3;
+	const u8  link = (mthd & NV50_DISP_SOR_MTHD_LINK) >> 2;
 	const u8    or = (mthd & NV50_DISP_SOR_MTHD_OR);
+	const u16 mask = (0x0100 << head) | (0x0040 << link) | (0x0001 << or);
+	struct nvkm_output *outp = NULL, *temp;
 	u32 data;
 	int ret = -EINVAL;
 
@@ -56,6 +60,13 @@ nv50_sor_mthd(struct nouveau_object *object, u32 mthd, void *args, u32 size)
 		return -EINVAL;
 	data = *(u32 *)args;
 
+	list_for_each_entry(temp, &priv->base.outp, head) {
+		if ((temp->info.hasht & 0xff) == type &&
+		    (temp->info.hashm & mask) == mask) {
+			outp = temp;
+			break;
+		}
+	}
 
 	switch (mthd & ~0x3f) {
 	case NV50_DISP_SOR_PWR:
@@ -71,6 +82,23 @@ nv50_sor_mthd(struct nouveau_object *object, u32 mthd, void *args, u32 size)
 		priv->sor.lvdsconf = data & NV50_DISP_SOR_LVDS_SCRIPT_ID;
 		ret = 0;
 		break;
+	case NV94_DISP_SOR_DP_PWR:
+		if (outp) {
+			struct nvkm_output_dp *outpdp = (void *)outp;
+			switch (data) {
+			case NV94_DISP_SOR_DP_PWR_STATE_OFF:
+				((struct nvkm_output_dp_impl *)nv_oclass(outp))
+					->lnk_pwr(outpdp, 0);
+				atomic_set(&outpdp->lt.done, 0);
+				break;
+			case NV94_DISP_SOR_DP_PWR_STATE_ON:
+				nvkm_output_dp_train(&outpdp->base, 0, true);
+				break;
+			default:
+				return -EINVAL;
+			}
+		}
+		break;
 	default:
 		BUG_ON(1);
 	}
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/sornv94.c b/drivers/gpu/drm/nouveau/core/engine/disp/sornv94.c
index eea3ef59693d..05487cda84a8 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/sornv94.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/sornv94.c
@@ -29,19 +29,21 @@
 #include <subdev/bios/dcb.h>
 #include <subdev/bios/dp.h>
 #include <subdev/bios/init.h>
+#include <subdev/timer.h>
 
 #include "nv50.h"
+#include "outpdp.h"
 
 static inline u32
-nv94_sor_soff(struct dcb_output *outp)
+nv94_sor_soff(struct nvkm_output_dp *outp)
 {
-	return (ffs(outp->or) - 1) * 0x800;
+	return (ffs(outp->base.info.or) - 1) * 0x800;
 }
 
 static inline u32
-nv94_sor_loff(struct dcb_output *outp)
+nv94_sor_loff(struct nvkm_output_dp *outp)
 {
-	return nv94_sor_soff(outp) + !(outp->sorconf.link & 1) * 0x80;
+	return nv94_sor_soff(outp) + !(outp->base.info.sorconf.link & 1) * 0x80;
 }
 
 static inline u32
@@ -55,77 +57,96 @@ nv94_sor_dp_lane_map(struct nv50_disp_priv *priv, u8 lane)
 }
 
 static int
-nv94_sor_dp_pattern(struct nouveau_disp *disp, struct dcb_output *outp,
-		    int head, int pattern)
+nv94_sor_dp_pattern(struct nvkm_output_dp *outp, int pattern)
 {
-	struct nv50_disp_priv *priv = (void *)disp;
+	struct nv50_disp_priv *priv = (void *)nouveau_disp(outp);
 	const u32 loff = nv94_sor_loff(outp);
 	nv_mask(priv, 0x61c10c + loff, 0x0f000000, pattern << 24);
 	return 0;
 }
 
+int
+nv94_sor_dp_lnk_pwr(struct nvkm_output_dp *outp, int nr)
+{
+	struct nv50_disp_priv *priv = (void *)nouveau_disp(outp);
+	const u32 soff = nv94_sor_soff(outp);
+	const u32 loff = nv94_sor_loff(outp);
+	u32 mask = 0, i;
+
+	for (i = 0; i < nr; i++)
+		mask |= 1 << (nv94_sor_dp_lane_map(priv, i) >> 3);
+
+	nv_mask(priv, 0x61c130 + loff, 0x0000000f, mask);
+	nv_mask(priv, 0x61c034 + soff, 0x80000000, 0x80000000);
+	nv_wait(priv, 0x61c034 + soff, 0x80000000, 0x00000000);
+	return 0;
+}
+
 static int
-nv94_sor_dp_lnk_ctl(struct nouveau_disp *disp, struct dcb_output *outp,
-		    int head, int link_nr, int link_bw, bool enh_frame)
+nv94_sor_dp_lnk_ctl(struct nvkm_output_dp *outp, int nr, int bw, bool ef)
 {
-	struct nv50_disp_priv *priv = (void *)disp;
+	struct nv50_disp_priv *priv = (void *)nouveau_disp(outp);
 	const u32 soff = nv94_sor_soff(outp);
 	const u32 loff = nv94_sor_loff(outp);
 	u32 dpctrl = 0x00000000;
 	u32 clksor = 0x00000000;
-	u32 lane = 0;
-	int i;
 
-	dpctrl |= ((1 << link_nr) - 1) << 16;
-	if (enh_frame)
+	dpctrl |= ((1 << nr) - 1) << 16;
+	if (ef)
 		dpctrl |= 0x00004000;
-	if (link_bw > 0x06)
+	if (bw > 0x06)
 		clksor |= 0x00040000;
 
-	for (i = 0; i < link_nr; i++)
-		lane |= 1 << (nv94_sor_dp_lane_map(priv, i) >> 3);
-
 	nv_mask(priv, 0x614300 + soff, 0x000c0000, clksor);
 	nv_mask(priv, 0x61c10c + loff, 0x001f4000, dpctrl);
-	nv_mask(priv, 0x61c130 + loff, 0x0000000f, lane);
 	return 0;
 }
 
 static int
-nv94_sor_dp_drv_ctl(struct nouveau_disp *disp, struct dcb_output *outp,
-		    int head, int lane, int swing, int preem)
+nv94_sor_dp_drv_ctl(struct nvkm_output_dp *outp, int ln, int vs, int pe, int pc)
 {
-	struct nouveau_bios *bios = nouveau_bios(disp);
-	struct nv50_disp_priv *priv = (void *)disp;
-	const u32 shift = nv94_sor_dp_lane_map(priv, lane);
+	struct nv50_disp_priv *priv = (void *)nouveau_disp(outp);
+	struct nouveau_bios *bios = nouveau_bios(priv);
+	const u32 shift = nv94_sor_dp_lane_map(priv, ln);
 	const u32 loff = nv94_sor_loff(outp);
 	u32 addr, data[3];
 	u8  ver, hdr, cnt, len;
 	struct nvbios_dpout info;
 	struct nvbios_dpcfg ocfg;
 
-	addr = nvbios_dpout_match(bios, outp->hasht, outp->hashm,
+	addr = nvbios_dpout_match(bios, outp->base.info.hasht,
+					outp->base.info.hashm,
 				 &ver, &hdr, &cnt, &len, &info);
 	if (!addr)
 		return -ENODEV;
 
-	addr = nvbios_dpcfg_match(bios, addr, 0, swing, preem,
+	addr = nvbios_dpcfg_match(bios, addr, 0, vs, pe,
 				 &ver, &hdr, &cnt, &len, &ocfg);
 	if (!addr)
 		return -EINVAL;
 
 	data[0] = nv_rd32(priv, 0x61c118 + loff) & ~(0x000000ff << shift);
 	data[1] = nv_rd32(priv, 0x61c120 + loff) & ~(0x000000ff << shift);
-	data[2] = nv_rd32(priv, 0x61c130 + loff) & ~(0x0000ff00);
-	nv_wr32(priv, 0x61c118 + loff, data[0] | (ocfg.drv << shift));
-	nv_wr32(priv, 0x61c120 + loff, data[1] | (ocfg.pre << shift));
-	nv_wr32(priv, 0x61c130 + loff, data[2] | (ocfg.unk << 8));
+	data[2] = nv_rd32(priv, 0x61c130 + loff);
+	if ((data[2] & 0x0000ff00) < (ocfg.tx_pu << 8) || ln == 0)
+		data[2] = (data[2] & ~0x0000ff00) | (ocfg.tx_pu << 8);
+	nv_wr32(priv, 0x61c118 + loff, data[0] | (ocfg.dc << shift));
+	nv_wr32(priv, 0x61c120 + loff, data[1] | (ocfg.pe << shift));
+	nv_wr32(priv, 0x61c130 + loff, data[2] | (ocfg.tx_pu << 8));
 	return 0;
 }
 
-const struct nouveau_dp_func
-nv94_sor_dp_func = {
+struct nvkm_output_dp_impl
+nv94_sor_dp_impl = {
+	.base.base.handle = DCB_OUTPUT_DP,
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nvkm_output_dp_ctor,
+		.dtor = _nvkm_output_dp_dtor,
+		.init = _nvkm_output_dp_init,
+		.fini = _nvkm_output_dp_fini,
+	},
 	.pattern = nv94_sor_dp_pattern,
+	.lnk_pwr = nv94_sor_dp_lnk_pwr,
 	.lnk_ctl = nv94_sor_dp_lnk_ctl,
 	.drv_ctl = nv94_sor_dp_drv_ctl,
 };
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/sornvd0.c b/drivers/gpu/drm/nouveau/core/engine/disp/sornvd0.c
index d2df572f16a3..97f0e9cd3d40 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/sornvd0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/sornvd0.c
@@ -29,19 +29,20 @@
 #include <subdev/bios/dcb.h>
 #include <subdev/bios/dp.h>
 #include <subdev/bios/init.h>
+#include <subdev/timer.h>
 
 #include "nv50.h"
 
 static inline u32
-nvd0_sor_soff(struct dcb_output *outp)
+nvd0_sor_soff(struct nvkm_output_dp *outp)
 {
-	return (ffs(outp->or) - 1) * 0x800;
+	return (ffs(outp->base.info.or) - 1) * 0x800;
 }
 
 static inline u32
-nvd0_sor_loff(struct dcb_output *outp)
+nvd0_sor_loff(struct nvkm_output_dp *outp)
 {
-	return nvd0_sor_soff(outp) + !(outp->sorconf.link & 1) * 0x80;
+	return nvd0_sor_soff(outp) + !(outp->base.info.sorconf.link & 1) * 0x80;
 }
 
 static inline u32
@@ -52,77 +53,80 @@ nvd0_sor_dp_lane_map(struct nv50_disp_priv *priv, u8 lane)
 }
 
 static int
-nvd0_sor_dp_pattern(struct nouveau_disp *disp, struct dcb_output *outp,
-		    int head, int pattern)
+nvd0_sor_dp_pattern(struct nvkm_output_dp *outp, int pattern)
 {
-	struct nv50_disp_priv *priv = (void *)disp;
+	struct nv50_disp_priv *priv = (void *)nouveau_disp(outp);
 	const u32 loff = nvd0_sor_loff(outp);
 	nv_mask(priv, 0x61c110 + loff, 0x0f0f0f0f, 0x01010101 * pattern);
 	return 0;
 }
 
 static int
-nvd0_sor_dp_lnk_ctl(struct nouveau_disp *disp, struct dcb_output *outp,
-		    int head, int link_nr, int link_bw, bool enh_frame)
+nvd0_sor_dp_lnk_ctl(struct nvkm_output_dp *outp, int nr, int bw, bool ef)
 {
-	struct nv50_disp_priv *priv = (void *)disp;
+	struct nv50_disp_priv *priv = (void *)nouveau_disp(outp);
 	const u32 soff = nvd0_sor_soff(outp);
 	const u32 loff = nvd0_sor_loff(outp);
 	u32 dpctrl = 0x00000000;
 	u32 clksor = 0x00000000;
-	u32 lane = 0;
-	int i;
 
-	clksor |= link_bw << 18;
-	dpctrl |= ((1 << link_nr) - 1) << 16;
-	if (enh_frame)
+	clksor |= bw << 18;
+	dpctrl |= ((1 << nr) - 1) << 16;
+	if (ef)
 		dpctrl |= 0x00004000;
 
-	for (i = 0; i < link_nr; i++)
-		lane |= 1 << (nvd0_sor_dp_lane_map(priv, i) >> 3);
-
 	nv_mask(priv, 0x612300 + soff, 0x007c0000, clksor);
 	nv_mask(priv, 0x61c10c + loff, 0x001f4000, dpctrl);
-	nv_mask(priv, 0x61c130 + loff, 0x0000000f, lane);
 	return 0;
 }
 
 static int
-nvd0_sor_dp_drv_ctl(struct nouveau_disp *disp, struct dcb_output *outp,
-		    int head, int lane, int swing, int preem)
+nvd0_sor_dp_drv_ctl(struct nvkm_output_dp *outp, int ln, int vs, int pe, int pc)
 {
-	struct nouveau_bios *bios = nouveau_bios(disp);
-	struct nv50_disp_priv *priv = (void *)disp;
-	const u32 shift = nvd0_sor_dp_lane_map(priv, lane);
+	struct nv50_disp_priv *priv = (void *)nouveau_disp(outp);
+	struct nouveau_bios *bios = nouveau_bios(priv);
+	const u32 shift = nvd0_sor_dp_lane_map(priv, ln);
 	const u32 loff = nvd0_sor_loff(outp);
-	u32 addr, data[3];
+	u32 addr, data[4];
 	u8  ver, hdr, cnt, len;
 	struct nvbios_dpout info;
 	struct nvbios_dpcfg ocfg;
 
-	addr = nvbios_dpout_match(bios, outp->hasht, outp->hashm,
+	addr = nvbios_dpout_match(bios, outp->base.info.hasht,
+					outp->base.info.hashm,
 				 &ver, &hdr, &cnt, &len, &info);
 	if (!addr)
 		return -ENODEV;
 
-	addr = nvbios_dpcfg_match(bios, addr, 0, swing, preem,
+	addr = nvbios_dpcfg_match(bios, addr, pc, vs, pe,
 				 &ver, &hdr, &cnt, &len, &ocfg);
 	if (!addr)
 		return -EINVAL;
 
 	data[0] = nv_rd32(priv, 0x61c118 + loff) & ~(0x000000ff << shift);
 	data[1] = nv_rd32(priv, 0x61c120 + loff) & ~(0x000000ff << shift);
-	data[2] = nv_rd32(priv, 0x61c130 + loff) & ~(0x0000ff00);
-	nv_wr32(priv, 0x61c118 + loff, data[0] | (ocfg.drv << shift));
-	nv_wr32(priv, 0x61c120 + loff, data[1] | (ocfg.pre << shift));
-	nv_wr32(priv, 0x61c130 + loff, data[2] | (ocfg.unk << 8));
-	nv_mask(priv, 0x61c13c + loff, 0x00000000, 0x00000000);
+	data[2] = nv_rd32(priv, 0x61c130 + loff);
+	if ((data[2] & 0x0000ff00) < (ocfg.tx_pu << 8) || ln == 0)
+		data[2] = (data[2] & ~0x0000ff00) | (ocfg.tx_pu << 8);
+	nv_wr32(priv, 0x61c118 + loff, data[0] | (ocfg.dc << shift));
+	nv_wr32(priv, 0x61c120 + loff, data[1] | (ocfg.pe << shift));
+	nv_wr32(priv, 0x61c130 + loff, data[2] | (ocfg.tx_pu << 8));
+	data[3] = nv_rd32(priv, 0x61c13c + loff) & ~(0x000000ff << shift);
+	nv_wr32(priv, 0x61c13c + loff, data[3] | (ocfg.pc << shift));
 	return 0;
 }
 
-const struct nouveau_dp_func
-nvd0_sor_dp_func = {
+struct nvkm_output_dp_impl
+nvd0_sor_dp_impl = {
+	.base.base.handle = DCB_OUTPUT_DP,
+	.base.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nvkm_output_dp_ctor,
+		.dtor = _nvkm_output_dp_dtor,
+		.init = _nvkm_output_dp_init,
+		.fini = _nvkm_output_dp_fini,
+	},
 	.pattern = nvd0_sor_dp_pattern,
+	.lnk_pwr = nv94_sor_dp_lnk_pwr,
 	.lnk_ctl = nvd0_sor_dp_lnk_ctl,
 	.drv_ctl = nvd0_sor_dp_drv_ctl,
 };
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/base.c b/drivers/gpu/drm/nouveau/core/engine/fifo/base.c
index 6f9041ced9a2..56ed3d73bf8e 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/base.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/base.c
@@ -91,7 +91,7 @@ nouveau_fifo_channel_create_(struct nouveau_object *parent,
 	if (!chan->user)
 		return -EFAULT;
 
-	nouveau_event_trigger(priv->cevent, 0);
+	nouveau_event_trigger(priv->cevent, 1, 0);
 
 	chan->size = size;
 	return 0;
@@ -194,11 +194,11 @@ nouveau_fifo_create_(struct nouveau_object *parent,
 	if (!priv->channel)
 		return -ENOMEM;
 
-	ret = nouveau_event_create(1, &priv->cevent);
+	ret = nouveau_event_create(1, 1, &priv->cevent);
 	if (ret)
 		return ret;
 
-	ret = nouveau_event_create(1, &priv->uevent);
+	ret = nouveau_event_create(1, 1, &priv->uevent);
 	if (ret)
 		return ret;
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/gk20a.c b/drivers/gpu/drm/nouveau/core/engine/fifo/gk20a.c
new file mode 100644
index 000000000000..327456eae963
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/gk20a.c
@@ -0,0 +1,35 @@
+/*
+ * Copyright (c) 2014, NVIDIA CORPORATION. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#include "nve0.h"
+
+struct nouveau_oclass *
+gk20a_fifo_oclass = &(struct nve0_fifo_impl) {
+	.base.handle = NV_ENGINE(FIFO, 0xea),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nve0_fifo_ctor,
+		.dtor = nve0_fifo_dtor,
+		.init = nve0_fifo_init,
+		.fini = nve0_fifo_fini,
+	},
+	.channels = 128,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c
index 54f26cc801c7..c61b16a63884 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c
@@ -539,7 +539,7 @@ nv04_fifo_intr(struct nouveau_subdev *subdev)
 			}
 
 			if (status & 0x40000000) {
-				nouveau_event_trigger(priv->base.uevent, 0);
+				nouveau_event_trigger(priv->base.uevent, 1, 0);
 				nv_wr32(priv, 0x002100, 0x40000000);
 				status &= ~0x40000000;
 			}
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c
index fe0f41e65d9b..6e5ac16e5460 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c
@@ -389,14 +389,14 @@ nv84_fifo_cclass = {
  ******************************************************************************/
 
 static void
-nv84_fifo_uevent_enable(struct nouveau_event *event, int index)
+nv84_fifo_uevent_enable(struct nouveau_event *event, int type, int index)
 {
 	struct nv84_fifo_priv *priv = event->priv;
 	nv_mask(priv, 0x002140, 0x40000000, 0x40000000);
 }
 
 static void
-nv84_fifo_uevent_disable(struct nouveau_event *event, int index)
+nv84_fifo_uevent_disable(struct nouveau_event *event, int type, int index)
 {
 	struct nv84_fifo_priv *priv = event->priv;
 	nv_mask(priv, 0x002140, 0x40000000, 0x00000000);
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c
index fa1e719872b7..ae4a4dc5642a 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c
@@ -730,7 +730,7 @@ nvc0_fifo_intr_engine_unit(struct nvc0_fifo_priv *priv, int engn)
 	for (unkn = 0; unkn < 8; unkn++) {
 		u32 ints = (intr >> (unkn * 0x04)) & inte;
 		if (ints & 0x1) {
-			nouveau_event_trigger(priv->base.uevent, 0);
+			nouveau_event_trigger(priv->base.uevent, 1, 0);
 			ints &= ~1;
 		}
 		if (ints) {
@@ -827,14 +827,14 @@ nvc0_fifo_intr(struct nouveau_subdev *subdev)
 }
 
 static void
-nvc0_fifo_uevent_enable(struct nouveau_event *event, int index)
+nvc0_fifo_uevent_enable(struct nouveau_event *event, int type, int index)
 {
 	struct nvc0_fifo_priv *priv = event->priv;
 	nv_mask(priv, 0x002140, 0x80000000, 0x80000000);
 }
 
 static void
-nvc0_fifo_uevent_disable(struct nouveau_event *event, int index)
+nvc0_fifo_uevent_disable(struct nouveau_event *event, int type, int index)
 {
 	struct nvc0_fifo_priv *priv = event->priv;
 	nv_mask(priv, 0x002140, 0x80000000, 0x00000000);
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c
index a9a1a9c9f9f2..298063edb92d 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c
@@ -859,7 +859,7 @@ nve0_fifo_intr_runlist(struct nve0_fifo_priv *priv)
 static void
 nve0_fifo_intr_engine(struct nve0_fifo_priv *priv)
 {
-	nouveau_event_trigger(priv->base.uevent, 0);
+	nouveau_event_trigger(priv->base.uevent, 1, 0);
 }
 
 static void
@@ -952,14 +952,14 @@ nve0_fifo_intr(struct nouveau_subdev *subdev)
 }
 
 static void
-nve0_fifo_uevent_enable(struct nouveau_event *event, int index)
+nve0_fifo_uevent_enable(struct nouveau_event *event, int type, int index)
 {
 	struct nve0_fifo_priv *priv = event->priv;
 	nv_mask(priv, 0x002140, 0x80000000, 0x80000000);
 }
 
 static void
-nve0_fifo_uevent_disable(struct nouveau_event *event, int index)
+nve0_fifo_uevent_disable(struct nouveau_event *event, int type, int index)
 {
 	struct nve0_fifo_priv *priv = event->priv;
 	nv_mask(priv, 0x002140, 0x80000000, 0x00000000);
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.h b/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.h
index 014344ebee66..e96b32bb1bbc 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.h
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.h
@@ -8,6 +8,7 @@ int  nve0_fifo_ctor(struct nouveau_object *, struct nouveau_object *,
 		    struct nouveau_object **);
 void nve0_fifo_dtor(struct nouveau_object *);
 int  nve0_fifo_init(struct nouveau_object *);
+int  nve0_fifo_fini(struct nouveau_object *, bool);
 
 struct nve0_fifo_impl {
 	struct nouveau_oclass base;
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/ctxgk20a.c b/drivers/gpu/drm/nouveau/core/engine/graph/ctxgk20a.c
new file mode 100644
index 000000000000..224ee0287ab7
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/ctxgk20a.c
@@ -0,0 +1,53 @@
+/*
+ * Copyright (c) 2014, NVIDIA CORPORATION. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#include "ctxnvc0.h"
+
+static const struct nvc0_graph_pack
+gk20a_grctx_pack_mthd[] = {
+	{ nve4_grctx_init_a097_0, 0xa297 },
+	{ nvc0_grctx_init_902d_0, 0x902d },
+	{}
+};
+
+struct nouveau_oclass *
+gk20a_grctx_oclass = &(struct nvc0_grctx_oclass) {
+	.base.handle = NV_ENGCTX(GR, 0xea),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nvc0_graph_context_ctor,
+		.dtor = nvc0_graph_context_dtor,
+		.init = _nouveau_graph_context_init,
+		.fini = _nouveau_graph_context_fini,
+		.rd32 = _nouveau_graph_context_rd32,
+		.wr32 = _nouveau_graph_context_wr32,
+	},
+	.main  = nve4_grctx_generate_main,
+	.mods  = nve4_grctx_generate_mods,
+	.unkn  = nve4_grctx_generate_unkn,
+	.hub   = nve4_grctx_pack_hub,
+	.gpc   = nve4_grctx_pack_gpc,
+	.zcull = nvc0_grctx_pack_zcull,
+	.tpc   = nve4_grctx_pack_tpc,
+	.ppc   = nve4_grctx_pack_ppc,
+	.icmd  = nve4_grctx_pack_icmd,
+	.mthd  = gk20a_grctx_pack_mthd,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnv108.c b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnv108.c
index 48351b4d6d6b..8de4a4291548 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnv108.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnv108.c
@@ -545,10 +545,12 @@ nv108_grctx_generate_mods(struct nvc0_graph_priv *priv, struct nvc0_grctx *info)
 	mmio_list(0x408010, 0x80000000,  0, 0);
 	mmio_list(0x419004, 0x00000000,  8, 1);
 	mmio_list(0x419008, 0x00000000,  0, 0);
+	mmio_list(0x4064cc, 0x80000000,  0, 0);
 	mmio_list(0x408004, 0x00000000,  8, 0);
 	mmio_list(0x408008, 0x80000030,  0, 0);
 	mmio_list(0x418808, 0x00000000,  8, 0);
 	mmio_list(0x41880c, 0x80000030,  0, 0);
+	mmio_list(0x4064c8, 0x00c20200,  0, 0);
 	mmio_list(0x418810, 0x80000000, 12, 2);
 	mmio_list(0x419848, 0x10000000, 12, 2);
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvc0.h b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvc0.h
index 9c815d1f99ef..8da8b627b9d0 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvc0.h
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvc0.h
@@ -69,7 +69,9 @@ extern struct nouveau_oclass *nvd7_grctx_oclass;
 extern struct nouveau_oclass *nvd9_grctx_oclass;
 
 extern struct nouveau_oclass *nve4_grctx_oclass;
+extern struct nouveau_oclass *gk20a_grctx_oclass;
 void nve4_grctx_generate_main(struct nvc0_graph_priv *, struct nvc0_grctx *);
+void nve4_grctx_generate_mods(struct nvc0_graph_priv *, struct nvc0_grctx *);
 void nve4_grctx_generate_unkn(struct nvc0_graph_priv *);
 void nve4_grctx_generate_r418bb8(struct nvc0_graph_priv *);
 
@@ -151,6 +153,13 @@ extern const struct nvc0_graph_init nve4_grctx_init_gpm_0[];
 
 extern const struct nvc0_graph_init nve4_grctx_init_pes_0[];
 
+extern const struct nvc0_graph_pack nve4_grctx_pack_hub[];
+extern const struct nvc0_graph_pack nve4_grctx_pack_gpc[];
+extern const struct nvc0_graph_pack nve4_grctx_pack_tpc[];
+extern const struct nvc0_graph_pack nve4_grctx_pack_ppc[];
+extern const struct nvc0_graph_pack nve4_grctx_pack_icmd[];
+extern const struct nvc0_graph_init nve4_grctx_init_a097_0[];
+
 extern const struct nvc0_graph_pack nvf0_grctx_pack_mthd[];
 
 extern const struct nvc0_graph_init nvf0_grctx_init_pri_0[];
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnve4.c b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnve4.c
index 49a14b116a5f..c5b249238587 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnve4.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnve4.c
@@ -272,13 +272,13 @@ nve4_grctx_init_icmd_0[] = {
 	{}
 };
 
-static const struct nvc0_graph_pack
+const struct nvc0_graph_pack
 nve4_grctx_pack_icmd[] = {
 	{ nve4_grctx_init_icmd_0 },
 	{}
 };
 
-static const struct nvc0_graph_init
+const struct nvc0_graph_init
 nve4_grctx_init_a097_0[] = {
 	{ 0x000800,   8, 0x40, 0x00000000 },
 	{ 0x000804,   8, 0x40, 0x00000000 },
@@ -697,7 +697,7 @@ nve4_grctx_init_be_0[] = {
 	{}
 };
 
-static const struct nvc0_graph_pack
+const struct nvc0_graph_pack
 nve4_grctx_pack_hub[] = {
 	{ nvc0_grctx_init_main_0 },
 	{ nve4_grctx_init_fe_0 },
@@ -737,7 +737,7 @@ nve4_grctx_init_gpm_0[] = {
 	{}
 };
 
-static const struct nvc0_graph_pack
+const struct nvc0_graph_pack
 nve4_grctx_pack_gpc[] = {
 	{ nvc0_grctx_init_gpc_unk_0 },
 	{ nvd9_grctx_init_prop_0 },
@@ -802,7 +802,7 @@ nve4_grctx_init_sm_0[] = {
 	{}
 };
 
-static const struct nvc0_graph_pack
+const struct nvc0_graph_pack
 nve4_grctx_pack_tpc[] = {
 	{ nvd7_grctx_init_pe_0 },
 	{ nve4_grctx_init_tex_0 },
@@ -826,7 +826,7 @@ nve4_grctx_init_cbm_0[] = {
 	{}
 };
 
-static const struct nvc0_graph_pack
+const struct nvc0_graph_pack
 nve4_grctx_pack_ppc[] = {
 	{ nve4_grctx_init_pes_0 },
 	{ nve4_grctx_init_cbm_0 },
@@ -838,7 +838,7 @@ nve4_grctx_pack_ppc[] = {
  * PGRAPH context implementation
  ******************************************************************************/
 
-static void
+void
 nve4_grctx_generate_mods(struct nvc0_graph_priv *priv, struct nvc0_grctx *info)
 {
 	u32 magic[GPC_MAX][2];
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvf0.c b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvf0.c
index 0fab95e49f53..dec03f04114d 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvf0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/ctxnvf0.c
@@ -842,7 +842,7 @@ nvf0_grctx_generate_mods(struct nvc0_graph_priv *priv, struct nvc0_grctx *info)
 		u16 magic3 = 0x0648;
 		magic[gpc][0]  = 0x10000000 | (magic0 << 16) | offset;
 		magic[gpc][1]  = 0x00000000 | (magic1 << 16);
-		offset += 0x0324 * (priv->tpc_nr[gpc] - 1);;
+		offset += 0x0324 * (priv->tpc_nr[gpc] - 1);
 		magic[gpc][2]  = 0x10000000 | (magic2 << 16) | offset;
 		magic[gpc][3]  = 0x00000000 | (magic3 << 16);
 		offset += 0x0324;
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/gk20a.c b/drivers/gpu/drm/nouveau/core/engine/graph/gk20a.c
new file mode 100644
index 000000000000..83048a56430d
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/gk20a.c
@@ -0,0 +1,47 @@
+/*
+ * Copyright (c) 2014, NVIDIA CORPORATION. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#include "nvc0.h"
+#include "ctxnvc0.h"
+
+static struct nouveau_oclass
+gk20a_graph_sclass[] = {
+	{ 0x902d, &nouveau_object_ofuncs },
+	{ 0xa040, &nouveau_object_ofuncs },
+	{ 0xa297, &nouveau_object_ofuncs },
+	{ 0xa0c0, &nouveau_object_ofuncs },
+	{}
+};
+
+struct nouveau_oclass *
+gk20a_graph_oclass = &(struct nvc0_graph_oclass) {
+	.base.handle = NV_ENGINE(GR, 0xea),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nvc0_graph_ctor,
+		.dtor = nvc0_graph_dtor,
+		.init = nve4_graph_init,
+		.fini = nve4_graph_fini,
+	},
+	.cclass = &gk20a_grctx_oclass,
+	.sclass = gk20a_graph_sclass,
+	.mmio = nve4_graph_pack_mmio,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/nv50.c b/drivers/gpu/drm/nouveau/core/engine/graph/nv50.c
index 2c7809e1a09b..1a2d56493cf6 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/nv50.c
@@ -901,7 +901,7 @@ nv50_graph_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 		nv_engine(priv)->sclass = nvaf_graph_sclass;
 		break;
 
-	};
+	}
 
 	/* unfortunate hw bug workaround... */
 	if (nv_device(priv)->chipset != 0x50 &&
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.c b/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.c
index f3c7329da0a0..bf7bdb1f291e 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.c
@@ -894,6 +894,10 @@ nvc0_graph_init_fw(struct nvc0_graph_priv *priv, u32 fuc_base,
 			nv_wr32(priv, fuc_base + 0x0188, i >> 6);
 		nv_wr32(priv, fuc_base + 0x0184, code->data[i]);
 	}
+
+	/* code must be padded to 0x40 words */
+	for (; i & 0x3f; i++)
+		nv_wr32(priv, fuc_base + 0x0184, 0);
 }
 
 static void
@@ -1259,10 +1263,14 @@ nvc0_graph_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nvc0_graph_oclass *oclass = (void *)bclass;
 	struct nouveau_device *device = nv_device(parent);
 	struct nvc0_graph_priv *priv;
+	bool use_ext_fw, enable;
 	int ret, i;
 
-	ret = nouveau_graph_create(parent, engine, bclass,
-				   (oclass->fecs.ucode != NULL), &priv);
+	use_ext_fw = nouveau_boolopt(device->cfgopt, "NvGrUseFW",
+				     oclass->fecs.ucode == NULL);
+	enable = use_ext_fw || oclass->fecs.ucode != NULL;
+
+	ret = nouveau_graph_create(parent, engine, bclass, enable, &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
@@ -1272,7 +1280,7 @@ nvc0_graph_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 
 	priv->base.units = nvc0_graph_units;
 
-	if (nouveau_boolopt(device->cfgopt, "NvGrUseFW", false)) {
+	if (use_ext_fw) {
 		nv_info(priv, "using external firmware\n");
 		if (nvc0_graph_ctor_fw(priv, "fuc409c", &priv->fuc409c) ||
 		    nvc0_graph_ctor_fw(priv, "fuc409d", &priv->fuc409d) ||
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.h b/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.h
index 90d44616c876..75203a99d902 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.h
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/nvc0.h
@@ -116,6 +116,7 @@ int  nvc0_graph_ctor(struct nouveau_object *, struct nouveau_object *,
 		     struct nouveau_object **);
 void nvc0_graph_dtor(struct nouveau_object *);
 int  nvc0_graph_init(struct nouveau_object *);
+int  nve4_graph_fini(struct nouveau_object *, bool);
 int  nve4_graph_init(struct nouveau_object *);
 
 extern struct nouveau_oclass nvc0_graph_sclass[];
@@ -217,6 +218,7 @@ extern const struct nvc0_graph_init nve4_graph_init_main_0[];
 extern const struct nvc0_graph_init nve4_graph_init_tpccs_0[];
 extern const struct nvc0_graph_init nve4_graph_init_pe_0[];
 extern const struct nvc0_graph_init nve4_graph_init_be_0[];
+extern const struct nvc0_graph_pack nve4_graph_pack_mmio[];
 
 extern const struct nvc0_graph_init nvf0_graph_init_fe_0[];
 extern const struct nvc0_graph_init nvf0_graph_init_sked_0[];
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/nve4.c b/drivers/gpu/drm/nouveau/core/engine/graph/nve4.c
index f7c011217175..51e0c075ad34 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/nve4.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/nve4.c
@@ -151,7 +151,7 @@ nve4_graph_init_be_0[] = {
 	{}
 };
 
-static const struct nvc0_graph_pack
+const struct nvc0_graph_pack
 nve4_graph_pack_mmio[] = {
 	{ nve4_graph_init_main_0 },
 	{ nvc0_graph_init_fe_0 },
@@ -189,7 +189,7 @@ nve4_graph_pack_mmio[] = {
  * PGRAPH engine/subdev functions
  ******************************************************************************/
 
-static int
+int
 nve4_graph_fini(struct nouveau_object *object, bool suspend)
 {
 	struct nvc0_graph_priv *priv = (void *)object;
diff --git a/drivers/gpu/drm/nouveau/core/engine/software/nv50.c b/drivers/gpu/drm/nouveau/core/engine/software/nv50.c
index 5ce686ee729e..f3b4d9dbf23c 100644
--- a/drivers/gpu/drm/nouveau/core/engine/software/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/software/nv50.c
@@ -124,7 +124,7 @@ nv50_software_sclass[] = {
  ******************************************************************************/
 
 static int
-nv50_software_vblsem_release(void *data, int head)
+nv50_software_vblsem_release(void *data, u32 type, int head)
 {
 	struct nv50_software_chan *chan = data;
 	struct nv50_software_priv *priv = (void *)nv_object(chan)->engine;
@@ -183,7 +183,7 @@ nv50_software_context_ctor(struct nouveau_object *parent,
 		return -ENOMEM;
 
 	for (i = 0; i < chan->vblank.nr_event; i++) {
-		ret = nouveau_event_new(pdisp->vblank, i, pclass->vblank,
+		ret = nouveau_event_new(pdisp->vblank, 1, i, pclass->vblank,
 					chan, &chan->vblank.event[i]);
 		if (ret)
 			return ret;
diff --git a/drivers/gpu/drm/nouveau/core/engine/software/nv50.h b/drivers/gpu/drm/nouveau/core/engine/software/nv50.h
index 2de370c21279..bb49a7a20857 100644
--- a/drivers/gpu/drm/nouveau/core/engine/software/nv50.h
+++ b/drivers/gpu/drm/nouveau/core/engine/software/nv50.h
@@ -19,7 +19,7 @@ int  nv50_software_ctor(struct nouveau_object *, struct nouveau_object *,
 
 struct nv50_software_cclass {
 	struct nouveau_oclass base;
-	int (*vblank)(void *, int);
+	int (*vblank)(void *, u32, int);
 };
 
 struct nv50_software_chan {
diff --git a/drivers/gpu/drm/nouveau/core/engine/software/nvc0.c b/drivers/gpu/drm/nouveau/core/engine/software/nvc0.c
index f9430c1bf3e5..135c20f38356 100644
--- a/drivers/gpu/drm/nouveau/core/engine/software/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/software/nvc0.c
@@ -104,7 +104,7 @@ nvc0_software_sclass[] = {
  ******************************************************************************/
 
 static int
-nvc0_software_vblsem_release(void *data, int head)
+nvc0_software_vblsem_release(void *data, u32 type, int head)
 {
 	struct nv50_software_chan *chan = data;
 	struct nv50_software_priv *priv = (void *)nv_object(chan)->engine;
diff --git a/drivers/gpu/drm/nouveau/core/include/core/class.h b/drivers/gpu/drm/nouveau/core/include/core/class.h
index 9c0cd73462d9..e0c812bc884f 100644
--- a/drivers/gpu/drm/nouveau/core/include/core/class.h
+++ b/drivers/gpu/drm/nouveau/core/include/core/class.h
@@ -295,6 +295,10 @@ struct nv04_display_scanoutpos {
 #define NV84_DISP_SOR_HDMI_PWR_REKEY                                 0x0000007f
 #define NV50_DISP_SOR_LVDS_SCRIPT                                    0x00013000
 #define NV50_DISP_SOR_LVDS_SCRIPT_ID                                 0x0000ffff
+#define NV94_DISP_SOR_DP_PWR                                         0x00016000
+#define NV94_DISP_SOR_DP_PWR_STATE                                   0x00000001
+#define NV94_DISP_SOR_DP_PWR_STATE_OFF                               0x00000000
+#define NV94_DISP_SOR_DP_PWR_STATE_ON                                0x00000001
 
 #define NV50_DISP_DAC_MTHD                                           0x00020000
 #define NV50_DISP_DAC_MTHD_TYPE                                      0x0000f000
diff --git a/drivers/gpu/drm/nouveau/core/include/core/event.h b/drivers/gpu/drm/nouveau/core/include/core/event.h
index 5d539ebff3ed..ba3f1a76a815 100644
--- a/drivers/gpu/drm/nouveau/core/include/core/event.h
+++ b/drivers/gpu/drm/nouveau/core/include/core/event.h
@@ -12,32 +12,33 @@ struct nouveau_eventh {
 	struct nouveau_event *event;
 	struct list_head head;
 	unsigned long flags;
+	u32 types;
 	int index;
-	int (*func)(void *, int);
+	int (*func)(void *, u32, int);
 	void *priv;
 };
 
 struct nouveau_event {
-	spinlock_t list_lock;
-	spinlock_t refs_lock;
-
 	void *priv;
-	void (*enable)(struct nouveau_event *, int index);
-	void (*disable)(struct nouveau_event *, int index);
+	int (*check)(struct nouveau_event *, u32 type, int index);
+	void (*enable)(struct nouveau_event *, int type, int index);
+	void (*disable)(struct nouveau_event *, int type, int index);
 
+	int types_nr;
 	int index_nr;
-	struct {
-		struct list_head list;
-		int refs;
-	} index[];
+
+	spinlock_t list_lock;
+	struct list_head *list;
+	spinlock_t refs_lock;
+	int refs[];
 };
 
-int  nouveau_event_create(int index_nr, struct nouveau_event **);
+int  nouveau_event_create(int types_nr, int index_nr, struct nouveau_event **);
 void nouveau_event_destroy(struct nouveau_event **);
-void nouveau_event_trigger(struct nouveau_event *, int index);
+void nouveau_event_trigger(struct nouveau_event *, u32 types, int index);
 
-int  nouveau_event_new(struct nouveau_event *, int index,
-		       int (*func)(void *, int), void *,
+int  nouveau_event_new(struct nouveau_event *, u32 types, int index,
+		       int (*func)(void *, u32, int), void *,
 		       struct nouveau_eventh **);
 void nouveau_event_ref(struct nouveau_eventh *, struct nouveau_eventh **);
 void nouveau_event_get(struct nouveau_eventh *);
diff --git a/drivers/gpu/drm/nouveau/core/include/engine/disp.h b/drivers/gpu/drm/nouveau/core/include/engine/disp.h
index fd0c68804de3..fde842896806 100644
--- a/drivers/gpu/drm/nouveau/core/include/engine/disp.h
+++ b/drivers/gpu/drm/nouveau/core/include/engine/disp.h
@@ -6,8 +6,19 @@
 #include <core/device.h>
 #include <core/event.h>
 
+enum nvkm_hpd_event {
+	NVKM_HPD_PLUG = 1,
+	NVKM_HPD_UNPLUG = 2,
+	NVKM_HPD_IRQ = 4,
+	NVKM_HPD = (NVKM_HPD_PLUG | NVKM_HPD_UNPLUG | NVKM_HPD_IRQ)
+};
+
 struct nouveau_disp {
 	struct nouveau_engine base;
+
+	struct list_head outp;
+	struct nouveau_event *hpd;
+
 	struct nouveau_event *vblank;
 };
 
@@ -17,25 +28,6 @@ nouveau_disp(void *obj)
 	return (void *)nv_device(obj)->subdev[NVDEV_ENGINE_DISP];
 }
 
-#define nouveau_disp_create(p,e,c,h,i,x,d)                                     \
-	nouveau_disp_create_((p), (e), (c), (h), (i), (x),                     \
-			     sizeof(**d), (void **)d)
-#define nouveau_disp_destroy(d) ({                                             \
-	struct nouveau_disp *disp = (d);                                       \
-	_nouveau_disp_dtor(nv_object(disp));                                   \
-})
-#define nouveau_disp_init(d)                                                   \
-	nouveau_engine_init(&(d)->base)
-#define nouveau_disp_fini(d,s)                                                 \
-	nouveau_engine_fini(&(d)->base, (s))
-
-int  nouveau_disp_create_(struct nouveau_object *, struct nouveau_object *,
-			  struct nouveau_oclass *, int heads,
-			  const char *, const char *, int, void **);
-void _nouveau_disp_dtor(struct nouveau_object *);
-#define _nouveau_disp_init _nouveau_engine_init
-#define _nouveau_disp_fini _nouveau_engine_fini
-
 extern struct nouveau_oclass *nv04_disp_oclass;
 extern struct nouveau_oclass *nv50_disp_oclass;
 extern struct nouveau_oclass *nv84_disp_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/include/engine/fifo.h b/drivers/gpu/drm/nouveau/core/include/engine/fifo.h
index 26b6b2bb1112..b639eb2c74ff 100644
--- a/drivers/gpu/drm/nouveau/core/include/engine/fifo.h
+++ b/drivers/gpu/drm/nouveau/core/include/engine/fifo.h
@@ -109,6 +109,7 @@ extern struct nouveau_oclass *nv50_fifo_oclass;
 extern struct nouveau_oclass *nv84_fifo_oclass;
 extern struct nouveau_oclass *nvc0_fifo_oclass;
 extern struct nouveau_oclass *nve0_fifo_oclass;
+extern struct nouveau_oclass *gk20a_fifo_oclass;
 extern struct nouveau_oclass *nv108_fifo_oclass;
 
 void nv04_fifo_intr(struct nouveau_subdev *);
diff --git a/drivers/gpu/drm/nouveau/core/include/engine/graph.h b/drivers/gpu/drm/nouveau/core/include/engine/graph.h
index 871edfdf3d5b..8c1d4772da0c 100644
--- a/drivers/gpu/drm/nouveau/core/include/engine/graph.h
+++ b/drivers/gpu/drm/nouveau/core/include/engine/graph.h
@@ -68,6 +68,7 @@ extern struct nouveau_oclass *nvc8_graph_oclass;
 extern struct nouveau_oclass *nvd7_graph_oclass;
 extern struct nouveau_oclass *nvd9_graph_oclass;
 extern struct nouveau_oclass *nve4_graph_oclass;
+extern struct nouveau_oclass *gk20a_graph_oclass;
 extern struct nouveau_oclass *nvf0_graph_oclass;
 extern struct nouveau_oclass *nv108_graph_oclass;
 extern struct nouveau_oclass *gm107_graph_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/bios/conn.h b/drivers/gpu/drm/nouveau/core/include/subdev/bios/conn.h
index a32feb3f3fb6..f3930c27cb7a 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/bios/conn.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/bios/conn.h
@@ -22,7 +22,25 @@ enum dcb_connector_type {
 	DCB_CONNECTOR_NONE = 0xff
 };
 
-u16 dcb_conntab(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len);
-u16 dcb_conn(struct nouveau_bios *bios, u8 idx, u8 *ver, u8 *len);
+struct nvbios_connT {
+};
+
+u32 nvbios_connTe(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len);
+u32 nvbios_connTp(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
+		  struct nvbios_connT *info);
+
+struct nvbios_connE {
+	u8 type;
+	u8 location;
+	u8 hpd;
+	u8 dp;
+	u8 di;
+	u8 sr;
+	u8 lcdid;
+};
+
+u32 nvbios_connEe(struct nouveau_bios *bios, u8 idx, u8 *ver, u8 *hdr);
+u32 nvbios_connEp(struct nouveau_bios *bios, u8 idx, u8 *ver, u8 *hdr,
+		  struct nvbios_connE *info);
 
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/bios/dp.h b/drivers/gpu/drm/nouveau/core/include/subdev/bios/dp.h
index 6e54218b55fc..728206e21777 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/bios/dp.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/bios/dp.h
@@ -17,9 +17,10 @@ u16 nvbios_dpout_match(struct nouveau_bios *, u16 type, u16 mask,
 		       struct nvbios_dpout *);
 
 struct nvbios_dpcfg {
-	u8 drv;
-	u8 pre;
-	u8 unk;
+	u8 pc;
+	u8 dc;
+	u8 pe;
+	u8 tx_pu;
 };
 
 u16
@@ -27,7 +28,7 @@ nvbios_dpcfg_parse(struct nouveau_bios *, u16 outp, u8 idx,
 		   u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
 		   struct nvbios_dpcfg *);
 u16
-nvbios_dpcfg_match(struct nouveau_bios *, u16 outp, u8 un, u8 vs, u8 pe,
+nvbios_dpcfg_match(struct nouveau_bios *, u16 outp, u8 pc, u8 vs, u8 pe,
 		   u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
 		   struct nvbios_dpcfg *);
 
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/clock.h b/drivers/gpu/drm/nouveau/core/include/subdev/clock.h
index 8f4ced75444a..c01e29c9f89a 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/clock.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/clock.h
@@ -77,6 +77,8 @@ struct nouveau_clock {
 	int tstate; /* thermal adjustment (max-) */
 	int dstate; /* display adjustment (min+) */
 
+	bool allow_reclock;
+
 	int  (*read)(struct nouveau_clock *, enum nv_clk_src);
 	int  (*calc)(struct nouveau_clock *, struct nouveau_cstate *);
 	int  (*prog)(struct nouveau_clock *);
@@ -106,8 +108,8 @@ struct nouveau_clocks {
 	int mdiv;
 };
 
-#define nouveau_clock_create(p,e,o,i,d)                                        \
-	nouveau_clock_create_((p), (e), (o), (i), sizeof(**d), (void **)d)
+#define nouveau_clock_create(p,e,o,i,r,d)                                      \
+	nouveau_clock_create_((p), (e), (o), (i), (r), sizeof(**d), (void **)d)
 #define nouveau_clock_destroy(p) ({                                            \
 	struct nouveau_clock *clk = (p);                                       \
 	_nouveau_clock_dtor(nv_object(clk));                                   \
@@ -121,7 +123,7 @@ struct nouveau_clocks {
 
 int  nouveau_clock_create_(struct nouveau_object *, struct nouveau_object *,
 			   struct nouveau_oclass *,
-			   struct nouveau_clocks *, int, void **);
+			   struct nouveau_clocks *, bool, int, void **);
 void _nouveau_clock_dtor(struct nouveau_object *);
 int _nouveau_clock_init(struct nouveau_object *);
 #define _nouveau_clock_fini _nouveau_subdev_fini
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/fb.h b/drivers/gpu/drm/nouveau/core/include/subdev/fb.h
index 58c7ccdebb01..871e73914b24 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/fb.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/fb.h
@@ -105,6 +105,7 @@ extern struct nouveau_oclass *nvaa_fb_oclass;
 extern struct nouveau_oclass *nvaf_fb_oclass;
 extern struct nouveau_oclass *nvc0_fb_oclass;
 extern struct nouveau_oclass *nve0_fb_oclass;
+extern struct nouveau_oclass *gk20a_fb_oclass;
 extern struct nouveau_oclass *gm107_fb_oclass;
 
 #include <subdev/bios/ramcfg.h>
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/gpio.h b/drivers/gpu/drm/nouveau/core/include/subdev/gpio.h
index c85b9f1579ad..612d82ab683d 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/gpio.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/gpio.h
@@ -8,17 +8,18 @@
 #include <subdev/bios.h>
 #include <subdev/bios/gpio.h>
 
+enum nvkm_gpio_event {
+	NVKM_GPIO_HI = 1,
+	NVKM_GPIO_LO = 2,
+	NVKM_GPIO_TOGGLED = (NVKM_GPIO_HI | NVKM_GPIO_LO),
+};
+
 struct nouveau_gpio {
 	struct nouveau_subdev base;
 
 	struct nouveau_event *events;
 
-	/* hardware interfaces */
 	void (*reset)(struct nouveau_gpio *, u8 func);
-	int  (*drive)(struct nouveau_gpio *, int line, int dir, int out);
-	int  (*sense)(struct nouveau_gpio *, int line);
-
-	/* software interfaces */
 	int  (*find)(struct nouveau_gpio *, int idx, u8 tag, u8 line,
 		     struct dcb_gpio_func *);
 	int  (*set)(struct nouveau_gpio *, int idx, u8 tag, u8 line, int state);
@@ -31,23 +32,10 @@ nouveau_gpio(void *obj)
 	return (void *)nv_device(obj)->subdev[NVDEV_SUBDEV_GPIO];
 }
 
-#define nouveau_gpio_create(p,e,o,l,d)                                         \
-	nouveau_gpio_create_((p), (e), (o), (l), sizeof(**d), (void **)d)
-#define nouveau_gpio_destroy(p) ({                                             \
-	struct nouveau_gpio *gpio = (p);                                       \
-	_nouveau_gpio_dtor(nv_object(gpio));                                   \
-})
-#define nouveau_gpio_fini(p,s)                                                 \
-	nouveau_subdev_fini(&(p)->base, (s))
-
-int  nouveau_gpio_create_(struct nouveau_object *, struct nouveau_object *,
-			  struct nouveau_oclass *, int, int, void **);
-void _nouveau_gpio_dtor(struct nouveau_object *);
-int  nouveau_gpio_init(struct nouveau_gpio *);
-
-extern struct nouveau_oclass nv10_gpio_oclass;
-extern struct nouveau_oclass nv50_gpio_oclass;
-extern struct nouveau_oclass nvd0_gpio_oclass;
-extern struct nouveau_oclass nve0_gpio_oclass;
+extern struct nouveau_oclass *nv10_gpio_oclass;
+extern struct nouveau_oclass *nv50_gpio_oclass;
+extern struct nouveau_oclass *nv92_gpio_oclass;
+extern struct nouveau_oclass *nvd0_gpio_oclass;
+extern struct nouveau_oclass *nve0_gpio_oclass;
 
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/i2c.h b/drivers/gpu/drm/nouveau/core/include/subdev/i2c.h
index 7f50a858b16f..db1b39d08013 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/i2c.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/i2c.h
@@ -14,52 +14,41 @@
 #define NV_I2C_TYPE_EXTDDC(e) (0x0005 | (e) << 8)
 #define NV_I2C_TYPE_EXTAUX(e) (0x0006 | (e) << 8)
 
+enum nvkm_i2c_event {
+	NVKM_I2C_PLUG = 1,
+	NVKM_I2C_UNPLUG = 2,
+	NVKM_I2C_IRQ = 4,
+	NVKM_I2C_DONE = 8,
+	NVKM_I2C_ANY = (NVKM_I2C_PLUG |
+			NVKM_I2C_UNPLUG |
+			NVKM_I2C_IRQ |
+			NVKM_I2C_DONE),
+};
+
 struct nouveau_i2c_port {
 	struct nouveau_object base;
 	struct i2c_adapter adapter;
+	struct mutex mutex;
 
 	struct list_head head;
 	u8  index;
+	int aux;
 
 	const struct nouveau_i2c_func *func;
 };
 
 struct nouveau_i2c_func {
-	void (*acquire)(struct nouveau_i2c_port *);
-	void (*release)(struct nouveau_i2c_port *);
-
 	void (*drive_scl)(struct nouveau_i2c_port *, int);
 	void (*drive_sda)(struct nouveau_i2c_port *, int);
 	int  (*sense_scl)(struct nouveau_i2c_port *);
 	int  (*sense_sda)(struct nouveau_i2c_port *);
 
-	int  (*aux)(struct nouveau_i2c_port *, u8, u32, u8 *, u8);
+	int  (*aux)(struct nouveau_i2c_port *, bool, u8, u32, u8 *, u8);
 	int  (*pattern)(struct nouveau_i2c_port *, int pattern);
 	int  (*lnk_ctl)(struct nouveau_i2c_port *, int nr, int bw, bool enh);
 	int  (*drv_ctl)(struct nouveau_i2c_port *, int lane, int sw, int pe);
 };
 
-#define nouveau_i2c_port_create(p,e,o,i,a,f,d)                                 \
-	nouveau_i2c_port_create_((p), (e), (o), (i), (a), (f),                 \
-				 sizeof(**d), (void **)d)
-#define nouveau_i2c_port_destroy(p) ({                                         \
-	struct nouveau_i2c_port *port = (p);                                   \
-	_nouveau_i2c_port_dtor(nv_object(i2c));                                \
-})
-#define nouveau_i2c_port_init(p)                                               \
-	nouveau_object_init(&(p)->base)
-#define nouveau_i2c_port_fini(p,s)                                             \
-	nouveau_object_fini(&(p)->base, (s))
-
-int nouveau_i2c_port_create_(struct nouveau_object *, struct nouveau_object *,
-			     struct nouveau_oclass *, u8,
-			     const struct i2c_algorithm *,
-			     const struct nouveau_i2c_func *,
-			     int, void **);
-void _nouveau_i2c_port_dtor(struct nouveau_object *);
-#define _nouveau_i2c_port_init nouveau_object_init
-#define _nouveau_i2c_port_fini nouveau_object_fini
-
 struct nouveau_i2c_board_info {
 	struct i2c_board_info dev;
 	u8 udelay; /* set to 0 to use the standard delay */
@@ -67,13 +56,20 @@ struct nouveau_i2c_board_info {
 
 struct nouveau_i2c {
 	struct nouveau_subdev base;
+	struct nouveau_event *ntfy;
 
 	struct nouveau_i2c_port *(*find)(struct nouveau_i2c *, u8 index);
 	struct nouveau_i2c_port *(*find_type)(struct nouveau_i2c *, u16 type);
+	int  (*acquire_pad)(struct nouveau_i2c_port *, unsigned long timeout);
+	void (*release_pad)(struct nouveau_i2c_port *);
+	int  (*acquire)(struct nouveau_i2c_port *, unsigned long timeout);
+	void (*release)(struct nouveau_i2c_port *);
 	int (*identify)(struct nouveau_i2c *, int index,
 			const char *what, struct nouveau_i2c_board_info *,
 			bool (*match)(struct nouveau_i2c_port *,
 				      struct i2c_board_info *, void *), void *);
+
+	wait_queue_head_t wait;
 	struct list_head ports;
 };
 
@@ -83,37 +79,12 @@ nouveau_i2c(void *obj)
 	return (void *)nv_device(obj)->subdev[NVDEV_SUBDEV_I2C];
 }
 
-#define nouveau_i2c_create(p,e,o,s,d)                                          \
-	nouveau_i2c_create_((p), (e), (o), (s), sizeof(**d), (void **)d)
-#define nouveau_i2c_destroy(p) ({                                              \
-	struct nouveau_i2c *i2c = (p);                                         \
-	_nouveau_i2c_dtor(nv_object(i2c));                                     \
-})
-#define nouveau_i2c_init(p) ({                                                 \
-	struct nouveau_i2c *i2c = (p);                                         \
-	_nouveau_i2c_init(nv_object(i2c));                                     \
-})
-#define nouveau_i2c_fini(p,s) ({                                               \
-	struct nouveau_i2c *i2c = (p);                                         \
-	_nouveau_i2c_fini(nv_object(i2c), (s));                                \
-})
-
-int nouveau_i2c_create_(struct nouveau_object *, struct nouveau_object *,
-			struct nouveau_oclass *, struct nouveau_oclass *,
-			int, void **);
-void _nouveau_i2c_dtor(struct nouveau_object *);
-int  _nouveau_i2c_init(struct nouveau_object *);
-int  _nouveau_i2c_fini(struct nouveau_object *, bool);
-
-extern struct nouveau_oclass nv04_i2c_oclass;
-extern struct nouveau_oclass nv4e_i2c_oclass;
-extern struct nouveau_oclass nv50_i2c_oclass;
-extern struct nouveau_oclass nv94_i2c_oclass;
-extern struct nouveau_oclass nvd0_i2c_oclass;
-extern struct nouveau_oclass nouveau_anx9805_sclass[];
-
-extern const struct i2c_algorithm nouveau_i2c_bit_algo;
-extern const struct i2c_algorithm nouveau_i2c_aux_algo;
+extern struct nouveau_oclass *nv04_i2c_oclass;
+extern struct nouveau_oclass *nv4e_i2c_oclass;
+extern struct nouveau_oclass *nv50_i2c_oclass;
+extern struct nouveau_oclass *nv94_i2c_oclass;
+extern struct nouveau_oclass *nvd0_i2c_oclass;
+extern struct nouveau_oclass *nve0_i2c_oclass;
 
 static inline int
 nv_rdi2cr(struct nouveau_i2c_port *port, u8 addr, u8 reg)
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/ibus.h b/drivers/gpu/drm/nouveau/core/include/subdev/ibus.h
index 88814f159d89..31df634c0fdc 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/ibus.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/ibus.h
@@ -30,5 +30,6 @@ nouveau_ibus(void *obj)
 
 extern struct nouveau_oclass nvc0_ibus_oclass;
 extern struct nouveau_oclass nve0_ibus_oclass;
+extern struct nouveau_oclass gk20a_ibus_oclass;
 
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bar/base.c b/drivers/gpu/drm/nouveau/core/subdev/bar/base.c
index bdf594116f3f..73b1ed20c8d5 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bar/base.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bar/base.c
@@ -118,8 +118,10 @@ nouveau_bar_create_(struct nouveau_object *parent,
 	if (ret)
 		return ret;
 
-	bar->iomem = ioremap(nv_device_resource_start(device, 3),
-			     nv_device_resource_len(device, 3));
+	if (nv_device_resource_len(device, 3) != 0)
+		bar->iomem = ioremap(nv_device_resource_start(device, 3),
+				     nv_device_resource_len(device, 3));
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bar/nvc0.c b/drivers/gpu/drm/nouveau/core/subdev/bar/nvc0.c
index 3f30db62e656..ca8139b9ab27 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bar/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bar/nvc0.c
@@ -30,14 +30,16 @@
 
 #include "priv.h"
 
+struct nvc0_bar_priv_vm {
+	struct nouveau_gpuobj *mem;
+	struct nouveau_gpuobj *pgd;
+	struct nouveau_vm *vm;
+};
+
 struct nvc0_bar_priv {
 	struct nouveau_bar base;
 	spinlock_t lock;
-	struct {
-		struct nouveau_gpuobj *mem;
-		struct nouveau_gpuobj *pgd;
-		struct nouveau_vm *vm;
-	} bar[2];
+	struct nvc0_bar_priv_vm bar[2];
 };
 
 static int
@@ -79,87 +81,87 @@ nvc0_bar_unmap(struct nouveau_bar *bar, struct nouveau_vma *vma)
 }
 
 static int
-nvc0_bar_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	      struct nouveau_oclass *oclass, void *data, u32 size,
-	      struct nouveau_object **pobject)
+nvc0_bar_init_vm(struct nvc0_bar_priv *priv, struct nvc0_bar_priv_vm *bar_vm,
+		 int bar_nr)
 {
-	struct nouveau_device *device = nv_device(parent);
-	struct nvc0_bar_priv *priv;
-	struct nouveau_gpuobj *mem;
+	struct nouveau_device *device = nv_device(&priv->base);
 	struct nouveau_vm *vm;
+	resource_size_t bar_len;
 	int ret;
 
-	ret = nouveau_bar_create(parent, engine, oclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	/* BAR3 */
 	ret = nouveau_gpuobj_new(nv_object(priv), NULL, 0x1000, 0, 0,
-				&priv->bar[0].mem);
-	mem = priv->bar[0].mem;
+				&bar_vm->mem);
 	if (ret)
 		return ret;
 
 	ret = nouveau_gpuobj_new(nv_object(priv), NULL, 0x8000, 0, 0,
-				&priv->bar[0].pgd);
+				&bar_vm->pgd);
 	if (ret)
 		return ret;
 
-	ret = nouveau_vm_new(device, 0, nv_device_resource_len(device, 3), 0, &vm);
+	bar_len = nv_device_resource_len(device, bar_nr);
+
+	ret = nouveau_vm_new(device, 0, bar_len, 0, &vm);
 	if (ret)
 		return ret;
 
 	atomic_inc(&vm->engref[NVDEV_SUBDEV_BAR]);
 
-	ret = nouveau_gpuobj_new(nv_object(priv), NULL,
-				 (nv_device_resource_len(device, 3) >> 12) * 8,
-				 0x1000, NVOBJ_FLAG_ZERO_ALLOC,
-				 &vm->pgt[0].obj[0]);
-	vm->pgt[0].refcount[0] = 1;
-	if (ret)
-		return ret;
+	/*
+	 * Bootstrap page table lookup.
+	 */
+	if (bar_nr == 3) {
+		ret = nouveau_gpuobj_new(nv_object(priv), NULL,
+					 (bar_len >> 12) * 8, 0x1000,
+					 NVOBJ_FLAG_ZERO_ALLOC,
+					&vm->pgt[0].obj[0]);
+		vm->pgt[0].refcount[0] = 1;
+		if (ret)
+			return ret;
+	}
 
-	ret = nouveau_vm_ref(vm, &priv->bar[0].vm, priv->bar[0].pgd);
+	ret = nouveau_vm_ref(vm, &bar_vm->vm, bar_vm->pgd);
 	nouveau_vm_ref(NULL, &vm, NULL);
 	if (ret)
 		return ret;
 
-	nv_wo32(mem, 0x0200, lower_32_bits(priv->bar[0].pgd->addr));
-	nv_wo32(mem, 0x0204, upper_32_bits(priv->bar[0].pgd->addr));
-	nv_wo32(mem, 0x0208, lower_32_bits(nv_device_resource_len(device, 3) - 1));
-	nv_wo32(mem, 0x020c, upper_32_bits(nv_device_resource_len(device, 3) - 1));
+	nv_wo32(bar_vm->mem, 0x0200, lower_32_bits(bar_vm->pgd->addr));
+	nv_wo32(bar_vm->mem, 0x0204, upper_32_bits(bar_vm->pgd->addr));
+	nv_wo32(bar_vm->mem, 0x0208, lower_32_bits(bar_len - 1));
+	nv_wo32(bar_vm->mem, 0x020c, upper_32_bits(bar_len - 1));
 
-	/* BAR1 */
-	ret = nouveau_gpuobj_new(nv_object(priv), NULL, 0x1000, 0, 0,
-				&priv->bar[1].mem);
-	mem = priv->bar[1].mem;
-	if (ret)
-		return ret;
+	return 0;
+}
 
-	ret = nouveau_gpuobj_new(nv_object(priv), NULL, 0x8000, 0, 0,
-				&priv->bar[1].pgd);
-	if (ret)
-		return ret;
+static int
+nvc0_bar_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	      struct nouveau_oclass *oclass, void *data, u32 size,
+	      struct nouveau_object **pobject)
+{
+	struct nouveau_device *device = nv_device(parent);
+	struct nvc0_bar_priv *priv;
+	bool has_bar3 = nv_device_resource_len(device, 3) != 0;
+	int ret;
 
-	ret = nouveau_vm_new(device, 0, nv_device_resource_len(device, 1), 0, &vm);
+	ret = nouveau_bar_create(parent, engine, oclass, &priv);
+	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
 
-	atomic_inc(&vm->engref[NVDEV_SUBDEV_BAR]);
+	/* BAR3 */
+	if (has_bar3) {
+		ret = nvc0_bar_init_vm(priv, &priv->bar[0], 3);
+		if (ret)
+			return ret;
+		priv->base.alloc = nouveau_bar_alloc;
+		priv->base.kmap = nvc0_bar_kmap;
+	}
 
-	ret = nouveau_vm_ref(vm, &priv->bar[1].vm, priv->bar[1].pgd);
-	nouveau_vm_ref(NULL, &vm, NULL);
+	/* BAR1 */
+	ret = nvc0_bar_init_vm(priv, &priv->bar[1], 1);
 	if (ret)
 		return ret;
 
-	nv_wo32(mem, 0x0200, lower_32_bits(priv->bar[1].pgd->addr));
-	nv_wo32(mem, 0x0204, upper_32_bits(priv->bar[1].pgd->addr));
-	nv_wo32(mem, 0x0208, lower_32_bits(nv_device_resource_len(device, 1) - 1));
-	nv_wo32(mem, 0x020c, upper_32_bits(nv_device_resource_len(device, 1) - 1));
-
-	priv->base.alloc = nouveau_bar_alloc;
-	priv->base.kmap = nvc0_bar_kmap;
 	priv->base.umap = nvc0_bar_umap;
 	priv->base.unmap = nvc0_bar_unmap;
 	priv->base.flush = nv84_bar_flush;
@@ -201,7 +203,9 @@ nvc0_bar_init(struct nouveau_object *object)
 	nv_mask(priv, 0x100c80, 0x00000001, 0x00000000);
 
 	nv_wr32(priv, 0x001704, 0x80000000 | priv->bar[1].mem->addr >> 12);
-	nv_wr32(priv, 0x001714, 0xc0000000 | priv->bar[0].mem->addr >> 12);
+	if (priv->bar[0].mem)
+		nv_wr32(priv, 0x001714,
+			0xc0000000 | priv->bar[0].mem->addr >> 12);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/base.c b/drivers/gpu/drm/nouveau/core/subdev/bios/base.c
index 222e8ebb669d..d45704a2c2df 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bios/base.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/base.c
@@ -183,10 +183,11 @@ nouveau_bios_shadow_prom(struct nouveau_bios *bios)
 		goto out;
 
 	bios->data = kmalloc(bios->size, GFP_KERNEL);
-	if (bios->data) {
-		for (i = 0; i < bios->size; i += 4)
-			((u32 *)bios->data)[i/4] = nv_rd32(bios, 0x300000 + i);
-	}
+	if (!bios->data)
+		goto out;
+
+	for (i = 0; i < bios->size; i += 4)
+		((u32 *)bios->data)[i/4] = nv_rd32(bios, 0x300000 + i);
 
 	/* check the PCI record header */
 	pcir = nv_ro16(bios, 0x0018);
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/conn.c b/drivers/gpu/drm/nouveau/core/subdev/bios/conn.c
index 5ac010efd959..2ede3bcd96a1 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bios/conn.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/conn.c
@@ -28,12 +28,12 @@
 #include <subdev/bios/dcb.h>
 #include <subdev/bios/conn.h>
 
-u16
-dcb_conntab(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
+u32
+nvbios_connTe(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
 {
-	u16 dcb = dcb_table(bios, ver, hdr, cnt, len);
+	u32 dcb = dcb_table(bios, ver, hdr, cnt, len);
 	if (dcb && *ver >= 0x30 && *hdr >= 0x16) {
-		u16 data = nv_ro16(bios, dcb + 0x14);
+		u32 data = nv_ro16(bios, dcb + 0x14);
 		if (data) {
 			*ver = nv_ro08(bios, data + 0);
 			*hdr = nv_ro08(bios, data + 1);
@@ -42,15 +42,59 @@ dcb_conntab(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
 			return data;
 		}
 	}
-	return 0x0000;
+	return 0x00000000;
 }
 
-u16
-dcb_conn(struct nouveau_bios *bios, u8 idx, u8 *ver, u8 *len)
+u32
+nvbios_connTp(struct nouveau_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
+	      struct nvbios_connT *info)
+{
+	u32 data = nvbios_connTe(bios, ver, hdr, cnt, len);
+	memset(info, 0x00, sizeof(*info));
+	switch (!!data * *ver) {
+	case 0x30:
+	case 0x40:
+		return data;
+	default:
+		break;
+	}
+	return 0x00000000;
+}
+
+u32
+nvbios_connEe(struct nouveau_bios *bios, u8 idx, u8 *ver, u8 *len)
 {
 	u8  hdr, cnt;
-	u16 data = dcb_conntab(bios, ver, &hdr, &cnt, len);
+	u32 data = nvbios_connTe(bios, ver, &hdr, &cnt, len);
 	if (data && idx < cnt)
 		return data + hdr + (idx * *len);
-	return 0x0000;
+	return 0x00000000;
+}
+
+u32
+nvbios_connEp(struct nouveau_bios *bios, u8 idx, u8 *ver, u8 *len,
+	      struct nvbios_connE *info)
+{
+	u32 data = nvbios_connEe(bios, idx, ver, len);
+	memset(info, 0x00, sizeof(*info));
+	switch (!!data * *ver) {
+	case 0x30:
+	case 0x40:
+		info->type     =  nv_ro08(bios, data + 0x00);
+		info->location =  nv_ro08(bios, data + 0x01) & 0x0f;
+		info->hpd      = (nv_ro08(bios, data + 0x01) & 0x30) >> 4;
+		info->dp       = (nv_ro08(bios, data + 0x01) & 0xc0) >> 6;
+		if (*len < 4)
+			return data;
+		info->hpd     |= (nv_ro08(bios, data + 0x02) & 0x03) << 2;
+		info->dp      |=  nv_ro08(bios, data + 0x02) & 0x0c;
+		info->di       = (nv_ro08(bios, data + 0x02) & 0xf0) >> 4;
+		info->hpd     |= (nv_ro08(bios, data + 0x03) & 0x07) << 4;
+		info->sr       = (nv_ro08(bios, data + 0x03) & 0x08) >> 3;
+		info->lcdid    = (nv_ro08(bios, data + 0x03) & 0x70) >> 4;
+		return data;
+	default:
+		break;
+	}
+	return 0x00000000;
 }
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/dp.c b/drivers/gpu/drm/nouveau/core/subdev/bios/dp.c
index 7628fe759220..f309dd657250 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bios/dp.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/dp.c
@@ -162,18 +162,20 @@ nvbios_dpcfg_parse(struct nouveau_bios *bios, u16 outp, u8 idx,
 		   struct nvbios_dpcfg *info)
 {
 	u16 data = nvbios_dpcfg_entry(bios, outp, idx, ver, hdr, cnt, len);
+	memset(info, 0x00, sizeof(*info));
 	if (data) {
 		switch (*ver) {
 		case 0x21:
-			info->drv = nv_ro08(bios, data + 0x02);
-			info->pre = nv_ro08(bios, data + 0x03);
-			info->unk = nv_ro08(bios, data + 0x04);
+			info->dc    = nv_ro08(bios, data + 0x02);
+			info->pe    = nv_ro08(bios, data + 0x03);
+			info->tx_pu = nv_ro08(bios, data + 0x04);
 			break;
 		case 0x30:
 		case 0x40:
-			info->drv = nv_ro08(bios, data + 0x01);
-			info->pre = nv_ro08(bios, data + 0x02);
-			info->unk = nv_ro08(bios, data + 0x03);
+			info->pc    = nv_ro08(bios, data + 0x00);
+			info->dc    = nv_ro08(bios, data + 0x01);
+			info->pe    = nv_ro08(bios, data + 0x02);
+			info->tx_pu = nv_ro08(bios, data + 0x03);
 			break;
 		default:
 			data = 0x0000;
@@ -184,7 +186,7 @@ nvbios_dpcfg_parse(struct nouveau_bios *bios, u16 outp, u8 idx,
 }
 
 u16
-nvbios_dpcfg_match(struct nouveau_bios *bios, u16 outp, u8 un, u8 vs, u8 pe,
+nvbios_dpcfg_match(struct nouveau_bios *bios, u16 outp, u8 pc, u8 vs, u8 pe,
 		   u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
 		   struct nvbios_dpcfg *info)
 {
@@ -193,16 +195,15 @@ nvbios_dpcfg_match(struct nouveau_bios *bios, u16 outp, u8 un, u8 vs, u8 pe,
 
 	if (*ver >= 0x30) {
 		const u8 vsoff[] = { 0, 4, 7, 9 };
-		idx = (un * 10) + vsoff[vs] + pe;
+		idx = (pc * 10) + vsoff[vs] + pe;
 	} else {
-		while ((data = nvbios_dpcfg_entry(bios, outp, idx,
+		while ((data = nvbios_dpcfg_entry(bios, outp, ++idx,
 						  ver, hdr, cnt, len))) {
 			if (nv_ro08(bios, data + 0x00) == vs &&
 			    nv_ro08(bios, data + 0x01) == pe)
 				break;
-			idx++;
 		}
 	}
 
-	return nvbios_dpcfg_parse(bios, outp, pe, ver, hdr, cnt, len, info);
+	return nvbios_dpcfg_parse(bios, outp, idx, ver, hdr, cnt, len, info);
 }
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/init.c b/drivers/gpu/drm/nouveau/core/subdev/bios/init.c
index acaeaf79e3f0..626380f9e4c0 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bios/init.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/init.c
@@ -98,15 +98,16 @@ static u8
 init_conn(struct nvbios_init *init)
 {
 	struct nouveau_bios *bios = init->bios;
-	u8  ver, len;
-	u16 conn;
+	struct nvbios_connE connE;
+	u8  ver, hdr;
+	u32 conn;
 
 	if (init_exec(init)) {
 		if (init->outp) {
 			conn = init->outp->connector;
-			conn = dcb_conn(bios, conn, &ver, &len);
+			conn = nvbios_connEp(bios, conn, &ver, &hdr, &connE);
 			if (conn)
-				return nv_ro08(bios, conn);
+				return connE.type;
 		}
 
 		error("script needs connector type\n");
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/base.c b/drivers/gpu/drm/nouveau/core/subdev/clock/base.c
index dd62baead39c..22351f594d2a 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/base.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/base.c
@@ -346,8 +346,8 @@ nouveau_clock_ustate_update(struct nouveau_clock *clk, int req)
 	struct nouveau_pstate *pstate;
 	int i = 0;
 
-	/* YKW repellant */
-	return -ENOSYS;
+	if (!clk->allow_reclock)
+		return -ENOSYS;
 
 	if (req != -1 && req != -2) {
 		list_for_each_entry(pstate, &clk->states, head) {
@@ -456,6 +456,7 @@ nouveau_clock_create_(struct nouveau_object *parent,
 		      struct nouveau_object *engine,
 		      struct nouveau_oclass *oclass,
 		      struct nouveau_clocks *clocks,
+		      bool allow_reclock,
 		      int length, void **object)
 {
 	struct nouveau_device *device = nv_device(parent);
@@ -478,6 +479,8 @@ nouveau_clock_create_(struct nouveau_object *parent,
 		ret = nouveau_pstate_new(clk, idx++);
 	} while (ret == 0);
 
+	clk->allow_reclock = allow_reclock;
+
 	mode = nouveau_stropt(device->cfgopt, "NvClkMode", &arglen);
 	if (mode) {
 		if (!strncasecmpz(mode, "disabled", arglen)) {
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nv04.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nv04.c
index b74db6cfc4e2..eb2d4425a49e 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nv04.c
@@ -82,7 +82,8 @@ nv04_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nv04_clock_priv *priv;
 	int ret;
 
-	ret = nouveau_clock_create(parent, engine, oclass, nv04_domain, &priv);
+	ret = nouveau_clock_create(parent, engine, oclass, nv04_domain, false,
+				   &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nv40.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nv40.c
index db7346f79080..8a9e16839791 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/nv40.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nv40.c
@@ -213,7 +213,8 @@ nv40_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nv40_clock_priv *priv;
 	int ret;
 
-	ret = nouveau_clock_create(parent, engine, oclass, nv40_domain, &priv);
+	ret = nouveau_clock_create(parent, engine, oclass, nv40_domain, true,
+				   &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nv50.c
index 250a6d96016b..8c132772ba9e 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nv50.c
@@ -507,7 +507,7 @@ nv50_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	int ret;
 
 	ret = nouveau_clock_create(parent, engine, oclass, pclass->domains,
-				  &priv);
+				   false, &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nva3.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nva3.c
index 4f5a1373f002..9fb58354a80b 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/nva3.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nva3.c
@@ -302,7 +302,8 @@ nva3_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nva3_clock_priv *priv;
 	int ret;
 
-	ret = nouveau_clock_create(parent, engine, oclass, nva3_domain, &priv);
+	ret = nouveau_clock_create(parent, engine, oclass, nva3_domain, false,
+				   &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nvaa.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nvaa.c
index 7a723b4f564d..6a65fc9e9663 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/nvaa.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nvaa.c
@@ -421,7 +421,8 @@ nvaa_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nvaa_clock_priv *priv;
 	int ret;
 
-	ret = nouveau_clock_create(parent, engine, oclass, nvaa_domains, &priv);
+	ret = nouveau_clock_create(parent, engine, oclass, nvaa_domains, true,
+				   &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nvc0.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nvc0.c
index c3105720ed24..dbf8517f54da 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nvc0.c
@@ -437,7 +437,8 @@ nvc0_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nvc0_clock_priv *priv;
 	int ret;
 
-	ret = nouveau_clock_create(parent, engine, oclass, nvc0_domain, &priv);
+	ret = nouveau_clock_create(parent, engine, oclass, nvc0_domain, false,
+				   &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nve0.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nve0.c
index d3c37c96f0e7..4ac1aa30ea11 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/nve0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nve0.c
@@ -473,7 +473,8 @@ nve0_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nve0_clock_priv *priv;
 	int ret;
 
-	ret = nouveau_clock_create(parent, engine, oclass, nve0_domain, &priv);
+	ret = nouveau_clock_create(parent, engine, oclass, nve0_domain, true,
+				   &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/gk20a.c b/drivers/gpu/drm/nouveau/core/subdev/fb/gk20a.c
new file mode 100644
index 000000000000..a16024a74771
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/gk20a.c
@@ -0,0 +1,56 @@
+/*
+ * Copyright (c) 2014, NVIDIA CORPORATION. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#include "nvc0.h"
+
+struct gk20a_fb_priv {
+	struct nouveau_fb base;
+};
+
+static int
+gk20a_fb_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	     struct nouveau_oclass *oclass, void *data, u32 size,
+	     struct nouveau_object **pobject)
+{
+	struct gk20a_fb_priv *priv;
+	int ret;
+
+	ret = nouveau_fb_create(parent, engine, oclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+struct nouveau_oclass *
+gk20a_fb_oclass = &(struct nouveau_fb_impl) {
+	.base.handle = NV_SUBDEV(FB, 0xea),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = gk20a_fb_ctor,
+		.dtor = _nouveau_fb_dtor,
+		.init = _nouveau_fb_init,
+		.fini = _nouveau_fb_fini,
+	},
+	.memtype = nvc0_fb_memtype_valid,
+	.ram = &gk20a_ram_oclass,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/priv.h b/drivers/gpu/drm/nouveau/core/subdev/fb/priv.h
index da74c889aed4..82273f832e42 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/priv.h
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/priv.h
@@ -32,6 +32,7 @@ extern struct nouveau_oclass nva3_ram_oclass;
 extern struct nouveau_oclass nvaa_ram_oclass;
 extern struct nouveau_oclass nvc0_ram_oclass;
 extern struct nouveau_oclass nve0_ram_oclass;
+extern struct nouveau_oclass gk20a_ram_oclass;
 extern struct nouveau_oclass gm107_ram_oclass;
 
 int nouveau_sddr3_calc(struct nouveau_ram *ram);
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramgk20a.c b/drivers/gpu/drm/nouveau/core/subdev/fb/ramgk20a.c
new file mode 100644
index 000000000000..4d77d75e4673
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramgk20a.c
@@ -0,0 +1,152 @@
+/*
+ * Copyright (c) 2014, NVIDIA CORPORATION. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#include "priv.h"
+
+#include <subdev/fb.h>
+
+struct gk20a_mem {
+	struct nouveau_mem base;
+	void *cpuaddr;
+	dma_addr_t handle;
+};
+#define to_gk20a_mem(m) container_of(m, struct gk20a_mem, base)
+
+static void
+gk20a_ram_put(struct nouveau_fb *pfb, struct nouveau_mem **pmem)
+{
+	struct device *dev = nv_device_base(nv_device(pfb));
+	struct gk20a_mem *mem = to_gk20a_mem(*pmem);
+
+	*pmem = NULL;
+	if (unlikely(mem == NULL))
+		return;
+
+	if (likely(mem->cpuaddr))
+		dma_free_coherent(dev, mem->base.size << PAGE_SHIFT,
+				  mem->cpuaddr, mem->handle);
+
+	kfree(mem->base.pages);
+	kfree(mem);
+}
+
+static int
+gk20a_ram_get(struct nouveau_fb *pfb, u64 size, u32 align, u32 ncmin,
+	     u32 memtype, struct nouveau_mem **pmem)
+{
+	struct device *dev = nv_device_base(nv_device(pfb));
+	struct gk20a_mem *mem;
+	u32 type = memtype & 0xff;
+	u32 npages, order;
+	int i;
+
+	nv_debug(pfb, "%s: size: %llx align: %x, ncmin: %x\n", __func__, size,
+		 align, ncmin);
+
+	npages = size >> PAGE_SHIFT;
+	if (npages == 0)
+		npages = 1;
+
+	if (align == 0)
+		align = PAGE_SIZE;
+	align >>= PAGE_SHIFT;
+
+	/* round alignment to the next power of 2, if needed */
+	order = fls(align);
+	if ((align & (align - 1)) == 0)
+		order--;
+	align = BIT(order);
+
+	/* ensure returned address is correctly aligned */
+	npages = max(align, npages);
+
+	mem = kzalloc(sizeof(*mem), GFP_KERNEL);
+	if (!mem)
+		return -ENOMEM;
+
+	mem->base.size = npages;
+	mem->base.memtype = type;
+
+	mem->base.pages = kzalloc(sizeof(dma_addr_t) * npages, GFP_KERNEL);
+	if (!mem->base.pages) {
+		kfree(mem);
+		return -ENOMEM;
+	}
+
+	*pmem = &mem->base;
+
+	mem->cpuaddr = dma_alloc_coherent(dev, npages << PAGE_SHIFT,
+					  &mem->handle, GFP_KERNEL);
+	if (!mem->cpuaddr) {
+		nv_error(pfb, "%s: cannot allocate memory!\n", __func__);
+		gk20a_ram_put(pfb, pmem);
+		return -ENOMEM;
+	}
+
+	align <<= PAGE_SHIFT;
+
+	/* alignment check */
+	if (unlikely(mem->handle & (align - 1)))
+		nv_warn(pfb, "memory not aligned as requested: %pad (0x%x)\n",
+			&mem->handle, align);
+
+	nv_debug(pfb, "alloc size: 0x%x, align: 0x%x, paddr: %pad, vaddr: %p\n",
+		 npages << PAGE_SHIFT, align, &mem->handle, mem->cpuaddr);
+
+	for (i = 0; i < npages; i++)
+		mem->base.pages[i] = mem->handle + (PAGE_SIZE * i);
+
+	mem->base.offset = (u64)mem->base.pages[0];
+
+	return 0;
+}
+
+static int
+gk20a_ram_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	      struct nouveau_oclass *oclass, void *data, u32 datasize,
+	      struct nouveau_object **pobject)
+{
+	struct nouveau_ram *ram;
+	int ret;
+
+	ret = nouveau_ram_create(parent, engine, oclass, &ram);
+	*pobject = nv_object(ram);
+	if (ret)
+		return ret;
+	ram->type = NV_MEM_TYPE_STOLEN;
+	ram->size = get_num_physpages() << PAGE_SHIFT;
+
+	ram->get = gk20a_ram_get;
+	ram->put = gk20a_ram_put;
+
+	return 0;
+}
+
+struct nouveau_oclass
+gk20a_ram_oclass = {
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = gk20a_ram_ctor,
+		.dtor = _nouveau_ram_dtor,
+		.init = _nouveau_ram_init,
+		.fini = _nouveau_ram_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv50.c b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv50.c
index ef91b6e893af..e5d12c24cc43 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnv50.c
@@ -211,7 +211,7 @@ nv50_ram_prog(struct nouveau_fb *pfb)
 	struct nv50_ram *ram = (void *)pfb->ram;
 	struct nv50_ramseq *hwsq = &ram->hwsq;
 
-	ram_exec(hwsq, nouveau_boolopt(device->cfgopt, "NvMemExec", false));
+	ram_exec(hwsq, nouveau_boolopt(device->cfgopt, "NvMemExec", true));
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnva3.c b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnva3.c
index 6eb97f16fbda..8076fb195dd5 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnva3.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnva3.c
@@ -309,7 +309,7 @@ nva3_ram_prog(struct nouveau_fb *pfb)
 	struct nouveau_device *device = nv_device(pfb);
 	struct nva3_ram *ram = (void *)pfb->ram;
 	struct nva3_ramfuc *fuc = &ram->fuc;
-	ram_exec(fuc, nouveau_boolopt(device->cfgopt, "NvMemExec", false));
+	ram_exec(fuc, nouveau_boolopt(device->cfgopt, "NvMemExec", true));
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnvc0.c b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnvc0.c
index 8edc92224c84..5a6a5027f749 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnvc0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnvc0.c
@@ -408,7 +408,7 @@ nvc0_ram_prog(struct nouveau_fb *pfb)
 	struct nouveau_device *device = nv_device(pfb);
 	struct nvc0_ram *ram = (void *)pfb->ram;
 	struct nvc0_ramfuc *fuc = &ram->fuc;
-	ram_exec(fuc, nouveau_boolopt(device->cfgopt, "NvMemExec", false));
+	ram_exec(fuc, nouveau_boolopt(device->cfgopt, "NvMemExec", true));
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnve0.c b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnve0.c
index 16752192cf87..84c7efbc4f38 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnve0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnve0.c
@@ -1111,7 +1111,7 @@ nve0_ram_prog(struct nouveau_fb *pfb)
 	struct nouveau_device *device = nv_device(pfb);
 	struct nve0_ram *ram = (void *)pfb->ram;
 	struct nve0_ramfuc *fuc = &ram->fuc;
-	ram_exec(fuc, nouveau_boolopt(device->cfgopt, "NvMemExec", false));
+	ram_exec(fuc, nouveau_boolopt(device->cfgopt, "NvMemExec", true));
 	return (ram->base.next == &ram->base.xition);
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/gpio/base.c b/drivers/gpu/drm/nouveau/core/subdev/gpio/base.c
index f572c2804c32..45e0202f3151 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/gpio/base.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/gpio/base.c
@@ -22,21 +22,24 @@
  * Authors: Ben Skeggs
  */
 
-#include <subdev/gpio.h>
 #include <subdev/bios.h>
 #include <subdev/bios/gpio.h>
 
+#include "priv.h"
+
 static int
 nouveau_gpio_drive(struct nouveau_gpio *gpio,
 		   int idx, int line, int dir, int out)
 {
-	return gpio->drive ? gpio->drive(gpio, line, dir, out) : -ENODEV;
+	const struct nouveau_gpio_impl *impl = (void *)nv_object(gpio)->oclass;
+	return impl->drive ? impl->drive(gpio, line, dir, out) : -ENODEV;
 }
 
 static int
 nouveau_gpio_sense(struct nouveau_gpio *gpio, int idx, int line)
 {
-	return gpio->sense ? gpio->sense(gpio, line) : -ENODEV;
+	const struct nouveau_gpio_impl *impl = (void *)nv_object(gpio)->oclass;
+	return impl->sense ? impl->sense(gpio, line) : -ENODEV;
 }
 
 static int
@@ -102,6 +105,80 @@ nouveau_gpio_get(struct nouveau_gpio *gpio, int idx, u8 tag, u8 line)
 	return ret;
 }
 
+static void
+nouveau_gpio_intr_disable(struct nouveau_event *event, int type, int index)
+{
+	struct nouveau_gpio *gpio = nouveau_gpio(event->priv);
+	const struct nouveau_gpio_impl *impl = (void *)nv_object(gpio)->oclass;
+	impl->intr_mask(gpio, type, 1 << index, 0);
+}
+
+static void
+nouveau_gpio_intr_enable(struct nouveau_event *event, int type, int index)
+{
+	struct nouveau_gpio *gpio = nouveau_gpio(event->priv);
+	const struct nouveau_gpio_impl *impl = (void *)nv_object(gpio)->oclass;
+	impl->intr_mask(gpio, type, 1 << index, 1 << index);
+}
+
+static void
+nouveau_gpio_intr(struct nouveau_subdev *subdev)
+{
+	struct nouveau_gpio *gpio = nouveau_gpio(subdev);
+	const struct nouveau_gpio_impl *impl = (void *)nv_object(gpio)->oclass;
+	u32 hi, lo, e, i;
+
+	impl->intr_stat(gpio, &hi, &lo);
+
+	for (i = 0; e = 0, (hi | lo) && i < impl->lines; i++) {
+		if (hi & (1 << i))
+			e |= NVKM_GPIO_HI;
+		if (lo & (1 << i))
+			e |= NVKM_GPIO_LO;
+		nouveau_event_trigger(gpio->events, e, i);
+	}
+}
+
+int
+_nouveau_gpio_fini(struct nouveau_object *object, bool suspend)
+{
+	const struct nouveau_gpio_impl *impl = (void *)object->oclass;
+	struct nouveau_gpio *gpio = nouveau_gpio(object);
+	u32 mask = (1 << impl->lines) - 1;
+
+	impl->intr_mask(gpio, NVKM_GPIO_TOGGLED, mask, 0);
+	impl->intr_stat(gpio, &mask, &mask);
+
+	return nouveau_subdev_fini(&gpio->base, suspend);
+}
+
+static struct dmi_system_id gpio_reset_ids[] = {
+	{
+		.ident = "Apple Macbook 10,1",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "MacBookPro10,1"),
+		}
+	},
+	{ }
+};
+
+int
+_nouveau_gpio_init(struct nouveau_object *object)
+{
+	struct nouveau_gpio *gpio = nouveau_gpio(object);
+	int ret;
+
+	ret = nouveau_subdev_init(&gpio->base);
+	if (ret)
+		return ret;
+
+	if (gpio->reset && dmi_check_system(gpio_reset_ids))
+		gpio->reset(gpio, DCB_GPIO_UNUSED);
+
+	return ret;
+}
+
 void
 _nouveau_gpio_dtor(struct nouveau_object *object)
 {
@@ -113,9 +190,10 @@ _nouveau_gpio_dtor(struct nouveau_object *object)
 int
 nouveau_gpio_create_(struct nouveau_object *parent,
 		     struct nouveau_object *engine,
-		     struct nouveau_oclass *oclass, int lines,
+		     struct nouveau_oclass *oclass,
 		     int length, void **pobject)
 {
+	const struct nouveau_gpio_impl *impl = (void *)oclass;
 	struct nouveau_gpio *gpio;
 	int ret;
 
@@ -125,34 +203,34 @@ nouveau_gpio_create_(struct nouveau_object *parent,
 	if (ret)
 		return ret;
 
-	ret = nouveau_event_create(lines, &gpio->events);
-	if (ret)
-		return ret;
-
 	gpio->find = nouveau_gpio_find;
 	gpio->set  = nouveau_gpio_set;
 	gpio->get  = nouveau_gpio_get;
+	gpio->reset = impl->reset;
+
+	ret = nouveau_event_create(2, impl->lines, &gpio->events);
+	if (ret)
+		return ret;
+
+	gpio->events->priv = gpio;
+	gpio->events->enable = nouveau_gpio_intr_enable;
+	gpio->events->disable = nouveau_gpio_intr_disable;
+	nv_subdev(gpio)->intr = nouveau_gpio_intr;
 	return 0;
 }
 
-static struct dmi_system_id gpio_reset_ids[] = {
-	{
-		.ident = "Apple Macbook 10,1",
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."),
-			DMI_MATCH(DMI_PRODUCT_NAME, "MacBookPro10,1"),
-		}
-	},
-	{ }
-};
-
 int
-nouveau_gpio_init(struct nouveau_gpio *gpio)
+_nouveau_gpio_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+		   struct nouveau_oclass *oclass, void *data, u32 size,
+		   struct nouveau_object **pobject)
 {
-	int ret = nouveau_subdev_init(&gpio->base);
-	if (ret == 0 && gpio->reset) {
-		if (dmi_check_system(gpio_reset_ids))
-			gpio->reset(gpio, DCB_GPIO_UNUSED);
-	}
-	return ret;
+	struct nouveau_gpio *gpio;
+	int ret;
+
+	ret = nouveau_gpio_create(parent, engine, oclass, &gpio);
+	*pobject = nv_object(gpio);
+	if (ret)
+		return ret;
+
+	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/core/subdev/gpio/nv10.c b/drivers/gpu/drm/nouveau/core/subdev/gpio/nv10.c
index 76d5d5465ddd..27ad23eaf185 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/gpio/nv10.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/gpio/nv10.c
@@ -26,10 +26,6 @@
 
 #include "priv.h"
 
-struct nv10_gpio_priv {
-	struct nouveau_gpio base;
-};
-
 static int
 nv10_gpio_sense(struct nouveau_gpio *gpio, int line)
 {
@@ -83,95 +79,38 @@ nv10_gpio_drive(struct nouveau_gpio *gpio, int line, int dir, int out)
 }
 
 static void
-nv10_gpio_intr(struct nouveau_subdev *subdev)
-{
-	struct nv10_gpio_priv *priv = (void *)subdev;
-	u32 intr = nv_rd32(priv, 0x001104);
-	u32 hi = (intr & 0x0000ffff) >> 0;
-	u32 lo = (intr & 0xffff0000) >> 16;
-	int i;
-
-	for (i = 0; (hi | lo) && i < 32; i++) {
-		if ((hi | lo) & (1 << i))
-			nouveau_event_trigger(priv->base.events, i);
-	}
-
-	nv_wr32(priv, 0x001104, intr);
-}
-
-static void
-nv10_gpio_intr_enable(struct nouveau_event *event, int line)
-{
-	nv_wr32(event->priv, 0x001104, 0x00010001 << line);
-	nv_mask(event->priv, 0x001144, 0x00010001 << line, 0x00010001 << line);
-}
-
-static void
-nv10_gpio_intr_disable(struct nouveau_event *event, int line)
-{
-	nv_wr32(event->priv, 0x001104, 0x00010001 << line);
-	nv_mask(event->priv, 0x001144, 0x00010001 << line, 0x00000000);
-}
-
-static int
-nv10_gpio_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	       struct nouveau_oclass *oclass, void *data, u32 size,
-	       struct nouveau_object **pobject)
+nv10_gpio_intr_stat(struct nouveau_gpio *gpio, u32 *hi, u32 *lo)
 {
-	struct nv10_gpio_priv *priv;
-	int ret;
-
-	ret = nouveau_gpio_create(parent, engine, oclass, 16, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	priv->base.drive = nv10_gpio_drive;
-	priv->base.sense = nv10_gpio_sense;
-	priv->base.events->priv = priv;
-	priv->base.events->enable = nv10_gpio_intr_enable;
-	priv->base.events->disable = nv10_gpio_intr_disable;
-	nv_subdev(priv)->intr = nv10_gpio_intr;
-	return 0;
+	u32 intr = nv_rd32(gpio, 0x001104);
+	u32 stat = nv_rd32(gpio, 0x001144) & intr;
+	*lo = (stat & 0xffff0000) >> 16;
+	*hi = (stat & 0x0000ffff);
+	nv_wr32(gpio, 0x001104, intr);
 }
 
 static void
-nv10_gpio_dtor(struct nouveau_object *object)
-{
-	struct nv10_gpio_priv *priv = (void *)object;
-	nouveau_gpio_destroy(&priv->base);
-}
-
-static int
-nv10_gpio_init(struct nouveau_object *object)
-{
-	struct nv10_gpio_priv *priv = (void *)object;
-	int ret;
-
-	ret = nouveau_gpio_init(&priv->base);
-	if (ret)
-		return ret;
-
-	nv_wr32(priv, 0x001144, 0x00000000);
-	nv_wr32(priv, 0x001104, 0xffffffff);
-	return 0;
-}
-
-static int
-nv10_gpio_fini(struct nouveau_object *object, bool suspend)
+nv10_gpio_intr_mask(struct nouveau_gpio *gpio, u32 type, u32 mask, u32 data)
 {
-	struct nv10_gpio_priv *priv = (void *)object;
-	nv_wr32(priv, 0x001144, 0x00000000);
-	return nouveau_gpio_fini(&priv->base, suspend);
+	u32 inte = nv_rd32(gpio, 0x001144);
+	if (type & NVKM_GPIO_LO)
+		inte = (inte & ~(mask << 16)) | (data << 16);
+	if (type & NVKM_GPIO_HI)
+		inte = (inte & ~mask) | data;
+	nv_wr32(gpio, 0x001144, inte);
 }
 
-struct nouveau_oclass
-nv10_gpio_oclass = {
-	.handle = NV_SUBDEV(GPIO, 0x10),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv10_gpio_ctor,
-		.dtor = nv10_gpio_dtor,
-		.init = nv10_gpio_init,
-		.fini = nv10_gpio_fini,
+struct nouveau_oclass *
+nv10_gpio_oclass = &(struct nouveau_gpio_impl) {
+	.base.handle = NV_SUBDEV(GPIO, 0x10),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_gpio_ctor,
+		.dtor = _nouveau_gpio_dtor,
+		.init = _nouveau_gpio_init,
+		.fini = _nouveau_gpio_fini,
 	},
-};
+	.lines = 16,
+	.intr_stat = nv10_gpio_intr_stat,
+	.intr_mask = nv10_gpio_intr_mask,
+	.drive = nv10_gpio_drive,
+	.sense = nv10_gpio_sense,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/gpio/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/gpio/nv50.c
index 2ef774731629..1864fa98e6b1 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/gpio/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/gpio/nv50.c
@@ -24,15 +24,10 @@
 
 #include "priv.h"
 
-struct nv50_gpio_priv {
-	struct nouveau_gpio base;
-};
-
-static void
+void
 nv50_gpio_reset(struct nouveau_gpio *gpio, u8 match)
 {
 	struct nouveau_bios *bios = nouveau_bios(gpio);
-	struct nv50_gpio_priv *priv = (void *)gpio;
 	u8 ver, len;
 	u16 entry;
 	int ent = -1;
@@ -55,7 +50,7 @@ nv50_gpio_reset(struct nouveau_gpio *gpio, u8 match)
 
 		gpio->set(gpio, 0, func, line, defs);
 
-		nv_mask(priv, reg, 0x00010001 << lsh, val << lsh);
+		nv_mask(gpio, reg, 0x00010001 << lsh, val << lsh);
 	}
 }
 
@@ -72,7 +67,7 @@ nv50_gpio_location(int line, u32 *reg, u32 *shift)
 	return 0;
 }
 
-static int
+int
 nv50_gpio_drive(struct nouveau_gpio *gpio, int line, int dir, int out)
 {
 	u32 reg, shift;
@@ -84,7 +79,7 @@ nv50_gpio_drive(struct nouveau_gpio *gpio, int line, int dir, int out)
 	return 0;
 }
 
-static int
+int
 nv50_gpio_sense(struct nouveau_gpio *gpio, int line)
 {
 	u32 reg, shift;
@@ -95,119 +90,40 @@ nv50_gpio_sense(struct nouveau_gpio *gpio, int line)
 	return !!(nv_rd32(gpio, reg) & (4 << shift));
 }
 
-void
-nv50_gpio_intr(struct nouveau_subdev *subdev)
-{
-	struct nv50_gpio_priv *priv = (void *)subdev;
-	u32 intr0, intr1 = 0;
-	u32 hi, lo;
-	int i;
-
-	intr0 = nv_rd32(priv, 0xe054) & nv_rd32(priv, 0xe050);
-	if (nv_device(priv)->chipset > 0x92)
-		intr1 = nv_rd32(priv, 0xe074) & nv_rd32(priv, 0xe070);
-
-	hi = (intr0 & 0x0000ffff) | (intr1 << 16);
-	lo = (intr0 >> 16) | (intr1 & 0xffff0000);
-
-	for (i = 0; (hi | lo) && i < 32; i++) {
-		if ((hi | lo) & (1 << i))
-			nouveau_event_trigger(priv->base.events, i);
-	}
-
-	nv_wr32(priv, 0xe054, intr0);
-	if (nv_device(priv)->chipset > 0x92)
-		nv_wr32(priv, 0xe074, intr1);
-}
-
-void
-nv50_gpio_intr_enable(struct nouveau_event *event, int line)
-{
-	const u32 addr = line < 16 ? 0xe050 : 0xe070;
-	const u32 mask = 0x00010001 << (line & 0xf);
-	nv_wr32(event->priv, addr + 0x04, mask);
-	nv_mask(event->priv, addr + 0x00, mask, mask);
-}
-
-void
-nv50_gpio_intr_disable(struct nouveau_event *event, int line)
-{
-	const u32 addr = line < 16 ? 0xe050 : 0xe070;
-	const u32 mask = 0x00010001 << (line & 0xf);
-	nv_wr32(event->priv, addr + 0x04, mask);
-	nv_mask(event->priv, addr + 0x00, mask, 0x00000000);
-}
-
-static int
-nv50_gpio_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	       struct nouveau_oclass *oclass, void *data, u32 size,
-	       struct nouveau_object **pobject)
-{
-	struct nv50_gpio_priv *priv;
-	int ret;
-
-	ret = nouveau_gpio_create(parent, engine, oclass,
-				  nv_device(parent)->chipset > 0x92 ? 32 : 16,
-				  &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	priv->base.reset = nv50_gpio_reset;
-	priv->base.drive = nv50_gpio_drive;
-	priv->base.sense = nv50_gpio_sense;
-	priv->base.events->priv = priv;
-	priv->base.events->enable = nv50_gpio_intr_enable;
-	priv->base.events->disable = nv50_gpio_intr_disable;
-	nv_subdev(priv)->intr = nv50_gpio_intr;
-	return 0;
-}
-
-void
-nv50_gpio_dtor(struct nouveau_object *object)
-{
-	struct nv50_gpio_priv *priv = (void *)object;
-	nouveau_gpio_destroy(&priv->base);
-}
-
-int
-nv50_gpio_init(struct nouveau_object *object)
+static void
+nv50_gpio_intr_stat(struct nouveau_gpio *gpio, u32 *hi, u32 *lo)
 {
-	struct nv50_gpio_priv *priv = (void *)object;
-	int ret;
-
-	ret = nouveau_gpio_init(&priv->base);
-	if (ret)
-		return ret;
-
-	/* disable, and ack any pending gpio interrupts */
-	nv_wr32(priv, 0xe050, 0x00000000);
-	nv_wr32(priv, 0xe054, 0xffffffff);
-	if (nv_device(priv)->chipset > 0x92) {
-		nv_wr32(priv, 0xe070, 0x00000000);
-		nv_wr32(priv, 0xe074, 0xffffffff);
-	}
-
-	return 0;
+	u32 intr = nv_rd32(gpio, 0x00e054);
+	u32 stat = nv_rd32(gpio, 0x00e050) & intr;
+	*lo = (stat & 0xffff0000) >> 16;
+	*hi = (stat & 0x0000ffff);
+	nv_wr32(gpio, 0x00e054, intr);
 }
 
-int
-nv50_gpio_fini(struct nouveau_object *object, bool suspend)
+static void
+nv50_gpio_intr_mask(struct nouveau_gpio *gpio, u32 type, u32 mask, u32 data)
 {
-	struct nv50_gpio_priv *priv = (void *)object;
-	nv_wr32(priv, 0xe050, 0x00000000);
-	if (nv_device(priv)->chipset > 0x92)
-		nv_wr32(priv, 0xe070, 0x00000000);
-	return nouveau_gpio_fini(&priv->base, suspend);
+	u32 inte = nv_rd32(gpio, 0x00e050);
+	if (type & NVKM_GPIO_LO)
+		inte = (inte & ~(mask << 16)) | (data << 16);
+	if (type & NVKM_GPIO_HI)
+		inte = (inte & ~mask) | data;
+	nv_wr32(gpio, 0x00e050, inte);
 }
 
-struct nouveau_oclass
-nv50_gpio_oclass = {
-	.handle = NV_SUBDEV(GPIO, 0x50),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv50_gpio_ctor,
-		.dtor = nv50_gpio_dtor,
-		.init = nv50_gpio_init,
-		.fini = nv50_gpio_fini,
+struct nouveau_oclass *
+nv50_gpio_oclass = &(struct nouveau_gpio_impl) {
+	.base.handle = NV_SUBDEV(GPIO, 0x50),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_gpio_ctor,
+		.dtor = _nouveau_gpio_dtor,
+		.init = _nouveau_gpio_init,
+		.fini = _nouveau_gpio_fini,
 	},
-};
+	.lines = 16,
+	.intr_stat = nv50_gpio_intr_stat,
+	.intr_mask = nv50_gpio_intr_mask,
+	.drive = nv50_gpio_drive,
+	.sense = nv50_gpio_sense,
+	.reset = nv50_gpio_reset,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/gpio/nv92.c b/drivers/gpu/drm/nouveau/core/subdev/gpio/nv92.c
new file mode 100644
index 000000000000..252083d376f5
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/gpio/nv92.c
@@ -0,0 +1,74 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "priv.h"
+
+void
+nv92_gpio_intr_stat(struct nouveau_gpio *gpio, u32 *hi, u32 *lo)
+{
+	u32 intr0 = nv_rd32(gpio, 0x00e054);
+	u32 intr1 = nv_rd32(gpio, 0x00e074);
+	u32 stat0 = nv_rd32(gpio, 0x00e050) & intr0;
+	u32 stat1 = nv_rd32(gpio, 0x00e070) & intr1;
+	*lo = (stat1 & 0xffff0000) | (stat0 >> 16);
+	*hi = (stat1 << 16) | (stat0 & 0x0000ffff);
+	nv_wr32(gpio, 0x00e054, intr0);
+	nv_wr32(gpio, 0x00e074, intr1);
+}
+
+void
+nv92_gpio_intr_mask(struct nouveau_gpio *gpio, u32 type, u32 mask, u32 data)
+{
+	u32 inte0 = nv_rd32(gpio, 0x00e050);
+	u32 inte1 = nv_rd32(gpio, 0x00e070);
+	if (type & NVKM_GPIO_LO)
+		inte0 = (inte0 & ~(mask << 16)) | (data << 16);
+	if (type & NVKM_GPIO_HI)
+		inte0 = (inte0 & ~(mask & 0xffff)) | (data & 0xffff);
+	mask >>= 16;
+	data >>= 16;
+	if (type & NVKM_GPIO_LO)
+		inte1 = (inte1 & ~(mask << 16)) | (data << 16);
+	if (type & NVKM_GPIO_HI)
+		inte1 = (inte1 & ~mask) | data;
+	nv_wr32(gpio, 0x00e050, inte0);
+	nv_wr32(gpio, 0x00e070, inte1);
+}
+
+struct nouveau_oclass *
+nv92_gpio_oclass = &(struct nouveau_gpio_impl) {
+	.base.handle = NV_SUBDEV(GPIO, 0x92),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_gpio_ctor,
+		.dtor = _nouveau_gpio_dtor,
+		.init = _nouveau_gpio_init,
+		.fini = _nouveau_gpio_fini,
+	},
+	.lines = 32,
+	.intr_stat = nv92_gpio_intr_stat,
+	.intr_mask = nv92_gpio_intr_mask,
+	.drive = nv50_gpio_drive,
+	.sense = nv50_gpio_sense,
+	.reset = nv50_gpio_reset,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/gpio/nvd0.c b/drivers/gpu/drm/nouveau/core/subdev/gpio/nvd0.c
index 010431e3acec..a4682b0956ad 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/gpio/nvd0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/gpio/nvd0.c
@@ -24,15 +24,10 @@
 
 #include "priv.h"
 
-struct nvd0_gpio_priv {
-	struct nouveau_gpio base;
-};
-
 void
 nvd0_gpio_reset(struct nouveau_gpio *gpio, u8 match)
 {
 	struct nouveau_bios *bios = nouveau_bios(gpio);
-	struct nvd0_gpio_priv *priv = (void *)gpio;
 	u8 ver, len;
 	u16 entry;
 	int ent = -1;
@@ -51,9 +46,9 @@ nvd0_gpio_reset(struct nouveau_gpio *gpio, u8 match)
 
 		gpio->set(gpio, 0, func, line, defs);
 
-		nv_mask(priv, 0x00d610 + (line * 4), 0xff, unk0);
+		nv_mask(gpio, 0x00d610 + (line * 4), 0xff, unk0);
 		if (unk1--)
-			nv_mask(priv, 0x00d740 + (unk1 * 4), 0xff, line);
+			nv_mask(gpio, 0x00d740 + (unk1 * 4), 0xff, line);
 	}
 }
 
@@ -72,36 +67,19 @@ nvd0_gpio_sense(struct nouveau_gpio *gpio, int line)
 	return !!(nv_rd32(gpio, 0x00d610 + (line * 4)) & 0x00004000);
 }
 
-static int
-nvd0_gpio_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	       struct nouveau_oclass *oclass, void *data, u32 size,
-	       struct nouveau_object **pobject)
-{
-	struct nvd0_gpio_priv *priv;
-	int ret;
-
-	ret = nouveau_gpio_create(parent, engine, oclass, 32, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	priv->base.reset = nvd0_gpio_reset;
-	priv->base.drive = nvd0_gpio_drive;
-	priv->base.sense = nvd0_gpio_sense;
-	priv->base.events->priv = priv;
-	priv->base.events->enable = nv50_gpio_intr_enable;
-	priv->base.events->disable = nv50_gpio_intr_disable;
-	nv_subdev(priv)->intr = nv50_gpio_intr;
-	return 0;
-}
-
-struct nouveau_oclass
-nvd0_gpio_oclass = {
-	.handle = NV_SUBDEV(GPIO, 0xd0),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nvd0_gpio_ctor,
-		.dtor = nv50_gpio_dtor,
-		.init = nv50_gpio_init,
-		.fini = nv50_gpio_fini,
+struct nouveau_oclass *
+nvd0_gpio_oclass = &(struct nouveau_gpio_impl) {
+	.base.handle = NV_SUBDEV(GPIO, 0xd0),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_gpio_ctor,
+		.dtor = _nouveau_gpio_dtor,
+		.init = _nouveau_gpio_init,
+		.fini = _nouveau_gpio_fini,
 	},
-};
+	.lines = 32,
+	.intr_stat = nv92_gpio_intr_stat,
+	.intr_mask = nv92_gpio_intr_mask,
+	.drive = nvd0_gpio_drive,
+	.sense = nvd0_gpio_sense,
+	.reset = nvd0_gpio_reset,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/gpio/nve0.c b/drivers/gpu/drm/nouveau/core/subdev/gpio/nve0.c
index 16b8c5bf5efa..e1145b48c76c 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/gpio/nve0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/gpio/nve0.c
@@ -24,108 +24,51 @@
 
 #include "priv.h"
 
-struct nve0_gpio_priv {
-	struct nouveau_gpio base;
-};
-
-void
-nve0_gpio_intr(struct nouveau_subdev *subdev)
+static void
+nve0_gpio_intr_stat(struct nouveau_gpio *gpio, u32 *hi, u32 *lo)
 {
-	struct nve0_gpio_priv *priv = (void *)subdev;
-	u32 intr0 = nv_rd32(priv, 0xdc00) & nv_rd32(priv, 0xdc08);
-	u32 intr1 = nv_rd32(priv, 0xdc80) & nv_rd32(priv, 0xdc88);
-	u32 hi = (intr0 & 0x0000ffff) | (intr1 << 16);
-	u32 lo = (intr0 >> 16) | (intr1 & 0xffff0000);
-	int i;
-
-	for (i = 0; (hi | lo) && i < 32; i++) {
-		if ((hi | lo) & (1 << i))
-			nouveau_event_trigger(priv->base.events, i);
-	}
-
-	nv_wr32(priv, 0xdc00, intr0);
-	nv_wr32(priv, 0xdc88, intr1);
+	u32 intr0 = nv_rd32(gpio, 0x00dc00);
+	u32 intr1 = nv_rd32(gpio, 0x00dc80);
+	u32 stat0 = nv_rd32(gpio, 0x00dc08) & intr0;
+	u32 stat1 = nv_rd32(gpio, 0x00dc88) & intr1;
+	*lo = (stat1 & 0xffff0000) | (stat0 >> 16);
+	*hi = (stat1 << 16) | (stat0 & 0x0000ffff);
+	nv_wr32(gpio, 0x00dc00, intr0);
+	nv_wr32(gpio, 0x00dc80, intr1);
 }
 
 void
-nve0_gpio_intr_enable(struct nouveau_event *event, int line)
+nve0_gpio_intr_mask(struct nouveau_gpio *gpio, u32 type, u32 mask, u32 data)
 {
-	const u32 addr = line < 16 ? 0xdc00 : 0xdc80;
-	const u32 mask = 0x00010001 << (line & 0xf);
-	nv_wr32(event->priv, addr + 0x08, mask);
-	nv_mask(event->priv, addr + 0x00, mask, mask);
-}
-
-void
-nve0_gpio_intr_disable(struct nouveau_event *event, int line)
-{
-	const u32 addr = line < 16 ? 0xdc00 : 0xdc80;
-	const u32 mask = 0x00010001 << (line & 0xf);
-	nv_wr32(event->priv, addr + 0x08, mask);
-	nv_mask(event->priv, addr + 0x00, mask, 0x00000000);
-}
-
-int
-nve0_gpio_fini(struct nouveau_object *object, bool suspend)
-{
-	struct nve0_gpio_priv *priv = (void *)object;
-	nv_wr32(priv, 0xdc08, 0x00000000);
-	nv_wr32(priv, 0xdc88, 0x00000000);
-	return nouveau_gpio_fini(&priv->base, suspend);
-}
-
-int
-nve0_gpio_init(struct nouveau_object *object)
-{
-	struct nve0_gpio_priv *priv = (void *)object;
-	int ret;
-
-	ret = nouveau_gpio_init(&priv->base);
-	if (ret)
-		return ret;
-
-	nv_wr32(priv, 0xdc00, 0xffffffff);
-	nv_wr32(priv, 0xdc80, 0xffffffff);
-	return 0;
-}
-
-void
-nve0_gpio_dtor(struct nouveau_object *object)
-{
-	struct nve0_gpio_priv *priv = (void *)object;
-	nouveau_gpio_destroy(&priv->base);
-}
-
-static int
-nve0_gpio_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	       struct nouveau_oclass *oclass, void *data, u32 size,
-	       struct nouveau_object **pobject)
-{
-	struct nve0_gpio_priv *priv;
-	int ret;
-
-	ret = nouveau_gpio_create(parent, engine, oclass, 32, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	priv->base.reset = nvd0_gpio_reset;
-	priv->base.drive = nvd0_gpio_drive;
-	priv->base.sense = nvd0_gpio_sense;
-	priv->base.events->priv = priv;
-	priv->base.events->enable = nve0_gpio_intr_enable;
-	priv->base.events->disable = nve0_gpio_intr_disable;
-	nv_subdev(priv)->intr = nve0_gpio_intr;
-	return 0;
+	u32 inte0 = nv_rd32(gpio, 0x00dc08);
+	u32 inte1 = nv_rd32(gpio, 0x00dc88);
+	if (type & NVKM_GPIO_LO)
+		inte0 = (inte0 & ~(mask << 16)) | (data << 16);
+	if (type & NVKM_GPIO_HI)
+		inte0 = (inte0 & ~(mask & 0xffff)) | (data & 0xffff);
+	mask >>= 16;
+	data >>= 16;
+	if (type & NVKM_GPIO_LO)
+		inte1 = (inte1 & ~(mask << 16)) | (data << 16);
+	if (type & NVKM_GPIO_HI)
+		inte1 = (inte1 & ~mask) | data;
+	nv_wr32(gpio, 0x00dc08, inte0);
+	nv_wr32(gpio, 0x00dc88, inte1);
 }
 
-struct nouveau_oclass
-nve0_gpio_oclass = {
-	.handle = NV_SUBDEV(GPIO, 0xe0),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nve0_gpio_ctor,
-		.dtor = nv50_gpio_dtor,
-		.init = nve0_gpio_init,
-		.fini = nve0_gpio_fini,
+struct nouveau_oclass *
+nve0_gpio_oclass = &(struct nouveau_gpio_impl) {
+	.base.handle = NV_SUBDEV(GPIO, 0xe0),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_gpio_ctor,
+		.dtor = _nouveau_gpio_dtor,
+		.init = _nouveau_gpio_init,
+		.fini = _nouveau_gpio_fini,
 	},
-};
+	.lines = 32,
+	.intr_stat = nve0_gpio_intr_stat,
+	.intr_mask = nve0_gpio_intr_mask,
+	.drive = nvd0_gpio_drive,
+	.sense = nvd0_gpio_sense,
+	.reset = nvd0_gpio_reset,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/gpio/priv.h b/drivers/gpu/drm/nouveau/core/subdev/gpio/priv.h
index 2ee1c895c782..e1724dfc86ae 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/gpio/priv.h
+++ b/drivers/gpu/drm/nouveau/core/subdev/gpio/priv.h
@@ -3,15 +3,65 @@
 
 #include <subdev/gpio.h>
 
-void nv50_gpio_dtor(struct nouveau_object *);
-int  nv50_gpio_init(struct nouveau_object *);
-int  nv50_gpio_fini(struct nouveau_object *, bool);
-void nv50_gpio_intr(struct nouveau_subdev *);
-void nv50_gpio_intr_enable(struct nouveau_event *, int line);
-void nv50_gpio_intr_disable(struct nouveau_event *, int line);
+#define nouveau_gpio_create(p,e,o,d)                                           \
+	nouveau_gpio_create_((p), (e), (o), sizeof(**d), (void **)d)
+#define nouveau_gpio_destroy(p) ({                                             \
+	struct nouveau_gpio *gpio = (p);                                       \
+	_nouveau_gpio_dtor(nv_object(gpio));                                   \
+})
+#define nouveau_gpio_init(p) ({                                                \
+	struct nouveau_gpio *gpio = (p);                                       \
+	_nouveau_gpio_init(nv_object(gpio));                                   \
+})
+#define nouveau_gpio_fini(p,s) ({                                              \
+	struct nouveau_gpio *gpio = (p);                                       \
+	_nouveau_gpio_fini(nv_object(gpio), (s));                              \
+})
+
+int  nouveau_gpio_create_(struct nouveau_object *, struct nouveau_object *,
+			  struct nouveau_oclass *, int, void **);
+int  _nouveau_gpio_ctor(struct nouveau_object *, struct nouveau_object *,
+			struct nouveau_oclass *, void *, u32,
+			struct nouveau_object **);
+void _nouveau_gpio_dtor(struct nouveau_object *);
+int  _nouveau_gpio_init(struct nouveau_object *);
+int  _nouveau_gpio_fini(struct nouveau_object *, bool);
+
+struct nouveau_gpio_impl {
+	struct nouveau_oclass base;
+	int lines;
+
+	/* read and ack pending interrupts, returning only data
+	 * for lines that have not been masked off, while still
+	 * performing the ack for anything that was pending.
+	 */
+	void (*intr_stat)(struct nouveau_gpio *, u32 *, u32 *);
+
+	/* mask on/off interrupts for hi/lo transitions on a
+	 * given set of gpio lines
+	 */
+	void (*intr_mask)(struct nouveau_gpio *, u32, u32, u32);
+
+	/* configure gpio direction and output value */
+	int  (*drive)(struct nouveau_gpio *, int line, int dir, int out);
+
+	/* sense current state of given gpio line */
+	int  (*sense)(struct nouveau_gpio *, int line);
+
+	/*XXX*/
+	void (*reset)(struct nouveau_gpio *, u8);
+};
+
+void nv50_gpio_reset(struct nouveau_gpio *, u8);
+int  nv50_gpio_drive(struct nouveau_gpio *, int, int, int);
+int  nv50_gpio_sense(struct nouveau_gpio *, int);
+
+void nv92_gpio_intr_stat(struct nouveau_gpio *, u32 *, u32 *);
+void nv92_gpio_intr_mask(struct nouveau_gpio *, u32, u32, u32);
 
 void nvd0_gpio_reset(struct nouveau_gpio *, u8);
 int  nvd0_gpio_drive(struct nouveau_gpio *, int, int, int);
 int  nvd0_gpio_sense(struct nouveau_gpio *, int);
 
+
 #endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/anx9805.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/anx9805.c
index 4b195ac4da66..2c2731a6cf91 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/anx9805.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/anx9805.c
@@ -22,7 +22,7 @@
  * Authors: Ben Skeggs <bskeggs@redhat.com>
  */
 
-#include <subdev/i2c.h>
+#include "port.h"
 
 struct anx9805_i2c_port {
 	struct nouveau_i2c_port base;
@@ -37,6 +37,8 @@ anx9805_train(struct nouveau_i2c_port *port, int link_nr, int link_bw, bool enh)
 	struct nouveau_i2c_port *mast = (void *)nv_object(chan)->parent;
 	u8 tmp, i;
 
+	DBG("ANX9805 train %d 0x%02x %d\n", link_nr, link_bw, enh);
+
 	nv_wri2cr(mast, chan->addr, 0xa0, link_bw);
 	nv_wri2cr(mast, chan->addr, 0xa1, link_nr | (enh ? 0x80 : 0x00));
 	nv_wri2cr(mast, chan->addr, 0xa2, 0x01);
@@ -60,21 +62,29 @@ anx9805_train(struct nouveau_i2c_port *port, int link_nr, int link_bw, bool enh)
 }
 
 static int
-anx9805_aux(struct nouveau_i2c_port *port, u8 type, u32 addr, u8 *data, u8 size)
+anx9805_aux(struct nouveau_i2c_port *port, bool retry,
+	    u8 type, u32 addr, u8 *data, u8 size)
 {
 	struct anx9805_i2c_port *chan = (void *)port;
 	struct nouveau_i2c_port *mast = (void *)nv_object(chan)->parent;
 	int i, ret = -ETIMEDOUT;
+	u8 buf[16] = {};
 	u8 tmp;
 
+	DBG("%02x %05x %d\n", type, addr, size);
+
 	tmp = nv_rdi2cr(mast, chan->ctrl, 0x07) & ~0x04;
 	nv_wri2cr(mast, chan->ctrl, 0x07, tmp | 0x04);
 	nv_wri2cr(mast, chan->ctrl, 0x07, tmp);
 	nv_wri2cr(mast, chan->ctrl, 0xf7, 0x01);
 
 	nv_wri2cr(mast, chan->addr, 0xe4, 0x80);
-	for (i = 0; !(type & 1) && i < size; i++)
-		nv_wri2cr(mast, chan->addr, 0xf0 + i, data[i]);
+	if (!(type & 1)) {
+		memcpy(buf, data, size);
+		DBG("%16ph", buf);
+		for (i = 0; i < size; i++)
+			nv_wri2cr(mast, chan->addr, 0xf0 + i, buf[i]);
+	}
 	nv_wri2cr(mast, chan->addr, 0xe5, ((size - 1) << 4) | type);
 	nv_wri2cr(mast, chan->addr, 0xe6, (addr & 0x000ff) >>  0);
 	nv_wri2cr(mast, chan->addr, 0xe7, (addr & 0x0ff00) >>  8);
@@ -93,8 +103,13 @@ anx9805_aux(struct nouveau_i2c_port *port, u8 type, u32 addr, u8 *data, u8 size)
 		goto done;
 	}
 
-	for (i = 0; (type & 1) && i < size; i++)
-		data[i] = nv_rdi2cr(mast, chan->addr, 0xf0 + i);
+	if (type & 1) {
+		for (i = 0; i < size; i++)
+			buf[i] = nv_rdi2cr(mast, chan->addr, 0xf0 + i);
+		DBG("%16ph", buf);
+		memcpy(data, buf, size);
+	}
+
 	ret = 0;
 done:
 	nv_wri2cr(mast, chan->ctrl, 0xf7, 0x01);
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/aux.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/aux.c
index 5de074ad170b..02eb42be2e9e 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/aux.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/aux.c
@@ -22,15 +22,19 @@
  * Authors: Ben Skeggs
  */
 
-#include <subdev/i2c.h>
+#include "priv.h"
 
 int
 nv_rdaux(struct nouveau_i2c_port *port, u32 addr, u8 *data, u8 size)
 {
+	struct nouveau_i2c *i2c = nouveau_i2c(port);
 	if (port->func->aux) {
-		if (port->func->acquire)
-			port->func->acquire(port);
-		return port->func->aux(port, 9, addr, data, size);
+		int ret = i2c->acquire(port, 0);
+		if (ret == 0) {
+			ret = port->func->aux(port, true, 9, addr, data, size);
+			i2c->release(port);
+		}
+		return ret;
 	}
 	return -ENODEV;
 }
@@ -38,10 +42,14 @@ nv_rdaux(struct nouveau_i2c_port *port, u32 addr, u8 *data, u8 size)
 int
 nv_wraux(struct nouveau_i2c_port *port, u32 addr, u8 *data, u8 size)
 {
+	struct nouveau_i2c *i2c = nouveau_i2c(port);
 	if (port->func->aux) {
-		if (port->func->acquire)
-			port->func->acquire(port);
-		return port->func->aux(port, 8, addr, data, size);
+		int ret = i2c->acquire(port, 0);
+		if (ret == 0) {
+			ret = port->func->aux(port, true, 8, addr, data, size);
+			i2c->release(port);
+		}
+		return ret;
 	}
 	return -ENODEV;
 }
@@ -50,13 +58,16 @@ static int
 aux_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs, int num)
 {
 	struct nouveau_i2c_port *port = adap->algo_data;
+	struct nouveau_i2c *i2c = nouveau_i2c(port);
 	struct i2c_msg *msg = msgs;
 	int ret, mcnt = num;
 
 	if (!port->func->aux)
 		return -ENODEV;
-	if ( port->func->acquire)
-		port->func->acquire(port);
+
+	ret = i2c->acquire(port, 0);
+	if (ret)
+		return ret;
 
 	while (mcnt--) {
 		u8 remaining = msg->len;
@@ -74,9 +85,11 @@ aux_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs, int num)
 			if (mcnt || remaining > 16)
 				cmd |= 4; /* MOT */
 
-			ret = port->func->aux(port, cmd, msg->addr, ptr, cnt);
-			if (ret < 0)
+			ret = port->func->aux(port, true, cmd, msg->addr, ptr, cnt);
+			if (ret < 0) {
+				i2c->release(port);
 				return ret;
+			}
 
 			ptr += cnt;
 			remaining -= cnt;
@@ -85,6 +98,7 @@ aux_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs, int num)
 		msg++;
 	}
 
+	i2c->release(port);
 	return num;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/base.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/base.c
index 378e05b88e6f..09ba2cc851cf 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/base.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/base.c
@@ -23,13 +23,16 @@
  */
 
 #include <core/option.h>
+#include <core/event.h>
 
 #include <subdev/bios.h>
 #include <subdev/bios/dcb.h>
 #include <subdev/bios/i2c.h>
-#include <subdev/i2c.h>
 #include <subdev/vga.h>
 
+#include "priv.h"
+#include "pad.h"
+
 /******************************************************************************
  * interface to linux i2c bit-banging algorithm
  *****************************************************************************/
@@ -45,9 +48,15 @@ nouveau_i2c_pre_xfer(struct i2c_adapter *adap)
 {
 	struct i2c_algo_bit_data *bit = adap->algo_data;
 	struct nouveau_i2c_port *port = bit->data;
-	if (port->func->acquire)
-		port->func->acquire(port);
-	return 0;
+	return nouveau_i2c(port)->acquire(port, bit->timeout);
+}
+
+static void
+nouveau_i2c_post_xfer(struct i2c_adapter *adap)
+{
+	struct i2c_algo_bit_data *bit = adap->algo_data;
+	struct nouveau_i2c_port *port = bit->data;
+	return nouveau_i2c(port)->release(port);
 }
 
 static void
@@ -82,6 +91,15 @@ nouveau_i2c_getsda(void *data)
  * base i2c "port" class implementation
  *****************************************************************************/
 
+int
+_nouveau_i2c_port_fini(struct nouveau_object *object, bool suspend)
+{
+	struct nouveau_i2c_port *port = (void *)object;
+	struct nvkm_i2c_pad *pad = nvkm_i2c_pad(port);
+	nv_ofuncs(pad)->fini(nv_object(pad), suspend);
+	return nouveau_object_fini(&port->base, suspend);
+}
+
 void
 _nouveau_i2c_port_dtor(struct nouveau_object *object)
 {
@@ -98,7 +116,7 @@ nouveau_i2c_port_create_(struct nouveau_object *parent,
 			 const struct nouveau_i2c_func *func,
 			 int size, void **pobject)
 {
-	struct nouveau_device *device = nv_device(parent);
+	struct nouveau_device *device = nv_device(engine);
 	struct nouveau_i2c *i2c = (void *)engine;
 	struct nouveau_i2c_port *port;
 	int ret;
@@ -113,8 +131,9 @@ nouveau_i2c_port_create_(struct nouveau_object *parent,
 	port->adapter.owner = THIS_MODULE;
 	port->adapter.dev.parent = nv_device_base(device);
 	port->index = index;
+	port->aux = -1;
 	port->func = func;
-	i2c_set_adapdata(&port->adapter, i2c);
+	mutex_init(&port->mutex);
 
 	if ( algo == &nouveau_i2c_bit_algo &&
 	    !nouveau_boolopt(device->cfgopt, "NvI2C", CSTMSEL)) {
@@ -128,6 +147,7 @@ nouveau_i2c_port_create_(struct nouveau_object *parent,
 		bit->timeout = usecs_to_jiffies(2200);
 		bit->data = port;
 		bit->pre_xfer = nouveau_i2c_pre_xfer;
+		bit->post_xfer = nouveau_i2c_post_xfer;
 		bit->setsda = nouveau_i2c_setsda;
 		bit->setscl = nouveau_i2c_setscl;
 		bit->getsda = nouveau_i2c_getsda;
@@ -141,7 +161,6 @@ nouveau_i2c_port_create_(struct nouveau_object *parent,
 		ret = i2c_add_adapter(&port->adapter);
 	}
 
-	/* drop port's i2c subdev refcount, i2c handles this itself */
 	if (ret == 0)
 		list_add_tail(&port->head, &i2c->ports);
 	return ret;
@@ -193,6 +212,75 @@ nouveau_i2c_find_type(struct nouveau_i2c *i2c, u16 type)
 	return NULL;
 }
 
+static void
+nouveau_i2c_release_pad(struct nouveau_i2c_port *port)
+{
+	struct nvkm_i2c_pad *pad = nvkm_i2c_pad(port);
+	struct nouveau_i2c *i2c = nouveau_i2c(port);
+
+	if (atomic_dec_and_test(&nv_object(pad)->usecount)) {
+		nv_ofuncs(pad)->fini(nv_object(pad), false);
+		wake_up_all(&i2c->wait);
+	}
+}
+
+static int
+nouveau_i2c_try_acquire_pad(struct nouveau_i2c_port *port)
+{
+	struct nvkm_i2c_pad *pad = nvkm_i2c_pad(port);
+
+	if (atomic_add_return(1, &nv_object(pad)->usecount) != 1) {
+		struct nouveau_object *owner = (void *)pad->port;
+		do {
+			if (owner == (void *)port)
+				return 0;
+			owner = owner->parent;
+		} while(owner);
+		nouveau_i2c_release_pad(port);
+		return -EBUSY;
+	}
+
+	pad->next = port;
+	nv_ofuncs(pad)->init(nv_object(pad));
+	return 0;
+}
+
+static int
+nouveau_i2c_acquire_pad(struct nouveau_i2c_port *port, unsigned long timeout)
+{
+	struct nouveau_i2c *i2c = nouveau_i2c(port);
+
+	if (timeout) {
+		if (wait_event_timeout(i2c->wait,
+				       nouveau_i2c_try_acquire_pad(port) == 0,
+				       timeout) == 0)
+			return -EBUSY;
+	} else {
+		wait_event(i2c->wait, nouveau_i2c_try_acquire_pad(port) == 0);
+	}
+
+	return 0;
+}
+
+static void
+nouveau_i2c_release(struct nouveau_i2c_port *port)
+__releases(pad->mutex)
+{
+	nouveau_i2c(port)->release_pad(port);
+	mutex_unlock(&port->mutex);
+}
+
+static int
+nouveau_i2c_acquire(struct nouveau_i2c_port *port, unsigned long timeout)
+__acquires(pad->mutex)
+{
+	int ret;
+	mutex_lock(&port->mutex);
+	if ((ret = nouveau_i2c(port)->acquire_pad(port, timeout)))
+		mutex_unlock(&port->mutex);
+	return ret;
+}
+
 static int
 nouveau_i2c_identify(struct nouveau_i2c *i2c, int index, const char *what,
 		     struct nouveau_i2c_board_info *info,
@@ -237,11 +325,59 @@ nouveau_i2c_identify(struct nouveau_i2c *i2c, int index, const char *what,
 	return -ENODEV;
 }
 
+static void
+nouveau_i2c_intr_disable(struct nouveau_event *event, int type, int index)
+{
+	struct nouveau_i2c *i2c = nouveau_i2c(event->priv);
+	struct nouveau_i2c_port *port = i2c->find(i2c, index);
+	const struct nouveau_i2c_impl *impl = (void *)nv_object(i2c)->oclass;
+	if (port && port->aux >= 0)
+		impl->aux_mask(i2c, type, 1 << port->aux, 0);
+}
+
+static void
+nouveau_i2c_intr_enable(struct nouveau_event *event, int type, int index)
+{
+	struct nouveau_i2c *i2c = nouveau_i2c(event->priv);
+	struct nouveau_i2c_port *port = i2c->find(i2c, index);
+	const struct nouveau_i2c_impl *impl = (void *)nv_object(i2c)->oclass;
+	if (port && port->aux >= 0)
+		impl->aux_mask(i2c, type, 1 << port->aux, 1 << port->aux);
+}
+
+static void
+nouveau_i2c_intr(struct nouveau_subdev *subdev)
+{
+	struct nouveau_i2c_impl *impl = (void *)nv_oclass(subdev);
+	struct nouveau_i2c *i2c = nouveau_i2c(subdev);
+	struct nouveau_i2c_port *port;
+	u32 hi, lo, rq, tx, e;
+
+	if (impl->aux_stat) {
+		impl->aux_stat(i2c, &hi, &lo, &rq, &tx);
+		if (hi || lo || rq || tx) {
+			list_for_each_entry(port, &i2c->ports, head) {
+				if (e = 0, port->aux < 0)
+					continue;
+
+				if (hi & (1 << port->aux)) e |= NVKM_I2C_PLUG;
+				if (lo & (1 << port->aux)) e |= NVKM_I2C_UNPLUG;
+				if (rq & (1 << port->aux)) e |= NVKM_I2C_IRQ;
+				if (tx & (1 << port->aux)) e |= NVKM_I2C_DONE;
+
+				nouveau_event_trigger(i2c->ntfy, e, port->index);
+			}
+		}
+	}
+}
+
 int
 _nouveau_i2c_fini(struct nouveau_object *object, bool suspend)
 {
+	struct nouveau_i2c_impl *impl = (void *)nv_oclass(object);
 	struct nouveau_i2c *i2c = (void *)object;
 	struct nouveau_i2c_port *port;
+	u32 mask;
 	int ret;
 
 	list_for_each_entry(port, &i2c->ports, head) {
@@ -250,6 +386,11 @@ _nouveau_i2c_fini(struct nouveau_object *object, bool suspend)
 			goto fail;
 	}
 
+	if ((mask = (1 << impl->aux) - 1), impl->aux_stat) {
+		impl->aux_mask(i2c, NVKM_I2C_ANY, mask, 0);
+		impl->aux_stat(i2c, &mask, &mask, &mask, &mask);
+	}
+
 	return nouveau_subdev_fini(&i2c->base, suspend);
 fail:
 	list_for_each_entry_continue_reverse(port, &i2c->ports, head) {
@@ -290,6 +431,8 @@ _nouveau_i2c_dtor(struct nouveau_object *object)
 	struct nouveau_i2c *i2c = (void *)object;
 	struct nouveau_i2c_port *port, *temp;
 
+	nouveau_event_destroy(&i2c->ntfy);
+
 	list_for_each_entry_safe(port, temp, &i2c->ports, head) {
 		nouveau_object_ref(NULL, (struct nouveau_object **)&port);
 	}
@@ -306,14 +449,14 @@ int
 nouveau_i2c_create_(struct nouveau_object *parent,
 		    struct nouveau_object *engine,
 		    struct nouveau_oclass *oclass,
-		    struct nouveau_oclass *sclass,
 		    int length, void **pobject)
 {
+	const struct nouveau_i2c_impl *impl = (void *)oclass;
 	struct nouveau_bios *bios = nouveau_bios(parent);
 	struct nouveau_i2c *i2c;
 	struct nouveau_object *object;
 	struct dcb_i2c_entry info;
-	int ret, i, j, index = -1;
+	int ret, i, j, index = -1, pad;
 	struct dcb_output outp;
 	u8  ver, hdr;
 	u32 data;
@@ -324,24 +467,48 @@ nouveau_i2c_create_(struct nouveau_object *parent,
 	if (ret)
 		return ret;
 
+	nv_subdev(i2c)->intr = nouveau_i2c_intr;
 	i2c->find = nouveau_i2c_find;
 	i2c->find_type = nouveau_i2c_find_type;
+	i2c->acquire_pad = nouveau_i2c_acquire_pad;
+	i2c->release_pad = nouveau_i2c_release_pad;
+	i2c->acquire = nouveau_i2c_acquire;
+	i2c->release = nouveau_i2c_release;
 	i2c->identify = nouveau_i2c_identify;
+	init_waitqueue_head(&i2c->wait);
 	INIT_LIST_HEAD(&i2c->ports);
 
 	while (!dcb_i2c_parse(bios, ++index, &info)) {
 		if (info.type == DCB_I2C_UNUSED)
 			continue;
 
-		oclass = sclass;
+		if (info.share != DCB_I2C_UNUSED) {
+			if (info.type == DCB_I2C_NVIO_AUX)
+				pad = info.drive;
+			else
+				pad = info.share;
+			oclass = impl->pad_s;
+		} else {
+			pad = 0x100 + info.drive;
+			oclass = impl->pad_x;
+		}
+
+		ret = nouveau_object_ctor(NULL, *pobject, oclass,
+					  NULL, pad, &parent);
+		if (ret < 0)
+			continue;
+
+		oclass = impl->sclass;
 		do {
 			ret = -EINVAL;
 			if (oclass->handle == info.type) {
-				ret = nouveau_object_ctor(*pobject, *pobject,
+				ret = nouveau_object_ctor(parent, *pobject,
 							  oclass, &info,
 							  index, &object);
 			}
 		} while (ret && (++oclass)->handle);
+
+		nouveau_object_ref(NULL, &parent);
 	}
 
 	/* in addition to the busses specified in the i2c table, there
@@ -380,5 +547,28 @@ nouveau_i2c_create_(struct nouveau_object *parent,
 		}
 	}
 
+	ret = nouveau_event_create(4, index, &i2c->ntfy);
+	if (ret)
+		return ret;
+
+	i2c->ntfy->priv = i2c;
+	i2c->ntfy->enable = nouveau_i2c_intr_enable;
+	i2c->ntfy->disable = nouveau_i2c_intr_disable;
+	return 0;
+}
+
+int
+_nouveau_i2c_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+		  struct nouveau_oclass *oclass, void *data, u32 size,
+		  struct nouveau_object **pobject)
+{
+	struct nouveau_i2c *i2c;
+	int ret;
+
+	ret = nouveau_i2c_create(parent, engine, oclass, &i2c);
+	*pobject = nv_object(i2c);
+	if (ret)
+		return ret;
+
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/bit.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/bit.c
index a6e72d3b06b5..813ffc96e864 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/bit.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/bit.c
@@ -22,7 +22,7 @@
  * Authors: Ben Skeggs
  */
 
-#include "subdev/i2c.h"
+#include "priv.h"
 
 #ifdef CONFIG_NOUVEAU_I2C_INTERNAL
 #define T_TIMEOUT  2200000
@@ -187,8 +187,9 @@ i2c_bit_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs, int num)
 	struct i2c_msg *msg = msgs;
 	int ret = 0, mcnt = num;
 
-	if (port->func->acquire)
-		port->func->acquire(port);
+	ret = nouveau_i2c(port)->acquire(port, nsecs_to_jiffies(T_TIMEOUT));
+	if (ret)
+		return ret;
 
 	while (!ret && mcnt--) {
 		u8 remaining = msg->len;
@@ -210,6 +211,7 @@ i2c_bit_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs, int num)
 	}
 
 	i2c_stop(port);
+	nouveau_i2c(port)->release(port);
 	return (ret < 0) ? ret : num;
 }
 #else
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv04.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv04.c
index 860d5d2365da..b1725bdea967 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv04.c
@@ -22,9 +22,10 @@
  * Authors: Ben Skeggs
  */
 
-#include <subdev/i2c.h>
 #include <subdev/vga.h>
 
+#include "priv.h"
+
 struct nv04_i2c_priv {
 	struct nouveau_i2c base;
 };
@@ -115,29 +116,15 @@ nv04_i2c_sclass[] = {
 	{}
 };
 
-static int
-nv04_i2c_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	      struct nouveau_oclass *oclass, void *data, u32 size,
-	      struct nouveau_object **pobject)
-{
-	struct nv04_i2c_priv *priv;
-	int ret;
-
-	ret = nouveau_i2c_create(parent, engine, oclass, nv04_i2c_sclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	return 0;
-}
-
-struct nouveau_oclass
-nv04_i2c_oclass = {
-	.handle = NV_SUBDEV(I2C, 0x04),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv04_i2c_ctor,
+struct nouveau_oclass *
+nv04_i2c_oclass = &(struct nouveau_i2c_impl) {
+	.base.handle = NV_SUBDEV(I2C, 0x04),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_i2c_ctor,
 		.dtor = _nouveau_i2c_dtor,
 		.init = _nouveau_i2c_init,
 		.fini = _nouveau_i2c_fini,
 	},
-};
+	.sclass = nv04_i2c_sclass,
+	.pad_x = &nv04_i2c_pad_oclass,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv4e.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv4e.c
index 0c2655a03bb4..f16c87ce5ba1 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv4e.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv4e.c
@@ -22,9 +22,10 @@
  * Authors: Ben Skeggs
  */
 
-#include <subdev/i2c.h>
 #include <subdev/vga.h>
 
+#include "priv.h"
+
 struct nv4e_i2c_priv {
 	struct nouveau_i2c base;
 };
@@ -107,29 +108,15 @@ nv4e_i2c_sclass[] = {
 	{}
 };
 
-static int
-nv4e_i2c_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	      struct nouveau_oclass *oclass, void *data, u32 size,
-	      struct nouveau_object **pobject)
-{
-	struct nv4e_i2c_priv *priv;
-	int ret;
-
-	ret = nouveau_i2c_create(parent, engine, oclass, nv4e_i2c_sclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	return 0;
-}
-
-struct nouveau_oclass
-nv4e_i2c_oclass = {
-	.handle = NV_SUBDEV(I2C, 0x4e),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv4e_i2c_ctor,
+struct nouveau_oclass *
+nv4e_i2c_oclass = &(struct nouveau_i2c_impl) {
+	.base.handle = NV_SUBDEV(I2C, 0x4e),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_i2c_ctor,
 		.dtor = _nouveau_i2c_dtor,
 		.init = _nouveau_i2c_init,
 		.fini = _nouveau_i2c_fini,
 	},
-};
+	.sclass = nv4e_i2c_sclass,
+	.pad_x = &nv04_i2c_pad_oclass,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.c
index a8d67a287704..7b8756d4df08 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.c
@@ -121,29 +121,15 @@ nv50_i2c_sclass[] = {
 	{}
 };
 
-static int
-nv50_i2c_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	      struct nouveau_oclass *oclass, void *data, u32 size,
-	      struct nouveau_object **pobject)
-{
-	struct nv50_i2c_priv *priv;
-	int ret;
-
-	ret = nouveau_i2c_create(parent, engine, oclass, nv50_i2c_sclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	return 0;
-}
-
-struct nouveau_oclass
-nv50_i2c_oclass = {
-	.handle = NV_SUBDEV(I2C, 0x50),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv50_i2c_ctor,
+struct nouveau_oclass *
+nv50_i2c_oclass = &(struct nouveau_i2c_impl) {
+	.base.handle = NV_SUBDEV(I2C, 0x50),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_i2c_ctor,
 		.dtor = _nouveau_i2c_dtor,
 		.init = _nouveau_i2c_init,
 		.fini = _nouveau_i2c_fini,
 	},
-};
+	.sclass = nv50_i2c_sclass,
+	.pad_x = &nv04_i2c_pad_oclass,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.h b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.h
index 4e5ba48ebf5a..5d2a77421c74 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.h
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv50.h
@@ -1,7 +1,7 @@
 #ifndef __NV50_I2C_H__
 #define __NV50_I2C_H__
 
-#include <subdev/i2c.h>
+#include "priv.h"
 
 struct nv50_i2c_priv {
 	struct nouveau_i2c base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv94.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv94.c
index df6d3e4b68be..f59c3a255462 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/nv94.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nv94.c
@@ -24,6 +24,36 @@
 
 #include "nv50.h"
 
+void
+nv94_aux_stat(struct nouveau_i2c *i2c, u32 *hi, u32 *lo, u32 *rq, u32 *tx)
+{
+	u32 intr = nv_rd32(i2c, 0x00e06c);
+	u32 stat = nv_rd32(i2c, 0x00e068) & intr, i;
+	for (i = 0, *hi = *lo = *rq = *tx = 0; i < 8; i++) {
+		if ((stat & (1 << (i * 4)))) *hi |= 1 << i;
+		if ((stat & (2 << (i * 4)))) *lo |= 1 << i;
+		if ((stat & (4 << (i * 4)))) *rq |= 1 << i;
+		if ((stat & (8 << (i * 4)))) *tx |= 1 << i;
+	}
+	nv_wr32(i2c, 0x00e06c, intr);
+}
+
+void
+nv94_aux_mask(struct nouveau_i2c *i2c, u32 type, u32 mask, u32 data)
+{
+	u32 temp = nv_rd32(i2c, 0x00e068), i;
+	for (i = 0; i < 8; i++) {
+		if (mask & (1 << i)) {
+			if (!(data & (1 << i))) {
+				temp &= ~(type << (i * 4));
+				continue;
+			}
+			temp |= type << (i * 4);
+		}
+	}
+	nv_wr32(i2c, 0x00e068, temp);
+}
+
 #define AUX_DBG(fmt, args...) nv_debug(aux, "AUXCH(%d): " fmt, ch, ##args)
 #define AUX_ERR(fmt, args...) nv_error(aux, "AUXCH(%d): " fmt, ch, ##args)
 
@@ -69,7 +99,8 @@ auxch_init(struct nouveau_i2c *aux, int ch)
 }
 
 int
-nv94_aux(struct nouveau_i2c_port *base, u8 type, u32 addr, u8 *data, u8 size)
+nv94_aux(struct nouveau_i2c_port *base, bool retry,
+	 u8 type, u32 addr, u8 *data, u8 size)
 {
 	struct nouveau_i2c *aux = nouveau_i2c(base);
 	struct nv50_i2c_port *port = (void *)base;
@@ -105,9 +136,8 @@ nv94_aux(struct nouveau_i2c_port *base, u8 type, u32 addr, u8 *data, u8 size)
 	ctrl |= size - 1;
 	nv_wr32(aux, 0x00e4e0 + (ch * 0x50), addr);
 
-	/* retry transaction a number of times on failure... */
-	ret = -EREMOTEIO;
-	for (retries = 0; retries < 32; retries++) {
+	/* (maybe) retry transaction a number of times on failure... */
+	for (retries = 0; !ret && retries < 32; retries++) {
 		/* reset, and delay a while if this is a retry */
 		nv_wr32(aux, 0x00e4e4 + (ch * 0x50), 0x80000000 | ctrl);
 		nv_wr32(aux, 0x00e4e4 + (ch * 0x50), 0x00000000 | ctrl);
@@ -123,16 +153,21 @@ nv94_aux(struct nouveau_i2c_port *base, u8 type, u32 addr, u8 *data, u8 size)
 			udelay(1);
 			if (!timeout--) {
 				AUX_ERR("tx req timeout 0x%08x\n", ctrl);
+				ret = -EIO;
 				goto out;
 			}
 		} while (ctrl & 0x00010000);
+		ret = 1;
 
 		/* read status, and check if transaction completed ok */
 		stat = nv_mask(aux, 0x00e4e8 + (ch * 0x50), 0, 0);
-		if (!(stat & 0x000f0f00)) {
-			ret = 0;
-			break;
-		}
+		if ((stat & 0x000f0000) == 0x00080000 ||
+		    (stat & 0x000f0000) == 0x00020000)
+			ret = retry ? 0 : 1;
+		if ((stat & 0x00000100))
+			ret = -ETIMEDOUT;
+		if ((stat & 0x00000e00))
+			ret = -EIO;
 
 		AUX_DBG("%02d 0x%08x 0x%08x\n", retries, ctrl, stat);
 	}
@@ -147,29 +182,11 @@ nv94_aux(struct nouveau_i2c_port *base, u8 type, u32 addr, u8 *data, u8 size)
 
 out:
 	auxch_fini(aux, ch);
-	return ret;
-}
-
-void
-nv94_i2c_acquire(struct nouveau_i2c_port *base)
-{
-	struct nv50_i2c_priv *priv = (void *)nv_object(base)->engine;
-	struct nv50_i2c_port *port = (void *)base;
-	if (port->ctrl) {
-		nv_mask(priv, port->ctrl + 0x0c, 0x00000001, 0x00000000);
-		nv_mask(priv, port->ctrl + 0x00, 0x0000f003, port->data);
-	}
-}
-
-void
-nv94_i2c_release(struct nouveau_i2c_port *base)
-{
+	return ret < 0 ? ret : (stat & 0x000f0000) >> 16;
 }
 
 static const struct nouveau_i2c_func
 nv94_i2c_func = {
-	.acquire   = nv94_i2c_acquire,
-	.release   = nv94_i2c_release,
 	.drive_scl = nv50_i2c_drive_scl,
 	.drive_sda = nv50_i2c_drive_sda,
 	.sense_scl = nv50_i2c_sense_scl,
@@ -206,8 +223,6 @@ nv94_i2c_port_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 
 static const struct nouveau_i2c_func
 nv94_aux_func = {
-	.acquire   = nv94_i2c_acquire,
-	.release   = nv94_i2c_release,
 	.aux       = nv94_aux,
 };
 
@@ -227,6 +242,7 @@ nv94_aux_port_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	if (ret)
 		return ret;
 
+	port->base.aux = info->drive;
 	port->addr = info->drive;
 	if (info->share != DCB_I2C_UNUSED) {
 		port->ctrl = 0x00e500 + (info->drive * 0x50);
@@ -257,29 +273,19 @@ nv94_i2c_sclass[] = {
 	{}
 };
 
-static int
-nv94_i2c_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	      struct nouveau_oclass *oclass, void *data, u32 size,
-	      struct nouveau_object **pobject)
-{
-	struct nv50_i2c_priv *priv;
-	int ret;
-
-	ret = nouveau_i2c_create(parent, engine, oclass, nv94_i2c_sclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	return 0;
-}
-
-struct nouveau_oclass
-nv94_i2c_oclass = {
-	.handle = NV_SUBDEV(I2C, 0x94),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nv94_i2c_ctor,
+struct nouveau_oclass *
+nv94_i2c_oclass = &(struct nouveau_i2c_impl) {
+	.base.handle = NV_SUBDEV(I2C, 0x94),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_i2c_ctor,
 		.dtor = _nouveau_i2c_dtor,
 		.init = _nouveau_i2c_init,
 		.fini = _nouveau_i2c_fini,
 	},
-};
+	.sclass = nv94_i2c_sclass,
+	.pad_x = &nv04_i2c_pad_oclass,
+	.pad_s = &nv94_i2c_pad_oclass,
+	.aux = 4,
+	.aux_stat = nv94_aux_stat,
+	.aux_mask = nv94_aux_mask,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nvd0.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/nvd0.c
index 29967d30f97c..364ddb1c5f03 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/i2c/nvd0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nvd0.c
@@ -42,8 +42,6 @@ nvd0_i2c_sense_sda(struct nouveau_i2c_port *base)
 
 static const struct nouveau_i2c_func
 nvd0_i2c_func = {
-	.acquire   = nv94_i2c_acquire,
-	.release   = nv94_i2c_release,
 	.drive_scl = nv50_i2c_drive_scl,
 	.drive_sda = nv50_i2c_drive_sda,
 	.sense_scl = nvd0_i2c_sense_scl,
@@ -75,7 +73,7 @@ nvd0_i2c_port_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	return 0;
 }
 
-static struct nouveau_oclass
+struct nouveau_oclass
 nvd0_i2c_sclass[] = {
 	{ .handle = NV_I2C_TYPE_DCBI2C(DCB_I2C_NVIO_BIT),
 	  .ofuncs = &(struct nouveau_ofuncs) {
@@ -96,29 +94,19 @@ nvd0_i2c_sclass[] = {
 	{}
 };
 
-static int
-nvd0_i2c_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
-	      struct nouveau_oclass *oclass, void *data, u32 size,
-	      struct nouveau_object **pobject)
-{
-	struct nv50_i2c_priv *priv;
-	int ret;
-
-	ret = nouveau_i2c_create(parent, engine, oclass, nvd0_i2c_sclass, &priv);
-	*pobject = nv_object(priv);
-	if (ret)
-		return ret;
-
-	return 0;
-}
-
-struct nouveau_oclass
-nvd0_i2c_oclass = {
-	.handle = NV_SUBDEV(I2C, 0xd0),
-	.ofuncs = &(struct nouveau_ofuncs) {
-		.ctor = nvd0_i2c_ctor,
+struct nouveau_oclass *
+nvd0_i2c_oclass = &(struct nouveau_i2c_impl) {
+	.base.handle = NV_SUBDEV(I2C, 0xd0),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_i2c_ctor,
 		.dtor = _nouveau_i2c_dtor,
 		.init = _nouveau_i2c_init,
 		.fini = _nouveau_i2c_fini,
 	},
-};
+	.sclass = nvd0_i2c_sclass,
+	.pad_x = &nv04_i2c_pad_oclass,
+	.pad_s = &nv94_i2c_pad_oclass,
+	.aux = 4,
+	.aux_stat = nv94_aux_stat,
+	.aux_mask = nv94_aux_mask,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/nve0.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/nve0.c
new file mode 100644
index 000000000000..cae77e1ad8dc
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/nve0.c
@@ -0,0 +1,72 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "nv50.h"
+
+static void
+nve0_aux_stat(struct nouveau_i2c *i2c, u32 *hi, u32 *lo, u32 *rq, u32 *tx)
+{
+	u32 intr = nv_rd32(i2c, 0x00dc60);
+	u32 stat = nv_rd32(i2c, 0x00dc68) & intr, i;
+	for (i = 0, *hi = *lo = *rq = *tx = 0; i < 8; i++) {
+		if ((stat & (1 << (i * 4)))) *hi |= 1 << i;
+		if ((stat & (2 << (i * 4)))) *lo |= 1 << i;
+		if ((stat & (4 << (i * 4)))) *rq |= 1 << i;
+		if ((stat & (8 << (i * 4)))) *tx |= 1 << i;
+	}
+	nv_wr32(i2c, 0x00dc60, intr);
+}
+
+static void
+nve0_aux_mask(struct nouveau_i2c *i2c, u32 type, u32 mask, u32 data)
+{
+	u32 temp = nv_rd32(i2c, 0x00dc68), i;
+	for (i = 0; i < 8; i++) {
+		if (mask & (1 << i)) {
+			if (!(data & (1 << i))) {
+				temp &= ~(type << (i * 4));
+				continue;
+			}
+			temp |= type << (i * 4);
+		}
+	}
+	nv_wr32(i2c, 0x00dc68, temp);
+}
+
+struct nouveau_oclass *
+nve0_i2c_oclass = &(struct nouveau_i2c_impl) {
+	.base.handle = NV_SUBDEV(I2C, 0xe0),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nouveau_i2c_ctor,
+		.dtor = _nouveau_i2c_dtor,
+		.init = _nouveau_i2c_init,
+		.fini = _nouveau_i2c_fini,
+	},
+	.sclass = nvd0_i2c_sclass,
+	.pad_x = &nv04_i2c_pad_oclass,
+	.pad_s = &nv94_i2c_pad_oclass,
+	.aux = 4,
+	.aux_stat = nve0_aux_stat,
+	.aux_mask = nve0_aux_mask,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/pad.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/pad.c
new file mode 100644
index 000000000000..e9e412477c12
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/pad.c
@@ -0,0 +1,84 @@
+/*
+ * Copyright 2014 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "pad.h"
+
+int
+_nvkm_i2c_pad_fini(struct nouveau_object *object, bool suspend)
+{
+	struct nvkm_i2c_pad *pad = (void *)object;
+	DBG("-> NULL\n");
+	pad->port = NULL;
+	return nouveau_object_fini(&pad->base, suspend);
+}
+
+int
+_nvkm_i2c_pad_init(struct nouveau_object *object)
+{
+	struct nvkm_i2c_pad *pad = (void *)object;
+	DBG("-> PORT:%02x\n", pad->next->index);
+	pad->port = pad->next;
+	return nouveau_object_init(&pad->base);
+}
+
+int
+nvkm_i2c_pad_create_(struct nouveau_object *parent,
+		     struct nouveau_object *engine,
+		     struct nouveau_oclass *oclass, int index,
+		     int size, void **pobject)
+{
+	struct nouveau_i2c *i2c = (void *)engine;
+	struct nouveau_i2c_port *port;
+	struct nvkm_i2c_pad *pad;
+	int ret;
+
+	list_for_each_entry(port, &i2c->ports, head) {
+		pad = nvkm_i2c_pad(port);
+		if (pad->index == index) {
+			atomic_inc(&nv_object(pad)->refcount);
+			*pobject = pad;
+			return 1;
+		}
+	}
+
+	ret = nouveau_object_create_(parent, engine, oclass, 0, size, pobject);
+	pad = *pobject;
+	if (ret)
+		return ret;
+
+	pad->index = index;
+	return 0;
+}
+
+int
+_nvkm_i2c_pad_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+		   struct nouveau_oclass *oclass, void *data, u32 index,
+		   struct nouveau_object **pobject)
+{
+	struct nvkm_i2c_pad *pad;
+	int ret;
+	ret = nvkm_i2c_pad_create(parent, engine, oclass, index, &pad);
+	*pobject = nv_object(pad);
+	return ret;
+}
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/pad.h b/drivers/gpu/drm/nouveau/core/subdev/i2c/pad.h
new file mode 100644
index 000000000000..452ac10c3004
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/pad.h
@@ -0,0 +1,58 @@
+#ifndef __NVKM_I2C_PAD_H__
+#define __NVKM_I2C_PAD_H__
+
+#include "priv.h"
+
+struct nvkm_i2c_pad {
+	struct nouveau_object base;
+	int index;
+	struct nouveau_i2c_port *port;
+	struct nouveau_i2c_port *next;
+};
+
+static inline struct nvkm_i2c_pad *
+nvkm_i2c_pad(struct nouveau_i2c_port *port)
+{
+	struct nouveau_object *pad = nv_object(port);
+	while (pad->parent)
+		pad = pad->parent;
+	return (void *)pad;
+}
+
+#define nvkm_i2c_pad_create(p,e,o,i,d)                                         \
+	nvkm_i2c_pad_create_((p), (e), (o), (i), sizeof(**d), (void **)d)
+#define nvkm_i2c_pad_destroy(p) ({                                             \
+	struct nvkm_i2c_pad *_p = (p);                                         \
+	_nvkm_i2c_pad_dtor(nv_object(_p));                                     \
+})
+#define nvkm_i2c_pad_init(p) ({                                                \
+	struct nvkm_i2c_pad *_p = (p);                                         \
+	_nvkm_i2c_pad_init(nv_object(_p));                                     \
+})
+#define nvkm_i2c_pad_fini(p,s) ({                                              \
+	struct nvkm_i2c_pad *_p = (p);                                         \
+	_nvkm_i2c_pad_fini(nv_object(_p), (s));                                \
+})
+
+int nvkm_i2c_pad_create_(struct nouveau_object *, struct nouveau_object *,
+			 struct nouveau_oclass *, int index, int, void **);
+
+int _nvkm_i2c_pad_ctor(struct nouveau_object *, struct nouveau_object *,
+		       struct nouveau_oclass *, void *, u32,
+		       struct nouveau_object **);
+#define _nvkm_i2c_pad_dtor nouveau_object_destroy
+int _nvkm_i2c_pad_init(struct nouveau_object *);
+int _nvkm_i2c_pad_fini(struct nouveau_object *, bool);
+
+#ifndef MSG
+#define MSG(l,f,a...) do {                                                     \
+	struct nvkm_i2c_pad *_pad = (void *)pad;                               \
+	nv_##l(nv_object(_pad)->engine, "PAD:%c:%02x: "f,                      \
+	       _pad->index >= 0x100 ? 'X' : 'S',                               \
+	       _pad->index >= 0x100 ? _pad->index - 0x100 : _pad->index, ##a); \
+} while(0)
+#define DBG(f,a...) MSG(debug, f, ##a)
+#define ERR(f,a...) MSG(error, f, ##a)
+#endif
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/padnv04.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/padnv04.c
new file mode 100644
index 000000000000..2c4b61296dd1
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/padnv04.c
@@ -0,0 +1,35 @@
+/*
+ * Copyright 2014 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "pad.h"
+
+struct nouveau_oclass
+nv04_i2c_pad_oclass = {
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = _nvkm_i2c_pad_ctor,
+		.dtor = _nvkm_i2c_pad_dtor,
+		.init = _nvkm_i2c_pad_init,
+		.fini = _nvkm_i2c_pad_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/padnv94.c b/drivers/gpu/drm/nouveau/core/subdev/i2c/padnv94.c
new file mode 100644
index 000000000000..0dc6753014f0
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/padnv94.c
@@ -0,0 +1,86 @@
+/*
+ * Copyright 2014 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include "pad.h"
+
+struct nv94_i2c_pad {
+	struct nvkm_i2c_pad base;
+	int addr;
+};
+
+static int
+nv94_i2c_pad_fini(struct nouveau_object *object, bool suspend)
+{
+	struct nouveau_i2c *i2c = (void *)object->engine;
+	struct nv94_i2c_pad *pad = (void *)object;
+	nv_mask(i2c, 0x00e50c + pad->addr, 0x00000001, 0x00000001);
+	return nvkm_i2c_pad_fini(&pad->base, suspend);
+}
+
+static int
+nv94_i2c_pad_init(struct nouveau_object *object)
+{
+	struct nouveau_i2c *i2c = (void *)object->engine;
+	struct nv94_i2c_pad *pad = (void *)object;
+
+	switch (nv_oclass(pad->base.next)->handle) {
+	case NV_I2C_TYPE_DCBI2C(DCB_I2C_NVIO_AUX):
+		nv_mask(i2c, 0x00e500 + pad->addr, 0x0000c003, 0x00000002);
+		break;
+	case NV_I2C_TYPE_DCBI2C(DCB_I2C_NVIO_BIT):
+	default:
+		nv_mask(i2c, 0x00e500 + pad->addr, 0x0000c003, 0x0000c001);
+		break;
+	}
+
+	nv_mask(i2c, 0x00e50c + pad->addr, 0x00000001, 0x00000000);
+	return nvkm_i2c_pad_init(&pad->base);
+}
+
+static int
+nv94_i2c_pad_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+		  struct nouveau_oclass *oclass, void *data, u32 index,
+		  struct nouveau_object **pobject)
+{
+	struct nv94_i2c_pad *pad;
+	int ret;
+
+	ret = nvkm_i2c_pad_create(parent, engine, oclass, index, &pad);
+	*pobject = nv_object(pad);
+	if (ret)
+		return ret;
+
+	pad->addr = index * 0x50;;
+	return 0;
+}
+
+struct nouveau_oclass
+nv94_i2c_pad_oclass = {
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv94_i2c_pad_ctor,
+		.dtor = _nvkm_i2c_pad_dtor,
+		.init = nv94_i2c_pad_init,
+		.fini = nv94_i2c_pad_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/port.h b/drivers/gpu/drm/nouveau/core/subdev/i2c/port.h
new file mode 100644
index 000000000000..a8ff6e077af5
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/port.h
@@ -0,0 +1,15 @@
+#ifndef __NVKM_I2C_PORT_H__
+#define __NVKM_I2C_PORT_H__
+
+#include "priv.h"
+
+#ifndef MSG
+#define MSG(l,f,a...) do {                                                     \
+	struct nouveau_i2c_port *_port = (void *)port;                         \
+	nv_##l(nv_object(_port)->engine, "PORT:%02x: "f, _port->index, ##a);   \
+} while(0)
+#define DBG(f,a...) MSG(debug, f, ##a)
+#define ERR(f,a...) MSG(error, f, ##a)
+#endif
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/i2c/priv.h b/drivers/gpu/drm/nouveau/core/subdev/i2c/priv.h
new file mode 100644
index 000000000000..780090b6425a
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/i2c/priv.h
@@ -0,0 +1,85 @@
+#ifndef __NVKM_I2C_H__
+#define __NVKM_I2C_H__
+
+#include <subdev/i2c.h>
+
+extern struct nouveau_oclass nv04_i2c_pad_oclass;
+extern struct nouveau_oclass nv94_i2c_pad_oclass;
+
+#define nouveau_i2c_port_create(p,e,o,i,a,f,d)                                 \
+	nouveau_i2c_port_create_((p), (e), (o), (i), (a), (f),                 \
+				 sizeof(**d), (void **)d)
+#define nouveau_i2c_port_destroy(p) ({                                         \
+	struct nouveau_i2c_port *port = (p);                                   \
+	_nouveau_i2c_port_dtor(nv_object(i2c));                                \
+})
+#define nouveau_i2c_port_init(p)                                               \
+	nouveau_object_init(&(p)->base)
+#define nouveau_i2c_port_fini(p,s)                                             \
+	nouveau_object_fini(&(p)->base, (s))
+
+int nouveau_i2c_port_create_(struct nouveau_object *, struct nouveau_object *,
+			     struct nouveau_oclass *, u8,
+			     const struct i2c_algorithm *,
+			     const struct nouveau_i2c_func *,
+			     int, void **);
+void _nouveau_i2c_port_dtor(struct nouveau_object *);
+#define _nouveau_i2c_port_init nouveau_object_init
+int  _nouveau_i2c_port_fini(struct nouveau_object *, bool);
+
+#define nouveau_i2c_create(p,e,o,d)                                            \
+	nouveau_i2c_create_((p), (e), (o), sizeof(**d), (void **)d)
+#define nouveau_i2c_destroy(p) ({                                              \
+	struct nouveau_i2c *i2c = (p);                                         \
+	_nouveau_i2c_dtor(nv_object(i2c));                                     \
+})
+#define nouveau_i2c_init(p) ({                                                 \
+	struct nouveau_i2c *i2c = (p);                                         \
+	_nouveau_i2c_init(nv_object(i2c));                                     \
+})
+#define nouveau_i2c_fini(p,s) ({                                               \
+	struct nouveau_i2c *i2c = (p);                                         \
+	_nouveau_i2c_fini(nv_object(i2c), (s));                                \
+})
+
+int nouveau_i2c_create_(struct nouveau_object *, struct nouveau_object *,
+			struct nouveau_oclass *, int, void **);
+int  _nouveau_i2c_ctor(struct nouveau_object *, struct nouveau_object *,
+		       struct nouveau_oclass *, void *, u32,
+		       struct nouveau_object **);
+void _nouveau_i2c_dtor(struct nouveau_object *);
+int  _nouveau_i2c_init(struct nouveau_object *);
+int  _nouveau_i2c_fini(struct nouveau_object *, bool);
+
+extern struct nouveau_oclass nouveau_anx9805_sclass[];
+extern struct nouveau_oclass nvd0_i2c_sclass[];
+
+extern const struct i2c_algorithm nouveau_i2c_bit_algo;
+extern const struct i2c_algorithm nouveau_i2c_aux_algo;
+
+struct nouveau_i2c_impl {
+	struct nouveau_oclass base;
+
+	/* supported i2c port classes */
+	struct nouveau_oclass *sclass;
+	struct nouveau_oclass *pad_x;
+	struct nouveau_oclass *pad_s;
+
+	/* number of native dp aux channels present */
+	int aux;
+
+	/* read and ack pending interrupts, returning only data
+	 * for ports that have not been masked off, while still
+	 * performing the ack for anything that was pending.
+	 */
+	void (*aux_stat)(struct nouveau_i2c *, u32 *, u32 *, u32 *, u32 *);
+
+	/* mask on/off interrupt types for a given set of auxch
+	 */
+	void (*aux_mask)(struct nouveau_i2c *, u32, u32, u32);
+};
+
+void nv94_aux_stat(struct nouveau_i2c *, u32 *, u32 *, u32 *, u32 *);
+void nv94_aux_mask(struct nouveau_i2c *, u32, u32, u32);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/core/subdev/ibus/gk20a.c b/drivers/gpu/drm/nouveau/core/subdev/ibus/gk20a.c
new file mode 100644
index 000000000000..245f0ebaa6af
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/ibus/gk20a.c
@@ -0,0 +1,103 @@
+/*
+ * Copyright (c) 2014, NVIDIA CORPORATION. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#include <subdev/ibus.h>
+#include <subdev/timer.h>
+
+struct gk20a_ibus_priv {
+	struct nouveau_ibus base;
+};
+
+static void
+gk20a_ibus_init_priv_ring(struct gk20a_ibus_priv *priv)
+{
+	nv_mask(priv, 0x137250, 0x3f, 0);
+
+	nv_mask(priv, 0x000200, 0x20, 0);
+	usleep_range(20, 30);
+	nv_mask(priv, 0x000200, 0x20, 0x20);
+
+	nv_wr32(priv, 0x12004c, 0x4);
+	nv_wr32(priv, 0x122204, 0x2);
+	nv_rd32(priv, 0x122204);
+}
+
+static void
+gk20a_ibus_intr(struct nouveau_subdev *subdev)
+{
+	struct gk20a_ibus_priv *priv = (void *)subdev;
+	u32 status0 = nv_rd32(priv, 0x120058);
+
+	if (status0 & 0x7) {
+		nv_debug(priv, "resetting priv ring\n");
+		gk20a_ibus_init_priv_ring(priv);
+	}
+
+	/* Acknowledge interrupt */
+	nv_mask(priv, 0x12004c, 0x2, 0x2);
+
+	if (!nv_wait(subdev, 0x12004c, 0x3f, 0x00))
+		nv_warn(priv, "timeout waiting for ringmaster ack\n");
+}
+
+static int
+gk20a_ibus_init(struct nouveau_object *object)
+{
+	struct gk20a_ibus_priv *priv = (void *)object;
+	int ret;
+
+	ret = _nouveau_ibus_init(object);
+	if (ret)
+		return ret;
+
+	gk20a_ibus_init_priv_ring(priv);
+
+	return 0;
+}
+
+static int
+gk20a_ibus_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+	       struct nouveau_oclass *oclass, void *data, u32 size,
+	       struct nouveau_object **pobject)
+{
+	struct gk20a_ibus_priv *priv;
+	int ret;
+
+	ret = nouveau_ibus_create(parent, engine, oclass, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	nv_subdev(priv)->intr = gk20a_ibus_intr;
+	return 0;
+}
+
+struct nouveau_oclass
+gk20a_ibus_oclass = {
+	.handle = NV_SUBDEV(IBUS, 0xea),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = gk20a_ibus_ctor,
+		.dtor = _nouveau_ibus_dtor,
+		.init = gk20a_ibus_init,
+		.fini = _nouveau_ibus_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/mc/nv50.c
index e8822a934c48..9ca93e2718f7 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/mc/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nv50.c
@@ -26,6 +26,7 @@
 
 const struct nouveau_mc_intr
 nv50_mc_intr[] = {
+	{ 0x04000000, NVDEV_ENGINE_DISP },  /* DISP before FIFO, so pageflip-timestamping works! */
 	{ 0x00000001, NVDEV_ENGINE_MPEG },
 	{ 0x00000100, NVDEV_ENGINE_FIFO },
 	{ 0x00001000, NVDEV_ENGINE_GR },
@@ -33,8 +34,8 @@ nv50_mc_intr[] = {
 	{ 0x00008000, NVDEV_ENGINE_BSP },	/* NV84- */
 	{ 0x00020000, NVDEV_ENGINE_VP },	/* NV84- */
 	{ 0x00100000, NVDEV_SUBDEV_TIMER },
-	{ 0x00200000, NVDEV_SUBDEV_GPIO },
-	{ 0x04000000, NVDEV_ENGINE_DISP },
+	{ 0x00200000, NVDEV_SUBDEV_GPIO },	/* PMGR->GPIO */
+	{ 0x00200000, NVDEV_SUBDEV_I2C }, 	/* PMGR->I2C/AUX */
 	{ 0x10000000, NVDEV_SUBDEV_BUS },
 	{ 0x80000000, NVDEV_ENGINE_SW },
 	{ 0x0002d101, NVDEV_SUBDEV_FB },
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c b/drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c
index f8a6f18e2d34..3c76d9038f38 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nv98.c
@@ -26,6 +26,7 @@
 
 static const struct nouveau_mc_intr
 nv98_mc_intr[] = {
+	{ 0x04000000, NVDEV_ENGINE_DISP },  /* DISP first, so pageflip timestamps work */
 	{ 0x00000001, NVDEV_ENGINE_PPP },
 	{ 0x00000100, NVDEV_ENGINE_FIFO },
 	{ 0x00001000, NVDEV_ENGINE_GR },
@@ -35,9 +36,9 @@ nv98_mc_intr[] = {
 	{ 0x00040000, NVDEV_SUBDEV_PWR },	/* NVA3:NVC0 */
 	{ 0x00080000, NVDEV_SUBDEV_THERM },	/* NVA3:NVC0 */
 	{ 0x00100000, NVDEV_SUBDEV_TIMER },
-	{ 0x00200000, NVDEV_SUBDEV_GPIO },
+	{ 0x00200000, NVDEV_SUBDEV_GPIO },	/* PMGR->GPIO */
+	{ 0x00200000, NVDEV_SUBDEV_I2C }, 	/* PMGR->I2C/AUX */
 	{ 0x00400000, NVDEV_ENGINE_COPY0 },	/* NVA3-     */
-	{ 0x04000000, NVDEV_ENGINE_DISP },
 	{ 0x10000000, NVDEV_SUBDEV_BUS },
 	{ 0x80000000, NVDEV_ENGINE_SW },
 	{ 0x0042d101, NVDEV_SUBDEV_FB },
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nvc0.c b/drivers/gpu/drm/nouveau/core/subdev/mc/nvc0.c
index 34472d317097..f9c6a678b47d 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/mc/nvc0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nvc0.c
@@ -26,6 +26,7 @@
 
 const struct nouveau_mc_intr
 nvc0_mc_intr[] = {
+	{ 0x04000000, NVDEV_ENGINE_DISP },  /* DISP first, so pageflip timestamps work. */
 	{ 0x00000001, NVDEV_ENGINE_PPP },
 	{ 0x00000020, NVDEV_ENGINE_COPY0 },
 	{ 0x00000040, NVDEV_ENGINE_COPY1 },
@@ -37,10 +38,10 @@ nvc0_mc_intr[] = {
 	{ 0x00040000, NVDEV_SUBDEV_THERM },
 	{ 0x00020000, NVDEV_ENGINE_VP },
 	{ 0x00100000, NVDEV_SUBDEV_TIMER },
-	{ 0x00200000, NVDEV_SUBDEV_GPIO },
+	{ 0x00200000, NVDEV_SUBDEV_GPIO },	/* PMGR->GPIO */
+	{ 0x00200000, NVDEV_SUBDEV_I2C },	/* PMGR->I2C/AUX */
 	{ 0x01000000, NVDEV_SUBDEV_PWR },
 	{ 0x02000000, NVDEV_SUBDEV_LTCG },
-	{ 0x04000000, NVDEV_ENGINE_DISP },
 	{ 0x08000000, NVDEV_SUBDEV_FB },
 	{ 0x10000000, NVDEV_SUBDEV_BUS },
 	{ 0x40000000, NVDEV_SUBDEV_IBUS },
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mxm/nv50.c b/drivers/gpu/drm/nouveau/core/subdev/mxm/nv50.c
index 64f8b4702bf7..fcaabe8456e3 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/mxm/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/mxm/nv50.c
@@ -150,7 +150,7 @@ mxm_dcb_sanitise_entry(struct nouveau_bios *bios, void *data, int idx, u16 pdcb)
 	 * common example is DP->eDP.
 	 */
 	conn  = bios->data;
-	conn += dcb_conn(bios, (ctx.outp[0] & 0x0000f000) >> 12, &ver, &len);
+	conn += nvbios_connEe(bios, (ctx.outp[0] & 0x0000f000) >> 12, &ver, &len);
 	type  = conn[0];
 	switch (ctx.desc.conn_type) {
 	case 0x01: /* LVDS */
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/ic.c b/drivers/gpu/drm/nouveau/core/subdev/therm/ic.c
index 7610fc5f8fa2..ca9ad9fd47be 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/therm/ic.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/ic.c
@@ -60,9 +60,9 @@ static struct nouveau_i2c_board_info
 nv_board_infos[] = {
 	{ { I2C_BOARD_INFO("w83l785ts", 0x2d) }, 0 },
 	{ { I2C_BOARD_INFO("w83781d", 0x2d) }, 0  },
-	{ { I2C_BOARD_INFO("adt7473", 0x2e) }, 20  },
-	{ { I2C_BOARD_INFO("adt7473", 0x2d) }, 20  },
-	{ { I2C_BOARD_INFO("adt7473", 0x2c) }, 20  },
+	{ { I2C_BOARD_INFO("adt7473", 0x2e) }, 40  },
+	{ { I2C_BOARD_INFO("adt7473", 0x2d) }, 40  },
+	{ { I2C_BOARD_INFO("adt7473", 0x2c) }, 40  },
 	{ { I2C_BOARD_INFO("f75375", 0x2e) }, 0  },
 	{ { I2C_BOARD_INFO("lm99", 0x4c) }, 0  },
 	{ { I2C_BOARD_INFO("lm90", 0x4c) }, 0  },
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/nva3.c b/drivers/gpu/drm/nouveau/core/subdev/therm/nva3.c
index 3b2c4580098b..0478b2e3fb1d 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/therm/nva3.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/nva3.c
@@ -36,7 +36,7 @@ nva3_therm_fan_sense(struct nouveau_therm *therm)
 	u32 tach = nv_rd32(therm, 0x00e728) & 0x0000ffff;
 	u32 ctrl = nv_rd32(therm, 0x00e720);
 	if (ctrl & 0x00000001)
-		return tach * 60;
+		return tach * 60 / 2;
 	return -ENODEV;
 }
 
diff --git a/drivers/gpu/drm/nouveau/dispnv04/dac.c b/drivers/gpu/drm/nouveau/dispnv04/dac.c
index 434b920f6bd4..a96dda48718e 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/dac.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/dac.c
@@ -414,7 +414,7 @@ static void nv04_dac_commit(struct drm_encoder *encoder)
 	helper->dpms(encoder, DRM_MODE_DPMS_ON);
 
 	NV_DEBUG(drm, "Output %s is running on CRTC %d using output %c\n",
-		 drm_get_connector_name(&nouveau_encoder_connector_get(nv_encoder)->base),
+		 nouveau_encoder_connector_get(nv_encoder)->base.name,
 		 nv_crtc->index, '@' + ffs(nv_encoder->dcb->or));
 }
 
diff --git a/drivers/gpu/drm/nouveau/dispnv04/dfp.c b/drivers/gpu/drm/nouveau/dispnv04/dfp.c
index a2d669b4acf2..e57babb206d3 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/dfp.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/dfp.c
@@ -477,7 +477,7 @@ static void nv04_dfp_commit(struct drm_encoder *encoder)
 	helper->dpms(encoder, DRM_MODE_DPMS_ON);
 
 	NV_DEBUG(drm, "Output %s is running on CRTC %d using output %c\n",
-		 drm_get_connector_name(&nouveau_encoder_connector_get(nv_encoder)->base),
+		 nouveau_encoder_connector_get(nv_encoder)->base.name,
 		 nv_crtc->index, '@' + ffs(nv_encoder->dcb->or));
 }
 
diff --git a/drivers/gpu/drm/nouveau/dispnv04/disp.c b/drivers/gpu/drm/nouveau/dispnv04/disp.c
index 2f1ed61f7c8c..4342fdaee707 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/disp.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/disp.c
@@ -115,7 +115,7 @@ nv04_display_create(struct drm_device *dev)
 				 &dev->mode_config.connector_list, head) {
 		if (!connector->encoder_ids[0]) {
 			NV_WARN(drm, "%s has no encoders, removing\n",
-				drm_get_connector_name(connector));
+				connector->name);
 			connector->funcs->destroy(connector);
 		}
 	}
diff --git a/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c b/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c
index 244822df8ffc..8667620b703a 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/tvnv04.c
@@ -171,7 +171,8 @@ static void nv04_tv_commit(struct drm_encoder *encoder)
 	helper->dpms(encoder, DRM_MODE_DPMS_ON);
 
 	NV_DEBUG(drm, "Output %s is running on CRTC %d using output %c\n",
-		 drm_get_connector_name(&nouveau_encoder_connector_get(nv_encoder)->base), nv_crtc->index, '@' + ffs(nv_encoder->dcb->or));
+		 nouveau_encoder_connector_get(nv_encoder)->base.name,
+		 nv_crtc->index, '@' + ffs(nv_encoder->dcb->or));
 }
 
 static void nv04_tv_destroy(struct drm_encoder *encoder)
diff --git a/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c b/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c
index acef48f4a4ea..195bd8e86c6a 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/tvnv17.c
@@ -612,8 +612,7 @@ static void nv17_tv_commit(struct drm_encoder *encoder)
 	helper->dpms(encoder, DRM_MODE_DPMS_ON);
 
 	NV_INFO(drm, "Output %s is running on CRTC %d using output %c\n",
-		drm_get_connector_name(
-			&nouveau_encoder_connector_get(nv_encoder)->base),
+		nouveau_encoder_connector_get(nv_encoder)->base.name,
 		nv_crtc->index, '@' + ffs(nv_encoder->dcb->or));
 }
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c
index d07ce028af51..1fa222e8f007 100644
--- a/drivers/gpu/drm/nouveau/nouveau_connector.c
+++ b/drivers/gpu/drm/nouveau/nouveau_connector.c
@@ -44,6 +44,7 @@
 
 #include <subdev/i2c.h>
 #include <subdev/gpio.h>
+#include <engine/disp.h>
 
 MODULE_PARM_DESC(tv_disable, "Disable TV-out detection");
 static int nouveau_tv_disable = 0;
@@ -75,7 +76,8 @@ find_encoder(struct drm_connector *connector, int type)
 			continue;
 		nv_encoder = nouveau_encoder(obj_to_encoder(obj));
 
-		if (type == DCB_OUTPUT_ANY || nv_encoder->dcb->type == type)
+		if (type == DCB_OUTPUT_ANY ||
+		    (nv_encoder->dcb && nv_encoder->dcb->type == type))
 			return nv_encoder;
 	}
 
@@ -100,22 +102,24 @@ static void
 nouveau_connector_destroy(struct drm_connector *connector)
 {
 	struct nouveau_connector *nv_connector = nouveau_connector(connector);
-	nouveau_event_ref(NULL, &nv_connector->hpd_func);
+	nouveau_event_ref(NULL, &nv_connector->hpd);
 	kfree(nv_connector->edid);
 	drm_sysfs_connector_remove(connector);
 	drm_connector_cleanup(connector);
+	if (nv_connector->aux.transfer)
+		drm_dp_aux_unregister(&nv_connector->aux);
 	kfree(connector);
 }
 
-static struct nouveau_i2c_port *
-nouveau_connector_ddc_detect(struct drm_connector *connector,
-			     struct nouveau_encoder **pnv_encoder)
+static struct nouveau_encoder *
+nouveau_connector_ddc_detect(struct drm_connector *connector)
 {
 	struct drm_device *dev = connector->dev;
 	struct nouveau_connector *nv_connector = nouveau_connector(connector);
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nouveau_gpio *gpio = nouveau_gpio(drm->device);
-	struct nouveau_i2c_port *port = NULL;
+	struct nouveau_encoder *nv_encoder;
+	struct drm_mode_object *obj;
 	int i, panel = -ENODEV;
 
 	/* eDP panels need powering on by us (if the VBIOS doesn't default it
@@ -130,13 +134,9 @@ nouveau_connector_ddc_detect(struct drm_connector *connector,
 		}
 	}
 
-	for (i = 0; i < DRM_CONNECTOR_MAX_ENCODER; i++) {
-		struct nouveau_encoder *nv_encoder;
-		struct drm_mode_object *obj;
-		int id;
-
-		id = connector->encoder_ids[i];
-		if (!id)
+	for (i = 0; nv_encoder = NULL, i < DRM_CONNECTOR_MAX_ENCODER; i++) {
+		int id = connector->encoder_ids[i];
+		if (id == 0)
 			break;
 
 		obj = drm_mode_object_find(dev, id, DRM_MODE_OBJECT_ENCODER);
@@ -144,22 +144,24 @@ nouveau_connector_ddc_detect(struct drm_connector *connector,
 			continue;
 		nv_encoder = nouveau_encoder(obj_to_encoder(obj));
 
-		port = nv_encoder->i2c;
-		if (port && nv_probe_i2c(port, 0x50)) {
-			*pnv_encoder = nv_encoder;
-			break;
+		if (nv_encoder->dcb->type == DCB_OUTPUT_DP) {
+			int ret = nouveau_dp_detect(nv_encoder);
+			if (ret == 0)
+				break;
+		} else
+		if (nv_encoder->i2c) {
+			if (nv_probe_i2c(nv_encoder->i2c, 0x50))
+				break;
 		}
-
-		port = NULL;
 	}
 
 	/* eDP panel not detected, restore panel power GPIO to previous
 	 * state to avoid confusing the SOR for other output types.
 	 */
-	if (!port && panel == 0)
+	if (!nv_encoder && panel == 0)
 		gpio->set(gpio, 0, DCB_GPIO_PANEL_POWER, 0xff, panel);
 
-	return port;
+	return nv_encoder;
 }
 
 static struct nouveau_encoder *
@@ -258,25 +260,17 @@ nouveau_connector_detect(struct drm_connector *connector, bool force)
 	if (ret < 0 && ret != -EACCES)
 		return conn_status;
 
-	i2c = nouveau_connector_ddc_detect(connector, &nv_encoder);
-	if (i2c) {
+	nv_encoder = nouveau_connector_ddc_detect(connector);
+	if (nv_encoder && (i2c = nv_encoder->i2c) != NULL) {
 		nv_connector->edid = drm_get_edid(connector, &i2c->adapter);
 		drm_mode_connector_update_edid_property(connector,
 							nv_connector->edid);
 		if (!nv_connector->edid) {
 			NV_ERROR(drm, "DDC responded, but no EDID for %s\n",
-				 drm_get_connector_name(connector));
+				 connector->name);
 			goto detect_analog;
 		}
 
-		if (nv_encoder->dcb->type == DCB_OUTPUT_DP &&
-		    !nouveau_dp_detect(to_drm_encoder(nv_encoder))) {
-			NV_ERROR(drm, "Detected %s, but failed init\n",
-				 drm_get_connector_name(connector));
-			conn_status = connector_status_disconnected;
-			goto out;
-		}
-
 		/* Override encoder type for DVI-I based on whether EDID
 		 * says the display is digital or analog, both use the
 		 * same i2c channel so the value returned from ddc_detect
@@ -437,7 +431,7 @@ nouveau_connector_force(struct drm_connector *connector)
 	nv_encoder = find_encoder(connector, type);
 	if (!nv_encoder) {
 		NV_ERROR(drm, "can't find encoder to force %s on!\n",
-			 drm_get_connector_name(connector));
+			 connector->name);
 		connector->status = connector_status_disconnected;
 		return;
 	}
@@ -912,33 +906,103 @@ nouveau_connector_funcs_lvds = {
 };
 
 static void
+nouveau_connector_dp_dpms(struct drm_connector *connector, int mode)
+{
+	struct nouveau_encoder *nv_encoder = NULL;
+
+	if (connector->encoder)
+		nv_encoder = nouveau_encoder(connector->encoder);
+	if (nv_encoder && nv_encoder->dcb &&
+	    nv_encoder->dcb->type == DCB_OUTPUT_DP) {
+		if (mode == DRM_MODE_DPMS_ON) {
+			u8 data = DP_SET_POWER_D0;
+			nv_wraux(nv_encoder->i2c, DP_SET_POWER, &data, 1);
+			usleep_range(1000, 2000);
+		} else {
+			u8 data = DP_SET_POWER_D3;
+			nv_wraux(nv_encoder->i2c, DP_SET_POWER, &data, 1);
+		}
+	}
+
+	drm_helper_connector_dpms(connector, mode);
+}
+
+static const struct drm_connector_funcs
+nouveau_connector_funcs_dp = {
+	.dpms = nouveau_connector_dp_dpms,
+	.save = NULL,
+	.restore = NULL,
+	.detect = nouveau_connector_detect,
+	.destroy = nouveau_connector_destroy,
+	.fill_modes = drm_helper_probe_single_connector_modes,
+	.set_property = nouveau_connector_set_property,
+	.force = nouveau_connector_force
+};
+
+static void
 nouveau_connector_hotplug_work(struct work_struct *work)
 {
 	struct nouveau_connector *nv_connector =
-		container_of(work, struct nouveau_connector, hpd_work);
+		container_of(work, typeof(*nv_connector), work);
 	struct drm_connector *connector = &nv_connector->base;
-	struct drm_device *dev = connector->dev;
-	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nouveau_gpio *gpio = nouveau_gpio(drm->device);
-	bool plugged = gpio->get(gpio, 0, nv_connector->hpd.func, 0xff);
+	struct nouveau_drm *drm = nouveau_drm(connector->dev);
+	const char *name = connector->name;
 
-	NV_DEBUG(drm, "%splugged %s\n", plugged ? "" : "un",
-		 drm_get_connector_name(connector));
+	if (nv_connector->status & NVKM_HPD_IRQ) {
+	} else {
+		bool plugged = (nv_connector->status != NVKM_HPD_UNPLUG);
 
-	if (plugged)
-		drm_helper_connector_dpms(connector, DRM_MODE_DPMS_ON);
-	else
-		drm_helper_connector_dpms(connector, DRM_MODE_DPMS_OFF);
+		NV_DEBUG(drm, "%splugged %s\n", plugged ? "" : "un", name);
 
-	drm_helper_hpd_irq_event(dev);
+		if (plugged)
+			drm_helper_connector_dpms(connector, DRM_MODE_DPMS_ON);
+		else
+			drm_helper_connector_dpms(connector, DRM_MODE_DPMS_OFF);
+		drm_helper_hpd_irq_event(connector->dev);
+	}
+
+	nouveau_event_get(nv_connector->hpd);
 }
 
 static int
-nouveau_connector_hotplug(void *data, int index)
+nouveau_connector_hotplug(void *data, u32 type, int index)
 {
 	struct nouveau_connector *nv_connector = data;
-	schedule_work(&nv_connector->hpd_work);
-	return NVKM_EVENT_KEEP;
+	nv_connector->status = type;
+	schedule_work(&nv_connector->work);
+	return NVKM_EVENT_DROP;
+}
+
+static ssize_t
+nouveau_connector_aux_xfer(struct drm_dp_aux *aux, struct drm_dp_aux_msg *msg)
+{
+	struct nouveau_connector *nv_connector =
+		container_of(aux, typeof(*nv_connector), aux);
+	struct nouveau_encoder *nv_encoder;
+	struct nouveau_i2c_port *port;
+	int ret;
+
+	nv_encoder = find_encoder(&nv_connector->base, DCB_OUTPUT_DP);
+	if (!nv_encoder || !(port = nv_encoder->i2c))
+		return -ENODEV;
+	if (WARN_ON(msg->size > 16))
+		return -E2BIG;
+	if (msg->size == 0)
+		return msg->size;
+
+	ret = nouveau_i2c(port)->acquire(port, 0);
+	if (ret)
+		return ret;
+
+	ret = port->func->aux(port, false, msg->request, msg->address,
+			      msg->buffer, msg->size);
+	nouveau_i2c(port)->release(port);
+	if (ret >= 0) {
+		msg->reply = ret;
+		return msg->size;
+	}
+
+	return ret;
 }
 
 static int
@@ -974,9 +1038,9 @@ nouveau_connector_create(struct drm_device *dev, int index)
 {
 	const struct drm_connector_funcs *funcs = &nouveau_connector_funcs;
 	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nouveau_gpio *gpio = nouveau_gpio(drm->device);
 	struct nouveau_display *disp = nouveau_display(dev);
 	struct nouveau_connector *nv_connector = NULL;
+	struct nouveau_disp *pdisp = nouveau_disp(drm->device);
 	struct drm_connector *connector;
 	int type, ret = 0;
 	bool dummy;
@@ -992,33 +1056,15 @@ nouveau_connector_create(struct drm_device *dev, int index)
 		return ERR_PTR(-ENOMEM);
 
 	connector = &nv_connector->base;
-	INIT_WORK(&nv_connector->hpd_work, nouveau_connector_hotplug_work);
 	nv_connector->index = index;
 
 	/* attempt to parse vbios connector type and hotplug gpio */
 	nv_connector->dcb = olddcb_conn(dev, index);
 	if (nv_connector->dcb) {
-		static const u8 hpd[16] = {
-			0xff, 0x07, 0x08, 0xff, 0xff, 0x51, 0x52, 0xff,
-			0xff, 0xff, 0xff, 0xff, 0xff, 0x5e, 0x5f, 0x60,
-		};
-
 		u32 entry = ROM16(nv_connector->dcb[0]);
 		if (olddcb_conntab(dev)[3] >= 4)
 			entry |= (u32)ROM16(nv_connector->dcb[2]) << 16;
 
-		ret = gpio->find(gpio, 0, hpd[ffs((entry & 0x07033000) >> 12)],
-				 DCB_GPIO_UNUSED, &nv_connector->hpd);
-		if (ret)
-			nv_connector->hpd.func = DCB_GPIO_UNUSED;
-
-		if (nv_connector->hpd.func != DCB_GPIO_UNUSED) {
-			nouveau_event_new(gpio->events, nv_connector->hpd.line,
-					  nouveau_connector_hotplug,
-					  nv_connector,
-					 &nv_connector->hpd_func);
-		}
-
 		nv_connector->type = nv_connector->dcb[0];
 		if (drm_conntype_from_dcb(nv_connector->type) ==
 					  DRM_MODE_CONNECTOR_Unknown) {
@@ -1040,7 +1086,6 @@ nouveau_connector_create(struct drm_device *dev, int index)
 		}
 	} else {
 		nv_connector->type = DCB_CONNECTOR_NONE;
-		nv_connector->hpd.func = DCB_GPIO_UNUSED;
 	}
 
 	/* no vbios data, or an unknown dcb connector type - attempt to
@@ -1080,8 +1125,8 @@ nouveau_connector_create(struct drm_device *dev, int index)
 		}
 	}
 
-	type = drm_conntype_from_dcb(nv_connector->type);
-	if (type == DRM_MODE_CONNECTOR_LVDS) {
+	switch ((type = drm_conntype_from_dcb(nv_connector->type))) {
+	case DRM_MODE_CONNECTOR_LVDS:
 		ret = nouveau_bios_parse_lvds_table(dev, 0, &dummy, &dummy);
 		if (ret) {
 			NV_ERROR(drm, "Error parsing LVDS table, disabling\n");
@@ -1090,8 +1135,23 @@ nouveau_connector_create(struct drm_device *dev, int index)
 		}
 
 		funcs = &nouveau_connector_funcs_lvds;
-	} else {
+		break;
+	case DRM_MODE_CONNECTOR_DisplayPort:
+	case DRM_MODE_CONNECTOR_eDP:
+		nv_connector->aux.dev = dev->dev;
+		nv_connector->aux.transfer = nouveau_connector_aux_xfer;
+		ret = drm_dp_aux_register(&nv_connector->aux);
+		if (ret) {
+			NV_ERROR(drm, "failed to register aux channel\n");
+			kfree(nv_connector);
+			return ERR_PTR(ret);
+		}
+
+		funcs = &nouveau_connector_funcs_dp;
+		break;
+	default:
 		funcs = &nouveau_connector_funcs;
+		break;
 	}
 
 	/* defaults, will get overridden in detect() */
@@ -1166,10 +1226,16 @@ nouveau_connector_create(struct drm_device *dev, int index)
 		break;
 	}
 
-	connector->polled = DRM_CONNECTOR_POLL_CONNECT;
-	if (nv_connector->hpd.func != DCB_GPIO_UNUSED)
+	ret = nouveau_event_new(pdisp->hpd, NVKM_HPD, index,
+				nouveau_connector_hotplug,
+				nv_connector, &nv_connector->hpd);
+	if (ret)
+		connector->polled = DRM_CONNECTOR_POLL_CONNECT;
+	else
 		connector->polled = DRM_CONNECTOR_POLL_HPD;
 
+	INIT_WORK(&nv_connector->work, nouveau_connector_hotplug_work);
+
 	drm_sysfs_connector_add(connector);
 	return connector;
 }
diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.h b/drivers/gpu/drm/nouveau/nouveau_connector.h
index 264a778f473b..8861b6c579ad 100644
--- a/drivers/gpu/drm/nouveau/nouveau_connector.h
+++ b/drivers/gpu/drm/nouveau/nouveau_connector.h
@@ -28,12 +28,12 @@
 #define __NOUVEAU_CONNECTOR_H__
 
 #include <drm/drm_edid.h>
+#include <drm/drm_dp_helper.h>
 #include "nouveau_crtc.h"
 
 #include <core/event.h>
 
 #include <subdev/bios.h>
-#include <subdev/bios/gpio.h>
 
 struct nouveau_i2c_port;
 
@@ -67,9 +67,11 @@ struct nouveau_connector {
 	u8 index;
 	u8 *dcb;
 
-	struct dcb_gpio_func hpd;
-	struct work_struct hpd_work;
-	struct nouveau_eventh *hpd_func;
+	struct nouveau_eventh *hpd;
+	u32 status;
+	struct work_struct work;
+
+	struct drm_dp_aux aux;
 
 	int dithering_mode;
 	int dithering_depth;
diff --git a/drivers/gpu/drm/nouveau/nouveau_crtc.h b/drivers/gpu/drm/nouveau/nouveau_crtc.h
index d1e5890784d7..a0534489d23f 100644
--- a/drivers/gpu/drm/nouveau/nouveau_crtc.h
+++ b/drivers/gpu/drm/nouveau/nouveau_crtc.h
@@ -74,7 +74,7 @@ struct nouveau_crtc {
 
 static inline struct nouveau_crtc *nouveau_crtc(struct drm_crtc *crtc)
 {
-	return container_of(crtc, struct nouveau_crtc, base);
+	return crtc ? container_of(crtc, struct nouveau_crtc, base) : NULL;
 }
 
 static inline struct drm_crtc *to_drm_crtc(struct nouveau_crtc *crtc)
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c
index da764a4ed958..26b5647188ef 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.c
+++ b/drivers/gpu/drm/nouveau/nouveau_display.c
@@ -42,7 +42,7 @@
 #include <core/class.h>
 
 static int
-nouveau_display_vblank_handler(void *data, int head)
+nouveau_display_vblank_handler(void *data, u32 type, int head)
 {
 	struct nouveau_drm *drm = data;
 	drm_handle_vblank(drm->dev, head);
@@ -178,7 +178,7 @@ nouveau_display_vblank_init(struct drm_device *dev)
 		return -ENOMEM;
 
 	for (i = 0; i < dev->mode_config.num_crtc; i++) {
-		ret = nouveau_event_new(pdisp->vblank, i,
+		ret = nouveau_event_new(pdisp->vblank, 1, i,
 					nouveau_display_vblank_handler,
 					drm, &disp->vblank[i]);
 		if (ret) {
@@ -393,7 +393,7 @@ nouveau_display_init(struct drm_device *dev)
 	/* enable hotplug interrupts */
 	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
 		struct nouveau_connector *conn = nouveau_connector(connector);
-		if (conn->hpd_func) nouveau_event_get(conn->hpd_func);
+		if (conn->hpd) nouveau_event_get(conn->hpd);
 	}
 
 	return ret;
@@ -408,7 +408,7 @@ nouveau_display_fini(struct drm_device *dev)
 	/* disable hotplug interrupts */
 	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
 		struct nouveau_connector *conn = nouveau_connector(connector);
-		if (conn->hpd_func) nouveau_event_put(conn->hpd_func);
+		if (conn->hpd) nouveau_event_put(conn->hpd);
 	}
 
 	drm_kms_helper_poll_disable(dev);
@@ -798,6 +798,7 @@ nouveau_finish_page_flip(struct nouveau_channel *chan,
 	struct drm_device *dev = drm->dev;
 	struct nouveau_page_flip_state *s;
 	unsigned long flags;
+	int crtcid = -1;
 
 	spin_lock_irqsave(&dev->event_lock, flags);
 
@@ -808,8 +809,13 @@ nouveau_finish_page_flip(struct nouveau_channel *chan,
 	}
 
 	s = list_first_entry(&fctx->flip, struct nouveau_page_flip_state, head);
-	if (s->event)
-		drm_send_vblank_event(dev, s->crtc, s->event);
+	if (s->event) {
+		/* Vblank timestamps/counts are only correct on >= NV-50 */
+		if (nv_device(drm->device)->card_type >= NV_50)
+			crtcid = s->crtc;
+
+		drm_send_vblank_event(dev, crtcid, s->event);
+	}
 
 	list_del(&s->head);
 	if (ps)
diff --git a/drivers/gpu/drm/nouveau/nouveau_dp.c b/drivers/gpu/drm/nouveau/nouveau_dp.c
index 36fd22500569..5675ffc175ae 100644
--- a/drivers/gpu/drm/nouveau/nouveau_dp.c
+++ b/drivers/gpu/drm/nouveau/nouveau_dp.c
@@ -55,11 +55,10 @@ nouveau_dp_probe_oui(struct drm_device *dev, struct nouveau_i2c_port *auxch,
 
 }
 
-bool
-nouveau_dp_detect(struct drm_encoder *encoder)
+int
+nouveau_dp_detect(struct nouveau_encoder *nv_encoder)
 {
-	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
-	struct drm_device *dev = encoder->dev;
+	struct drm_device *dev = nv_encoder->base.base.dev;
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nouveau_i2c_port *auxch;
 	u8 *dpcd = nv_encoder->dp.dpcd;
@@ -67,11 +66,11 @@ nouveau_dp_detect(struct drm_encoder *encoder)
 
 	auxch = nv_encoder->i2c;
 	if (!auxch)
-		return false;
+		return -ENODEV;
 
 	ret = nv_rdaux(auxch, DP_DPCD_REV, dpcd, 8);
 	if (ret)
-		return false;
+		return ret;
 
 	nv_encoder->dp.link_bw = 27000 * dpcd[1];
 	nv_encoder->dp.link_nr = dpcd[2] & DP_MAX_LANE_COUNT_MASK;
@@ -91,6 +90,5 @@ nouveau_dp_detect(struct drm_encoder *encoder)
 		     nv_encoder->dp.link_nr, nv_encoder->dp.link_bw);
 
 	nouveau_dp_probe_oui(dev, auxch, dpcd);
-
-	return true;
+	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nouveau_encoder.h b/drivers/gpu/drm/nouveau/nouveau_encoder.h
index 24660c0f713d..5f0e37fc2849 100644
--- a/drivers/gpu/drm/nouveau/nouveau_encoder.h
+++ b/drivers/gpu/drm/nouveau/nouveau_encoder.h
@@ -46,6 +46,7 @@ struct nouveau_encoder {
 	/* different to drm_encoder.crtc, this reflects what's
 	 * actually programmed on the hw, not the proposed crtc */
 	struct drm_crtc *crtc;
+	u32 ctrl;
 
 	struct drm_display_mode mode;
 	int last_dpms;
@@ -84,9 +85,7 @@ get_slave_funcs(struct drm_encoder *enc)
 }
 
 /* nouveau_dp.c */
-bool nouveau_dp_detect(struct drm_encoder *);
-void nouveau_dp_dpms(struct drm_encoder *, int mode, u32 datarate,
-		     struct nouveau_object *);
+int nouveau_dp_detect(struct nouveau_encoder *);
 
 struct nouveau_connector *
 nouveau_encoder_connector_get(struct nouveau_encoder *encoder);
diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.c b/drivers/gpu/drm/nouveau/nouveau_fence.c
index 90074d620e31..ab5ea3b0d666 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fence.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fence.c
@@ -166,7 +166,7 @@ nouveau_fence_done(struct nouveau_fence *fence)
 }
 
 static int
-nouveau_fence_wait_uevent_handler(void *data, int index)
+nouveau_fence_wait_uevent_handler(void *data, u32 type, int index)
 {
 	struct nouveau_fence_priv *priv = data;
 	wake_up_all(&priv->waiting);
@@ -183,7 +183,7 @@ nouveau_fence_wait_uevent(struct nouveau_fence *fence, bool intr)
 	struct nouveau_eventh *handler;
 	int ret = 0;
 
-	ret = nouveau_event_new(pfifo->uevent, 0,
+	ret = nouveau_event_new(pfifo->uevent, 1, 0,
 				nouveau_fence_wait_uevent_handler,
 				priv, &handler);
 	if (ret)
diff --git a/drivers/gpu/drm/nouveau/nouveau_ioc32.c b/drivers/gpu/drm/nouveau/nouveau_ioc32.c
index c1a7e5a73a26..462679a8fec5 100644
--- a/drivers/gpu/drm/nouveau/nouveau_ioc32.c
+++ b/drivers/gpu/drm/nouveau/nouveau_ioc32.c
@@ -57,7 +57,7 @@ long nouveau_compat_ioctl(struct file *filp, unsigned int cmd,
 		return drm_compat_ioctl(filp, cmd, arg);
 
 #if 0
-	if (nr < DRM_COMMAND_BASE + DRM_ARRAY_SIZE(mga_compat_ioctls))
+	if (nr < DRM_COMMAND_BASE + ARRAY_SIZE(mga_compat_ioctls))
 		fn = nouveau_compat_ioctls[nr - DRM_COMMAND_BASE];
 #endif
 	if (fn != NULL)
diff --git a/drivers/gpu/drm/nouveau/nouveau_vga.c b/drivers/gpu/drm/nouveau/nouveau_vga.c
index fb84da3cb50d..4f4c3fec6916 100644
--- a/drivers/gpu/drm/nouveau/nouveau_vga.c
+++ b/drivers/gpu/drm/nouveau/nouveau_vga.c
@@ -64,12 +64,13 @@ static bool
 nouveau_switcheroo_can_switch(struct pci_dev *pdev)
 {
 	struct drm_device *dev = pci_get_drvdata(pdev);
-	bool can_switch;
 
-	spin_lock(&dev->count_lock);
-	can_switch = (dev->open_count == 0);
-	spin_unlock(&dev->count_lock);
-	return can_switch;
+	/*
+	 * FIXME: open_count is protected by drm_global_mutex but that would lead to
+	 * locking inversion with the driver load path. And the access here is
+	 * completely racy anyway. So don't bother with locking for now.
+	 */
+	return dev->open_count == 0;
 }
 
 static const struct vga_switcheroo_client_ops
diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c
index 58af547b0b93..afdf607df3e6 100644
--- a/drivers/gpu/drm/nouveau/nv50_display.c
+++ b/drivers/gpu/drm/nouveau/nv50_display.c
@@ -1,4 +1,4 @@
-	/*
+/*
  * Copyright 2011 Red Hat Inc.
  *
  * Permission is hereby granted, free of charge, to any person obtaining a
@@ -26,6 +26,7 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
+#include <drm/drm_dp_helper.h>
 
 #include "nouveau_drm.h"
 #include "nouveau_dma.h"
@@ -957,7 +958,7 @@ nv50_crtc_prepare(struct drm_crtc *crtc)
 
 	nv50_display_flip_stop(crtc);
 
-	push = evo_wait(mast, 2);
+	push = evo_wait(mast, 6);
 	if (push) {
 		if (nv50_vers(mast) < NV84_DISP_MAST_CLASS) {
 			evo_mthd(push, 0x0874 + (nv_crtc->index * 0x400), 1);
@@ -1207,6 +1208,7 @@ static void
 nv50_crtc_disable(struct drm_crtc *crtc)
 {
 	struct nv50_head *head = nv50_head(crtc);
+	evo_sync(crtc->dev);
 	if (head->image)
 		nouveau_bo_unpin(head->image);
 	nouveau_bo_ref(NULL, &head->image);
@@ -1700,10 +1702,9 @@ nv50_hdmi_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode)
 }
 
 static void
-nv50_hdmi_disconnect(struct drm_encoder *encoder)
+nv50_hdmi_disconnect(struct drm_encoder *encoder, struct nouveau_crtc *nv_crtc)
 {
 	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
-	struct nouveau_crtc *nv_crtc = nouveau_crtc(nv_encoder->crtc);
 	struct nv50_disp *disp = nv50_disp(encoder->dev);
 	const u32 moff = (nv_crtc->index << 3) | nv_encoder->or;
 
@@ -1722,7 +1723,7 @@ nv50_sor_dpms(struct drm_encoder *encoder, int mode)
 	struct drm_device *dev = encoder->dev;
 	struct nv50_disp *disp = nv50_disp(dev);
 	struct drm_encoder *partner;
-	int or = nv_encoder->or;
+	u32 mthd;
 
 	nv_encoder->last_dpms = mode;
 
@@ -1740,7 +1741,17 @@ nv50_sor_dpms(struct drm_encoder *encoder, int mode)
 		}
 	}
 
-	nv_call(disp->core, NV50_DISP_SOR_PWR + or, (mode == DRM_MODE_DPMS_ON));
+	mthd  = (ffs(nv_encoder->dcb->sorconf.link) - 1) << 2;
+	mthd |= nv_encoder->or;
+
+	if (nv_encoder->dcb->type == DCB_OUTPUT_DP) {
+		nv_call(disp->core, NV50_DISP_SOR_PWR | mthd, 1);
+		mthd |= NV94_DISP_SOR_DP_PWR;
+	} else {
+		mthd |= NV50_DISP_SOR_PWR;
+	}
+
+	nv_call(disp->core, mthd, (mode == DRM_MODE_DPMS_ON));
 }
 
 static bool
@@ -1764,33 +1775,36 @@ nv50_sor_mode_fixup(struct drm_encoder *encoder,
 }
 
 static void
-nv50_sor_disconnect(struct drm_encoder *encoder)
+nv50_sor_ctrl(struct nouveau_encoder *nv_encoder, u32 mask, u32 data)
 {
-	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
-	struct nv50_mast *mast = nv50_mast(encoder->dev);
-	const int or = nv_encoder->or;
-	u32 *push;
-
-	if (nv_encoder->crtc) {
-		nv50_crtc_prepare(nv_encoder->crtc);
-
-		push = evo_wait(mast, 4);
-		if (push) {
-			if (nv50_vers(mast) < NVD0_DISP_MAST_CLASS) {
-				evo_mthd(push, 0x0600 + (or * 0x40), 1);
-				evo_data(push, 0x00000000);
-			} else {
-				evo_mthd(push, 0x0200 + (or * 0x20), 1);
-				evo_data(push, 0x00000000);
-			}
-			evo_kick(push, mast);
+	struct nv50_mast *mast = nv50_mast(nv_encoder->base.base.dev);
+	u32 temp = (nv_encoder->ctrl & ~mask) | (data & mask), *push;
+	if (temp != nv_encoder->ctrl && (push = evo_wait(mast, 2))) {
+		if (nv50_vers(mast) < NVD0_DISP_MAST_CLASS) {
+			evo_mthd(push, 0x0600 + (nv_encoder->or * 0x40), 1);
+			evo_data(push, (nv_encoder->ctrl = temp));
+		} else {
+			evo_mthd(push, 0x0200 + (nv_encoder->or * 0x20), 1);
+			evo_data(push, (nv_encoder->ctrl = temp));
 		}
-
-		nv50_hdmi_disconnect(encoder);
+		evo_kick(push, mast);
 	}
+}
+
+static void
+nv50_sor_disconnect(struct drm_encoder *encoder)
+{
+	struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
+	struct nouveau_crtc *nv_crtc = nouveau_crtc(nv_encoder->crtc);
 
 	nv_encoder->last_dpms = DRM_MODE_DPMS_OFF;
 	nv_encoder->crtc = NULL;
+
+	if (nv_crtc) {
+		nv50_crtc_prepare(&nv_crtc->base);
+		nv50_sor_ctrl(nv_encoder, 1 << nv_crtc->index, 0);
+		nv50_hdmi_disconnect(&nv_encoder->base.base, nv_crtc);
+	}
 }
 
 static void
@@ -1810,12 +1824,14 @@ nv50_sor_mode_set(struct drm_encoder *encoder, struct drm_display_mode *umode,
 	struct nouveau_crtc *nv_crtc = nouveau_crtc(encoder->crtc);
 	struct nouveau_connector *nv_connector;
 	struct nvbios *bios = &drm->vbios;
-	u32 *push, lvds = 0;
+	u32 lvds = 0, mask, ctrl;
 	u8 owner = 1 << nv_crtc->index;
 	u8 proto = 0xf;
 	u8 depth = 0x0;
 
 	nv_connector = nouveau_encoder_connector_get(nv_encoder);
+	nv_encoder->crtc = encoder->crtc;
+
 	switch (nv_encoder->dcb->type) {
 	case DCB_OUTPUT_TMDS:
 		if (nv_encoder->dcb->sorconf.link & 1) {
@@ -1827,7 +1843,7 @@ nv50_sor_mode_set(struct drm_encoder *encoder, struct drm_display_mode *umode,
 			proto = 0x2;
 		}
 
-		nv50_hdmi_mode_set(encoder, mode);
+		nv50_hdmi_mode_set(&nv_encoder->base.base, mode);
 		break;
 	case DCB_OUTPUT_LVDS:
 		proto = 0x0;
@@ -1883,19 +1899,11 @@ nv50_sor_mode_set(struct drm_encoder *encoder, struct drm_display_mode *umode,
 		break;
 	}
 
-	nv50_sor_dpms(encoder, DRM_MODE_DPMS_ON);
+	nv50_sor_dpms(&nv_encoder->base.base, DRM_MODE_DPMS_ON);
 
-	push = evo_wait(nv50_mast(dev), 8);
-	if (push) {
-		if (nv50_vers(mast) < NVD0_DISP_CLASS) {
-			u32 ctrl = (depth << 16) | (proto << 8) | owner;
-			if (mode->flags & DRM_MODE_FLAG_NHSYNC)
-				ctrl |= 0x00001000;
-			if (mode->flags & DRM_MODE_FLAG_NVSYNC)
-				ctrl |= 0x00002000;
-			evo_mthd(push, 0x0600 + (nv_encoder->or * 0x040), 1);
-			evo_data(push, ctrl);
-		} else {
+	if (nv50_vers(mast) >= NVD0_DISP_CLASS) {
+		u32 *push = evo_wait(mast, 3);
+		if (push) {
 			u32 magic = 0x31ec6000 | (nv_crtc->index << 25);
 			u32 syncs = 0x00000001;
 
@@ -1910,14 +1918,21 @@ nv50_sor_mode_set(struct drm_encoder *encoder, struct drm_display_mode *umode,
 			evo_mthd(push, 0x0404 + (nv_crtc->index * 0x300), 2);
 			evo_data(push, syncs | (depth << 6));
 			evo_data(push, magic);
-			evo_mthd(push, 0x0200 + (nv_encoder->or * 0x020), 1);
-			evo_data(push, owner | (proto << 8));
+			evo_kick(push, mast);
 		}
 
-		evo_kick(push, mast);
+		ctrl = proto << 8;
+		mask = 0x00000f00;
+	} else {
+		ctrl = (depth << 16) | (proto << 8);
+		if (mode->flags & DRM_MODE_FLAG_NHSYNC)
+			ctrl |= 0x00001000;
+		if (mode->flags & DRM_MODE_FLAG_NVSYNC)
+			ctrl |= 0x00002000;
+		mask = 0x000f3f00;
 	}
 
-	nv_encoder->crtc = encoder->crtc;
+	nv50_sor_ctrl(nv_encoder, mask | owner, ctrl | owner);
 }
 
 static void
@@ -2295,7 +2310,7 @@ nv50_display_create(struct drm_device *dev)
 			continue;
 
 		NV_WARN(drm, "%s has no encoders, removing\n",
-			drm_get_connector_name(connector));
+			connector->name);
 		connector->funcs->destroy(connector);
 	}
 
diff --git a/drivers/gpu/drm/omapdrm/omap_crtc.c b/drivers/gpu/drm/omapdrm/omap_crtc.c
index e3c47a8005ff..2d28dc337cfb 100644
--- a/drivers/gpu/drm/omapdrm/omap_crtc.c
+++ b/drivers/gpu/drm/omapdrm/omap_crtc.c
@@ -319,13 +319,13 @@ static void page_flip_worker(struct work_struct *work)
 	struct drm_display_mode *mode = &crtc->mode;
 	struct drm_gem_object *bo;
 
-	mutex_lock(&crtc->mutex);
+	drm_modeset_lock(&crtc->mutex, NULL);
 	omap_plane_mode_set(omap_crtc->plane, crtc, crtc->primary->fb,
 			0, 0, mode->hdisplay, mode->vdisplay,
 			crtc->x << 16, crtc->y << 16,
 			mode->hdisplay << 16, mode->vdisplay << 16,
 			vblank_cb, crtc);
-	mutex_unlock(&crtc->mutex);
+	drm_modeset_unlock(&crtc->mutex);
 
 	bo = omap_framebuffer_bo(crtc->primary->fb, 0);
 	drm_gem_object_unreference_unlocked(bo);
@@ -465,7 +465,7 @@ static void apply_worker(struct work_struct *work)
 	 * the callbacks and list modification all serialized
 	 * with respect to modesetting ioctls from userspace.
 	 */
-	mutex_lock(&crtc->mutex);
+	drm_modeset_lock(&crtc->mutex, NULL);
 	dispc_runtime_get();
 
 	/*
@@ -510,7 +510,7 @@ static void apply_worker(struct work_struct *work)
 
 out:
 	dispc_runtime_put();
-	mutex_unlock(&crtc->mutex);
+	drm_modeset_unlock(&crtc->mutex);
 }
 
 int omap_crtc_apply(struct drm_crtc *crtc,
@@ -518,7 +518,7 @@ int omap_crtc_apply(struct drm_crtc *crtc,
 {
 	struct omap_crtc *omap_crtc = to_omap_crtc(crtc);
 
-	WARN_ON(!mutex_is_locked(&crtc->mutex));
+	WARN_ON(!drm_modeset_is_locked(&crtc->mutex));
 
 	/* no need to queue it again if it is already queued: */
 	if (apply->queued)
diff --git a/drivers/gpu/drm/omapdrm/omap_drv.c b/drivers/gpu/drm/omapdrm/omap_drv.c
index c8270e4b26f3..002b9721e85a 100644
--- a/drivers/gpu/drm/omapdrm/omap_drv.c
+++ b/drivers/gpu/drm/omapdrm/omap_drv.c
@@ -588,9 +588,7 @@ static void dev_lastclose(struct drm_device *dev)
 		}
 	}
 
-	drm_modeset_lock_all(dev);
-	ret = drm_fb_helper_restore_fbdev_mode(priv->fbdev);
-	drm_modeset_unlock_all(dev);
+	ret = drm_fb_helper_restore_fbdev_mode_unlocked(priv->fbdev);
 	if (ret)
 		DBG("failed to restore crtc mode");
 }
diff --git a/drivers/gpu/drm/omapdrm/omap_fb.c b/drivers/gpu/drm/omapdrm/omap_fb.c
index 8b019602ffe6..2a5cacdc344b 100644
--- a/drivers/gpu/drm/omapdrm/omap_fb.c
+++ b/drivers/gpu/drm/omapdrm/omap_fb.c
@@ -346,6 +346,7 @@ void omap_framebuffer_flush(struct drm_framebuffer *fb,
 
 	VERB("flush: %d,%d %dx%d, fb=%p", x, y, w, h, fb);
 
+	/* FIXME: This is racy - no protection against modeset config changes. */
 	while ((connector = omap_framebuffer_get_next_connector(fb, connector))) {
 		/* only consider connectors that are part of a chain */
 		if (connector->encoder && connector->encoder->crtc) {
diff --git a/drivers/gpu/drm/panel/panel-ld9040.c b/drivers/gpu/drm/panel/panel-ld9040.c
index 1f1f8371a199..db1601fdbe29 100644
--- a/drivers/gpu/drm/panel/panel-ld9040.c
+++ b/drivers/gpu/drm/panel/panel-ld9040.c
@@ -27,6 +27,7 @@
 #define MCS_ELVSS_ON		0xb1
 #define MCS_USER_SETTING	0xf0
 #define MCS_DISPCTL		0xf2
+#define MCS_POWER_CTRL		0xf4
 #define MCS_GTCON		0xf7
 #define MCS_PANEL_CONDITION	0xf8
 #define MCS_GAMMA_SET1		0xf9
@@ -182,6 +183,8 @@ static void ld9040_init(struct ld9040 *ctx)
 	ld9040_dcs_write_seq_static(ctx, MCS_DISPCTL,
 		0x02, 0x08, 0x08, 0x10, 0x10);
 	ld9040_dcs_write_seq_static(ctx, MCS_MANPWR, 0x04);
+	ld9040_dcs_write_seq_static(ctx, MCS_POWER_CTRL,
+		0x0a, 0x87, 0x25, 0x6a, 0x44, 0x02, 0x88);
 	ld9040_dcs_write_seq_static(ctx, MCS_ELVSS_ON, 0x0d, 0x00, 0x16);
 	ld9040_dcs_write_seq_static(ctx, MCS_GTCON, 0x09, 0x00, 0x00);
 	ld9040_brightness_set(ctx);
diff --git a/drivers/gpu/drm/panel/panel-s6e8aa0.c b/drivers/gpu/drm/panel/panel-s6e8aa0.c
index 35941d2412b8..06e57a26db7a 100644
--- a/drivers/gpu/drm/panel/panel-s6e8aa0.c
+++ b/drivers/gpu/drm/panel/panel-s6e8aa0.c
@@ -847,6 +847,7 @@ static void s6e8aa0_read_mtp_id(struct s6e8aa0 *ctx)
 	if (i >= ARRAY_SIZE(s6e8aa0_variants)) {
 		dev_err(ctx->dev, "unsupported display version %d\n", id[1]);
 		ctx->error = -EINVAL;
+		return;
 	}
 
 	ctx->variant = &s6e8aa0_variants[i];
diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c
index 309f29e9234a..a25136132c31 100644
--- a/drivers/gpu/drm/panel/panel-simple.c
+++ b/drivers/gpu/drm/panel/panel-simple.c
@@ -262,6 +262,13 @@ static int panel_simple_remove(struct device *dev)
 	return 0;
 }
 
+static void panel_simple_shutdown(struct device *dev)
+{
+	struct panel_simple *panel = dev_get_drvdata(dev);
+
+	panel_simple_disable(&panel->base);
+}
+
 static const struct drm_display_mode auo_b101aw03_mode = {
 	.clock = 51450,
 	.hdisplay = 1024,
@@ -284,6 +291,28 @@ static const struct panel_desc auo_b101aw03 = {
 	},
 };
 
+static const struct drm_display_mode auo_b133xtn01_mode = {
+	.clock = 69500,
+	.hdisplay = 1366,
+	.hsync_start = 1366 + 48,
+	.hsync_end = 1366 + 48 + 32,
+	.htotal = 1366 + 48 + 32 + 20,
+	.vdisplay = 768,
+	.vsync_start = 768 + 3,
+	.vsync_end = 768 + 3 + 6,
+	.vtotal = 768 + 3 + 6 + 13,
+	.vrefresh = 60,
+};
+
+static const struct panel_desc auo_b133xtn01 = {
+	.modes = &auo_b133xtn01_mode,
+	.num_modes = 1,
+	.size = {
+		.width = 293,
+		.height = 165,
+	},
+};
+
 static const struct drm_display_mode chunghwa_claa101wa01a_mode = {
 	.clock = 72070,
 	.hdisplay = 1366,
@@ -328,6 +357,52 @@ static const struct panel_desc chunghwa_claa101wb01 = {
 	},
 };
 
+static const struct drm_display_mode edt_et057090dhu_mode = {
+	.clock = 25175,
+	.hdisplay = 640,
+	.hsync_start = 640 + 16,
+	.hsync_end = 640 + 16 + 30,
+	.htotal = 640 + 16 + 30 + 114,
+	.vdisplay = 480,
+	.vsync_start = 480 + 10,
+	.vsync_end = 480 + 10 + 3,
+	.vtotal = 480 + 10 + 3 + 32,
+	.vrefresh = 60,
+	.flags = DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_NHSYNC,
+};
+
+static const struct panel_desc edt_et057090dhu = {
+	.modes = &edt_et057090dhu_mode,
+	.num_modes = 1,
+	.size = {
+		.width = 115,
+		.height = 86,
+	},
+};
+
+static const struct drm_display_mode edt_etm0700g0dh6_mode = {
+	.clock = 33260,
+	.hdisplay = 800,
+	.hsync_start = 800 + 40,
+	.hsync_end = 800 + 40 + 128,
+	.htotal = 800 + 40 + 128 + 88,
+	.vdisplay = 480,
+	.vsync_start = 480 + 10,
+	.vsync_end = 480 + 10 + 2,
+	.vtotal = 480 + 10 + 2 + 33,
+	.vrefresh = 60,
+	.flags = DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC,
+};
+
+static const struct panel_desc edt_etm0700g0dh6 = {
+	.modes = &edt_etm0700g0dh6_mode,
+	.num_modes = 1,
+	.size = {
+		.width = 152,
+		.height = 91,
+	},
+};
+
 static const struct drm_display_mode lg_lp129qe_mode = {
 	.clock = 285250,
 	.hdisplay = 2560,
@@ -377,12 +452,24 @@ static const struct of_device_id platform_of_match[] = {
 		.compatible = "auo,b101aw03",
 		.data = &auo_b101aw03,
 	}, {
+		.compatible = "auo,b133xtn01",
+		.data = &auo_b133xtn01,
+	}, {
 		.compatible = "chunghwa,claa101wa01a",
 		.data = &chunghwa_claa101wa01a
 	}, {
 		.compatible = "chunghwa,claa101wb01",
 		.data = &chunghwa_claa101wb01
 	}, {
+		.compatible = "edt,et057090dhu",
+		.data = &edt_et057090dhu,
+	}, {
+		.compatible = "edt,et070080dh6",
+		.data = &edt_etm0700g0dh6,
+	}, {
+		.compatible = "edt,etm0700g0dh6",
+		.data = &edt_etm0700g0dh6,
+	}, {
 		.compatible = "lg,lp129qe",
 		.data = &lg_lp129qe,
 	}, {
@@ -412,6 +499,11 @@ static int panel_simple_platform_remove(struct platform_device *pdev)
 	return panel_simple_remove(&pdev->dev);
 }
 
+static void panel_simple_platform_shutdown(struct platform_device *pdev)
+{
+	panel_simple_shutdown(&pdev->dev);
+}
+
 static struct platform_driver panel_simple_platform_driver = {
 	.driver = {
 		.name = "panel-simple",
@@ -420,6 +512,7 @@ static struct platform_driver panel_simple_platform_driver = {
 	},
 	.probe = panel_simple_platform_probe,
 	.remove = panel_simple_platform_remove,
+	.shutdown = panel_simple_platform_shutdown,
 };
 
 struct panel_desc_dsi {
@@ -561,6 +654,11 @@ static int panel_simple_dsi_remove(struct mipi_dsi_device *dsi)
 	return panel_simple_remove(&dsi->dev);
 }
 
+static void panel_simple_dsi_shutdown(struct mipi_dsi_device *dsi)
+{
+	panel_simple_shutdown(&dsi->dev);
+}
+
 static struct mipi_dsi_driver panel_simple_dsi_driver = {
 	.driver = {
 		.name = "panel-simple-dsi",
@@ -569,6 +667,7 @@ static struct mipi_dsi_driver panel_simple_dsi_driver = {
 	},
 	.probe = panel_simple_dsi_probe,
 	.remove = panel_simple_dsi_remove,
+	.shutdown = panel_simple_dsi_shutdown,
 };
 
 static int __init panel_simple_init(void)
diff --git a/drivers/gpu/drm/qxl/qxl_display.c b/drivers/gpu/drm/qxl/qxl_display.c
index 41bdd174657e..5d7ea2461852 100644
--- a/drivers/gpu/drm/qxl/qxl_display.c
+++ b/drivers/gpu/drm/qxl/qxl_display.c
@@ -574,6 +574,10 @@ static int qxl_crtc_mode_set(struct drm_crtc *crtc,
 			   bo->surf.height, bo->surf.stride, bo->surf.format);
 		qxl_io_create_primary(qdev, base_offset, bo);
 		bo->is_primary = true;
+	}
+
+	if (bo->is_primary) {
+		DRM_DEBUG_KMS("setting surface_id to 0 for primary surface %d on crtc %d\n", bo->surface_id, qcrtc->index);
 		surf_id = 0;
 	} else {
 		surf_id = bo->surface_id;
@@ -841,7 +845,7 @@ static const struct drm_connector_funcs qxl_connector_funcs = {
 	.save = qxl_conn_save,
 	.restore = qxl_conn_restore,
 	.detect = qxl_conn_detect,
-	.fill_modes = drm_helper_probe_single_connector_modes,
+	.fill_modes = drm_helper_probe_single_connector_modes_nomerge,
 	.set_property = qxl_conn_set_property,
 	.destroy = qxl_conn_destroy,
 };
diff --git a/drivers/gpu/drm/qxl/qxl_drv.c b/drivers/gpu/drm/qxl/qxl_drv.c
index fee8748bdca5..6e936634d65c 100644
--- a/drivers/gpu/drm/qxl/qxl_drv.c
+++ b/drivers/gpu/drm/qxl/qxl_drv.c
@@ -214,7 +214,6 @@ static struct pci_driver qxl_pci_driver = {
 static struct drm_driver qxl_driver = {
 	.driver_features = DRIVER_GEM | DRIVER_MODESET |
 			   DRIVER_HAVE_IRQ | DRIVER_IRQ_SHARED,
-	.dev_priv_size = 0,
 	.load = qxl_driver_load,
 	.unload = qxl_driver_unload,
 
diff --git a/drivers/gpu/drm/qxl/qxl_ioctl.c b/drivers/gpu/drm/qxl/qxl_ioctl.c
index 0bb86e6d41b4..b110883f8253 100644
--- a/drivers/gpu/drm/qxl/qxl_ioctl.c
+++ b/drivers/gpu/drm/qxl/qxl_ioctl.c
@@ -451,4 +451,4 @@ const struct drm_ioctl_desc qxl_ioctls[] = {
 			  DRM_AUTH|DRM_UNLOCKED),
 };
 
-int qxl_max_ioctls = DRM_ARRAY_SIZE(qxl_ioctls);
+int qxl_max_ioctls = ARRAY_SIZE(qxl_ioctls);
diff --git a/drivers/gpu/drm/qxl/qxl_irq.c b/drivers/gpu/drm/qxl/qxl_irq.c
index 28f84b4fce32..34d6a85e9023 100644
--- a/drivers/gpu/drm/qxl/qxl_irq.c
+++ b/drivers/gpu/drm/qxl/qxl_irq.c
@@ -87,7 +87,7 @@ int qxl_irq_init(struct qxl_device *qdev)
 	atomic_set(&qdev->irq_received_cursor, 0);
 	atomic_set(&qdev->irq_received_io_cmd, 0);
 	qdev->irq_received_error = 0;
-	ret = drm_irq_install(qdev->ddev);
+	ret = drm_irq_install(qdev->ddev, qdev->ddev->pdev->irq);
 	qdev->ram_header->int_mask = QXL_INTERRUPT_MASK;
 	if (unlikely(ret != 0)) {
 		DRM_ERROR("Failed installing irq: %d\n", ret);
diff --git a/drivers/gpu/drm/qxl/qxl_ttm.c b/drivers/gpu/drm/qxl/qxl_ttm.c
index d52c27527b9a..71a1baeac14e 100644
--- a/drivers/gpu/drm/qxl/qxl_ttm.c
+++ b/drivers/gpu/drm/qxl/qxl_ttm.c
@@ -109,13 +109,11 @@ static const struct vm_operations_struct *ttm_vm_ops;
 static int qxl_ttm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 {
 	struct ttm_buffer_object *bo;
-	struct qxl_device *qdev;
 	int r;
 
 	bo = (struct ttm_buffer_object *)vma->vm_private_data;
 	if (bo == NULL)
 		return VM_FAULT_NOPAGE;
-	qdev = qxl_get_qdev(bo->bdev);
 	r = ttm_vm_ops->fault(vma, vmf);
 	return r;
 }
@@ -162,10 +160,6 @@ static int qxl_invalidate_caches(struct ttm_bo_device *bdev, uint32_t flags)
 static int qxl_init_mem_type(struct ttm_bo_device *bdev, uint32_t type,
 			     struct ttm_mem_type_manager *man)
 {
-	struct qxl_device *qdev;
-
-	qdev = qxl_get_qdev(bdev);
-
 	switch (type) {
 	case TTM_PL_SYSTEM:
 		/* System memory */
diff --git a/drivers/gpu/drm/r128/r128_ioc32.c b/drivers/gpu/drm/r128/r128_ioc32.c
index b0d0fd3e4376..663f38c63ba6 100644
--- a/drivers/gpu/drm/r128/r128_ioc32.c
+++ b/drivers/gpu/drm/r128/r128_ioc32.c
@@ -203,7 +203,7 @@ long r128_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 	if (nr < DRM_COMMAND_BASE)
 		return drm_compat_ioctl(filp, cmd, arg);
 
-	if (nr < DRM_COMMAND_BASE + DRM_ARRAY_SIZE(r128_compat_ioctls))
+	if (nr < DRM_COMMAND_BASE + ARRAY_SIZE(r128_compat_ioctls))
 		fn = r128_compat_ioctls[nr - DRM_COMMAND_BASE];
 
 	if (fn != NULL)
diff --git a/drivers/gpu/drm/r128/r128_state.c b/drivers/gpu/drm/r128/r128_state.c
index e806dacd452f..575e986f82a7 100644
--- a/drivers/gpu/drm/r128/r128_state.c
+++ b/drivers/gpu/drm/r128/r128_state.c
@@ -1594,7 +1594,7 @@ static int r128_getparam(struct drm_device *dev, void *data, struct drm_file *fi
 
 	switch (param->param) {
 	case R128_PARAM_IRQ_NR:
-		value = drm_dev_to_irq(dev);
+		value = dev->pdev->irq;
 		break;
 	default:
 		return -EINVAL;
@@ -1641,4 +1641,4 @@ const struct drm_ioctl_desc r128_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(R128_GETPARAM, r128_getparam, DRM_AUTH),
 };
 
-int r128_max_ioctl = DRM_ARRAY_SIZE(r128_ioctls);
+int r128_max_ioctl = ARRAY_SIZE(r128_ioctls);
diff --git a/drivers/gpu/drm/radeon/Makefile b/drivers/gpu/drm/radeon/Makefile
index 09433534dc47..dbcbfe80aac0 100644
--- a/drivers/gpu/drm/radeon/Makefile
+++ b/drivers/gpu/drm/radeon/Makefile
@@ -72,7 +72,7 @@ radeon-y += radeon_device.o radeon_asic.o radeon_kms.o \
 	radeon_cs.o radeon_bios.o radeon_benchmark.o r100.o r300.o r420.o \
 	rs400.o rs600.o rs690.o rv515.o r520.o r600.o rv770.o radeon_test.o \
 	r200.o radeon_legacy_tv.o r600_cs.o r600_blit_shaders.o \
-	radeon_pm.o atombios_dp.o r600_audio.o r600_hdmi.o \
+	radeon_pm.o atombios_dp.o r600_audio.o r600_hdmi.o dce3_1_afmt.o \
 	evergreen.o evergreen_cs.o evergreen_blit_shaders.o \
 	evergreen_hdmi.o radeon_trace_points.o ni.o cayman_blit_shaders.o \
 	atombios_encoders.o radeon_semaphore.o radeon_sa.o atombios_i2c.o si.o \
diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c
index e911898348f8..26c12a3fe430 100644
--- a/drivers/gpu/drm/radeon/atombios_crtc.c
+++ b/drivers/gpu/drm/radeon/atombios_crtc.c
@@ -557,6 +557,7 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc,
 	u32 adjusted_clock = mode->clock;
 	int encoder_mode = atombios_get_encoder_mode(encoder);
 	u32 dp_clock = mode->clock;
+	u32 clock = mode->clock;
 	int bpc = radeon_crtc->bpc;
 	bool is_duallink = radeon_dig_monitor_is_duallink(encoder, mode->clock);
 
@@ -632,6 +633,24 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc,
 			radeon_crtc->pll_flags |= RADEON_PLL_USE_REF_DIV;
 	}
 
+	/* adjust pll for deep color modes */
+	if (encoder_mode == ATOM_ENCODER_MODE_HDMI) {
+		switch (bpc) {
+		case 8:
+		default:
+			break;
+		case 10:
+			clock = (clock * 5) / 4;
+			break;
+		case 12:
+			clock = (clock * 3) / 2;
+			break;
+		case 16:
+			clock = clock * 2;
+			break;
+		}
+	}
+
 	/* DCE3+ has an AdjustDisplayPll that will adjust the pixel clock
 	 * accordingly based on the encoder/transmitter to work around
 	 * special hw requirements.
@@ -653,7 +672,7 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc,
 			switch (crev) {
 			case 1:
 			case 2:
-				args.v1.usPixelClock = cpu_to_le16(mode->clock / 10);
+				args.v1.usPixelClock = cpu_to_le16(clock / 10);
 				args.v1.ucTransmitterID = radeon_encoder->encoder_id;
 				args.v1.ucEncodeMode = encoder_mode;
 				if (radeon_crtc->ss_enabled && radeon_crtc->ss.percentage)
@@ -665,7 +684,7 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc,
 				adjusted_clock = le16_to_cpu(args.v1.usPixelClock) * 10;
 				break;
 			case 3:
-				args.v3.sInput.usPixelClock = cpu_to_le16(mode->clock / 10);
+				args.v3.sInput.usPixelClock = cpu_to_le16(clock / 10);
 				args.v3.sInput.ucTransmitterID = radeon_encoder->encoder_id;
 				args.v3.sInput.ucEncodeMode = encoder_mode;
 				args.v3.sInput.ucDispPllConfig = 0;
@@ -679,10 +698,6 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc,
 					args.v3.sInput.usPixelClock = cpu_to_le16(dp_clock / 10);
 				} else if (radeon_encoder->devices & (ATOM_DEVICE_DFP_SUPPORT)) {
 					struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
-					if (encoder_mode == ATOM_ENCODER_MODE_HDMI)
-						/* deep color support */
-						args.v3.sInput.usPixelClock =
-							cpu_to_le16((mode->clock * bpc / 8) / 10);
 					if (dig->coherent_mode)
 						args.v3.sInput.ucDispPllConfig |=
 							DISPPLL_CONFIG_COHERENT_MODE;
@@ -862,14 +877,21 @@ static void atombios_crtc_program_pll(struct drm_crtc *crtc,
 			args.v5.ucMiscInfo = 0; /* HDMI depth, etc. */
 			if (ss_enabled && (ss->type & ATOM_EXTERNAL_SS_MASK))
 				args.v5.ucMiscInfo |= PIXEL_CLOCK_V5_MISC_REF_DIV_SRC;
-			switch (bpc) {
-			case 8:
-			default:
-				args.v5.ucMiscInfo |= PIXEL_CLOCK_V5_MISC_HDMI_24BPP;
-				break;
-			case 10:
-				args.v5.ucMiscInfo |= PIXEL_CLOCK_V5_MISC_HDMI_30BPP;
-				break;
+			if (encoder_mode == ATOM_ENCODER_MODE_HDMI) {
+				switch (bpc) {
+				case 8:
+				default:
+					args.v5.ucMiscInfo |= PIXEL_CLOCK_V5_MISC_HDMI_24BPP;
+					break;
+				case 10:
+					/* yes this is correct, the atom define is wrong */
+					args.v5.ucMiscInfo |= PIXEL_CLOCK_V5_MISC_HDMI_32BPP;
+					break;
+				case 12:
+					/* yes this is correct, the atom define is wrong */
+					args.v5.ucMiscInfo |= PIXEL_CLOCK_V5_MISC_HDMI_30BPP;
+					break;
+				}
 			}
 			args.v5.ucTransmitterID = encoder_id;
 			args.v5.ucEncoderMode = encoder_mode;
@@ -884,20 +906,22 @@ static void atombios_crtc_program_pll(struct drm_crtc *crtc,
 			args.v6.ucMiscInfo = 0; /* HDMI depth, etc. */
 			if (ss_enabled && (ss->type & ATOM_EXTERNAL_SS_MASK))
 				args.v6.ucMiscInfo |= PIXEL_CLOCK_V6_MISC_REF_DIV_SRC;
-			switch (bpc) {
-			case 8:
-			default:
-				args.v6.ucMiscInfo |= PIXEL_CLOCK_V6_MISC_HDMI_24BPP;
-				break;
-			case 10:
-				args.v6.ucMiscInfo |= PIXEL_CLOCK_V6_MISC_HDMI_30BPP;
-				break;
-			case 12:
-				args.v6.ucMiscInfo |= PIXEL_CLOCK_V6_MISC_HDMI_36BPP;
-				break;
-			case 16:
-				args.v6.ucMiscInfo |= PIXEL_CLOCK_V6_MISC_HDMI_48BPP;
-				break;
+			if (encoder_mode == ATOM_ENCODER_MODE_HDMI) {
+				switch (bpc) {
+				case 8:
+				default:
+					args.v6.ucMiscInfo |= PIXEL_CLOCK_V6_MISC_HDMI_24BPP;
+					break;
+				case 10:
+					args.v6.ucMiscInfo |= PIXEL_CLOCK_V6_MISC_HDMI_30BPP_V6;
+					break;
+				case 12:
+					args.v6.ucMiscInfo |= PIXEL_CLOCK_V6_MISC_HDMI_36BPP_V6;
+					break;
+				case 16:
+					args.v6.ucMiscInfo |= PIXEL_CLOCK_V6_MISC_HDMI_48BPP;
+					break;
+				}
 			}
 			args.v6.ucTransmitterID = encoder_id;
 			args.v6.ucEncoderMode = encoder_mode;
@@ -938,6 +962,9 @@ static bool atombios_crtc_prepare_pll(struct drm_crtc *crtc, struct drm_display_
 		struct radeon_connector_atom_dig *dig_connector =
 			radeon_connector->con_priv;
 		int dp_clock;
+
+		/* Assign mode clock for hdmi deep color max clock limit check */
+		radeon_connector->pixelclock_for_modeset = mode->clock;
 		radeon_crtc->bpc = radeon_get_monitor_bpc(connector);
 
 		switch (encoder_mode) {
@@ -1019,10 +1046,17 @@ static void atombios_crtc_set_pll(struct drm_crtc *crtc, struct drm_display_mode
 	struct radeon_encoder *radeon_encoder =
 		to_radeon_encoder(radeon_crtc->encoder);
 	u32 pll_clock = mode->clock;
+	u32 clock = mode->clock;
 	u32 ref_div = 0, fb_div = 0, frac_fb_div = 0, post_div = 0;
 	struct radeon_pll *pll;
 	int encoder_mode = atombios_get_encoder_mode(radeon_crtc->encoder);
 
+	/* pass the actual clock to atombios_crtc_program_pll for DCE5,6 for HDMI */
+	if (ASIC_IS_DCE5(rdev) && !ASIC_IS_DCE8(rdev) &&
+	    (encoder_mode == ATOM_ENCODER_MODE_HDMI) &&
+	    (radeon_crtc->bpc > 8))
+		clock = radeon_crtc->adjusted_clock;
+
 	switch (radeon_crtc->pll_id) {
 	case ATOM_PPLL1:
 		pll = &rdev->clock.p1pll;
@@ -1057,7 +1091,7 @@ static void atombios_crtc_set_pll(struct drm_crtc *crtc, struct drm_display_mode
 				 radeon_crtc->crtc_id, &radeon_crtc->ss);
 
 	atombios_crtc_program_pll(crtc, radeon_crtc->crtc_id, radeon_crtc->pll_id,
-				  encoder_mode, radeon_encoder->encoder_id, mode->clock,
+				  encoder_mode, radeon_encoder->encoder_id, clock,
 				  ref_div, fb_div, frac_fb_div, post_div,
 				  radeon_crtc->bpc, radeon_crtc->ss_enabled, &radeon_crtc->ss);
 
diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c
index 54e4f52549af..c5b1f2da3954 100644
--- a/drivers/gpu/drm/radeon/atombios_dp.c
+++ b/drivers/gpu/drm/radeon/atombios_dp.c
@@ -95,9 +95,12 @@ static int radeon_process_aux_ch(struct radeon_i2c_chan *chan,
 	int index = GetIndexIntoMasterTable(COMMAND, ProcessAuxChannelTransaction);
 	unsigned char *base;
 	int recv_bytes;
+	int r = 0;
 
 	memset(&args, 0, sizeof(args));
 
+	mutex_lock(&chan->mutex);
+
 	base = (unsigned char *)(rdev->mode_info.atom_context->scratch + 1);
 
 	radeon_atom_copy_swap(base, send, send_bytes, true);
@@ -117,19 +120,22 @@ static int radeon_process_aux_ch(struct radeon_i2c_chan *chan,
 	/* timeout */
 	if (args.v1.ucReplyStatus == 1) {
 		DRM_DEBUG_KMS("dp_aux_ch timeout\n");
-		return -ETIMEDOUT;
+		r = -ETIMEDOUT;
+		goto done;
 	}
 
 	/* flags not zero */
 	if (args.v1.ucReplyStatus == 2) {
 		DRM_DEBUG_KMS("dp_aux_ch flags not zero\n");
-		return -EBUSY;
+		r = -EBUSY;
+		goto done;
 	}
 
 	/* error */
 	if (args.v1.ucReplyStatus == 3) {
 		DRM_DEBUG_KMS("dp_aux_ch error\n");
-		return -EIO;
+		r = -EIO;
+		goto done;
 	}
 
 	recv_bytes = args.v1.ucDataOutLen;
@@ -139,7 +145,11 @@ static int radeon_process_aux_ch(struct radeon_i2c_chan *chan,
 	if (recv && recv_size)
 		radeon_atom_copy_swap(recv, base + 16, recv_bytes, false);
 
-	return recv_bytes;
+	r = recv_bytes;
+done:
+	mutex_unlock(&chan->mutex);
+
+	return r;
 }
 
 #define BARE_ADDRESS_SIZE 3
@@ -212,11 +222,12 @@ void radeon_dp_aux_init(struct radeon_connector *radeon_connector)
 	radeon_connector->ddc_bus->rec.hpd = radeon_connector->hpd.hpd;
 	radeon_connector->ddc_bus->aux.dev = radeon_connector->base.kdev;
 	radeon_connector->ddc_bus->aux.transfer = radeon_dp_aux_transfer;
-	ret = drm_dp_aux_register_i2c_bus(&radeon_connector->ddc_bus->aux);
+
+	ret = drm_dp_aux_register(&radeon_connector->ddc_bus->aux);
 	if (!ret)
 		radeon_connector->ddc_bus->has_aux = true;
 
-	WARN(ret, "drm_dp_aux_register_i2c_bus() failed with error %d\n", ret);
+	WARN(ret, "drm_dp_aux_register() failed with error %d\n", ret);
 }
 
 /***** general DP utility functions *****/
@@ -281,6 +292,19 @@ static int dp_get_max_dp_pix_clock(int link_rate,
 
 /***** radeon specific DP functions *****/
 
+static int radeon_dp_get_max_link_rate(struct drm_connector *connector,
+				       u8 dpcd[DP_DPCD_SIZE])
+{
+	int max_link_rate;
+
+	if (radeon_connector_is_dp12_capable(connector))
+		max_link_rate = min(drm_dp_max_link_rate(dpcd), 540000);
+	else
+		max_link_rate = min(drm_dp_max_link_rate(dpcd), 270000);
+
+	return max_link_rate;
+}
+
 /* First get the min lane# when low rate is used according to pixel clock
  * (prefer low rate), second check max lane# supported by DP panel,
  * if the max lane# < low rate lane# then use max lane# instead.
@@ -290,7 +314,7 @@ static int radeon_dp_get_dp_lane_number(struct drm_connector *connector,
 					int pix_clock)
 {
 	int bpp = convert_bpc_to_bpp(radeon_get_monitor_bpc(connector));
-	int max_link_rate = drm_dp_max_link_rate(dpcd);
+	int max_link_rate = radeon_dp_get_max_link_rate(connector, dpcd);
 	int max_lane_num = drm_dp_max_lane_count(dpcd);
 	int lane_num;
 	int max_dp_pix_clock;
@@ -328,7 +352,7 @@ static int radeon_dp_get_dp_link_clock(struct drm_connector *connector,
 			return 540000;
 	}
 
-	return drm_dp_max_link_rate(dpcd);
+	return radeon_dp_get_max_link_rate(connector, dpcd);
 }
 
 static u8 radeon_dp_encoder_service(struct radeon_device *rdev,
diff --git a/drivers/gpu/drm/radeon/atombios_encoders.c b/drivers/gpu/drm/radeon/atombios_encoders.c
index e6eb5097597f..2b2908440644 100644
--- a/drivers/gpu/drm/radeon/atombios_encoders.c
+++ b/drivers/gpu/drm/radeon/atombios_encoders.c
@@ -1884,8 +1884,11 @@ atombios_set_encoder_crtc_source(struct drm_encoder *encoder)
 					args.v2.ucEncodeMode = ATOM_ENCODER_MODE_CRT;
 				else
 					args.v2.ucEncodeMode = atombios_get_encoder_mode(encoder);
-			} else
+			} else if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) {
+				args.v2.ucEncodeMode = ATOM_ENCODER_MODE_LVDS;
+			} else {
 				args.v2.ucEncodeMode = atombios_get_encoder_mode(encoder);
+			}
 			switch (radeon_encoder->encoder_id) {
 			case ENCODER_OBJECT_ID_INTERNAL_UNIPHY:
 			case ENCODER_OBJECT_ID_INTERNAL_UNIPHY1:
diff --git a/drivers/gpu/drm/radeon/atombios_i2c.c b/drivers/gpu/drm/radeon/atombios_i2c.c
index b5162c3b6111..9c570fb15b8c 100644
--- a/drivers/gpu/drm/radeon/atombios_i2c.c
+++ b/drivers/gpu/drm/radeon/atombios_i2c.c
@@ -43,15 +43,19 @@ static int radeon_process_i2c_ch(struct radeon_i2c_chan *chan,
 	int index = GetIndexIntoMasterTable(COMMAND, ProcessI2cChannelTransaction);
 	unsigned char *base;
 	u16 out = cpu_to_le16(0);
+	int r = 0;
 
 	memset(&args, 0, sizeof(args));
 
+	mutex_lock(&chan->mutex);
+
 	base = (unsigned char *)rdev->mode_info.atom_context->scratch;
 
 	if (flags & HW_I2C_WRITE) {
 		if (num > ATOM_MAX_HW_I2C_WRITE) {
 			DRM_ERROR("hw i2c: tried to write too many bytes (%d vs 3)\n", num);
-			return -EINVAL;
+			r = -EINVAL;
+			goto done;
 		}
 		if (buf == NULL)
 			args.ucRegIndex = 0;
@@ -65,7 +69,8 @@ static int radeon_process_i2c_ch(struct radeon_i2c_chan *chan,
 	} else {
 		if (num > ATOM_MAX_HW_I2C_READ) {
 			DRM_ERROR("hw i2c: tried to read too many bytes (%d vs 255)\n", num);
-			return -EINVAL;
+			r = -EINVAL;
+			goto done;
 		}
 		args.ucRegIndex = 0;
 		args.lpI2CDataOut = 0;
@@ -82,13 +87,17 @@ static int radeon_process_i2c_ch(struct radeon_i2c_chan *chan,
 	/* error */
 	if (args.ucStatus != HW_ASSISTED_I2C_STATUS_SUCCESS) {
 		DRM_DEBUG_KMS("hw_i2c error\n");
-		return -EIO;
+		r = -EIO;
+		goto done;
 	}
 
 	if (!(flags & HW_I2C_WRITE))
 		radeon_atom_copy_swap(buf, base, num, false);
 
-	return 0;
+done:
+	mutex_unlock(&chan->mutex);
+
+	return r;
 }
 
 int radeon_atom_hw_i2c_xfer(struct i2c_adapter *i2c_adap,
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index d2fd98968085..dcd4518a9b08 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -80,6 +80,7 @@ extern int sumo_rlc_init(struct radeon_device *rdev);
 extern void si_vram_gtt_location(struct radeon_device *rdev, struct radeon_mc *mc);
 extern void si_rlc_reset(struct radeon_device *rdev);
 extern void si_init_uvd_internal_cg(struct radeon_device *rdev);
+static u32 cik_get_cu_active_bitmap(struct radeon_device *rdev, u32 se, u32 sh);
 extern int cik_sdma_resume(struct radeon_device *rdev);
 extern void cik_sdma_enable(struct radeon_device *rdev, bool enable);
 extern void cik_sdma_fini(struct radeon_device *rdev);
@@ -3257,7 +3258,7 @@ static void cik_gpu_init(struct radeon_device *rdev)
 	u32 mc_shared_chmap, mc_arb_ramcfg;
 	u32 hdp_host_path_cntl;
 	u32 tmp;
-	int i, j;
+	int i, j, k;
 
 	switch (rdev->family) {
 	case CHIP_BONAIRE:
@@ -3446,6 +3447,15 @@ static void cik_gpu_init(struct radeon_device *rdev)
 		     rdev->config.cik.max_sh_per_se,
 		     rdev->config.cik.max_backends_per_se);
 
+	for (i = 0; i < rdev->config.cik.max_shader_engines; i++) {
+		for (j = 0; j < rdev->config.cik.max_sh_per_se; j++) {
+			for (k = 0; k < rdev->config.cik.max_cu_per_sh; k++) {
+				rdev->config.cik.active_cus +=
+					hweight32(cik_get_cu_active_bitmap(rdev, i, j));
+			}
+		}
+	}
+
 	/* set HW defaults for 3D engine */
 	WREG32(CP_MEQ_THRESHOLDS, MEQ1_START(0x30) | MEQ2_START(0x60));
 
@@ -3698,7 +3708,7 @@ bool cik_semaphore_ring_emit(struct radeon_device *rdev,
 	unsigned sel = emit_wait ? PACKET3_SEM_SEL_WAIT : PACKET3_SEM_SEL_SIGNAL;
 
 	radeon_ring_write(ring, PACKET3(PACKET3_MEM_SEMAPHORE, 1));
-	radeon_ring_write(ring, addr & 0xffffffff);
+	radeon_ring_write(ring, lower_32_bits(addr));
 	radeon_ring_write(ring, (upper_32_bits(addr) & 0xffff) | sel);
 
 	return true;
@@ -3818,7 +3828,7 @@ void cik_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib)
 			radeon_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
 			radeon_ring_write(ring, WRITE_DATA_DST_SEL(1));
 			radeon_ring_write(ring, ring->next_rptr_gpu_addr & 0xfffffffc);
-			radeon_ring_write(ring, upper_32_bits(ring->next_rptr_gpu_addr) & 0xffffffff);
+			radeon_ring_write(ring, upper_32_bits(ring->next_rptr_gpu_addr));
 			radeon_ring_write(ring, next_rptr);
 		}
 
@@ -5396,6 +5406,7 @@ static int cik_pcie_gart_enable(struct radeon_device *rdev)
 	WREG32(MC_VM_MX_L1_TLB_CNTL,
 	       (0xA << 7) |
 	       ENABLE_L1_TLB |
+	       ENABLE_L1_FRAGMENT_PROCESSING |
 	       SYSTEM_ACCESS_MODE_NOT_IN_SYS |
 	       ENABLE_ADVANCED_DRIVER_MODEL |
 	       SYSTEM_APERTURE_UNMAPPED_ACCESS_PASS_THRU);
@@ -5408,7 +5419,8 @@ static int cik_pcie_gart_enable(struct radeon_device *rdev)
 	       CONTEXT1_IDENTITY_ACCESS_MODE(1));
 	WREG32(VM_L2_CNTL2, INVALIDATE_ALL_L1_TLBS | INVALIDATE_L2_CACHE);
 	WREG32(VM_L2_CNTL3, L2_CACHE_BIGK_ASSOCIATIVITY |
-	       L2_CACHE_BIGK_FRAGMENT_SIZE(6));
+	       BANK_SELECT(4) |
+	       L2_CACHE_BIGK_FRAGMENT_SIZE(4));
 	/* setup context0 */
 	WREG32(VM_CONTEXT0_PAGE_TABLE_START_ADDR, rdev->mc.gtt_start >> 12);
 	WREG32(VM_CONTEXT0_PAGE_TABLE_END_ADDR, rdev->mc.gtt_end >> 12);
@@ -5444,6 +5456,7 @@ static int cik_pcie_gart_enable(struct radeon_device *rdev)
 	       (u32)(rdev->dummy_page.addr >> 12));
 	WREG32(VM_CONTEXT1_CNTL2, 4);
 	WREG32(VM_CONTEXT1_CNTL, ENABLE_CONTEXT | PAGE_TABLE_DEPTH(1) |
+				PAGE_TABLE_BLOCK_SIZE(radeon_vm_block_size - 9) |
 				RANGE_PROTECTION_FAULT_ENABLE_INTERRUPT |
 				RANGE_PROTECTION_FAULT_ENABLE_DEFAULT |
 				DUMMY_PAGE_PROTECTION_FAULT_ENABLE_INTERRUPT |
@@ -7450,7 +7463,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[0]))
-						radeon_crtc_handle_flip(rdev, 0);
+						radeon_crtc_handle_vblank(rdev, 0);
 					rdev->irq.stat_regs.cik.disp_int &= ~LB_D1_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D1 vblank\n");
 				}
@@ -7476,7 +7489,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[1]))
-						radeon_crtc_handle_flip(rdev, 1);
+						radeon_crtc_handle_vblank(rdev, 1);
 					rdev->irq.stat_regs.cik.disp_int_cont &= ~LB_D2_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D2 vblank\n");
 				}
@@ -7502,7 +7515,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[2]))
-						radeon_crtc_handle_flip(rdev, 2);
+						radeon_crtc_handle_vblank(rdev, 2);
 					rdev->irq.stat_regs.cik.disp_int_cont2 &= ~LB_D3_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D3 vblank\n");
 				}
@@ -7528,7 +7541,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[3]))
-						radeon_crtc_handle_flip(rdev, 3);
+						radeon_crtc_handle_vblank(rdev, 3);
 					rdev->irq.stat_regs.cik.disp_int_cont3 &= ~LB_D4_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D4 vblank\n");
 				}
@@ -7554,7 +7567,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[4]))
-						radeon_crtc_handle_flip(rdev, 4);
+						radeon_crtc_handle_vblank(rdev, 4);
 					rdev->irq.stat_regs.cik.disp_int_cont4 &= ~LB_D5_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D5 vblank\n");
 				}
@@ -7580,7 +7593,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[5]))
-						radeon_crtc_handle_flip(rdev, 5);
+						radeon_crtc_handle_vblank(rdev, 5);
 					rdev->irq.stat_regs.cik.disp_int_cont5 &= ~LB_D6_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D6 vblank\n");
 				}
diff --git a/drivers/gpu/drm/radeon/cik_sdma.c b/drivers/gpu/drm/radeon/cik_sdma.c
index 72e464c79a88..8e9d0f1d858e 100644
--- a/drivers/gpu/drm/radeon/cik_sdma.c
+++ b/drivers/gpu/drm/radeon/cik_sdma.c
@@ -141,7 +141,7 @@ void cik_sdma_ring_ib_execute(struct radeon_device *rdev,
 		next_rptr += 4;
 		radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_WRITE, SDMA_WRITE_SUB_OPCODE_LINEAR, 0));
 		radeon_ring_write(ring, ring->next_rptr_gpu_addr & 0xfffffffc);
-		radeon_ring_write(ring, upper_32_bits(ring->next_rptr_gpu_addr) & 0xffffffff);
+		radeon_ring_write(ring, upper_32_bits(ring->next_rptr_gpu_addr));
 		radeon_ring_write(ring, 1); /* number of DWs to follow */
 		radeon_ring_write(ring, next_rptr);
 	}
@@ -151,7 +151,7 @@ void cik_sdma_ring_ib_execute(struct radeon_device *rdev,
 		radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_NOP, 0, 0));
 	radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_INDIRECT_BUFFER, 0, extra_bits));
 	radeon_ring_write(ring, ib->gpu_addr & 0xffffffe0); /* base must be 32 byte aligned */
-	radeon_ring_write(ring, upper_32_bits(ib->gpu_addr) & 0xffffffff);
+	radeon_ring_write(ring, upper_32_bits(ib->gpu_addr));
 	radeon_ring_write(ring, ib->length_dw);
 
 }
@@ -203,8 +203,8 @@ void cik_sdma_fence_ring_emit(struct radeon_device *rdev,
 
 	/* write the fence */
 	radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_FENCE, 0, 0));
-	radeon_ring_write(ring, addr & 0xffffffff);
-	radeon_ring_write(ring, upper_32_bits(addr) & 0xffffffff);
+	radeon_ring_write(ring, lower_32_bits(addr));
+	radeon_ring_write(ring, upper_32_bits(addr));
 	radeon_ring_write(ring, fence->seq);
 	/* generate an interrupt */
 	radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_TRAP, 0, 0));
@@ -233,7 +233,7 @@ bool cik_sdma_semaphore_ring_emit(struct radeon_device *rdev,
 
 	radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_SEMAPHORE, 0, extra_bits));
 	radeon_ring_write(ring, addr & 0xfffffff8);
-	radeon_ring_write(ring, upper_32_bits(addr) & 0xffffffff);
+	radeon_ring_write(ring, upper_32_bits(addr));
 
 	return true;
 }
@@ -551,10 +551,10 @@ int cik_copy_dma(struct radeon_device *rdev,
 		radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_COPY, SDMA_COPY_SUB_OPCODE_LINEAR, 0));
 		radeon_ring_write(ring, cur_size_in_bytes);
 		radeon_ring_write(ring, 0); /* src/dst endian swap */
-		radeon_ring_write(ring, src_offset & 0xffffffff);
-		radeon_ring_write(ring, upper_32_bits(src_offset) & 0xffffffff);
-		radeon_ring_write(ring, dst_offset & 0xffffffff);
-		radeon_ring_write(ring, upper_32_bits(dst_offset) & 0xffffffff);
+		radeon_ring_write(ring, lower_32_bits(src_offset));
+		radeon_ring_write(ring, upper_32_bits(src_offset));
+		radeon_ring_write(ring, lower_32_bits(dst_offset));
+		radeon_ring_write(ring, upper_32_bits(dst_offset));
 		src_offset += cur_size_in_bytes;
 		dst_offset += cur_size_in_bytes;
 	}
@@ -605,7 +605,7 @@ int cik_sdma_ring_test(struct radeon_device *rdev,
 	}
 	radeon_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_WRITE, SDMA_WRITE_SUB_OPCODE_LINEAR, 0));
 	radeon_ring_write(ring, rdev->vram_scratch.gpu_addr & 0xfffffffc);
-	radeon_ring_write(ring, upper_32_bits(rdev->vram_scratch.gpu_addr) & 0xffffffff);
+	radeon_ring_write(ring, upper_32_bits(rdev->vram_scratch.gpu_addr));
 	radeon_ring_write(ring, 1); /* number of DWs to follow */
 	radeon_ring_write(ring, 0xDEADBEEF);
 	radeon_ring_unlock_commit(rdev, ring);
@@ -660,7 +660,7 @@ int cik_sdma_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
 
 	ib.ptr[0] = SDMA_PACKET(SDMA_OPCODE_WRITE, SDMA_WRITE_SUB_OPCODE_LINEAR, 0);
 	ib.ptr[1] = rdev->vram_scratch.gpu_addr & 0xfffffffc;
-	ib.ptr[2] = upper_32_bits(rdev->vram_scratch.gpu_addr) & 0xffffffff;
+	ib.ptr[2] = upper_32_bits(rdev->vram_scratch.gpu_addr);
 	ib.ptr[3] = 1;
 	ib.ptr[4] = 0xDEADBEEF;
 	ib.length_dw = 5;
@@ -742,7 +742,26 @@ void cik_sdma_vm_set_page(struct radeon_device *rdev,
 
 	trace_radeon_vm_set_page(pe, addr, count, incr, flags);
 
-	if (flags & R600_PTE_SYSTEM) {
+	if (flags == R600_PTE_GART) {
+		uint64_t src = rdev->gart.table_addr + (addr >> 12) * 8;
+		while (count) {
+			unsigned bytes = count * 8;
+			if (bytes > 0x1FFFF8)
+				bytes = 0x1FFFF8;
+
+			ib->ptr[ib->length_dw++] = SDMA_PACKET(SDMA_OPCODE_COPY, SDMA_WRITE_SUB_OPCODE_LINEAR, 0);
+			ib->ptr[ib->length_dw++] = bytes;
+			ib->ptr[ib->length_dw++] = 0; /* src/dst endian swap */
+			ib->ptr[ib->length_dw++] = lower_32_bits(src);
+			ib->ptr[ib->length_dw++] = upper_32_bits(src);
+			ib->ptr[ib->length_dw++] = lower_32_bits(pe);
+			ib->ptr[ib->length_dw++] = upper_32_bits(pe);
+
+			pe += bytes;
+			src += bytes;
+			count -= bytes / 8;
+		}
+	} else if (flags & R600_PTE_SYSTEM) {
 		while (count) {
 			ndw = count * 2;
 			if (ndw > 0xFFFFE)
diff --git a/drivers/gpu/drm/radeon/cikd.h b/drivers/gpu/drm/radeon/cikd.h
index dd7926394a8f..ae88660f34ea 100644
--- a/drivers/gpu/drm/radeon/cikd.h
+++ b/drivers/gpu/drm/radeon/cikd.h
@@ -482,6 +482,7 @@
 #define		READ_PROTECTION_FAULT_ENABLE_DEFAULT		(1 << 16)
 #define		WRITE_PROTECTION_FAULT_ENABLE_INTERRUPT		(1 << 18)
 #define		WRITE_PROTECTION_FAULT_ENABLE_DEFAULT		(1 << 19)
+#define		PAGE_TABLE_BLOCK_SIZE(x)			(((x) & 0xF) << 24)
 #define VM_CONTEXT1_CNTL				0x1414
 #define VM_CONTEXT0_CNTL2				0x1430
 #define VM_CONTEXT1_CNTL2				0x1434
diff --git a/drivers/gpu/drm/radeon/clearstate_cayman.h b/drivers/gpu/drm/radeon/clearstate_cayman.h
index aa908c55a513..e48a14037b76 100644
--- a/drivers/gpu/drm/radeon/clearstate_cayman.h
+++ b/drivers/gpu/drm/radeon/clearstate_cayman.h
@@ -1050,7 +1050,7 @@ static const struct cs_extent_def SECT_CONTEXT_defs[] =
     {SECT_CONTEXT_def_5, 0x0000a29e, 5 },
     {SECT_CONTEXT_def_6, 0x0000a2a5, 56 },
     {SECT_CONTEXT_def_7, 0x0000a2de, 290 },
-    { 0, 0, 0 }
+    { NULL, 0, 0 }
 };
 static const u32 SECT_CLEAR_def_1[] =
 {
@@ -1061,7 +1061,7 @@ static const u32 SECT_CLEAR_def_1[] =
 static const struct cs_extent_def SECT_CLEAR_defs[] =
 {
     {SECT_CLEAR_def_1, 0x0000ffc0, 3 },
-    { 0, 0, 0 }
+    { NULL, 0, 0 }
 };
 static const u32 SECT_CTRLCONST_def_1[] =
 {
@@ -1071,11 +1071,11 @@ static const u32 SECT_CTRLCONST_def_1[] =
 static const struct cs_extent_def SECT_CTRLCONST_defs[] =
 {
     {SECT_CTRLCONST_def_1, 0x0000f3fc, 2 },
-    { 0, 0, 0 }
+    { NULL, 0, 0 }
 };
 static const struct cs_section_def cayman_cs_data[] = {
     { SECT_CONTEXT_defs, SECT_CONTEXT },
     { SECT_CLEAR_defs, SECT_CLEAR },
     { SECT_CTRLCONST_defs, SECT_CTRLCONST },
-    { 0, SECT_NONE }
+    { NULL, SECT_NONE }
 };
diff --git a/drivers/gpu/drm/radeon/clearstate_ci.h b/drivers/gpu/drm/radeon/clearstate_ci.h
index c3982f9475fb..f55d06664e31 100644
--- a/drivers/gpu/drm/radeon/clearstate_ci.h
+++ b/drivers/gpu/drm/radeon/clearstate_ci.h
@@ -936,9 +936,9 @@ static const struct cs_extent_def ci_SECT_CONTEXT_defs[] =
     {ci_SECT_CONTEXT_def_5, 0x0000a2a0, 2 },
     {ci_SECT_CONTEXT_def_6, 0x0000a2a3, 1 },
     {ci_SECT_CONTEXT_def_7, 0x0000a2a5, 233 },
-    { 0, 0, 0 }
+    { NULL, 0, 0 }
 };
 static const struct cs_section_def ci_cs_data[] = {
     { ci_SECT_CONTEXT_defs, SECT_CONTEXT },
-    { 0, SECT_NONE }
+    { NULL, SECT_NONE }
 };
diff --git a/drivers/gpu/drm/radeon/clearstate_si.h b/drivers/gpu/drm/radeon/clearstate_si.h
index b994cb2a35a0..66e39cdb5cb0 100644
--- a/drivers/gpu/drm/radeon/clearstate_si.h
+++ b/drivers/gpu/drm/radeon/clearstate_si.h
@@ -933,9 +933,9 @@ static const struct cs_extent_def si_SECT_CONTEXT_defs[] =
     {si_SECT_CONTEXT_def_5, 0x0000a2a1, 1 },
     {si_SECT_CONTEXT_def_6, 0x0000a2a3, 1 },
     {si_SECT_CONTEXT_def_7, 0x0000a2a5, 233 },
-    { 0, 0, 0 }
+    { NULL, 0, 0 }
 };
 static const struct cs_section_def si_cs_data[] = {
     { si_SECT_CONTEXT_defs, SECT_CONTEXT },
-    { 0, SECT_NONE }
+    { NULL, SECT_NONE }
 };
diff --git a/drivers/gpu/drm/radeon/dce3_1_afmt.c b/drivers/gpu/drm/radeon/dce3_1_afmt.c
new file mode 100644
index 000000000000..51800e340a57
--- /dev/null
+++ b/drivers/gpu/drm/radeon/dce3_1_afmt.c
@@ -0,0 +1,244 @@
+/*
+ * Copyright 2013 Advanced Micro Devices, Inc.
+ * Copyright 2014 Rafał Miłecki
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+#include <linux/hdmi.h>
+#include <drm/drmP.h>
+#include "radeon.h"
+#include "radeon_asic.h"
+#include "r600d.h"
+
+static void dce3_2_afmt_write_speaker_allocation(struct drm_encoder *encoder)
+{
+	struct radeon_device *rdev = encoder->dev->dev_private;
+	struct drm_connector *connector;
+	struct radeon_connector *radeon_connector = NULL;
+	u32 tmp;
+	u8 *sadb;
+	int sad_count;
+
+	list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
+		if (connector->encoder == encoder) {
+			radeon_connector = to_radeon_connector(connector);
+			break;
+		}
+	}
+
+	if (!radeon_connector) {
+		DRM_ERROR("Couldn't find encoder's connector\n");
+		return;
+	}
+
+	sad_count = drm_edid_to_speaker_allocation(radeon_connector->edid, &sadb);
+	if (sad_count < 0) {
+		DRM_ERROR("Couldn't read Speaker Allocation Data Block: %d\n", sad_count);
+		return;
+	}
+
+	/* program the speaker allocation */
+	tmp = RREG32(AZ_F0_CODEC_PIN0_CONTROL_CHANNEL_SPEAKER);
+	tmp &= ~(DP_CONNECTION | SPEAKER_ALLOCATION_MASK);
+	/* set HDMI mode */
+	tmp |= HDMI_CONNECTION;
+	if (sad_count)
+		tmp |= SPEAKER_ALLOCATION(sadb[0]);
+	else
+		tmp |= SPEAKER_ALLOCATION(5); /* stereo */
+	WREG32(AZ_F0_CODEC_PIN0_CONTROL_CHANNEL_SPEAKER, tmp);
+
+	kfree(sadb);
+}
+
+static void dce3_2_afmt_write_sad_regs(struct drm_encoder *encoder)
+{
+	struct radeon_device *rdev = encoder->dev->dev_private;
+	struct drm_connector *connector;
+	struct radeon_connector *radeon_connector = NULL;
+	struct cea_sad *sads;
+	int i, sad_count;
+
+	static const u16 eld_reg_to_type[][2] = {
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR0, HDMI_AUDIO_CODING_TYPE_PCM },
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR1, HDMI_AUDIO_CODING_TYPE_AC3 },
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR2, HDMI_AUDIO_CODING_TYPE_MPEG1 },
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR3, HDMI_AUDIO_CODING_TYPE_MP3 },
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR4, HDMI_AUDIO_CODING_TYPE_MPEG2 },
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR5, HDMI_AUDIO_CODING_TYPE_AAC_LC },
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR6, HDMI_AUDIO_CODING_TYPE_DTS },
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR7, HDMI_AUDIO_CODING_TYPE_ATRAC },
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR9, HDMI_AUDIO_CODING_TYPE_EAC3 },
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR10, HDMI_AUDIO_CODING_TYPE_DTS_HD },
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR11, HDMI_AUDIO_CODING_TYPE_MLP },
+		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR13, HDMI_AUDIO_CODING_TYPE_WMA_PRO },
+	};
+
+	list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
+		if (connector->encoder == encoder) {
+			radeon_connector = to_radeon_connector(connector);
+			break;
+		}
+	}
+
+	if (!radeon_connector) {
+		DRM_ERROR("Couldn't find encoder's connector\n");
+		return;
+	}
+
+	sad_count = drm_edid_to_sad(radeon_connector->edid, &sads);
+	if (sad_count < 0) {
+		DRM_ERROR("Couldn't read SADs: %d\n", sad_count);
+		return;
+	}
+	BUG_ON(!sads);
+
+	for (i = 0; i < ARRAY_SIZE(eld_reg_to_type); i++) {
+		u32 value = 0;
+		u8 stereo_freqs = 0;
+		int max_channels = -1;
+		int j;
+
+		for (j = 0; j < sad_count; j++) {
+			struct cea_sad *sad = &sads[j];
+
+			if (sad->format == eld_reg_to_type[i][1]) {
+				if (sad->channels > max_channels) {
+					value = MAX_CHANNELS(sad->channels) |
+						DESCRIPTOR_BYTE_2(sad->byte2) |
+						SUPPORTED_FREQUENCIES(sad->freq);
+					max_channels = sad->channels;
+				}
+
+				if (sad->format == HDMI_AUDIO_CODING_TYPE_PCM)
+					stereo_freqs |= sad->freq;
+				else
+					break;
+			}
+		}
+
+		value |= SUPPORTED_FREQUENCIES_STEREO(stereo_freqs);
+
+		WREG32(eld_reg_to_type[i][0], value);
+	}
+
+	kfree(sads);
+}
+
+/*
+ * update the info frames with the data from the current display mode
+ */
+void dce3_1_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode *mode)
+{
+	struct drm_device *dev = encoder->dev;
+	struct radeon_device *rdev = dev->dev_private;
+	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
+	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
+	u8 buffer[HDMI_INFOFRAME_HEADER_SIZE + HDMI_AVI_INFOFRAME_SIZE];
+	struct hdmi_avi_infoframe frame;
+	uint32_t offset;
+	ssize_t err;
+
+	if (!dig || !dig->afmt)
+		return;
+
+	/* Silent, r600_hdmi_enable will raise WARN for us */
+	if (!dig->afmt->enabled)
+		return;
+	offset = dig->afmt->offset;
+
+	/* disable audio prior to setting up hw */
+	dig->afmt->pin = r600_audio_get_pin(rdev);
+	r600_audio_enable(rdev, dig->afmt->pin, false);
+
+	r600_audio_set_dto(encoder, mode->clock);
+
+	WREG32(HDMI0_VBI_PACKET_CONTROL + offset,
+	       HDMI0_NULL_SEND); /* send null packets when required */
+
+	WREG32(HDMI0_AUDIO_CRC_CONTROL + offset, 0x1000);
+
+	if (ASIC_IS_DCE32(rdev)) {
+		WREG32(HDMI0_AUDIO_PACKET_CONTROL + offset,
+		       HDMI0_AUDIO_DELAY_EN(1) | /* default audio delay */
+		       HDMI0_AUDIO_PACKETS_PER_LINE(3)); /* should be suffient for all audio modes and small enough for all hblanks */
+		WREG32(AFMT_AUDIO_PACKET_CONTROL + offset,
+		       AFMT_AUDIO_SAMPLE_SEND | /* send audio packets */
+		       AFMT_60958_CS_UPDATE); /* allow 60958 channel status fields to be updated */
+	} else {
+		WREG32(HDMI0_AUDIO_PACKET_CONTROL + offset,
+		       HDMI0_AUDIO_SAMPLE_SEND | /* send audio packets */
+		       HDMI0_AUDIO_DELAY_EN(1) | /* default audio delay */
+		       HDMI0_AUDIO_PACKETS_PER_LINE(3) | /* should be suffient for all audio modes and small enough for all hblanks */
+		       HDMI0_60958_CS_UPDATE); /* allow 60958 channel status fields to be updated */
+	}
+
+	if (ASIC_IS_DCE32(rdev)) {
+		dce3_2_afmt_write_speaker_allocation(encoder);
+		dce3_2_afmt_write_sad_regs(encoder);
+	}
+
+	WREG32(HDMI0_ACR_PACKET_CONTROL + offset,
+	       HDMI0_ACR_SOURCE | /* select SW CTS value - XXX verify that hw CTS works on all families */
+	       HDMI0_ACR_AUTO_SEND); /* allow hw to sent ACR packets when required */
+
+	WREG32(HDMI0_VBI_PACKET_CONTROL + offset,
+	       HDMI0_NULL_SEND | /* send null packets when required */
+	       HDMI0_GC_SEND | /* send general control packets */
+	       HDMI0_GC_CONT); /* send general control packets every frame */
+
+	/* TODO: HDMI0_AUDIO_INFO_UPDATE */
+	WREG32(HDMI0_INFOFRAME_CONTROL0 + offset,
+	       HDMI0_AVI_INFO_SEND | /* enable AVI info frames */
+	       HDMI0_AVI_INFO_CONT | /* send AVI info frames every frame/field */
+	       HDMI0_AUDIO_INFO_SEND | /* enable audio info frames (frames won't be set until audio is enabled) */
+	       HDMI0_AUDIO_INFO_CONT); /* send audio info frames every frame/field */
+
+	WREG32(HDMI0_INFOFRAME_CONTROL1 + offset,
+	       HDMI0_AVI_INFO_LINE(2) | /* anything other than 0 */
+	       HDMI0_AUDIO_INFO_LINE(2)); /* anything other than 0 */
+
+	WREG32(HDMI0_GC + offset, 0); /* unset HDMI0_GC_AVMUTE */
+
+	err = drm_hdmi_avi_infoframe_from_display_mode(&frame, mode);
+	if (err < 0) {
+		DRM_ERROR("failed to setup AVI infoframe: %zd\n", err);
+		return;
+	}
+
+	err = hdmi_avi_infoframe_pack(&frame, buffer, sizeof(buffer));
+	if (err < 0) {
+		DRM_ERROR("failed to pack AVI infoframe: %zd\n", err);
+		return;
+	}
+
+	r600_hdmi_update_avi_infoframe(encoder, buffer, sizeof(buffer));
+	r600_hdmi_update_ACR(encoder, mode->clock);
+
+	/* it's unknown what these bits do excatly, but it's indeed quite useful for debugging */
+	WREG32(HDMI0_RAMP_CONTROL0 + offset, 0x00FFFFFF);
+	WREG32(HDMI0_RAMP_CONTROL1 + offset, 0x007FFFFF);
+	WREG32(HDMI0_RAMP_CONTROL2 + offset, 0x00000001);
+	WREG32(HDMI0_RAMP_CONTROL3 + offset, 0x00000001);
+
+	r600_hdmi_audio_workaround(encoder);
+
+	/* enable audio after to setting up hw */
+	r600_audio_enable(rdev, dig->afmt->pin, true);
+}
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 0f7a51a3694f..e2f605224e8c 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -1301,36 +1301,6 @@ void dce4_wait_for_vblank(struct radeon_device *rdev, int crtc)
 }
 
 /**
- * radeon_irq_kms_pflip_irq_get - pre-pageflip callback.
- *
- * @rdev: radeon_device pointer
- * @crtc: crtc to prepare for pageflip on
- *
- * Pre-pageflip callback (evergreen+).
- * Enables the pageflip irq (vblank irq).
- */
-void evergreen_pre_page_flip(struct radeon_device *rdev, int crtc)
-{
-	/* enable the pflip int */
-	radeon_irq_kms_pflip_irq_get(rdev, crtc);
-}
-
-/**
- * evergreen_post_page_flip - pos-pageflip callback.
- *
- * @rdev: radeon_device pointer
- * @crtc: crtc to cleanup pageflip on
- *
- * Post-pageflip callback (evergreen+).
- * Disables the pageflip irq (vblank irq).
- */
-void evergreen_post_page_flip(struct radeon_device *rdev, int crtc)
-{
-	/* disable the pflip int */
-	radeon_irq_kms_pflip_irq_put(rdev, crtc);
-}
-
-/**
  * evergreen_page_flip - pageflip callback.
  *
  * @rdev: radeon_device pointer
@@ -1343,7 +1313,7 @@ void evergreen_post_page_flip(struct radeon_device *rdev, int crtc)
  * double buffered update to take place.
  * Returns the current update pending status.
  */
-u32 evergreen_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
+void evergreen_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 {
 	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
 	u32 tmp = RREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset);
@@ -1375,9 +1345,23 @@ u32 evergreen_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 	/* Unlock the lock, so double-buffering can take place inside vblank */
 	tmp &= ~EVERGREEN_GRPH_UPDATE_LOCK;
 	WREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset, tmp);
+}
+
+/**
+ * evergreen_page_flip_pending - check if page flip is still pending
+ *
+ * @rdev: radeon_device pointer
+ * @crtc_id: crtc to check
+ *
+ * Returns the current update pending status.
+ */
+bool evergreen_page_flip_pending(struct radeon_device *rdev, int crtc_id)
+{
+	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
 
 	/* Return current update_pending status: */
-	return RREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset) & EVERGREEN_GRPH_SURFACE_UPDATE_PENDING;
+	return !!(RREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset) &
+		EVERGREEN_GRPH_SURFACE_UPDATE_PENDING);
 }
 
 /* get temperature in millidegrees */
@@ -3353,6 +3337,18 @@ static void evergreen_gpu_init(struct radeon_device *rdev)
 			disabled_rb_mask &= ~(1 << i);
 	}
 
+	for (i = 0; i < rdev->config.evergreen.num_ses; i++) {
+		u32 simd_disable_bitmap;
+
+		WREG32(GRBM_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_INDEX(i));
+		WREG32(RLC_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_INDEX(i));
+		simd_disable_bitmap = (RREG32(CC_GC_SHADER_PIPE_CONFIG) & 0xffff0000) >> 16;
+		simd_disable_bitmap |= 0xffffffff << rdev->config.evergreen.max_simds;
+		tmp <<= 16;
+		tmp |= simd_disable_bitmap;
+	}
+	rdev->config.evergreen.active_simds = hweight32(~tmp);
+
 	WREG32(GRBM_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_BROADCAST_WRITES);
 	WREG32(RLC_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_BROADCAST_WRITES);
 
@@ -4810,7 +4806,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[0]))
-						radeon_crtc_handle_flip(rdev, 0);
+						radeon_crtc_handle_vblank(rdev, 0);
 					rdev->irq.stat_regs.evergreen.disp_int &= ~LB_D1_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D1 vblank\n");
 				}
@@ -4836,7 +4832,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[1]))
-						radeon_crtc_handle_flip(rdev, 1);
+						radeon_crtc_handle_vblank(rdev, 1);
 					rdev->irq.stat_regs.evergreen.disp_int_cont &= ~LB_D2_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D2 vblank\n");
 				}
@@ -4862,7 +4858,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[2]))
-						radeon_crtc_handle_flip(rdev, 2);
+						radeon_crtc_handle_vblank(rdev, 2);
 					rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~LB_D3_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D3 vblank\n");
 				}
@@ -4888,7 +4884,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[3]))
-						radeon_crtc_handle_flip(rdev, 3);
+						radeon_crtc_handle_vblank(rdev, 3);
 					rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~LB_D4_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D4 vblank\n");
 				}
@@ -4914,7 +4910,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[4]))
-						radeon_crtc_handle_flip(rdev, 4);
+						radeon_crtc_handle_vblank(rdev, 4);
 					rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~LB_D5_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D5 vblank\n");
 				}
@@ -4940,7 +4936,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[5]))
-						radeon_crtc_handle_flip(rdev, 5);
+						radeon_crtc_handle_vblank(rdev, 5);
 					rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~LB_D6_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D6 vblank\n");
 				}
diff --git a/drivers/gpu/drm/radeon/evergreen_hdmi.c b/drivers/gpu/drm/radeon/evergreen_hdmi.c
index 05b0c95813fd..1ec0e6e83f9f 100644
--- a/drivers/gpu/drm/radeon/evergreen_hdmi.c
+++ b/drivers/gpu/drm/radeon/evergreen_hdmi.c
@@ -293,10 +293,13 @@ void evergreen_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode
 	struct radeon_device *rdev = dev->dev_private;
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
+	struct drm_connector *connector = radeon_get_connector_for_encoder(encoder);
 	u8 buffer[HDMI_INFOFRAME_HEADER_SIZE + HDMI_AVI_INFOFRAME_SIZE];
 	struct hdmi_avi_infoframe frame;
 	uint32_t offset;
 	ssize_t err;
+	uint32_t val;
+	int bpc = 8;
 
 	if (!dig || !dig->afmt)
 		return;
@@ -306,6 +309,12 @@ void evergreen_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode
 		return;
 	offset = dig->afmt->offset;
 
+	/* hdmi deep color mode general control packets setup, if bpc > 8 */
+	if (encoder->crtc) {
+		struct radeon_crtc *radeon_crtc = to_radeon_crtc(encoder->crtc);
+		bpc = radeon_crtc->bpc;
+	}
+
 	/* disable audio prior to setting up hw */
 	if (ASIC_IS_DCE6(rdev)) {
 		dig->afmt->pin = dce6_audio_get_pin(rdev);
@@ -322,6 +331,35 @@ void evergreen_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode
 
 	WREG32(AFMT_AUDIO_CRC_CONTROL + offset, 0x1000);
 
+	val = RREG32(HDMI_CONTROL + offset);
+	val &= ~HDMI_DEEP_COLOR_ENABLE;
+	val &= ~HDMI_DEEP_COLOR_DEPTH_MASK;
+
+	switch (bpc) {
+		case 0:
+		case 6:
+		case 8:
+		case 16:
+		default:
+			DRM_DEBUG("%s: Disabling hdmi deep color for %d bpc.\n",
+					 connector->name, bpc);
+			break;
+		case 10:
+			val |= HDMI_DEEP_COLOR_ENABLE;
+			val |= HDMI_DEEP_COLOR_DEPTH(HDMI_30BIT_DEEP_COLOR);
+			DRM_DEBUG("%s: Enabling hdmi deep color 30 for 10 bpc.\n",
+					 connector->name);
+			break;
+		case 12:
+			val |= HDMI_DEEP_COLOR_ENABLE;
+			val |= HDMI_DEEP_COLOR_DEPTH(HDMI_36BIT_DEEP_COLOR);
+			DRM_DEBUG("%s: Enabling hdmi deep color 36 for 12 bpc.\n",
+					 connector->name);
+			break;
+	}
+
+	WREG32(HDMI_CONTROL + offset, val);
+
 	WREG32(HDMI_VBI_PACKET_CONTROL + offset,
 	       HDMI_NULL_SEND | /* send null packets when required */
 	       HDMI_GC_SEND | /* send general control packets */
@@ -348,9 +386,13 @@ void evergreen_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode
 
 	/* fglrx clears sth in AFMT_AUDIO_PACKET_CONTROL2 here */
 
-	WREG32(HDMI_ACR_PACKET_CONTROL + offset,
-	       HDMI_ACR_SOURCE | /* select SW CTS value */
-	       HDMI_ACR_AUTO_SEND); /* allow hw to sent ACR packets when required */
+	if (bpc > 8)
+		WREG32(HDMI_ACR_PACKET_CONTROL + offset,
+		       HDMI_ACR_AUTO_SEND); /* allow hw to sent ACR packets when required */
+	else
+		WREG32(HDMI_ACR_PACKET_CONTROL + offset,
+		       HDMI_ACR_SOURCE | /* select SW CTS value */
+		       HDMI_ACR_AUTO_SEND); /* allow hw to sent ACR packets when required */
 
 	evergreen_hdmi_update_ACR(encoder, mode->clock);
 
diff --git a/drivers/gpu/drm/radeon/evergreend.h b/drivers/gpu/drm/radeon/evergreend.h
index f9c7963b3ee6..b066d6711b8d 100644
--- a/drivers/gpu/drm/radeon/evergreend.h
+++ b/drivers/gpu/drm/radeon/evergreend.h
@@ -517,10 +517,11 @@
 #       define HDMI_ERROR_ACK                (1 << 8)
 #       define HDMI_ERROR_MASK               (1 << 9)
 #       define HDMI_DEEP_COLOR_ENABLE        (1 << 24)
-#       define HDMI_DEEP_COLOR_DEPTH         (((x) & 3) << 28)
+#       define HDMI_DEEP_COLOR_DEPTH(x)      (((x) & 3) << 28)
 #       define HDMI_24BIT_DEEP_COLOR         0
 #       define HDMI_30BIT_DEEP_COLOR         1
 #       define HDMI_36BIT_DEEP_COLOR         2
+#       define HDMI_DEEP_COLOR_DEPTH_MASK    (3 << 28)
 #define HDMI_STATUS                          0x7034
 #       define HDMI_ACTIVE_AVMUTE            (1 << 0)
 #       define HDMI_AUDIO_PACKET_ERROR       (1 << 16)
diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c
index d246e043421a..5a33ca681867 100644
--- a/drivers/gpu/drm/radeon/ni.c
+++ b/drivers/gpu/drm/radeon/ni.c
@@ -1057,6 +1057,18 @@ static void cayman_gpu_init(struct radeon_device *rdev)
 			disabled_rb_mask &= ~(1 << i);
 	}
 
+	for (i = 0; i < rdev->config.cayman.max_shader_engines; i++) {
+		u32 simd_disable_bitmap;
+
+		WREG32(GRBM_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_INDEX(i));
+		WREG32(RLC_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_INDEX(i));
+		simd_disable_bitmap = (RREG32(CC_GC_SHADER_PIPE_CONFIG) & 0xffff0000) >> 16;
+		simd_disable_bitmap |= 0xffffffff << rdev->config.cayman.max_simds_per_se;
+		tmp <<= 16;
+		tmp |= simd_disable_bitmap;
+	}
+	rdev->config.cayman.active_simds = hweight32(~tmp);
+
 	WREG32(GRBM_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_BROADCAST_WRITES);
 	WREG32(RLC_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_BROADCAST_WRITES);
 
@@ -1228,12 +1240,14 @@ static int cayman_pcie_gart_enable(struct radeon_device *rdev)
 	       SYSTEM_APERTURE_UNMAPPED_ACCESS_PASS_THRU);
 	/* Setup L2 cache */
 	WREG32(VM_L2_CNTL, ENABLE_L2_CACHE |
+	       ENABLE_L2_FRAGMENT_PROCESSING |
 	       ENABLE_L2_PTE_CACHE_LRU_UPDATE_BY_WRITE |
 	       ENABLE_L2_PDE0_CACHE_LRU_UPDATE_BY_WRITE |
 	       EFFECTIVE_L2_QUEUE_SIZE(7) |
 	       CONTEXT1_IDENTITY_ACCESS_MODE(1));
 	WREG32(VM_L2_CNTL2, INVALIDATE_ALL_L1_TLBS | INVALIDATE_L2_CACHE);
 	WREG32(VM_L2_CNTL3, L2_CACHE_BIGK_ASSOCIATIVITY |
+	       BANK_SELECT(6) |
 	       L2_CACHE_BIGK_FRAGMENT_SIZE(6));
 	/* setup context0 */
 	WREG32(VM_CONTEXT0_PAGE_TABLE_START_ADDR, rdev->mc.gtt_start >> 12);
@@ -1266,6 +1280,7 @@ static int cayman_pcie_gart_enable(struct radeon_device *rdev)
 	       (u32)(rdev->dummy_page.addr >> 12));
 	WREG32(VM_CONTEXT1_CNTL2, 4);
 	WREG32(VM_CONTEXT1_CNTL, ENABLE_CONTEXT | PAGE_TABLE_DEPTH(1) |
+				PAGE_TABLE_BLOCK_SIZE(radeon_vm_block_size - 9) |
 				RANGE_PROTECTION_FAULT_ENABLE_INTERRUPT |
 				RANGE_PROTECTION_FAULT_ENABLE_DEFAULT |
 				DUMMY_PAGE_PROTECTION_FAULT_ENABLE_INTERRUPT |
@@ -1343,7 +1358,7 @@ void cayman_fence_ring_emit(struct radeon_device *rdev,
 	/* EVENT_WRITE_EOP - flush caches, send int */
 	radeon_ring_write(ring, PACKET3(PACKET3_EVENT_WRITE_EOP, 4));
 	radeon_ring_write(ring, EVENT_TYPE(CACHE_FLUSH_AND_INV_EVENT_TS) | EVENT_INDEX(5));
-	radeon_ring_write(ring, addr & 0xffffffff);
+	radeon_ring_write(ring, lower_32_bits(addr));
 	radeon_ring_write(ring, (upper_32_bits(addr) & 0xff) | DATA_SEL(1) | INT_SEL(2));
 	radeon_ring_write(ring, fence->seq);
 	radeon_ring_write(ring, 0);
diff --git a/drivers/gpu/drm/radeon/nid.h b/drivers/gpu/drm/radeon/nid.h
index d996033c243e..2e12e4d69253 100644
--- a/drivers/gpu/drm/radeon/nid.h
+++ b/drivers/gpu/drm/radeon/nid.h
@@ -128,6 +128,7 @@
 #define		READ_PROTECTION_FAULT_ENABLE_DEFAULT		(1 << 16)
 #define		WRITE_PROTECTION_FAULT_ENABLE_INTERRUPT		(1 << 18)
 #define		WRITE_PROTECTION_FAULT_ENABLE_DEFAULT		(1 << 19)
+#define		PAGE_TABLE_BLOCK_SIZE(x)			(((x) & 0xF) << 24)
 #define VM_CONTEXT1_CNTL				0x1414
 #define VM_CONTEXT0_CNTL2				0x1430
 #define VM_CONTEXT1_CNTL2				0x1434
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index b6c32640df20..1544efcf1c3a 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -142,36 +142,6 @@ void r100_wait_for_vblank(struct radeon_device *rdev, int crtc)
 }
 
 /**
- * r100_pre_page_flip - pre-pageflip callback.
- *
- * @rdev: radeon_device pointer
- * @crtc: crtc to prepare for pageflip on
- *
- * Pre-pageflip callback (r1xx-r4xx).
- * Enables the pageflip irq (vblank irq).
- */
-void r100_pre_page_flip(struct radeon_device *rdev, int crtc)
-{
-	/* enable the pflip int */
-	radeon_irq_kms_pflip_irq_get(rdev, crtc);
-}
-
-/**
- * r100_post_page_flip - pos-pageflip callback.
- *
- * @rdev: radeon_device pointer
- * @crtc: crtc to cleanup pageflip on
- *
- * Post-pageflip callback (r1xx-r4xx).
- * Disables the pageflip irq (vblank irq).
- */
-void r100_post_page_flip(struct radeon_device *rdev, int crtc)
-{
-	/* disable the pflip int */
-	radeon_irq_kms_pflip_irq_put(rdev, crtc);
-}
-
-/**
  * r100_page_flip - pageflip callback.
  *
  * @rdev: radeon_device pointer
@@ -182,9 +152,8 @@ void r100_post_page_flip(struct radeon_device *rdev, int crtc)
  * During vblank we take the crtc lock and wait for the update_pending
  * bit to go high, when it does, we release the lock, and allow the
  * double buffered update to take place.
- * Returns the current update pending status.
  */
-u32 r100_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
+void r100_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 {
 	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
 	u32 tmp = ((u32)crtc_base) | RADEON_CRTC_OFFSET__OFFSET_LOCK;
@@ -206,8 +175,24 @@ u32 r100_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 	tmp &= ~RADEON_CRTC_OFFSET__OFFSET_LOCK;
 	WREG32(RADEON_CRTC_OFFSET + radeon_crtc->crtc_offset, tmp);
 
+}
+
+/**
+ * r100_page_flip_pending - check if page flip is still pending
+ *
+ * @rdev: radeon_device pointer
+ * @crtc_id: crtc to check
+ *
+ * Check if the last pagefilp is still pending (r1xx-r4xx).
+ * Returns the current update pending status.
+ */
+bool r100_page_flip_pending(struct radeon_device *rdev, int crtc_id)
+{
+	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
+
 	/* Return current update_pending status: */
-	return RREG32(RADEON_CRTC_OFFSET + radeon_crtc->crtc_offset) & RADEON_CRTC_OFFSET__GUI_TRIG_OFFSET;
+	return !!(RREG32(RADEON_CRTC_OFFSET + radeon_crtc->crtc_offset) &
+		RADEON_CRTC_OFFSET__GUI_TRIG_OFFSET);
 }
 
 /**
@@ -697,15 +682,11 @@ void r100_pci_gart_disable(struct radeon_device *rdev)
 	WREG32(RADEON_AIC_HI_ADDR, 0);
 }
 
-int r100_pci_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr)
+void r100_pci_gart_set_page(struct radeon_device *rdev, unsigned i,
+			    uint64_t addr)
 {
 	u32 *gtt = rdev->gart.ptr;
-
-	if (i < 0 || i > rdev->gart.num_gpu_pages) {
-		return -EINVAL;
-	}
 	gtt[i] = cpu_to_le32(lower_32_bits(addr));
-	return 0;
 }
 
 void r100_pci_gart_fini(struct radeon_device *rdev)
@@ -794,7 +775,7 @@ int r100_irq_process(struct radeon_device *rdev)
 				wake_up(&rdev->irq.vblank_queue);
 			}
 			if (atomic_read(&rdev->irq.pflip[0]))
-				radeon_crtc_handle_flip(rdev, 0);
+				radeon_crtc_handle_vblank(rdev, 0);
 		}
 		if (status & RADEON_CRTC2_VBLANK_STAT) {
 			if (rdev->irq.crtc_vblank_int[1]) {
@@ -803,7 +784,7 @@ int r100_irq_process(struct radeon_device *rdev)
 				wake_up(&rdev->irq.vblank_queue);
 			}
 			if (atomic_read(&rdev->irq.pflip[1]))
-				radeon_crtc_handle_flip(rdev, 1);
+				radeon_crtc_handle_vblank(rdev, 1);
 		}
 		if (status & RADEON_FP_DETECT_STAT) {
 			queue_hotplug = true;
diff --git a/drivers/gpu/drm/radeon/r300.c b/drivers/gpu/drm/radeon/r300.c
index 206caf9700b7..3c21d77a483d 100644
--- a/drivers/gpu/drm/radeon/r300.c
+++ b/drivers/gpu/drm/radeon/r300.c
@@ -72,13 +72,11 @@ void rv370_pcie_gart_tlb_flush(struct radeon_device *rdev)
 #define R300_PTE_WRITEABLE (1 << 2)
 #define R300_PTE_READABLE  (1 << 3)
 
-int rv370_pcie_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr)
+void rv370_pcie_gart_set_page(struct radeon_device *rdev, unsigned i,
+			      uint64_t addr)
 {
 	void __iomem *ptr = rdev->gart.ptr;
 
-	if (i < 0 || i > rdev->gart.num_gpu_pages) {
-		return -EINVAL;
-	}
 	addr = (lower_32_bits(addr) >> 8) |
 	       ((upper_32_bits(addr) & 0xff) << 24) |
 	       R300_PTE_WRITEABLE | R300_PTE_READABLE;
@@ -86,7 +84,6 @@ int rv370_pcie_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr)
 	 * on powerpc without HW swappers, it'll get swapped on way
 	 * into VRAM - so no need for cpu_to_le32 on VRAM tables */
 	writel(addr, ((void __iomem *)ptr) + (i * 4));
-	return 0;
 }
 
 int rv370_pcie_gart_init(struct radeon_device *rdev)
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index bbc189fd3ddc..c66952d4b00c 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -1958,6 +1958,9 @@ static void r600_gpu_init(struct radeon_device *rdev)
 	if (tmp < rdev->config.r600.max_simds) {
 		rdev->config.r600.max_simds = tmp;
 	}
+	tmp = rdev->config.r600.max_simds -
+		r600_count_pipe_bits((cc_gc_shader_pipe_config >> 16) & R6XX_MAX_SIMDS_MASK);
+	rdev->config.r600.active_simds = tmp;
 
 	disabled_rb_mask = (RREG32(CC_RB_BACKEND_DISABLE) >> 16) & R6XX_MAX_BACKENDS_MASK;
 	tmp = (tiling_config & PIPE_TILING__MASK) >> PIPE_TILING__SHIFT;
@@ -2724,7 +2727,7 @@ void r600_fence_ring_emit(struct radeon_device *rdev,
 		/* EVENT_WRITE_EOP - flush caches, send int */
 		radeon_ring_write(ring, PACKET3(PACKET3_EVENT_WRITE_EOP, 4));
 		radeon_ring_write(ring, EVENT_TYPE(CACHE_FLUSH_AND_INV_EVENT_TS) | EVENT_INDEX(5));
-		radeon_ring_write(ring, addr & 0xffffffff);
+		radeon_ring_write(ring, lower_32_bits(addr));
 		radeon_ring_write(ring, (upper_32_bits(addr) & 0xff) | DATA_SEL(1) | INT_SEL(2));
 		radeon_ring_write(ring, fence->seq);
 		radeon_ring_write(ring, 0);
@@ -2763,7 +2766,7 @@ bool r600_semaphore_ring_emit(struct radeon_device *rdev,
 		sel |= PACKET3_SEM_WAIT_ON_SIGNAL;
 
 	radeon_ring_write(ring, PACKET3(PACKET3_MEM_SEMAPHORE, 1));
-	radeon_ring_write(ring, addr & 0xffffffff);
+	radeon_ring_write(ring, lower_32_bits(addr));
 	radeon_ring_write(ring, (upper_32_bits(addr) & 0xff) | sel);
 
 	return true;
@@ -2824,9 +2827,9 @@ int r600_copy_cpdma(struct radeon_device *rdev,
 		if (size_in_bytes == 0)
 			tmp |= PACKET3_CP_DMA_CP_SYNC;
 		radeon_ring_write(ring, PACKET3(PACKET3_CP_DMA, 4));
-		radeon_ring_write(ring, src_offset & 0xffffffff);
+		radeon_ring_write(ring, lower_32_bits(src_offset));
 		radeon_ring_write(ring, tmp);
-		radeon_ring_write(ring, dst_offset & 0xffffffff);
+		radeon_ring_write(ring, lower_32_bits(dst_offset));
 		radeon_ring_write(ring, upper_32_bits(dst_offset) & 0xff);
 		radeon_ring_write(ring, cur_size_in_bytes);
 		src_offset += cur_size_in_bytes;
@@ -3876,7 +3879,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[0]))
-						radeon_crtc_handle_flip(rdev, 0);
+						radeon_crtc_handle_vblank(rdev, 0);
 					rdev->irq.stat_regs.r600.disp_int &= ~LB_D1_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D1 vblank\n");
 				}
@@ -3902,7 +3905,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[1]))
-						radeon_crtc_handle_flip(rdev, 1);
+						radeon_crtc_handle_vblank(rdev, 1);
 					rdev->irq.stat_regs.r600.disp_int &= ~LB_D2_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D2 vblank\n");
 				}
diff --git a/drivers/gpu/drm/radeon/r600_hdmi.c b/drivers/gpu/drm/radeon/r600_hdmi.c
index 85a2bb28aed2..26ef8ced6f89 100644
--- a/drivers/gpu/drm/radeon/r600_hdmi.c
+++ b/drivers/gpu/drm/radeon/r600_hdmi.c
@@ -133,7 +133,7 @@ struct radeon_hdmi_acr r600_hdmi_acr(uint32_t clock)
 /*
  * update the N and CTS parameters for a given pixel clock rate
  */
-static void r600_hdmi_update_ACR(struct drm_encoder *encoder, uint32_t clock)
+void r600_hdmi_update_ACR(struct drm_encoder *encoder, uint32_t clock)
 {
 	struct drm_device *dev = encoder->dev;
 	struct radeon_device *rdev = dev->dev_private;
@@ -142,21 +142,33 @@ static void r600_hdmi_update_ACR(struct drm_encoder *encoder, uint32_t clock)
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
 	uint32_t offset = dig->afmt->offset;
 
-	WREG32(HDMI0_ACR_32_0 + offset, HDMI0_ACR_CTS_32(acr.cts_32khz));
-	WREG32(HDMI0_ACR_32_1 + offset, acr.n_32khz);
-
-	WREG32(HDMI0_ACR_44_0 + offset, HDMI0_ACR_CTS_44(acr.cts_44_1khz));
-	WREG32(HDMI0_ACR_44_1 + offset, acr.n_44_1khz);
-
-	WREG32(HDMI0_ACR_48_0 + offset, HDMI0_ACR_CTS_48(acr.cts_48khz));
-	WREG32(HDMI0_ACR_48_1 + offset, acr.n_48khz);
+	WREG32_P(HDMI0_ACR_32_0 + offset,
+		 HDMI0_ACR_CTS_32(acr.cts_32khz),
+		 ~HDMI0_ACR_CTS_32_MASK);
+	WREG32_P(HDMI0_ACR_32_1 + offset,
+		 HDMI0_ACR_N_32(acr.n_32khz),
+		 ~HDMI0_ACR_N_32_MASK);
+
+	WREG32_P(HDMI0_ACR_44_0 + offset,
+		 HDMI0_ACR_CTS_44(acr.cts_44_1khz),
+		 ~HDMI0_ACR_CTS_44_MASK);
+	WREG32_P(HDMI0_ACR_44_1 + offset,
+		 HDMI0_ACR_N_44(acr.n_44_1khz),
+		 ~HDMI0_ACR_N_44_MASK);
+
+	WREG32_P(HDMI0_ACR_48_0 + offset,
+		 HDMI0_ACR_CTS_48(acr.cts_48khz),
+		 ~HDMI0_ACR_CTS_48_MASK);
+	WREG32_P(HDMI0_ACR_48_1 + offset,
+		 HDMI0_ACR_N_48(acr.n_48khz),
+		 ~HDMI0_ACR_N_48_MASK);
 }
 
 /*
  * build a HDMI Video Info Frame
  */
-static void r600_hdmi_update_avi_infoframe(struct drm_encoder *encoder,
-					   void *buffer, size_t size)
+void r600_hdmi_update_avi_infoframe(struct drm_encoder *encoder, void *buffer,
+				    size_t size)
 {
 	struct drm_device *dev = encoder->dev;
 	struct radeon_device *rdev = dev->dev_private;
@@ -231,7 +243,7 @@ int r600_hdmi_buffer_status_changed(struct drm_encoder *encoder)
 /*
  * write the audio workaround status to the hardware
  */
-static void r600_hdmi_audio_workaround(struct drm_encoder *encoder)
+void r600_hdmi_audio_workaround(struct drm_encoder *encoder)
 {
 	struct drm_device *dev = encoder->dev;
 	struct radeon_device *rdev = dev->dev_private;
@@ -250,7 +262,7 @@ static void r600_hdmi_audio_workaround(struct drm_encoder *encoder)
 		 value, ~HDMI0_AUDIO_TEST_EN);
 }
 
-static void r600_audio_set_dto(struct drm_encoder *encoder, u32 clock)
+void r600_audio_set_dto(struct drm_encoder *encoder, u32 clock)
 {
 	struct drm_device *dev = encoder->dev;
 	struct radeon_device *rdev = dev->dev_private;
@@ -320,121 +332,6 @@ static void r600_audio_set_dto(struct drm_encoder *encoder, u32 clock)
 	}
 }
 
-static void dce3_2_afmt_write_speaker_allocation(struct drm_encoder *encoder)
-{
-	struct radeon_device *rdev = encoder->dev->dev_private;
-	struct drm_connector *connector;
-	struct radeon_connector *radeon_connector = NULL;
-	u32 tmp;
-	u8 *sadb;
-	int sad_count;
-
-	list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
-		if (connector->encoder == encoder) {
-			radeon_connector = to_radeon_connector(connector);
-			break;
-		}
-	}
-
-	if (!radeon_connector) {
-		DRM_ERROR("Couldn't find encoder's connector\n");
-		return;
-	}
-
-	sad_count = drm_edid_to_speaker_allocation(radeon_connector->edid, &sadb);
-	if (sad_count < 0) {
-		DRM_ERROR("Couldn't read Speaker Allocation Data Block: %d\n", sad_count);
-		return;
-	}
-
-	/* program the speaker allocation */
-	tmp = RREG32(AZ_F0_CODEC_PIN0_CONTROL_CHANNEL_SPEAKER);
-	tmp &= ~(DP_CONNECTION | SPEAKER_ALLOCATION_MASK);
-	/* set HDMI mode */
-	tmp |= HDMI_CONNECTION;
-	if (sad_count)
-		tmp |= SPEAKER_ALLOCATION(sadb[0]);
-	else
-		tmp |= SPEAKER_ALLOCATION(5); /* stereo */
-	WREG32(AZ_F0_CODEC_PIN0_CONTROL_CHANNEL_SPEAKER, tmp);
-
-	kfree(sadb);
-}
-
-static void dce3_2_afmt_write_sad_regs(struct drm_encoder *encoder)
-{
-	struct radeon_device *rdev = encoder->dev->dev_private;
-	struct drm_connector *connector;
-	struct radeon_connector *radeon_connector = NULL;
-	struct cea_sad *sads;
-	int i, sad_count;
-
-	static const u16 eld_reg_to_type[][2] = {
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR0, HDMI_AUDIO_CODING_TYPE_PCM },
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR1, HDMI_AUDIO_CODING_TYPE_AC3 },
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR2, HDMI_AUDIO_CODING_TYPE_MPEG1 },
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR3, HDMI_AUDIO_CODING_TYPE_MP3 },
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR4, HDMI_AUDIO_CODING_TYPE_MPEG2 },
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR5, HDMI_AUDIO_CODING_TYPE_AAC_LC },
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR6, HDMI_AUDIO_CODING_TYPE_DTS },
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR7, HDMI_AUDIO_CODING_TYPE_ATRAC },
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR9, HDMI_AUDIO_CODING_TYPE_EAC3 },
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR10, HDMI_AUDIO_CODING_TYPE_DTS_HD },
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR11, HDMI_AUDIO_CODING_TYPE_MLP },
-		{ AZ_F0_CODEC_PIN0_CONTROL_AUDIO_DESCRIPTOR13, HDMI_AUDIO_CODING_TYPE_WMA_PRO },
-	};
-
-	list_for_each_entry(connector, &encoder->dev->mode_config.connector_list, head) {
-		if (connector->encoder == encoder) {
-			radeon_connector = to_radeon_connector(connector);
-			break;
-		}
-	}
-
-	if (!radeon_connector) {
-		DRM_ERROR("Couldn't find encoder's connector\n");
-		return;
-	}
-
-	sad_count = drm_edid_to_sad(radeon_connector->edid, &sads);
-	if (sad_count < 0) {
-		DRM_ERROR("Couldn't read SADs: %d\n", sad_count);
-		return;
-	}
-	BUG_ON(!sads);
-
-	for (i = 0; i < ARRAY_SIZE(eld_reg_to_type); i++) {
-		u32 value = 0;
-		u8 stereo_freqs = 0;
-		int max_channels = -1;
-		int j;
-
-		for (j = 0; j < sad_count; j++) {
-			struct cea_sad *sad = &sads[j];
-
-			if (sad->format == eld_reg_to_type[i][1]) {
-				if (sad->channels > max_channels) {
-					value = MAX_CHANNELS(sad->channels) |
-						DESCRIPTOR_BYTE_2(sad->byte2) |
-						SUPPORTED_FREQUENCIES(sad->freq);
-					max_channels = sad->channels;
-				}
-
-				if (sad->format == HDMI_AUDIO_CODING_TYPE_PCM)
-					stereo_freqs |= sad->freq;
-				else
-					break;
-			}
-		}
-
-		value |= SUPPORTED_FREQUENCIES_STEREO(stereo_freqs);
-
-		WREG32(eld_reg_to_type[i][0], value);
-	}
-
-	kfree(sads);
-}
-
 /*
  * update the info frames with the data from the current display mode
  */
@@ -447,6 +344,7 @@ void r600_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode *mod
 	u8 buffer[HDMI_INFOFRAME_HEADER_SIZE + HDMI_AVI_INFOFRAME_SIZE];
 	struct hdmi_avi_infoframe frame;
 	uint32_t offset;
+	uint32_t acr_ctl;
 	ssize_t err;
 
 	if (!dig || !dig->afmt)
@@ -463,52 +361,44 @@ void r600_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode *mod
 
 	r600_audio_set_dto(encoder, mode->clock);
 
-	WREG32(HDMI0_VBI_PACKET_CONTROL + offset,
-	       HDMI0_NULL_SEND); /* send null packets when required */
-
-	WREG32(HDMI0_AUDIO_CRC_CONTROL + offset, 0x1000);
-
-	if (ASIC_IS_DCE32(rdev)) {
-		WREG32(HDMI0_AUDIO_PACKET_CONTROL + offset,
-		       HDMI0_AUDIO_DELAY_EN(1) | /* default audio delay */
-		       HDMI0_AUDIO_PACKETS_PER_LINE(3)); /* should be suffient for all audio modes and small enough for all hblanks */
-		WREG32(AFMT_AUDIO_PACKET_CONTROL + offset,
-		       AFMT_AUDIO_SAMPLE_SEND | /* send audio packets */
-		       AFMT_60958_CS_UPDATE); /* allow 60958 channel status fields to be updated */
-	} else {
-		WREG32(HDMI0_AUDIO_PACKET_CONTROL + offset,
-		       HDMI0_AUDIO_SAMPLE_SEND | /* send audio packets */
-		       HDMI0_AUDIO_DELAY_EN(1) | /* default audio delay */
-		       HDMI0_AUDIO_PACKETS_PER_LINE(3) | /* should be suffient for all audio modes and small enough for all hblanks */
-		       HDMI0_60958_CS_UPDATE); /* allow 60958 channel status fields to be updated */
-	}
-
-	if (ASIC_IS_DCE32(rdev)) {
-		dce3_2_afmt_write_speaker_allocation(encoder);
-		dce3_2_afmt_write_sad_regs(encoder);
-	}
-
-	WREG32(HDMI0_ACR_PACKET_CONTROL + offset,
-	       HDMI0_ACR_SOURCE | /* select SW CTS value - XXX verify that hw CTS works on all families */
-	       HDMI0_ACR_AUTO_SEND); /* allow hw to sent ACR packets when required */
-
-	WREG32(HDMI0_VBI_PACKET_CONTROL + offset,
-	       HDMI0_NULL_SEND | /* send null packets when required */
-	       HDMI0_GC_SEND | /* send general control packets */
-	       HDMI0_GC_CONT); /* send general control packets every frame */
-
-	/* TODO: HDMI0_AUDIO_INFO_UPDATE */
-	WREG32(HDMI0_INFOFRAME_CONTROL0 + offset,
-	       HDMI0_AVI_INFO_SEND | /* enable AVI info frames */
-	       HDMI0_AVI_INFO_CONT | /* send AVI info frames every frame/field */
-	       HDMI0_AUDIO_INFO_SEND | /* enable audio info frames (frames won't be set until audio is enabled) */
-	       HDMI0_AUDIO_INFO_CONT); /* send audio info frames every frame/field */
-
-	WREG32(HDMI0_INFOFRAME_CONTROL1 + offset,
-	       HDMI0_AVI_INFO_LINE(2) | /* anything other than 0 */
-	       HDMI0_AUDIO_INFO_LINE(2)); /* anything other than 0 */
-
-	WREG32(HDMI0_GC + offset, 0); /* unset HDMI0_GC_AVMUTE */
+	WREG32_P(HDMI0_AUDIO_PACKET_CONTROL + offset,
+		 HDMI0_AUDIO_SAMPLE_SEND | /* send audio packets */
+		 HDMI0_AUDIO_DELAY_EN(1) | /* default audio delay */
+		 HDMI0_AUDIO_PACKETS_PER_LINE(3) | /* should be suffient for all audio modes and small enough for all hblanks */
+		 HDMI0_60958_CS_UPDATE, /* allow 60958 channel status fields to be updated */
+		 ~(HDMI0_AUDIO_SAMPLE_SEND |
+		   HDMI0_AUDIO_DELAY_EN_MASK |
+		   HDMI0_AUDIO_PACKETS_PER_LINE_MASK |
+		   HDMI0_60958_CS_UPDATE));
+
+	/* DCE 3.0 uses register that's normally for CRC_CONTROL */
+	acr_ctl = ASIC_IS_DCE3(rdev) ? DCE3_HDMI0_ACR_PACKET_CONTROL :
+				       HDMI0_ACR_PACKET_CONTROL;
+	WREG32_P(acr_ctl + offset,
+		 HDMI0_ACR_SOURCE | /* select SW CTS value - XXX verify that hw CTS works on all families */
+		 HDMI0_ACR_AUTO_SEND, /* allow hw to sent ACR packets when required */
+		 ~(HDMI0_ACR_SOURCE |
+		   HDMI0_ACR_AUTO_SEND));
+
+	WREG32_OR(HDMI0_VBI_PACKET_CONTROL + offset,
+		  HDMI0_NULL_SEND | /* send null packets when required */
+		  HDMI0_GC_SEND | /* send general control packets */
+		  HDMI0_GC_CONT); /* send general control packets every frame */
+
+	WREG32_OR(HDMI0_INFOFRAME_CONTROL0 + offset,
+		  HDMI0_AVI_INFO_SEND | /* enable AVI info frames */
+		  HDMI0_AVI_INFO_CONT | /* send AVI info frames every frame/field */
+		  HDMI0_AUDIO_INFO_SEND | /* enable audio info frames (frames won't be set until audio is enabled) */
+		  HDMI0_AUDIO_INFO_UPDATE); /* required for audio info values to be updated */
+
+	WREG32_P(HDMI0_INFOFRAME_CONTROL1 + offset,
+		 HDMI0_AVI_INFO_LINE(2) | /* anything other than 0 */
+		 HDMI0_AUDIO_INFO_LINE(2), /* anything other than 0 */
+		 ~(HDMI0_AVI_INFO_LINE_MASK |
+		   HDMI0_AUDIO_INFO_LINE_MASK));
+
+	WREG32_AND(HDMI0_GC + offset,
+		   ~HDMI0_GC_AVMUTE); /* unset HDMI0_GC_AVMUTE */
 
 	err = drm_hdmi_avi_infoframe_from_display_mode(&frame, mode);
 	if (err < 0) {
@@ -523,22 +413,45 @@ void r600_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode *mod
 	}
 
 	r600_hdmi_update_avi_infoframe(encoder, buffer, sizeof(buffer));
+
+	/* fglrx duplicates INFOFRAME_CONTROL0 & INFOFRAME_CONTROL1 ops here */
+
+	WREG32_AND(HDMI0_GENERIC_PACKET_CONTROL + offset,
+		   ~(HDMI0_GENERIC0_SEND |
+		     HDMI0_GENERIC0_CONT |
+		     HDMI0_GENERIC0_UPDATE |
+		     HDMI0_GENERIC1_SEND |
+		     HDMI0_GENERIC1_CONT |
+		     HDMI0_GENERIC0_LINE_MASK |
+		     HDMI0_GENERIC1_LINE_MASK));
+
 	r600_hdmi_update_ACR(encoder, mode->clock);
 
+	WREG32_P(HDMI0_60958_0 + offset,
+		 HDMI0_60958_CS_CHANNEL_NUMBER_L(1),
+		 ~(HDMI0_60958_CS_CHANNEL_NUMBER_L_MASK |
+		   HDMI0_60958_CS_CLOCK_ACCURACY_MASK));
+
+	WREG32_P(HDMI0_60958_1 + offset,
+		 HDMI0_60958_CS_CHANNEL_NUMBER_R(2),
+		 ~HDMI0_60958_CS_CHANNEL_NUMBER_R_MASK);
+
 	/* it's unknown what these bits do excatly, but it's indeed quite useful for debugging */
 	WREG32(HDMI0_RAMP_CONTROL0 + offset, 0x00FFFFFF);
 	WREG32(HDMI0_RAMP_CONTROL1 + offset, 0x007FFFFF);
 	WREG32(HDMI0_RAMP_CONTROL2 + offset, 0x00000001);
 	WREG32(HDMI0_RAMP_CONTROL3 + offset, 0x00000001);
 
-	r600_hdmi_audio_workaround(encoder);
-
 	/* enable audio after to setting up hw */
 	r600_audio_enable(rdev, dig->afmt->pin, true);
 }
 
-/*
- * update settings with current parameters from audio engine
+/**
+ * r600_hdmi_update_audio_settings - Update audio infoframe
+ *
+ * @encoder: drm encoder
+ *
+ * Gets info about current audio stream and updates audio infoframe.
  */
 void r600_hdmi_update_audio_settings(struct drm_encoder *encoder)
 {
@@ -550,7 +463,7 @@ void r600_hdmi_update_audio_settings(struct drm_encoder *encoder)
 	uint8_t buffer[HDMI_INFOFRAME_HEADER_SIZE + HDMI_AUDIO_INFOFRAME_SIZE];
 	struct hdmi_audio_infoframe frame;
 	uint32_t offset;
-	uint32_t iec;
+	uint32_t value;
 	ssize_t err;
 
 	if (!dig->afmt || !dig->afmt->enabled)
@@ -563,60 +476,6 @@ void r600_hdmi_update_audio_settings(struct drm_encoder *encoder)
 	DRM_DEBUG("0x%02X IEC60958 status bits and 0x%02X category code\n",
 		  (int)audio.status_bits, (int)audio.category_code);
 
-	iec = 0;
-	if (audio.status_bits & AUDIO_STATUS_PROFESSIONAL)
-		iec |= 1 << 0;
-	if (audio.status_bits & AUDIO_STATUS_NONAUDIO)
-		iec |= 1 << 1;
-	if (audio.status_bits & AUDIO_STATUS_COPYRIGHT)
-		iec |= 1 << 2;
-	if (audio.status_bits & AUDIO_STATUS_EMPHASIS)
-		iec |= 1 << 3;
-
-	iec |= HDMI0_60958_CS_CATEGORY_CODE(audio.category_code);
-
-	switch (audio.rate) {
-	case 32000:
-		iec |= HDMI0_60958_CS_SAMPLING_FREQUENCY(0x3);
-		break;
-	case 44100:
-		iec |= HDMI0_60958_CS_SAMPLING_FREQUENCY(0x0);
-		break;
-	case 48000:
-		iec |= HDMI0_60958_CS_SAMPLING_FREQUENCY(0x2);
-		break;
-	case 88200:
-		iec |= HDMI0_60958_CS_SAMPLING_FREQUENCY(0x8);
-		break;
-	case 96000:
-		iec |= HDMI0_60958_CS_SAMPLING_FREQUENCY(0xa);
-		break;
-	case 176400:
-		iec |= HDMI0_60958_CS_SAMPLING_FREQUENCY(0xc);
-		break;
-	case 192000:
-		iec |= HDMI0_60958_CS_SAMPLING_FREQUENCY(0xe);
-		break;
-	}
-
-	WREG32(HDMI0_60958_0 + offset, iec);
-
-	iec = 0;
-	switch (audio.bits_per_sample) {
-	case 16:
-		iec |= HDMI0_60958_CS_WORD_LENGTH(0x2);
-		break;
-	case 20:
-		iec |= HDMI0_60958_CS_WORD_LENGTH(0x3);
-		break;
-	case 24:
-		iec |= HDMI0_60958_CS_WORD_LENGTH(0xb);
-		break;
-	}
-	if (audio.status_bits & AUDIO_STATUS_V)
-		iec |= 0x5 << 16;
-	WREG32_P(HDMI0_60958_1 + offset, iec, ~0x5000f);
-
 	err = hdmi_audio_infoframe_init(&frame);
 	if (err < 0) {
 		DRM_ERROR("failed to setup audio infoframe\n");
@@ -631,8 +490,22 @@ void r600_hdmi_update_audio_settings(struct drm_encoder *encoder)
 		return;
 	}
 
+	value = RREG32(HDMI0_AUDIO_PACKET_CONTROL + offset);
+	if (value & HDMI0_AUDIO_TEST_EN)
+		WREG32(HDMI0_AUDIO_PACKET_CONTROL + offset,
+		       value & ~HDMI0_AUDIO_TEST_EN);
+
+	WREG32_OR(HDMI0_CONTROL + offset,
+		  HDMI0_ERROR_ACK);
+
+	WREG32_AND(HDMI0_INFOFRAME_CONTROL0 + offset,
+		   ~HDMI0_AUDIO_INFO_SOURCE);
+
 	r600_hdmi_update_audio_infoframe(encoder, buffer, sizeof(buffer));
-	r600_hdmi_audio_workaround(encoder);
+
+	WREG32_OR(HDMI0_INFOFRAME_CONTROL0 + offset,
+		  HDMI0_AUDIO_INFO_CONT |
+		  HDMI0_AUDIO_INFO_UPDATE);
 }
 
 /*
diff --git a/drivers/gpu/drm/radeon/r600d.h b/drivers/gpu/drm/radeon/r600d.h
index 37455f65107f..f94e7a9afe75 100644
--- a/drivers/gpu/drm/radeon/r600d.h
+++ b/drivers/gpu/drm/radeon/r600d.h
@@ -1029,15 +1029,18 @@
 #define HDMI0_AUDIO_PACKET_CONTROL   0x7408
 #       define HDMI0_AUDIO_SAMPLE_SEND  (1 << 0)
 #       define HDMI0_AUDIO_DELAY_EN(x)  (((x) & 3) << 4)
+#       define HDMI0_AUDIO_DELAY_EN_MASK	(3 << 4)
 #       define HDMI0_AUDIO_SEND_MAX_PACKETS  (1 << 8)
 #       define HDMI0_AUDIO_TEST_EN         (1 << 12)
 #       define HDMI0_AUDIO_PACKETS_PER_LINE(x)  (((x) & 0x1f) << 16)
+#       define HDMI0_AUDIO_PACKETS_PER_LINE_MASK	(0x1f << 16)
 #       define HDMI0_AUDIO_CHANNEL_SWAP    (1 << 24)
 #       define HDMI0_60958_CS_UPDATE       (1 << 26)
 #       define HDMI0_AZ_FORMAT_WTRIG_MASK  (1 << 28)
 #       define HDMI0_AZ_FORMAT_WTRIG_ACK   (1 << 29)
 #define HDMI0_AUDIO_CRC_CONTROL      0x740c
 #       define HDMI0_AUDIO_CRC_EN    (1 << 0)
+#define DCE3_HDMI0_ACR_PACKET_CONTROL	0x740c
 #define HDMI0_VBI_PACKET_CONTROL     0x7410
 #       define HDMI0_NULL_SEND       (1 << 0)
 #       define HDMI0_GC_SEND         (1 << 4)
@@ -1054,7 +1057,9 @@
 #       define HDMI0_MPEG_INFO_UPDATE  (1 << 10)
 #define HDMI0_INFOFRAME_CONTROL1     0x7418
 #       define HDMI0_AVI_INFO_LINE(x)  (((x) & 0x3f) << 0)
+#       define HDMI0_AVI_INFO_LINE_MASK		(0x3f << 0)
 #       define HDMI0_AUDIO_INFO_LINE(x)  (((x) & 0x3f) << 8)
+#       define HDMI0_AUDIO_INFO_LINE_MASK	(0x3f << 8)
 #       define HDMI0_MPEG_INFO_LINE(x)  (((x) & 0x3f) << 16)
 #define HDMI0_GENERIC_PACKET_CONTROL 0x741c
 #       define HDMI0_GENERIC0_SEND   (1 << 0)
@@ -1063,7 +1068,9 @@
 #       define HDMI0_GENERIC1_SEND   (1 << 4)
 #       define HDMI0_GENERIC1_CONT   (1 << 5)
 #       define HDMI0_GENERIC0_LINE(x)  (((x) & 0x3f) << 16)
+#       define HDMI0_GENERIC0_LINE_MASK		(0x3f << 16)
 #       define HDMI0_GENERIC1_LINE(x)  (((x) & 0x3f) << 24)
+#       define HDMI0_GENERIC1_LINE_MASK		(0x3f << 24)
 #define HDMI0_GC                     0x7428
 #       define HDMI0_GC_AVMUTE       (1 << 0)
 #define HDMI0_AVI_INFO0              0x7454
@@ -1119,16 +1126,22 @@
 #define HDMI0_GENERIC1_6             0x74a8
 #define HDMI0_ACR_32_0               0x74ac
 #       define HDMI0_ACR_CTS_32(x)   (((x) & 0xfffff) << 12)
+#       define HDMI0_ACR_CTS_32_MASK		(0xfffff << 12)
 #define HDMI0_ACR_32_1               0x74b0
 #       define HDMI0_ACR_N_32(x)   (((x) & 0xfffff) << 0)
+#       define HDMI0_ACR_N_32_MASK		(0xfffff << 0)
 #define HDMI0_ACR_44_0               0x74b4
 #       define HDMI0_ACR_CTS_44(x)   (((x) & 0xfffff) << 12)
+#       define HDMI0_ACR_CTS_44_MASK		(0xfffff << 12)
 #define HDMI0_ACR_44_1               0x74b8
 #       define HDMI0_ACR_N_44(x)   (((x) & 0xfffff) << 0)
+#       define HDMI0_ACR_N_44_MASK		(0xfffff << 0)
 #define HDMI0_ACR_48_0               0x74bc
 #       define HDMI0_ACR_CTS_48(x)   (((x) & 0xfffff) << 12)
+#       define HDMI0_ACR_CTS_48_MASK		(0xfffff << 12)
 #define HDMI0_ACR_48_1               0x74c0
 #       define HDMI0_ACR_N_48(x)   (((x) & 0xfffff) << 0)
+#       define HDMI0_ACR_N_48_MASK		(0xfffff << 0)
 #define HDMI0_ACR_STATUS_0           0x74c4
 #define HDMI0_ACR_STATUS_1           0x74c8
 #define HDMI0_AUDIO_INFO0            0x74cc
@@ -1148,14 +1161,17 @@
 #       define HDMI0_60958_CS_CATEGORY_CODE(x)      (((x) & 0xff) << 8)
 #       define HDMI0_60958_CS_SOURCE_NUMBER(x)      (((x) & 0xf) << 16)
 #       define HDMI0_60958_CS_CHANNEL_NUMBER_L(x)   (((x) & 0xf) << 20)
+#       define HDMI0_60958_CS_CHANNEL_NUMBER_L_MASK	(0xf << 20)
 #       define HDMI0_60958_CS_SAMPLING_FREQUENCY(x) (((x) & 0xf) << 24)
 #       define HDMI0_60958_CS_CLOCK_ACCURACY(x)     (((x) & 3) << 28)
+#       define HDMI0_60958_CS_CLOCK_ACCURACY_MASK	(3 << 28)
 #define HDMI0_60958_1                0x74d8
 #       define HDMI0_60958_CS_WORD_LENGTH(x)        (((x) & 0xf) << 0)
 #       define HDMI0_60958_CS_ORIGINAL_SAMPLING_FREQUENCY(x)   (((x) & 0xf) << 4)
 #       define HDMI0_60958_CS_VALID_L(x)   (((x) & 1) << 16)
 #       define HDMI0_60958_CS_VALID_R(x)   (((x) & 1) << 18)
 #       define HDMI0_60958_CS_CHANNEL_NUMBER_R(x)   (((x) & 0xf) << 20)
+#       define HDMI0_60958_CS_CHANNEL_NUMBER_R_MASK	(0xf << 20)
 #define HDMI0_ACR_PACKET_CONTROL     0x74dc
 #       define HDMI0_ACR_SEND        (1 << 0)
 #       define HDMI0_ACR_CONT        (1 << 1)
@@ -1166,6 +1182,7 @@
 #       define HDMI0_ACR_48          3
 #       define HDMI0_ACR_SOURCE      (1 << 8) /* 0 - hw; 1 - cts value */
 #       define HDMI0_ACR_AUTO_SEND   (1 << 12)
+#define DCE3_HDMI0_AUDIO_CRC_CONTROL	0x74dc
 #define HDMI0_RAMP_CONTROL0          0x74e0
 #       define HDMI0_RAMP_MAX_COUNT(x)   (((x) & 0xffffff) << 0)
 #define HDMI0_RAMP_CONTROL1          0x74e4
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 8149e7cf4303..4b0bbf88d5c0 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -100,6 +100,8 @@ extern int radeon_dpm;
 extern int radeon_aspm;
 extern int radeon_runtime_pm;
 extern int radeon_hard_reset;
+extern int radeon_vm_size;
+extern int radeon_vm_block_size;
 
 /*
  * Copy from radeon_drv.h so we don't have to include both and have conflicting
@@ -676,14 +678,16 @@ void radeon_doorbell_free(struct radeon_device *rdev, u32 doorbell);
  * IRQS.
  */
 
-struct radeon_unpin_work {
-	struct work_struct work;
-	struct radeon_device *rdev;
-	int crtc_id;
-	struct radeon_fence *fence;
+struct radeon_flip_work {
+	struct work_struct		flip_work;
+	struct work_struct		unpin_work;
+	struct radeon_device		*rdev;
+	int				crtc_id;
+	struct drm_framebuffer		*fb;
 	struct drm_pending_vblank_event *event;
-	struct radeon_bo *old_rbo;
-	u64 new_crtc_base;
+	struct radeon_bo		*old_rbo;
+	struct radeon_bo		*new_rbo;
+	struct radeon_fence		*fence;
 };
 
 struct r500_irq_stat_regs {
@@ -835,13 +839,8 @@ struct radeon_mec {
 /* maximum number of VMIDs */
 #define RADEON_NUM_VM	16
 
-/* defines number of bits in page table versus page directory,
- * a page is 4KB so we have 12 bits offset, 9 bits in the page
- * table and the remaining 19 bits are in the page directory */
-#define RADEON_VM_BLOCK_SIZE   9
-
 /* number of entries in page table */
-#define RADEON_VM_PTE_COUNT (1 << RADEON_VM_BLOCK_SIZE)
+#define RADEON_VM_PTE_COUNT (1 << radeon_vm_block_size)
 
 /* PTBs (Page Table Blocks) need to be aligned to 32K */
 #define RADEON_VM_PTB_ALIGN_SIZE   32768
@@ -854,6 +853,15 @@ struct radeon_mec {
 #define R600_PTE_READABLE	(1 << 5)
 #define R600_PTE_WRITEABLE	(1 << 6)
 
+/* PTE (Page Table Entry) fragment field for different page sizes */
+#define R600_PTE_FRAG_4KB	(0 << 7)
+#define R600_PTE_FRAG_64KB	(4 << 7)
+#define R600_PTE_FRAG_256KB	(6 << 7)
+
+/* flags used for GART page table entries on R600+ */
+#define R600_PTE_GART	( R600_PTE_VALID | R600_PTE_SYSTEM | R600_PTE_SNOOPED \
+			| R600_PTE_READABLE | R600_PTE_WRITEABLE)
+
 struct radeon_vm_pt {
 	struct radeon_bo		*bo;
 	uint64_t			addr;
@@ -986,8 +994,8 @@ struct radeon_cs_reloc {
 	struct radeon_bo		*robj;
 	struct ttm_validate_buffer	tv;
 	uint64_t			gpu_offset;
-	unsigned			domain;
-	unsigned			alt_domain;
+	unsigned			prefered_domains;
+	unsigned			allowed_domains;
 	uint32_t			tiling_flags;
 	uint32_t			handle;
 };
@@ -1771,7 +1779,8 @@ struct radeon_asic {
 	/* gart */
 	struct {
 		void (*tlb_flush)(struct radeon_device *rdev);
-		int (*set_page)(struct radeon_device *rdev, int i, uint64_t addr);
+		void (*set_page)(struct radeon_device *rdev, unsigned i,
+				 uint64_t addr);
 	} gart;
 	struct {
 		int (*init)(struct radeon_device *rdev);
@@ -1883,9 +1892,8 @@ struct radeon_asic {
 	} dpm;
 	/* pageflipping */
 	struct {
-		void (*pre_page_flip)(struct radeon_device *rdev, int crtc);
-		u32 (*page_flip)(struct radeon_device *rdev, int crtc, u64 crtc_base);
-		void (*post_page_flip)(struct radeon_device *rdev, int crtc);
+		void (*page_flip)(struct radeon_device *rdev, int crtc, u64 crtc_base);
+		bool (*page_flip_pending)(struct radeon_device *rdev, int crtc);
 	} pflip;
 };
 
@@ -1924,6 +1932,7 @@ struct r600_asic {
 	unsigned		tiling_group_size;
 	unsigned		tile_config;
 	unsigned		backend_map;
+	unsigned		active_simds;
 };
 
 struct rv770_asic {
@@ -1949,6 +1958,7 @@ struct rv770_asic {
 	unsigned		tiling_group_size;
 	unsigned		tile_config;
 	unsigned		backend_map;
+	unsigned		active_simds;
 };
 
 struct evergreen_asic {
@@ -1975,6 +1985,7 @@ struct evergreen_asic {
 	unsigned tiling_group_size;
 	unsigned tile_config;
 	unsigned backend_map;
+	unsigned active_simds;
 };
 
 struct cayman_asic {
@@ -2013,6 +2024,7 @@ struct cayman_asic {
 	unsigned multi_gpu_tile_size;
 
 	unsigned tile_config;
+	unsigned active_simds;
 };
 
 struct si_asic {
@@ -2043,6 +2055,7 @@ struct si_asic {
 
 	unsigned tile_config;
 	uint32_t tile_mode_array[32];
+	uint32_t active_cus;
 };
 
 struct cik_asic {
@@ -2074,6 +2087,7 @@ struct cik_asic {
 	unsigned tile_config;
 	uint32_t tile_mode_array[32];
 	uint32_t macrotile_mode_array[16];
+	uint32_t active_cus;
 };
 
 union radeon_asic_config {
@@ -2745,9 +2759,8 @@ void radeon_ring_write(struct radeon_ring *ring, uint32_t v);
 #define radeon_pm_finish(rdev) (rdev)->asic->pm.finish((rdev))
 #define radeon_pm_init_profile(rdev) (rdev)->asic->pm.init_profile((rdev))
 #define radeon_pm_get_dynpm_state(rdev) (rdev)->asic->pm.get_dynpm_state((rdev))
-#define radeon_pre_page_flip(rdev, crtc) (rdev)->asic->pflip.pre_page_flip((rdev), (crtc))
 #define radeon_page_flip(rdev, crtc, base) (rdev)->asic->pflip.page_flip((rdev), (crtc), (base))
-#define radeon_post_page_flip(rdev, crtc) (rdev)->asic->pflip.post_page_flip((rdev), (crtc))
+#define radeon_page_flip_pending(rdev, crtc) (rdev)->asic->pflip.page_flip_pending((rdev), (crtc))
 #define radeon_wait_for_vblank(rdev, crtc) (rdev)->asic->display.wait_for_vblank((rdev), (crtc))
 #define radeon_mc_wait_for_idle(rdev) (rdev)->asic->mc_wait_for_idle((rdev))
 #define radeon_get_xclk(rdev) (rdev)->asic->get_xclk((rdev))
diff --git a/drivers/gpu/drm/radeon/radeon_agp.c b/drivers/gpu/drm/radeon/radeon_agp.c
index 42433344cb1b..a9297b2c3524 100644
--- a/drivers/gpu/drm/radeon/radeon_agp.c
+++ b/drivers/gpu/drm/radeon/radeon_agp.c
@@ -117,9 +117,6 @@ static struct radeon_agpmode_quirk radeon_agpmode_quirk_list[] = {
 	/* ATI Host Bridge / RV280 [M9+] Needs AGPMode 1 (phoronix forum) */
 	{ PCI_VENDOR_ID_ATI, 0xcbb2, PCI_VENDOR_ID_ATI, 0x5c61,
 		PCI_VENDOR_ID_SONY, 0x8175, 1},
-	/* HP Host Bridge / R300 [FireGL X1] Needs AGPMode 2 (fdo #7770) */
-	{ PCI_VENDOR_ID_HP, 0x122e, PCI_VENDOR_ID_ATI, 0x4e47,
-		PCI_VENDOR_ID_ATI, 0x0152, 2},
 	{ 0, 0, 0, 0, 0, 0, 0 },
 };
 #endif
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index e5f0177bea1e..34b9aa9e3c06 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -248,9 +248,8 @@ static struct radeon_asic r100_asic = {
 		.set_clock_gating = &radeon_legacy_set_clock_gating,
 	},
 	.pflip = {
-		.pre_page_flip = &r100_pre_page_flip,
 		.page_flip = &r100_page_flip,
-		.post_page_flip = &r100_post_page_flip,
+		.page_flip_pending = &r100_page_flip_pending,
 	},
 };
 
@@ -315,9 +314,8 @@ static struct radeon_asic r200_asic = {
 		.set_clock_gating = &radeon_legacy_set_clock_gating,
 	},
 	.pflip = {
-		.pre_page_flip = &r100_pre_page_flip,
 		.page_flip = &r100_page_flip,
-		.post_page_flip = &r100_post_page_flip,
+		.page_flip_pending = &r100_page_flip_pending,
 	},
 };
 
@@ -396,9 +394,8 @@ static struct radeon_asic r300_asic = {
 		.set_clock_gating = &radeon_legacy_set_clock_gating,
 	},
 	.pflip = {
-		.pre_page_flip = &r100_pre_page_flip,
 		.page_flip = &r100_page_flip,
-		.post_page_flip = &r100_post_page_flip,
+		.page_flip_pending = &r100_page_flip_pending,
 	},
 };
 
@@ -463,9 +460,8 @@ static struct radeon_asic r300_asic_pcie = {
 		.set_clock_gating = &radeon_legacy_set_clock_gating,
 	},
 	.pflip = {
-		.pre_page_flip = &r100_pre_page_flip,
 		.page_flip = &r100_page_flip,
-		.post_page_flip = &r100_post_page_flip,
+		.page_flip_pending = &r100_page_flip_pending,
 	},
 };
 
@@ -530,9 +526,8 @@ static struct radeon_asic r420_asic = {
 		.set_clock_gating = &radeon_atom_set_clock_gating,
 	},
 	.pflip = {
-		.pre_page_flip = &r100_pre_page_flip,
 		.page_flip = &r100_page_flip,
-		.post_page_flip = &r100_post_page_flip,
+		.page_flip_pending = &r100_page_flip_pending,
 	},
 };
 
@@ -597,9 +592,8 @@ static struct radeon_asic rs400_asic = {
 		.set_clock_gating = &radeon_legacy_set_clock_gating,
 	},
 	.pflip = {
-		.pre_page_flip = &r100_pre_page_flip,
 		.page_flip = &r100_page_flip,
-		.post_page_flip = &r100_post_page_flip,
+		.page_flip_pending = &r100_page_flip_pending,
 	},
 };
 
@@ -666,9 +660,8 @@ static struct radeon_asic rs600_asic = {
 		.set_clock_gating = &radeon_atom_set_clock_gating,
 	},
 	.pflip = {
-		.pre_page_flip = &rs600_pre_page_flip,
 		.page_flip = &rs600_page_flip,
-		.post_page_flip = &rs600_post_page_flip,
+		.page_flip_pending = &rs600_page_flip_pending,
 	},
 };
 
@@ -735,9 +728,8 @@ static struct radeon_asic rs690_asic = {
 		.set_clock_gating = &radeon_atom_set_clock_gating,
 	},
 	.pflip = {
-		.pre_page_flip = &rs600_pre_page_flip,
 		.page_flip = &rs600_page_flip,
-		.post_page_flip = &rs600_post_page_flip,
+		.page_flip_pending = &rs600_page_flip_pending,
 	},
 };
 
@@ -802,9 +794,8 @@ static struct radeon_asic rv515_asic = {
 		.set_clock_gating = &radeon_atom_set_clock_gating,
 	},
 	.pflip = {
-		.pre_page_flip = &rs600_pre_page_flip,
 		.page_flip = &rs600_page_flip,
-		.post_page_flip = &rs600_post_page_flip,
+		.page_flip_pending = &rs600_page_flip_pending,
 	},
 };
 
@@ -869,9 +860,8 @@ static struct radeon_asic r520_asic = {
 		.set_clock_gating = &radeon_atom_set_clock_gating,
 	},
 	.pflip = {
-		.pre_page_flip = &rs600_pre_page_flip,
 		.page_flip = &rs600_page_flip,
-		.post_page_flip = &rs600_post_page_flip,
+		.page_flip_pending = &rs600_page_flip_pending,
 	},
 };
 
@@ -968,9 +958,8 @@ static struct radeon_asic r600_asic = {
 		.get_temperature = &rv6xx_get_temp,
 	},
 	.pflip = {
-		.pre_page_flip = &rs600_pre_page_flip,
 		.page_flip = &rs600_page_flip,
-		.post_page_flip = &rs600_post_page_flip,
+		.page_flip_pending = &rs600_page_flip_pending,
 	},
 };
 
@@ -1059,9 +1048,8 @@ static struct radeon_asic rv6xx_asic = {
 		.force_performance_level = &rv6xx_dpm_force_performance_level,
 	},
 	.pflip = {
-		.pre_page_flip = &rs600_pre_page_flip,
 		.page_flip = &rs600_page_flip,
-		.post_page_flip = &rs600_post_page_flip,
+		.page_flip_pending = &rs600_page_flip_pending,
 	},
 };
 
@@ -1150,9 +1138,8 @@ static struct radeon_asic rs780_asic = {
 		.force_performance_level = &rs780_dpm_force_performance_level,
 	},
 	.pflip = {
-		.pre_page_flip = &rs600_pre_page_flip,
 		.page_flip = &rs600_page_flip,
-		.post_page_flip = &rs600_post_page_flip,
+		.page_flip_pending = &rs600_page_flip_pending,
 	},
 };
 
@@ -1201,7 +1188,7 @@ static struct radeon_asic rv770_asic = {
 		.set_backlight_level = &atombios_set_backlight_level,
 		.get_backlight_level = &atombios_get_backlight_level,
 		.hdmi_enable = &r600_hdmi_enable,
-		.hdmi_setmode = &r600_hdmi_setmode,
+		.hdmi_setmode = &dce3_1_hdmi_setmode,
 	},
 	.copy = {
 		.blit = &r600_copy_cpdma,
@@ -1256,9 +1243,8 @@ static struct radeon_asic rv770_asic = {
 		.vblank_too_short = &rv770_dpm_vblank_too_short,
 	},
 	.pflip = {
-		.pre_page_flip = &rs600_pre_page_flip,
 		.page_flip = &rv770_page_flip,
-		.post_page_flip = &rs600_post_page_flip,
+		.page_flip_pending = &rv770_page_flip_pending,
 	},
 };
 
@@ -1375,9 +1361,8 @@ static struct radeon_asic evergreen_asic = {
 		.vblank_too_short = &cypress_dpm_vblank_too_short,
 	},
 	.pflip = {
-		.pre_page_flip = &evergreen_pre_page_flip,
 		.page_flip = &evergreen_page_flip,
-		.post_page_flip = &evergreen_post_page_flip,
+		.page_flip_pending = &evergreen_page_flip_pending,
 	},
 };
 
@@ -1467,9 +1452,8 @@ static struct radeon_asic sumo_asic = {
 		.force_performance_level = &sumo_dpm_force_performance_level,
 	},
 	.pflip = {
-		.pre_page_flip = &evergreen_pre_page_flip,
 		.page_flip = &evergreen_page_flip,
-		.post_page_flip = &evergreen_post_page_flip,
+		.page_flip_pending = &evergreen_page_flip_pending,
 	},
 };
 
@@ -1560,9 +1544,8 @@ static struct radeon_asic btc_asic = {
 		.vblank_too_short = &btc_dpm_vblank_too_short,
 	},
 	.pflip = {
-		.pre_page_flip = &evergreen_pre_page_flip,
 		.page_flip = &evergreen_page_flip,
-		.post_page_flip = &evergreen_post_page_flip,
+		.page_flip_pending = &evergreen_page_flip_pending,
 	},
 };
 
@@ -1704,9 +1687,8 @@ static struct radeon_asic cayman_asic = {
 		.vblank_too_short = &ni_dpm_vblank_too_short,
 	},
 	.pflip = {
-		.pre_page_flip = &evergreen_pre_page_flip,
 		.page_flip = &evergreen_page_flip,
-		.post_page_flip = &evergreen_post_page_flip,
+		.page_flip_pending = &evergreen_page_flip_pending,
 	},
 };
 
@@ -1805,9 +1787,8 @@ static struct radeon_asic trinity_asic = {
 		.enable_bapm = &trinity_dpm_enable_bapm,
 	},
 	.pflip = {
-		.pre_page_flip = &evergreen_pre_page_flip,
 		.page_flip = &evergreen_page_flip,
-		.post_page_flip = &evergreen_post_page_flip,
+		.page_flip_pending = &evergreen_page_flip_pending,
 	},
 };
 
@@ -1936,9 +1917,8 @@ static struct radeon_asic si_asic = {
 		.vblank_too_short = &ni_dpm_vblank_too_short,
 	},
 	.pflip = {
-		.pre_page_flip = &evergreen_pre_page_flip,
 		.page_flip = &evergreen_page_flip,
-		.post_page_flip = &evergreen_post_page_flip,
+		.page_flip_pending = &evergreen_page_flip_pending,
 	},
 };
 
@@ -2049,8 +2029,8 @@ static struct radeon_asic ci_asic = {
 		.blit_ring_index = RADEON_RING_TYPE_GFX_INDEX,
 		.dma = &cik_copy_dma,
 		.dma_ring_index = R600_RING_TYPE_DMA_INDEX,
-		.copy = &cik_copy_cpdma,
-		.copy_ring_index = RADEON_RING_TYPE_GFX_INDEX,
+		.copy = &cik_copy_dma,
+		.copy_ring_index = R600_RING_TYPE_DMA_INDEX,
 	},
 	.surface = {
 		.set_reg = r600_set_surface_reg,
@@ -2099,9 +2079,8 @@ static struct radeon_asic ci_asic = {
 		.powergate_uvd = &ci_dpm_powergate_uvd,
 	},
 	.pflip = {
-		.pre_page_flip = &evergreen_pre_page_flip,
 		.page_flip = &evergreen_page_flip,
-		.post_page_flip = &evergreen_post_page_flip,
+		.page_flip_pending = &evergreen_page_flip_pending,
 	},
 };
 
@@ -2204,9 +2183,8 @@ static struct radeon_asic kv_asic = {
 		.enable_bapm = &kv_dpm_enable_bapm,
 	},
 	.pflip = {
-		.pre_page_flip = &evergreen_pre_page_flip,
 		.page_flip = &evergreen_page_flip,
-		.post_page_flip = &evergreen_post_page_flip,
+		.page_flip_pending = &evergreen_page_flip_pending,
 	},
 };
 
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index 3d55a3a39e82..01e7c0ad8f01 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -67,7 +67,8 @@ bool r100_gpu_is_lockup(struct radeon_device *rdev, struct radeon_ring *cp);
 int r100_asic_reset(struct radeon_device *rdev);
 u32 r100_get_vblank_counter(struct radeon_device *rdev, int crtc);
 void r100_pci_gart_tlb_flush(struct radeon_device *rdev);
-int r100_pci_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr);
+void r100_pci_gart_set_page(struct radeon_device *rdev, unsigned i,
+			    uint64_t addr);
 void r100_ring_start(struct radeon_device *rdev, struct radeon_ring *ring);
 int r100_irq_set(struct radeon_device *rdev);
 int r100_irq_process(struct radeon_device *rdev);
@@ -135,9 +136,9 @@ extern void r100_pm_prepare(struct radeon_device *rdev);
 extern void r100_pm_finish(struct radeon_device *rdev);
 extern void r100_pm_init_profile(struct radeon_device *rdev);
 extern void r100_pm_get_dynpm_state(struct radeon_device *rdev);
-extern void r100_pre_page_flip(struct radeon_device *rdev, int crtc);
-extern u32 r100_page_flip(struct radeon_device *rdev, int crtc, u64 crtc_base);
-extern void r100_post_page_flip(struct radeon_device *rdev, int crtc);
+extern void r100_page_flip(struct radeon_device *rdev, int crtc,
+			   u64 crtc_base);
+extern bool r100_page_flip_pending(struct radeon_device *rdev, int crtc);
 extern void r100_wait_for_vblank(struct radeon_device *rdev, int crtc);
 extern int r100_mc_wait_for_idle(struct radeon_device *rdev);
 
@@ -171,7 +172,8 @@ extern void r300_fence_ring_emit(struct radeon_device *rdev,
 				struct radeon_fence *fence);
 extern int r300_cs_parse(struct radeon_cs_parser *p);
 extern void rv370_pcie_gart_tlb_flush(struct radeon_device *rdev);
-extern int rv370_pcie_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr);
+extern void rv370_pcie_gart_set_page(struct radeon_device *rdev, unsigned i,
+				     uint64_t addr);
 extern void rv370_set_pcie_lanes(struct radeon_device *rdev, int lanes);
 extern int rv370_get_pcie_lanes(struct radeon_device *rdev);
 extern void r300_set_reg_safe(struct radeon_device *rdev);
@@ -206,7 +208,8 @@ extern void rs400_fini(struct radeon_device *rdev);
 extern int rs400_suspend(struct radeon_device *rdev);
 extern int rs400_resume(struct radeon_device *rdev);
 void rs400_gart_tlb_flush(struct radeon_device *rdev);
-int rs400_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr);
+void rs400_gart_set_page(struct radeon_device *rdev, unsigned i,
+			 uint64_t addr);
 uint32_t rs400_mc_rreg(struct radeon_device *rdev, uint32_t reg);
 void rs400_mc_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v);
 int rs400_gart_init(struct radeon_device *rdev);
@@ -229,7 +232,8 @@ int rs600_irq_process(struct radeon_device *rdev);
 void rs600_irq_disable(struct radeon_device *rdev);
 u32 rs600_get_vblank_counter(struct radeon_device *rdev, int crtc);
 void rs600_gart_tlb_flush(struct radeon_device *rdev);
-int rs600_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr);
+void rs600_gart_set_page(struct radeon_device *rdev, unsigned i,
+			 uint64_t addr);
 uint32_t rs600_mc_rreg(struct radeon_device *rdev, uint32_t reg);
 void rs600_mc_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v);
 void rs600_bandwidth_update(struct radeon_device *rdev);
@@ -241,9 +245,9 @@ void rs600_hpd_set_polarity(struct radeon_device *rdev,
 extern void rs600_pm_misc(struct radeon_device *rdev);
 extern void rs600_pm_prepare(struct radeon_device *rdev);
 extern void rs600_pm_finish(struct radeon_device *rdev);
-extern void rs600_pre_page_flip(struct radeon_device *rdev, int crtc);
-extern u32 rs600_page_flip(struct radeon_device *rdev, int crtc, u64 crtc_base);
-extern void rs600_post_page_flip(struct radeon_device *rdev, int crtc);
+extern void rs600_page_flip(struct radeon_device *rdev, int crtc,
+			    u64 crtc_base);
+extern bool rs600_page_flip_pending(struct radeon_device *rdev, int crtc);
 void rs600_set_safe_registers(struct radeon_device *rdev);
 extern void avivo_wait_for_vblank(struct radeon_device *rdev, int crtc);
 extern int rs600_mc_wait_for_idle(struct radeon_device *rdev);
@@ -387,6 +391,11 @@ void r600_rlc_stop(struct radeon_device *rdev);
 int r600_audio_init(struct radeon_device *rdev);
 struct r600_audio_pin r600_audio_status(struct radeon_device *rdev);
 void r600_audio_fini(struct radeon_device *rdev);
+void r600_audio_set_dto(struct drm_encoder *encoder, u32 clock);
+void r600_hdmi_update_avi_infoframe(struct drm_encoder *encoder, void *buffer,
+				    size_t size);
+void r600_hdmi_update_ACR(struct drm_encoder *encoder, uint32_t clock);
+void r600_hdmi_audio_workaround(struct drm_encoder *encoder);
 int r600_hdmi_buffer_status_changed(struct drm_encoder *encoder);
 void r600_hdmi_update_audio_settings(struct drm_encoder *encoder);
 void r600_hdmi_enable(struct drm_encoder *encoder, bool enable);
@@ -447,7 +456,8 @@ void rv770_fini(struct radeon_device *rdev);
 int rv770_suspend(struct radeon_device *rdev);
 int rv770_resume(struct radeon_device *rdev);
 void rv770_pm_misc(struct radeon_device *rdev);
-u32 rv770_page_flip(struct radeon_device *rdev, int crtc, u64 crtc_base);
+void rv770_page_flip(struct radeon_device *rdev, int crtc, u64 crtc_base);
+bool rv770_page_flip_pending(struct radeon_device *rdev, int crtc);
 void r700_vram_gtt_location(struct radeon_device *rdev, struct radeon_mc *mc);
 void r700_cp_stop(struct radeon_device *rdev);
 void r700_cp_fini(struct radeon_device *rdev);
@@ -458,6 +468,8 @@ int rv770_copy_dma(struct radeon_device *rdev,
 u32 rv770_get_xclk(struct radeon_device *rdev);
 int rv770_set_uvd_clocks(struct radeon_device *rdev, u32 vclk, u32 dclk);
 int rv770_get_temp(struct radeon_device *rdev);
+/* hdmi */
+void dce3_1_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode *mode);
 /* rv7xx pm */
 int rv770_dpm_init(struct radeon_device *rdev);
 int rv770_dpm_enable(struct radeon_device *rdev);
@@ -513,9 +525,9 @@ extern void sumo_pm_init_profile(struct radeon_device *rdev);
 extern void btc_pm_init_profile(struct radeon_device *rdev);
 int sumo_set_uvd_clocks(struct radeon_device *rdev, u32 vclk, u32 dclk);
 int evergreen_set_uvd_clocks(struct radeon_device *rdev, u32 vclk, u32 dclk);
-extern void evergreen_pre_page_flip(struct radeon_device *rdev, int crtc);
-extern u32 evergreen_page_flip(struct radeon_device *rdev, int crtc, u64 crtc_base);
-extern void evergreen_post_page_flip(struct radeon_device *rdev, int crtc);
+extern void evergreen_page_flip(struct radeon_device *rdev, int crtc,
+				u64 crtc_base);
+extern bool evergreen_page_flip_pending(struct radeon_device *rdev, int crtc);
 extern void dce4_wait_for_vblank(struct radeon_device *rdev, int crtc);
 void evergreen_disable_interrupt_state(struct radeon_device *rdev);
 int evergreen_mc_wait_for_idle(struct radeon_device *rdev);
diff --git a/drivers/gpu/drm/radeon/radeon_bios.c b/drivers/gpu/drm/radeon/radeon_bios.c
index 9ab30976287d..6a03624fadaa 100644
--- a/drivers/gpu/drm/radeon/radeon_bios.c
+++ b/drivers/gpu/drm/radeon/radeon_bios.c
@@ -626,7 +626,7 @@ static bool radeon_acpi_vfct_bios(struct radeon_device *rdev)
 	    vhdr->DeviceID != rdev->pdev->device) {
 		DRM_INFO("ACPI VFCT table is not for this card\n");
 		goto out_unmap;
-	};
+	}
 
 	if (vfct->VBIOSImageOffset + sizeof(VFCT_IMAGE_HEADER) + vhdr->ImageLength > tbl_size) {
 		DRM_ERROR("ACPI VFCT image truncated\n");
diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c
index ea50e0ae7bf7..933c5c39654d 100644
--- a/drivers/gpu/drm/radeon/radeon_connectors.c
+++ b/drivers/gpu/drm/radeon/radeon_connectors.c
@@ -48,6 +48,7 @@ void radeon_connector_hotplug(struct drm_connector *connector)
 	radeon_hpd_set_polarity(rdev, radeon_connector->hpd.hpd);
 
 	/* if the connector is already off, don't turn it back on */
+	/* FIXME: This access isn't protected by any locks. */
 	if (connector->dpms != DRM_MODE_DPMS_ON)
 		return;
 
@@ -100,6 +101,7 @@ int radeon_get_monitor_bpc(struct drm_connector *connector)
 	struct radeon_connector *radeon_connector = to_radeon_connector(connector);
 	struct radeon_connector_atom_dig *dig_connector;
 	int bpc = 8;
+	int mode_clock, max_tmds_clock;
 
 	switch (connector->connector_type) {
 	case DRM_MODE_CONNECTOR_DVII:
@@ -145,6 +147,61 @@ int radeon_get_monitor_bpc(struct drm_connector *connector)
 		}
 		break;
 	}
+
+	if (drm_detect_hdmi_monitor(radeon_connector->edid)) {
+		/* hdmi deep color only implemented on DCE4+ */
+		if ((bpc > 8) && !ASIC_IS_DCE4(rdev)) {
+			DRM_DEBUG("%s: HDMI deep color %d bpc unsupported. Using 8 bpc.\n",
+					  connector->name, bpc);
+			bpc = 8;
+		}
+
+		/*
+		 * Pre DCE-8 hw can't handle > 12 bpc, and more than 12 bpc doesn't make
+		 * much sense without support for > 12 bpc framebuffers. RGB 4:4:4 at
+		 * 12 bpc is always supported on hdmi deep color sinks, as this is
+		 * required by the HDMI-1.3 spec. Clamp to a safe 12 bpc maximum.
+		 */
+		if (bpc > 12) {
+			DRM_DEBUG("%s: HDMI deep color %d bpc unsupported. Using 12 bpc.\n",
+					  connector->name, bpc);
+			bpc = 12;
+		}
+
+		/* Any defined maximum tmds clock limit we must not exceed? */
+		if (connector->max_tmds_clock > 0) {
+			/* mode_clock is clock in kHz for mode to be modeset on this connector */
+			mode_clock = radeon_connector->pixelclock_for_modeset;
+
+			/* Maximum allowable input clock in kHz */
+			max_tmds_clock = connector->max_tmds_clock * 1000;
+
+			DRM_DEBUG("%s: hdmi mode dotclock %d kHz, max tmds input clock %d kHz.\n",
+					  connector->name, mode_clock, max_tmds_clock);
+
+			/* Check if bpc is within clock limit. Try to degrade gracefully otherwise */
+			if ((bpc == 12) && (mode_clock * 3/2 > max_tmds_clock)) {
+				if ((connector->display_info.edid_hdmi_dc_modes & DRM_EDID_HDMI_DC_30) &&
+					(mode_clock * 5/4 <= max_tmds_clock))
+					bpc = 10;
+				else
+					bpc = 8;
+
+				DRM_DEBUG("%s: HDMI deep color 12 bpc exceeds max tmds clock. Using %d bpc.\n",
+						  connector->name, bpc);
+			}
+
+			if ((bpc == 10) && (mode_clock * 5/4 > max_tmds_clock)) {
+				bpc = 8;
+				DRM_DEBUG("%s: HDMI deep color 10 bpc exceeds max tmds clock. Using %d bpc.\n",
+						  connector->name, bpc);
+			}
+		}
+	}
+
+	DRM_DEBUG("%s: Display bpc=%d, returned bpc=%d\n",
+			  connector->name, connector->display_info.bpc, bpc);
+
 	return bpc;
 }
 
@@ -260,13 +317,17 @@ radeon_connector_analog_encoder_conflict_solve(struct drm_connector *connector,
 					continue;
 
 				if (priority == true) {
-					DRM_DEBUG_KMS("1: conflicting encoders switching off %s\n", drm_get_connector_name(conflict));
-					DRM_DEBUG_KMS("in favor of %s\n", drm_get_connector_name(connector));
+					DRM_DEBUG_KMS("1: conflicting encoders switching off %s\n",
+						      conflict->name);
+					DRM_DEBUG_KMS("in favor of %s\n",
+						      connector->name);
 					conflict->status = connector_status_disconnected;
 					radeon_connector_update_scratch_regs(conflict, connector_status_disconnected);
 				} else {
-					DRM_DEBUG_KMS("2: conflicting encoders switching off %s\n", drm_get_connector_name(connector));
-					DRM_DEBUG_KMS("in favor of %s\n", drm_get_connector_name(conflict));
+					DRM_DEBUG_KMS("2: conflicting encoders switching off %s\n",
+						      connector->name);
+					DRM_DEBUG_KMS("in favor of %s\n",
+						      conflict->name);
 					current_status = connector_status_disconnected;
 				}
 				break;
@@ -787,7 +848,7 @@ radeon_vga_detect(struct drm_connector *connector, bool force)
 
 		if (!radeon_connector->edid) {
 			DRM_ERROR("%s: probed a monitor but no|invalid EDID\n",
-					drm_get_connector_name(connector));
+					connector->name);
 			ret = connector_status_connected;
 		} else {
 			radeon_connector->use_digital = !!(radeon_connector->edid->input & DRM_EDID_INPUT_DIGITAL);
@@ -1010,12 +1071,13 @@ radeon_dvi_detect(struct drm_connector *connector, bool force)
 
 		if (!radeon_connector->edid) {
 			DRM_ERROR("%s: probed a monitor but no|invalid EDID\n",
-					drm_get_connector_name(connector));
+					connector->name);
 			/* rs690 seems to have a problem with connectors not existing and always
 			 * return a block of 0's. If we see this just stop polling on this output */
 			if ((rdev->family == CHIP_RS690 || rdev->family == CHIP_RS740) && radeon_connector->base.null_edid_counter) {
 				ret = connector_status_disconnected;
-				DRM_ERROR("%s: detected RS690 floating bus bug, stopping ddc detect\n", drm_get_connector_name(connector));
+				DRM_ERROR("%s: detected RS690 floating bus bug, stopping ddc detect\n",
+					  connector->name);
 				radeon_connector->ddc_bus = NULL;
 			} else {
 				ret = connector_status_connected;
@@ -1387,7 +1449,7 @@ bool radeon_connector_is_dp12_capable(struct drm_connector *connector)
 	struct radeon_device *rdev = dev->dev_private;
 
 	if (ASIC_IS_DCE5(rdev) &&
-	    (rdev->clock.dp_extclk >= 53900) &&
+	    (rdev->clock.default_dispclk >= 53900) &&
 	    radeon_connector_encoder_is_hbr2(connector)) {
 		return true;
 	}
diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c
index 41ecf8a60611..71a143461478 100644
--- a/drivers/gpu/drm/radeon/radeon_cs.c
+++ b/drivers/gpu/drm/radeon/radeon_cs.c
@@ -140,10 +140,10 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p)
 		if (p->ring == R600_RING_TYPE_UVD_INDEX &&
 		    (i == 0 || drm_pci_device_is_agp(p->rdev->ddev))) {
 			/* TODO: is this still needed for NI+ ? */
-			p->relocs[i].domain =
+			p->relocs[i].prefered_domains =
 				RADEON_GEM_DOMAIN_VRAM;
 
-			p->relocs[i].alt_domain =
+			p->relocs[i].allowed_domains =
 				RADEON_GEM_DOMAIN_VRAM;
 
 			/* prioritize this over any other relocation */
@@ -158,10 +158,10 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p)
 				return -EINVAL;
 			}
 
-			p->relocs[i].domain = domain;
+			p->relocs[i].prefered_domains = domain;
 			if (domain == RADEON_GEM_DOMAIN_VRAM)
 				domain |= RADEON_GEM_DOMAIN_GTT;
-			p->relocs[i].alt_domain = domain;
+			p->relocs[i].allowed_domains = domain;
 		}
 
 		p->relocs[i].tv.bo = &p->relocs[i].robj->tbo;
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index 2cd144c378d6..03686fab842d 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -1052,6 +1052,43 @@ static void radeon_check_arguments(struct radeon_device *rdev)
 		radeon_agpmode = 0;
 		break;
 	}
+
+	if (!radeon_check_pot_argument(radeon_vm_size)) {
+		dev_warn(rdev->dev, "VM size (%d) must be a power of 2\n",
+			 radeon_vm_size);
+		radeon_vm_size = 4096;
+	}
+
+	if (radeon_vm_size < 4) {
+		dev_warn(rdev->dev, "VM size (%d) to small, min is 4MB\n",
+			 radeon_vm_size);
+		radeon_vm_size = 4096;
+	}
+
+       /*
+        * Max GPUVM size for Cayman, SI and CI are 40 bits.
+        */
+	if (radeon_vm_size > 1024*1024) {
+		dev_warn(rdev->dev, "VM size (%d) to large, max is 1TB\n",
+			 radeon_vm_size);
+		radeon_vm_size = 4096;
+	}
+
+	/* defines number of bits in page table versus page directory,
+	 * a page is 4KB so we have 12 bits offset, minimum 9 bits in the
+	 * page table and the remaining bits are in the page directory */
+	if (radeon_vm_block_size < 9) {
+		dev_warn(rdev->dev, "VM page table size (%d) to small\n",
+			 radeon_vm_block_size);
+		radeon_vm_block_size = 9;
+	}
+
+	if (radeon_vm_block_size > 24 ||
+	    radeon_vm_size < (1ull << radeon_vm_block_size)) {
+		dev_warn(rdev->dev, "VM page table size (%d) to large\n",
+			 radeon_vm_block_size);
+		radeon_vm_block_size = 9;
+	}
 }
 
 /**
@@ -1126,12 +1163,13 @@ static void radeon_switcheroo_set_state(struct pci_dev *pdev, enum vga_switchero
 static bool radeon_switcheroo_can_switch(struct pci_dev *pdev)
 {
 	struct drm_device *dev = pci_get_drvdata(pdev);
-	bool can_switch;
 
-	spin_lock(&dev->count_lock);
-	can_switch = (dev->open_count == 0);
-	spin_unlock(&dev->count_lock);
-	return can_switch;
+	/*
+	 * FIXME: open_count is protected by drm_global_mutex but that would lead to
+	 * locking inversion with the driver load path. And the access here is
+	 * completely racy anyway. So don't bother with locking for now.
+	 */
+	return dev->open_count == 0;
 }
 
 static const struct vga_switcheroo_client_ops radeon_switcheroo_ops = {
@@ -1196,17 +1234,16 @@ int radeon_device_init(struct radeon_device *rdev,
 	if (r)
 		return r;
 
+	radeon_check_arguments(rdev);
 	/* Adjust VM size here.
-	 * Currently set to 4GB ((1 << 20) 4k pages).
-	 * Max GPUVM size for cayman and SI is 40 bits.
+	 * Max GPUVM size for cayman+ is 40 bits.
 	 */
-	rdev->vm_manager.max_pfn = 1 << 20;
+	rdev->vm_manager.max_pfn = radeon_vm_size << 8;
 
 	/* Set asic functions */
 	r = radeon_asic_init(rdev);
 	if (r)
 		return r;
-	radeon_check_arguments(rdev);
 
 	/* all of the newer IGP chips have an internal gart
 	 * However some rs4xx report as AGP, so remove that here.
diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c
index 356b733caafe..5ed617056b9c 100644
--- a/drivers/gpu/drm/radeon/radeon_display.c
+++ b/drivers/gpu/drm/radeon/radeon_display.c
@@ -249,16 +249,21 @@ static void radeon_crtc_destroy(struct drm_crtc *crtc)
 	struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc);
 
 	drm_crtc_cleanup(crtc);
+	destroy_workqueue(radeon_crtc->flip_queue);
 	kfree(radeon_crtc);
 }
 
-/*
- * Handle unpin events outside the interrupt handler proper.
+/**
+ * radeon_unpin_work_func - unpin old buffer object
+ *
+ * @__work - kernel work item
+ *
+ * Unpin the old frame buffer object outside of the interrupt handler
  */
 static void radeon_unpin_work_func(struct work_struct *__work)
 {
-	struct radeon_unpin_work *work =
-		container_of(__work, struct radeon_unpin_work, work);
+	struct radeon_flip_work *work =
+		container_of(__work, struct radeon_flip_work, unpin_work);
 	int r;
 
 	/* unpin of the old buffer */
@@ -276,10 +281,10 @@ static void radeon_unpin_work_func(struct work_struct *__work)
 	kfree(work);
 }
 
-void radeon_crtc_handle_flip(struct radeon_device *rdev, int crtc_id)
+void radeon_crtc_handle_vblank(struct radeon_device *rdev, int crtc_id)
 {
 	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
-	struct radeon_unpin_work *work;
+	struct radeon_flip_work *work;
 	unsigned long flags;
 	u32 update_pending;
 	int vpos, hpos;
@@ -289,24 +294,13 @@ void radeon_crtc_handle_flip(struct radeon_device *rdev, int crtc_id)
 		return;
 
 	spin_lock_irqsave(&rdev->ddev->event_lock, flags);
-	work = radeon_crtc->unpin_work;
-	if (work == NULL ||
-	    (work->fence && !radeon_fence_signaled(work->fence))) {
+	work = radeon_crtc->flip_work;
+	if (work == NULL) {
 		spin_unlock_irqrestore(&rdev->ddev->event_lock, flags);
 		return;
 	}
-	/* New pageflip, or just completion of a previous one? */
-	if (!radeon_crtc->deferred_flip_completion) {
-		/* do the flip (mmio) */
-		update_pending = radeon_page_flip(rdev, crtc_id, work->new_crtc_base);
-	} else {
-		/* This is just a completion of a flip queued in crtc
-		 * at last invocation. Make sure we go directly to
-		 * completion routine.
-		 */
-		update_pending = 0;
-		radeon_crtc->deferred_flip_completion = 0;
-	}
+
+	update_pending = radeon_page_flip_pending(rdev, crtc_id);
 
 	/* Has the pageflip already completed in crtc, or is it certain
 	 * to complete in this vblank?
@@ -324,19 +318,38 @@ void radeon_crtc_handle_flip(struct radeon_device *rdev, int crtc_id)
 		 */
 		update_pending = 0;
 	}
-	if (update_pending) {
-		/* crtc didn't flip in this target vblank interval,
-		 * but flip is pending in crtc. It will complete it
-		 * in next vblank interval, so complete the flip at
-		 * next vblank irq.
-		 */
-		radeon_crtc->deferred_flip_completion = 1;
+	spin_unlock_irqrestore(&rdev->ddev->event_lock, flags);
+	if (!update_pending)
+		radeon_crtc_handle_flip(rdev, crtc_id);
+}
+
+/**
+ * radeon_crtc_handle_flip - page flip completed
+ *
+ * @rdev: radeon device pointer
+ * @crtc_id: crtc number this event is for
+ *
+ * Called when we are sure that a page flip for this crtc is completed.
+ */
+void radeon_crtc_handle_flip(struct radeon_device *rdev, int crtc_id)
+{
+	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
+	struct radeon_flip_work *work;
+	unsigned long flags;
+
+	/* this can happen at init */
+	if (radeon_crtc == NULL)
+		return;
+
+	spin_lock_irqsave(&rdev->ddev->event_lock, flags);
+	work = radeon_crtc->flip_work;
+	if (work == NULL) {
 		spin_unlock_irqrestore(&rdev->ddev->event_lock, flags);
 		return;
 	}
 
-	/* Pageflip (will be) certainly completed in this vblank. Clean up. */
-	radeon_crtc->unpin_work = NULL;
+	/* Pageflip completed. Clean up. */
+	radeon_crtc->flip_work = NULL;
 
 	/* wakeup userspace */
 	if (work->event)
@@ -344,86 +357,71 @@ void radeon_crtc_handle_flip(struct radeon_device *rdev, int crtc_id)
 
 	spin_unlock_irqrestore(&rdev->ddev->event_lock, flags);
 
-	drm_vblank_put(rdev->ddev, radeon_crtc->crtc_id);
 	radeon_fence_unref(&work->fence);
-	radeon_post_page_flip(work->rdev, work->crtc_id);
-	schedule_work(&work->work);
+	radeon_irq_kms_pflip_irq_get(rdev, work->crtc_id);
+	queue_work(radeon_crtc->flip_queue, &work->unpin_work);
 }
 
-static int radeon_crtc_page_flip(struct drm_crtc *crtc,
-				 struct drm_framebuffer *fb,
-				 struct drm_pending_vblank_event *event,
-				 uint32_t page_flip_flags)
+/**
+ * radeon_flip_work_func - page flip framebuffer
+ *
+ * @work - kernel work item
+ *
+ * Wait for the buffer object to become idle and do the actual page flip
+ */
+static void radeon_flip_work_func(struct work_struct *__work)
 {
-	struct drm_device *dev = crtc->dev;
-	struct radeon_device *rdev = dev->dev_private;
-	struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc);
-	struct radeon_framebuffer *old_radeon_fb;
-	struct radeon_framebuffer *new_radeon_fb;
-	struct drm_gem_object *obj;
-	struct radeon_bo *rbo;
-	struct radeon_unpin_work *work;
-	unsigned long flags;
-	u32 tiling_flags, pitch_pixels;
-	u64 base;
-	int r;
+	struct radeon_flip_work *work =
+		container_of(__work, struct radeon_flip_work, flip_work);
+	struct radeon_device *rdev = work->rdev;
+	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[work->crtc_id];
 
-	work = kzalloc(sizeof *work, GFP_KERNEL);
-	if (work == NULL)
-		return -ENOMEM;
+	struct drm_crtc *crtc = &radeon_crtc->base;
+	struct drm_framebuffer *fb = work->fb;
 
-	work->event = event;
-	work->rdev = rdev;
-	work->crtc_id = radeon_crtc->crtc_id;
-	old_radeon_fb = to_radeon_framebuffer(crtc->primary->fb);
-	new_radeon_fb = to_radeon_framebuffer(fb);
-	/* schedule unpin of the old buffer */
-	obj = old_radeon_fb->obj;
-	/* take a reference to the old object */
-	drm_gem_object_reference(obj);
-	rbo = gem_to_radeon_bo(obj);
-	work->old_rbo = rbo;
-	obj = new_radeon_fb->obj;
-	rbo = gem_to_radeon_bo(obj);
+	uint32_t tiling_flags, pitch_pixels;
+	uint64_t base;
 
-	spin_lock(&rbo->tbo.bdev->fence_lock);
-	if (rbo->tbo.sync_obj)
-		work->fence = radeon_fence_ref(rbo->tbo.sync_obj);
-	spin_unlock(&rbo->tbo.bdev->fence_lock);
+	unsigned long flags;
+	int r;
 
-	INIT_WORK(&work->work, radeon_unpin_work_func);
+        down_read(&rdev->exclusive_lock);
+	while (work->fence) {
+		r = radeon_fence_wait(work->fence, false);
+		if (r == -EDEADLK) {
+			up_read(&rdev->exclusive_lock);
+			r = radeon_gpu_reset(rdev);
+			down_read(&rdev->exclusive_lock);
+		}
 
-	/* We borrow the event spin lock for protecting unpin_work */
-	spin_lock_irqsave(&dev->event_lock, flags);
-	if (radeon_crtc->unpin_work) {
-		DRM_DEBUG_DRIVER("flip queue: crtc already busy\n");
-		r = -EBUSY;
-		goto unlock_free;
+		if (r) {
+			DRM_ERROR("failed to wait on page flip fence (%d)!\n",
+				  r);
+			goto cleanup;
+		} else
+			radeon_fence_unref(&work->fence);
 	}
-	radeon_crtc->unpin_work = work;
-	radeon_crtc->deferred_flip_completion = 0;
-	spin_unlock_irqrestore(&dev->event_lock, flags);
 
 	/* pin the new buffer */
 	DRM_DEBUG_DRIVER("flip-ioctl() cur_fbo = %p, cur_bbo = %p\n",
-			 work->old_rbo, rbo);
+			 work->old_rbo, work->new_rbo);
 
-	r = radeon_bo_reserve(rbo, false);
+	r = radeon_bo_reserve(work->new_rbo, false);
 	if (unlikely(r != 0)) {
 		DRM_ERROR("failed to reserve new rbo buffer before flip\n");
-		goto pflip_cleanup;
+		goto cleanup;
 	}
 	/* Only 27 bit offset for legacy CRTC */
-	r = radeon_bo_pin_restricted(rbo, RADEON_GEM_DOMAIN_VRAM,
+	r = radeon_bo_pin_restricted(work->new_rbo, RADEON_GEM_DOMAIN_VRAM,
 				     ASIC_IS_AVIVO(rdev) ? 0 : 1 << 27, &base);
 	if (unlikely(r != 0)) {
-		radeon_bo_unreserve(rbo);
+		radeon_bo_unreserve(work->new_rbo);
 		r = -EINVAL;
 		DRM_ERROR("failed to pin new rbo buffer before flip\n");
-		goto pflip_cleanup;
+		goto cleanup;
 	}
-	radeon_bo_get_tiling_flags(rbo, &tiling_flags, NULL);
-	radeon_bo_unreserve(rbo);
+	radeon_bo_get_tiling_flags(work->new_rbo, &tiling_flags, NULL);
+	radeon_bo_unreserve(work->new_rbo);
 
 	if (!ASIC_IS_AVIVO(rdev)) {
 		/* crtc offset is from display base addr not FB location */
@@ -461,44 +459,91 @@ static int radeon_crtc_page_flip(struct drm_crtc *crtc,
 		base &= ~7;
 	}
 
-	spin_lock_irqsave(&dev->event_lock, flags);
-	work->new_crtc_base = base;
-	spin_unlock_irqrestore(&dev->event_lock, flags);
+	/* We borrow the event spin lock for protecting flip_work */
+	spin_lock_irqsave(&crtc->dev->event_lock, flags);
 
-	/* update crtc fb */
-	crtc->primary->fb = fb;
+	/* set the proper interrupt */
+	radeon_irq_kms_pflip_irq_get(rdev, radeon_crtc->crtc_id);
 
-	r = drm_vblank_get(dev, radeon_crtc->crtc_id);
-	if (r) {
-		DRM_ERROR("failed to get vblank before flip\n");
-		goto pflip_cleanup1;
-	}
+	/* do the flip (mmio) */
+	radeon_page_flip(rdev, radeon_crtc->crtc_id, base);
 
-	/* set the proper interrupt */
-	radeon_pre_page_flip(rdev, radeon_crtc->crtc_id);
+	spin_unlock_irqrestore(&crtc->dev->event_lock, flags);
+	up_read(&rdev->exclusive_lock);
 
-	return 0;
+	return;
 
-pflip_cleanup1:
-	if (unlikely(radeon_bo_reserve(rbo, false) != 0)) {
-		DRM_ERROR("failed to reserve new rbo in error path\n");
-		goto pflip_cleanup;
-	}
-	if (unlikely(radeon_bo_unpin(rbo) != 0)) {
-		DRM_ERROR("failed to unpin new rbo in error path\n");
-	}
-	radeon_bo_unreserve(rbo);
-
-pflip_cleanup:
-	spin_lock_irqsave(&dev->event_lock, flags);
-	radeon_crtc->unpin_work = NULL;
-unlock_free:
-	spin_unlock_irqrestore(&dev->event_lock, flags);
-	drm_gem_object_unreference_unlocked(old_radeon_fb->obj);
+cleanup:
+	drm_gem_object_unreference_unlocked(&work->old_rbo->gem_base);
 	radeon_fence_unref(&work->fence);
 	kfree(work);
+	up_read(&rdev->exclusive_lock);
+}
 
-	return r;
+static int radeon_crtc_page_flip(struct drm_crtc *crtc,
+				 struct drm_framebuffer *fb,
+				 struct drm_pending_vblank_event *event,
+				 uint32_t page_flip_flags)
+{
+	struct drm_device *dev = crtc->dev;
+	struct radeon_device *rdev = dev->dev_private;
+	struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc);
+	struct radeon_framebuffer *old_radeon_fb;
+	struct radeon_framebuffer *new_radeon_fb;
+	struct drm_gem_object *obj;
+	struct radeon_flip_work *work;
+	unsigned long flags;
+
+	work = kzalloc(sizeof *work, GFP_KERNEL);
+	if (work == NULL)
+		return -ENOMEM;
+
+	INIT_WORK(&work->flip_work, radeon_flip_work_func);
+	INIT_WORK(&work->unpin_work, radeon_unpin_work_func);
+
+	work->rdev = rdev;
+	work->crtc_id = radeon_crtc->crtc_id;
+	work->fb = fb;
+	work->event = event;
+
+	/* schedule unpin of the old buffer */
+	old_radeon_fb = to_radeon_framebuffer(crtc->primary->fb);
+	obj = old_radeon_fb->obj;
+
+	/* take a reference to the old object */
+	drm_gem_object_reference(obj);
+	work->old_rbo = gem_to_radeon_bo(obj);
+
+	new_radeon_fb = to_radeon_framebuffer(fb);
+	obj = new_radeon_fb->obj;
+	work->new_rbo = gem_to_radeon_bo(obj);
+
+	spin_lock(&work->new_rbo->tbo.bdev->fence_lock);
+	if (work->new_rbo->tbo.sync_obj)
+		work->fence = radeon_fence_ref(work->new_rbo->tbo.sync_obj);
+	spin_unlock(&work->new_rbo->tbo.bdev->fence_lock);
+
+	/* We borrow the event spin lock for protecting flip_work */
+	spin_lock_irqsave(&crtc->dev->event_lock, flags);
+
+	if (radeon_crtc->flip_work) {
+		DRM_DEBUG_DRIVER("flip queue: crtc already busy\n");
+		spin_unlock_irqrestore(&crtc->dev->event_lock, flags);
+		drm_gem_object_unreference_unlocked(&work->old_rbo->gem_base);
+		radeon_fence_unref(&work->fence);
+		kfree(work);
+		return -EBUSY;
+	}
+	radeon_crtc->flip_work = work;
+
+	/* update crtc fb */
+	crtc->primary->fb = fb;
+
+	spin_unlock_irqrestore(&crtc->dev->event_lock, flags);
+
+	queue_work(radeon_crtc->flip_queue, &work->flip_work);
+
+	return 0;
 }
 
 static int
@@ -568,6 +613,7 @@ static void radeon_crtc_init(struct drm_device *dev, int index)
 
 	drm_mode_crtc_set_gamma_size(&radeon_crtc->base, 256);
 	radeon_crtc->crtc_id = index;
+	radeon_crtc->flip_queue = create_singlethread_workqueue("radeon-crtc");
 	rdev->mode_info.crtcs[index] = radeon_crtc;
 
 	if (rdev->family >= CHIP_BONAIRE) {
@@ -661,7 +707,7 @@ static void radeon_print_display_setup(struct drm_device *dev)
 	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
 		radeon_connector = to_radeon_connector(connector);
 		DRM_INFO("Connector %d:\n", i);
-		DRM_INFO("  %s\n", drm_get_connector_name(connector));
+		DRM_INFO("  %s\n", connector->name);
 		if (radeon_connector->hpd.hpd != RADEON_HPD_NONE)
 			DRM_INFO("  %s\n", hpd_names[radeon_connector->hpd.hpd]);
 		if (radeon_connector->ddc_bus) {
diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c
index c00a2f585185..6e3017413386 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.c
+++ b/drivers/gpu/drm/radeon/radeon_drv.c
@@ -81,9 +81,10 @@
  *   2.37.0 - allow GS ring setup on r6xx/r7xx
  *   2.38.0 - RADEON_GEM_OP (GET_INITIAL_DOMAIN, SET_INITIAL_DOMAIN),
  *            CIK: 1D and linear tiling modes contain valid PIPE_CONFIG
+ *   2.39.0 - Add INFO query for number of active CUs
  */
 #define KMS_DRIVER_MAJOR	2
-#define KMS_DRIVER_MINOR	38
+#define KMS_DRIVER_MINOR	39
 #define KMS_DRIVER_PATCHLEVEL	0
 int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags);
 int radeon_driver_unload_kms(struct drm_device *dev);
@@ -172,6 +173,8 @@ int radeon_dpm = -1;
 int radeon_aspm = -1;
 int radeon_runtime_pm = -1;
 int radeon_hard_reset = 0;
+int radeon_vm_size = 4096;
+int radeon_vm_block_size = 9;
 
 MODULE_PARM_DESC(no_wb, "Disable AGP writeback for scratch registers");
 module_param_named(no_wb, radeon_no_wb, int, 0444);
@@ -239,6 +242,12 @@ module_param_named(runpm, radeon_runtime_pm, int, 0444);
 MODULE_PARM_DESC(hard_reset, "PCI config reset (1 = force enable, 0 = disable (default))");
 module_param_named(hard_reset, radeon_hard_reset, int, 0444);
 
+MODULE_PARM_DESC(vm_size, "VM address space size in megabytes (default 4GB)");
+module_param_named(vm_size, radeon_vm_size, int, 0444);
+
+MODULE_PARM_DESC(vm_block_size, "VM page table size in bits (default 9)");
+module_param_named(vm_block_size, radeon_vm_block_size, int, 0444);
+
 static struct pci_device_id pciidlist[] = {
 	radeon_PCI_IDS
 };
@@ -519,7 +528,6 @@ static struct drm_driver kms_driver = {
 	    DRIVER_USE_AGP |
 	    DRIVER_HAVE_IRQ | DRIVER_IRQ_SHARED | DRIVER_GEM |
 	    DRIVER_PRIME | DRIVER_RENDER,
-	.dev_priv_size = 0,
 	.load = radeon_driver_load_kms,
 	.open = radeon_driver_open_kms,
 	.preclose = radeon_driver_preclose_kms,
diff --git a/drivers/gpu/drm/radeon/radeon_fence.c b/drivers/gpu/drm/radeon/radeon_fence.c
index a77b1c13ea43..913787085dfa 100644
--- a/drivers/gpu/drm/radeon/radeon_fence.c
+++ b/drivers/gpu/drm/radeon/radeon_fence.c
@@ -819,15 +819,35 @@ static int radeon_debugfs_fence_info(struct seq_file *m, void *data)
 	return 0;
 }
 
+/**
+ * radeon_debugfs_gpu_reset - manually trigger a gpu reset
+ *
+ * Manually trigger a gpu reset at the next fence wait.
+ */
+static int radeon_debugfs_gpu_reset(struct seq_file *m, void *data)
+{
+	struct drm_info_node *node = (struct drm_info_node *) m->private;
+	struct drm_device *dev = node->minor->dev;
+	struct radeon_device *rdev = dev->dev_private;
+
+	down_read(&rdev->exclusive_lock);
+	seq_printf(m, "%d\n", rdev->needs_reset);
+	rdev->needs_reset = true;
+	up_read(&rdev->exclusive_lock);
+
+	return 0;
+}
+
 static struct drm_info_list radeon_debugfs_fence_list[] = {
 	{"radeon_fence_info", &radeon_debugfs_fence_info, 0, NULL},
+	{"radeon_gpu_reset", &radeon_debugfs_gpu_reset, 0, NULL}
 };
 #endif
 
 int radeon_debugfs_fence_init(struct radeon_device *rdev)
 {
 #if defined(CONFIG_DEBUG_FS)
-	return radeon_debugfs_add_files(rdev, radeon_debugfs_fence_list, 1);
+	return radeon_debugfs_add_files(rdev, radeon_debugfs_fence_list, 2);
 #else
 	return 0;
 #endif
diff --git a/drivers/gpu/drm/radeon/radeon_i2c.c b/drivers/gpu/drm/radeon/radeon_i2c.c
index 7b944142a9fd..add622008407 100644
--- a/drivers/gpu/drm/radeon/radeon_i2c.c
+++ b/drivers/gpu/drm/radeon/radeon_i2c.c
@@ -94,6 +94,8 @@ static int pre_xfer(struct i2c_adapter *i2c_adap)
 	struct radeon_i2c_bus_rec *rec = &i2c->rec;
 	uint32_t temp;
 
+	mutex_lock(&i2c->mutex);
+
 	/* RV410 appears to have a bug where the hw i2c in reset
 	 * holds the i2c port in a bad state - switch hw i2c away before
 	 * doing DDC - do this for all r200s/r300s/r400s for safety sake
@@ -170,6 +172,8 @@ static void post_xfer(struct i2c_adapter *i2c_adap)
 	temp = RREG32(rec->mask_data_reg) & ~rec->mask_data_mask;
 	WREG32(rec->mask_data_reg, temp);
 	temp = RREG32(rec->mask_data_reg);
+
+	mutex_unlock(&i2c->mutex);
 }
 
 static int get_clock(void *i2c_priv)
@@ -813,6 +817,8 @@ static int radeon_hw_i2c_xfer(struct i2c_adapter *i2c_adap,
 	struct radeon_i2c_bus_rec *rec = &i2c->rec;
 	int ret = 0;
 
+	mutex_lock(&i2c->mutex);
+
 	switch (rdev->family) {
 	case CHIP_R100:
 	case CHIP_RV100:
@@ -879,6 +885,8 @@ static int radeon_hw_i2c_xfer(struct i2c_adapter *i2c_adap,
 		break;
 	}
 
+	mutex_unlock(&i2c->mutex);
+
 	return ret;
 }
 
@@ -919,6 +927,7 @@ struct radeon_i2c_chan *radeon_i2c_create(struct drm_device *dev,
 	i2c->adapter.dev.parent = &dev->pdev->dev;
 	i2c->dev = dev;
 	i2c_set_adapdata(&i2c->adapter, i2c);
+	mutex_init(&i2c->mutex);
 	if (rec->mm_i2c ||
 	    (rec->hw_capable &&
 	     radeon_hw_i2c &&
@@ -979,7 +988,7 @@ void radeon_i2c_destroy(struct radeon_i2c_chan *i2c)
 		return;
 	i2c_del_adapter(&i2c->adapter);
 	if (i2c->has_aux)
-		drm_dp_aux_unregister_i2c_bus(&i2c->aux);
+		drm_dp_aux_unregister(&i2c->aux);
 	kfree(i2c);
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_ioc32.c b/drivers/gpu/drm/radeon/radeon_ioc32.c
index bdb0f93e73bc..0b98ea134579 100644
--- a/drivers/gpu/drm/radeon/radeon_ioc32.c
+++ b/drivers/gpu/drm/radeon/radeon_ioc32.c
@@ -399,7 +399,7 @@ long radeon_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 	if (nr < DRM_COMMAND_BASE)
 		return drm_compat_ioctl(filp, cmd, arg);
 
-	if (nr < DRM_COMMAND_BASE + DRM_ARRAY_SIZE(radeon_compat_ioctls))
+	if (nr < DRM_COMMAND_BASE + ARRAY_SIZE(radeon_compat_ioctls))
 		fn = radeon_compat_ioctls[nr - DRM_COMMAND_BASE];
 
 	if (fn != NULL)
diff --git a/drivers/gpu/drm/radeon/radeon_irq_kms.c b/drivers/gpu/drm/radeon/radeon_irq_kms.c
index 089c9ffb0aa9..16807afab362 100644
--- a/drivers/gpu/drm/radeon/radeon_irq_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_irq_kms.c
@@ -287,7 +287,7 @@ int radeon_irq_kms_init(struct radeon_device *rdev)
 	INIT_WORK(&rdev->reset_work, radeon_irq_reset_work_func);
 
 	rdev->irq.installed = true;
-	r = drm_irq_install(rdev->ddev);
+	r = drm_irq_install(rdev->ddev, rdev->ddev->pdev->irq);
 	if (r) {
 		rdev->irq.installed = false;
 		flush_work(&rdev->hotplug_work);
diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c
index eaaedba04675..35d931881b4b 100644
--- a/drivers/gpu/drm/radeon/radeon_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_kms.c
@@ -513,6 +513,22 @@ static int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 		value_size = sizeof(uint64_t);
 		value64 = atomic64_read(&rdev->gtt_usage);
 		break;
+	case RADEON_INFO_ACTIVE_CU_COUNT:
+		if (rdev->family >= CHIP_BONAIRE)
+			*value = rdev->config.cik.active_cus;
+		else if (rdev->family >= CHIP_TAHITI)
+			*value = rdev->config.si.active_cus;
+		else if (rdev->family >= CHIP_CAYMAN)
+			*value = rdev->config.cayman.active_simds;
+		else if (rdev->family >= CHIP_CEDAR)
+			*value = rdev->config.evergreen.active_simds;
+		else if (rdev->family >= CHIP_RV770)
+			*value = rdev->config.rv770.active_simds;
+		else if (rdev->family >= CHIP_R600)
+			*value = rdev->config.r600.active_simds;
+		else
+			*value = 1;
+		break;
 	default:
 		DRM_DEBUG_KMS("Invalid request %d\n", info->request);
 		return -EINVAL;
@@ -859,4 +875,4 @@ const struct drm_ioctl_desc radeon_ioctls_kms[] = {
 	DRM_IOCTL_DEF_DRV(RADEON_GEM_VA, radeon_gem_va_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(RADEON_GEM_OP, radeon_gem_op_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
 };
-int radeon_max_kms_ioctl = DRM_ARRAY_SIZE(radeon_ioctls_kms);
+int radeon_max_kms_ioctl = ARRAY_SIZE(radeon_ioctls_kms);
diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h
index 6ddf31a2d34e..ad0e4b8cc7e3 100644
--- a/drivers/gpu/drm/radeon/radeon_mode.h
+++ b/drivers/gpu/drm/radeon/radeon_mode.h
@@ -191,6 +191,7 @@ struct radeon_i2c_chan {
 	struct radeon_i2c_bus_rec rec;
 	struct drm_dp_aux aux;
 	bool has_aux;
+	struct mutex mutex;
 };
 
 /* mostly for macs, but really any system without connector tables */
@@ -324,8 +325,8 @@ struct radeon_crtc {
 	struct drm_display_mode native_mode;
 	int pll_id;
 	/* page flipping */
-	struct radeon_unpin_work *unpin_work;
-	int deferred_flip_completion;
+	struct workqueue_struct *flip_queue;
+	struct radeon_flip_work *flip_work;
 	/* pll sharing */
 	struct radeon_atom_ss ss;
 	bool ss_enabled;
@@ -505,6 +506,7 @@ struct radeon_connector {
 	struct radeon_i2c_chan *router_bus;
 	enum radeon_connector_audio audio;
 	enum radeon_connector_dither dither;
+	int pixelclock_for_modeset;
 };
 
 struct radeon_framebuffer {
@@ -906,6 +908,7 @@ bool radeon_fbdev_robj_is_fb(struct radeon_device *rdev, struct radeon_bo *robj)
 
 void radeon_fb_output_poll_changed(struct radeon_device *rdev);
 
+void radeon_crtc_handle_vblank(struct radeon_device *rdev, int crtc_id);
 void radeon_crtc_handle_flip(struct radeon_device *rdev, int crtc_id);
 
 int radeon_align_pitch(struct radeon_device *rdev, int width, int bpp, bool tiled);
diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c
index 4faa4d6f9bb4..6c717b257d6d 100644
--- a/drivers/gpu/drm/radeon/radeon_object.c
+++ b/drivers/gpu/drm/radeon/radeon_object.c
@@ -446,7 +446,7 @@ int radeon_bo_list_validate(struct radeon_device *rdev,
 	list_for_each_entry(lobj, head, tv.head) {
 		bo = lobj->robj;
 		if (!bo->pin_count) {
-			u32 domain = lobj->domain;
+			u32 domain = lobj->prefered_domains;
 			u32 current_domain =
 				radeon_mem_type_to_domain(bo->tbo.mem.mem_type);
 
@@ -458,7 +458,7 @@ int radeon_bo_list_validate(struct radeon_device *rdev,
 			 * into account. We don't want to disallow buffer moves
 			 * completely.
 			 */
-			if ((lobj->alt_domain & current_domain) != 0 &&
+			if ((lobj->allowed_domains & current_domain) != 0 &&
 			    (domain & current_domain) == 0 && /* will be moved */
 			    bytes_moved > bytes_moved_threshold) {
 				/* don't move it */
@@ -476,8 +476,9 @@ int radeon_bo_list_validate(struct radeon_device *rdev,
 				       initial_bytes_moved;
 
 			if (unlikely(r)) {
-				if (r != -ERESTARTSYS && domain != lobj->alt_domain) {
-					domain = lobj->alt_domain;
+				if (r != -ERESTARTSYS &&
+				    domain != lobj->allowed_domains) {
+					domain = lobj->allowed_domains;
 					goto retry;
 				}
 				ttm_eu_backoff_reservation(ticket, head);
@@ -730,7 +731,7 @@ int radeon_bo_wait(struct radeon_bo *bo, u32 *mem_type, bool no_wait)
 {
 	int r;
 
-	r = ttm_bo_reserve(&bo->tbo, true, no_wait, false, 0);
+	r = ttm_bo_reserve(&bo->tbo, true, no_wait, false, NULL);
 	if (unlikely(r != 0))
 		return r;
 	spin_lock(&bo->tbo.bdev->fence_lock);
diff --git a/drivers/gpu/drm/radeon/radeon_object.h b/drivers/gpu/drm/radeon/radeon_object.h
index 9e7b25a0629d..5a873f31a171 100644
--- a/drivers/gpu/drm/radeon/radeon_object.h
+++ b/drivers/gpu/drm/radeon/radeon_object.h
@@ -65,7 +65,7 @@ static inline int radeon_bo_reserve(struct radeon_bo *bo, bool no_intr)
 {
 	int r;
 
-	r = ttm_bo_reserve(&bo->tbo, !no_intr, false, false, 0);
+	r = ttm_bo_reserve(&bo->tbo, !no_intr, false, false, NULL);
 	if (unlikely(r != 0)) {
 		if (r != -ERESTARTSYS)
 			dev_err(bo->rdev->dev, "%p reserve failed\n", bo);
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index 2bdae61c0ac0..12c663e86ca1 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -984,6 +984,8 @@ void radeon_dpm_enable_uvd(struct radeon_device *rdev, bool enable)
 		if (enable) {
 			mutex_lock(&rdev->pm.mutex);
 			rdev->pm.dpm.uvd_active = true;
+			/* disable this for now */
+#if 0
 			if ((rdev->pm.dpm.sd == 1) && (rdev->pm.dpm.hd == 0))
 				dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_SD;
 			else if ((rdev->pm.dpm.sd == 2) && (rdev->pm.dpm.hd == 0))
@@ -993,6 +995,7 @@ void radeon_dpm_enable_uvd(struct radeon_device *rdev, bool enable)
 			else if ((rdev->pm.dpm.sd == 0) && (rdev->pm.dpm.hd == 2))
 				dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_HD2;
 			else
+#endif
 				dpm_state = POWER_STATE_TYPE_INTERNAL_UVD;
 			rdev->pm.dpm.state = dpm_state;
 			mutex_unlock(&rdev->pm.mutex);
diff --git a/drivers/gpu/drm/radeon/radeon_state.c b/drivers/gpu/drm/radeon/radeon_state.c
index 956ab7f14e16..23bb64fd775f 100644
--- a/drivers/gpu/drm/radeon/radeon_state.c
+++ b/drivers/gpu/drm/radeon/radeon_state.c
@@ -3054,7 +3054,7 @@ static int radeon_cp_getparam(struct drm_device *dev, void *data, struct drm_fil
 		if ((dev_priv->flags & RADEON_FAMILY_MASK) >= CHIP_R600)
 			value = 0;
 		else
-			value = drm_dev_to_irq(dev);
+			value = dev->pdev->irq;
 		break;
 	case RADEON_PARAM_GART_BASE:
 		value = dev_priv->gart_vm_start;
@@ -3258,4 +3258,4 @@ struct drm_ioctl_desc radeon_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(RADEON_CS, r600_cs_legacy_ioctl, DRM_AUTH)
 };
 
-int radeon_max_ioctl = DRM_ARRAY_SIZE(radeon_ioctls);
+int radeon_max_ioctl = ARRAY_SIZE(radeon_ioctls);
diff --git a/drivers/gpu/drm/radeon/radeon_uvd.c b/drivers/gpu/drm/radeon/radeon_uvd.c
index 1b65ae2433cd..a4ad270e8261 100644
--- a/drivers/gpu/drm/radeon/radeon_uvd.c
+++ b/drivers/gpu/drm/radeon/radeon_uvd.c
@@ -812,7 +812,8 @@ void radeon_uvd_note_usage(struct radeon_device *rdev)
 		    (rdev->pm.dpm.hd != hd)) {
 			rdev->pm.dpm.sd = sd;
 			rdev->pm.dpm.hd = hd;
-			streams_changed = true;
+			/* disable this for now */
+			/*streams_changed = true;*/
 		}
 	}
 
diff --git a/drivers/gpu/drm/radeon/radeon_vce.c b/drivers/gpu/drm/radeon/radeon_vce.c
index 3971d968af6c..aa21c31a846c 100644
--- a/drivers/gpu/drm/radeon/radeon_vce.c
+++ b/drivers/gpu/drm/radeon/radeon_vce.c
@@ -66,6 +66,7 @@ int radeon_vce_init(struct radeon_device *rdev)
 	case CHIP_BONAIRE:
 	case CHIP_KAVERI:
 	case CHIP_KABINI:
+	case CHIP_HAWAII:
 	case CHIP_MULLINS:
 		fw_name = FIRMWARE_BONAIRE;
 		break;
diff --git a/drivers/gpu/drm/radeon/radeon_vm.c b/drivers/gpu/drm/radeon/radeon_vm.c
index c11b71d249e3..899d9126cad6 100644
--- a/drivers/gpu/drm/radeon/radeon_vm.c
+++ b/drivers/gpu/drm/radeon/radeon_vm.c
@@ -59,7 +59,7 @@
  */
 static unsigned radeon_vm_num_pdes(struct radeon_device *rdev)
 {
-	return rdev->vm_manager.max_pfn >> RADEON_VM_BLOCK_SIZE;
+	return rdev->vm_manager.max_pfn >> radeon_vm_block_size;
 }
 
 /**
@@ -140,8 +140,8 @@ struct radeon_cs_reloc *radeon_vm_get_bos(struct radeon_device *rdev,
 	/* add the vm page table to the list */
 	list[0].gobj = NULL;
 	list[0].robj = vm->page_directory;
-	list[0].domain = RADEON_GEM_DOMAIN_VRAM;
-	list[0].alt_domain = RADEON_GEM_DOMAIN_VRAM;
+	list[0].prefered_domains = RADEON_GEM_DOMAIN_VRAM;
+	list[0].allowed_domains = RADEON_GEM_DOMAIN_VRAM;
 	list[0].tv.bo = &vm->page_directory->tbo;
 	list[0].tiling_flags = 0;
 	list[0].handle = 0;
@@ -153,8 +153,8 @@ struct radeon_cs_reloc *radeon_vm_get_bos(struct radeon_device *rdev,
 
 		list[idx].gobj = NULL;
 		list[idx].robj = vm->page_tables[i].bo;
-		list[idx].domain = RADEON_GEM_DOMAIN_VRAM;
-		list[idx].alt_domain = RADEON_GEM_DOMAIN_VRAM;
+		list[idx].prefered_domains = RADEON_GEM_DOMAIN_VRAM;
+		list[idx].allowed_domains = RADEON_GEM_DOMAIN_VRAM;
 		list[idx].tv.bo = &list[idx].robj->tbo;
 		list[idx].tiling_flags = 0;
 		list[idx].handle = 0;
@@ -474,8 +474,10 @@ int radeon_vm_bo_set_addr(struct radeon_device *rdev,
 	bo_va->valid = false;
 	list_move(&bo_va->vm_list, head);
 
-	soffset = (soffset / RADEON_GPU_PAGE_SIZE) >> RADEON_VM_BLOCK_SIZE;
-	eoffset = (eoffset / RADEON_GPU_PAGE_SIZE) >> RADEON_VM_BLOCK_SIZE;
+	soffset = (soffset / RADEON_GPU_PAGE_SIZE) >> radeon_vm_block_size;
+	eoffset = (eoffset / RADEON_GPU_PAGE_SIZE) >> radeon_vm_block_size;
+
+	BUG_ON(eoffset >= radeon_vm_num_pdes(rdev));
 
 	if (eoffset > vm->max_pde_used)
 		vm->max_pde_used = eoffset;
@@ -583,10 +585,9 @@ static uint32_t radeon_vm_page_flags(uint32_t flags)
 int radeon_vm_update_page_directory(struct radeon_device *rdev,
 				    struct radeon_vm *vm)
 {
-	static const uint32_t incr = RADEON_VM_PTE_COUNT * 8;
-
 	struct radeon_bo *pd = vm->page_directory;
 	uint64_t pd_addr = radeon_bo_gpu_offset(pd);
+	uint32_t incr = RADEON_VM_PTE_COUNT * 8;
 	uint64_t last_pde = ~0, last_pt = ~0;
 	unsigned count = 0, pt_idx, ndw;
 	struct radeon_ib ib;
@@ -660,6 +661,84 @@ int radeon_vm_update_page_directory(struct radeon_device *rdev,
 }
 
 /**
+ * radeon_vm_frag_ptes - add fragment information to PTEs
+ *
+ * @rdev: radeon_device pointer
+ * @ib: IB for the update
+ * @pe_start: first PTE to handle
+ * @pe_end: last PTE to handle
+ * @addr: addr those PTEs should point to
+ * @flags: hw mapping flags
+ *
+ * Global and local mutex must be locked!
+ */
+static void radeon_vm_frag_ptes(struct radeon_device *rdev,
+				struct radeon_ib *ib,
+				uint64_t pe_start, uint64_t pe_end,
+				uint64_t addr, uint32_t flags)
+{
+	/**
+	 * The MC L1 TLB supports variable sized pages, based on a fragment
+	 * field in the PTE. When this field is set to a non-zero value, page
+	 * granularity is increased from 4KB to (1 << (12 + frag)). The PTE
+	 * flags are considered valid for all PTEs within the fragment range
+	 * and corresponding mappings are assumed to be physically contiguous.
+	 *
+	 * The L1 TLB can store a single PTE for the whole fragment,
+	 * significantly increasing the space available for translation
+	 * caching. This leads to large improvements in throughput when the
+	 * TLB is under pressure.
+	 *
+	 * The L2 TLB distributes small and large fragments into two
+	 * asymmetric partitions. The large fragment cache is significantly
+	 * larger. Thus, we try to use large fragments wherever possible.
+	 * Userspace can support this by aligning virtual base address and
+	 * allocation size to the fragment size.
+	 */
+
+	/* NI is optimized for 256KB fragments, SI and newer for 64KB */
+	uint64_t frag_flags = rdev->family == CHIP_CAYMAN ?
+			R600_PTE_FRAG_256KB : R600_PTE_FRAG_64KB;
+	uint64_t frag_align = rdev->family == CHIP_CAYMAN ? 0x200 : 0x80;
+
+	uint64_t frag_start = ALIGN(pe_start, frag_align);
+	uint64_t frag_end = pe_end & ~(frag_align - 1);
+
+	unsigned count;
+
+	/* system pages are non continuously */
+	if ((flags & R600_PTE_SYSTEM) || !(flags & R600_PTE_VALID) ||
+	    (frag_start >= frag_end)) {
+
+		count = (pe_end - pe_start) / 8;
+		radeon_asic_vm_set_page(rdev, ib, pe_start, addr, count,
+					RADEON_GPU_PAGE_SIZE, flags);
+		return;
+	}
+
+	/* handle the 4K area at the beginning */
+	if (pe_start != frag_start) {
+		count = (frag_start - pe_start) / 8;
+		radeon_asic_vm_set_page(rdev, ib, pe_start, addr, count,
+					RADEON_GPU_PAGE_SIZE, flags);
+		addr += RADEON_GPU_PAGE_SIZE * count;
+	}
+
+	/* handle the area in the middle */
+	count = (frag_end - frag_start) / 8;
+	radeon_asic_vm_set_page(rdev, ib, frag_start, addr, count,
+				RADEON_GPU_PAGE_SIZE, flags | frag_flags);
+
+	/* handle the 4K area at the end */
+	if (frag_end != pe_end) {
+		addr += RADEON_GPU_PAGE_SIZE * count;
+		count = (pe_end - frag_end) / 8;
+		radeon_asic_vm_set_page(rdev, ib, frag_end, addr, count,
+					RADEON_GPU_PAGE_SIZE, flags);
+	}
+}
+
+/**
  * radeon_vm_update_ptes - make sure that page tables are valid
  *
  * @rdev: radeon_device pointer
@@ -679,8 +758,7 @@ static void radeon_vm_update_ptes(struct radeon_device *rdev,
 				  uint64_t start, uint64_t end,
 				  uint64_t dst, uint32_t flags)
 {
-	static const uint64_t mask = RADEON_VM_PTE_COUNT - 1;
-
+	uint64_t mask = RADEON_VM_PTE_COUNT - 1;
 	uint64_t last_pte = ~0, last_dst = ~0;
 	unsigned count = 0;
 	uint64_t addr;
@@ -690,7 +768,7 @@ static void radeon_vm_update_ptes(struct radeon_device *rdev,
 
 	/* walk over the address space and update the page tables */
 	for (addr = start; addr < end; ) {
-		uint64_t pt_idx = addr >> RADEON_VM_BLOCK_SIZE;
+		uint64_t pt_idx = addr >> radeon_vm_block_size;
 		struct radeon_bo *pt = vm->page_tables[pt_idx].bo;
 		unsigned nptes;
 		uint64_t pte;
@@ -708,10 +786,9 @@ static void radeon_vm_update_ptes(struct radeon_device *rdev,
 		if ((last_pte + 8 * count) != pte) {
 
 			if (count) {
-				radeon_asic_vm_set_page(rdev, ib, last_pte,
-							last_dst, count,
-							RADEON_GPU_PAGE_SIZE,
-							flags);
+				radeon_vm_frag_ptes(rdev, ib, last_pte,
+						    last_pte + 8 * count,
+						    last_dst, flags);
 			}
 
 			count = nptes;
@@ -726,9 +803,9 @@ static void radeon_vm_update_ptes(struct radeon_device *rdev,
 	}
 
 	if (count) {
-		radeon_asic_vm_set_page(rdev, ib, last_pte,
-					last_dst, count,
-					RADEON_GPU_PAGE_SIZE, flags);
+		radeon_vm_frag_ptes(rdev, ib, last_pte,
+				    last_pte + 8 * count,
+				    last_dst, flags);
 	}
 }
 
@@ -796,13 +873,13 @@ int radeon_vm_bo_update(struct radeon_device *rdev,
 	/* padding, etc. */
 	ndw = 64;
 
-	if (RADEON_VM_BLOCK_SIZE > 11)
+	if (radeon_vm_block_size > 11)
 		/* reserve space for one header for every 2k dwords */
 		ndw += (nptes >> 11) * 4;
 	else
 		/* reserve space for one header for
 		    every (1 << BLOCK_SIZE) entries */
-		ndw += (nptes >> RADEON_VM_BLOCK_SIZE) * 4;
+		ndw += (nptes >> radeon_vm_block_size) * 4;
 
 	/* reserve space for pte addresses */
 	ndw += nptes * 2;
@@ -892,6 +969,8 @@ void radeon_vm_bo_invalidate(struct radeon_device *rdev,
  */
 int radeon_vm_init(struct radeon_device *rdev, struct radeon_vm *vm)
 {
+	const unsigned align = min(RADEON_VM_PTB_ALIGN_SIZE,
+		RADEON_VM_PTE_COUNT * 8);
 	unsigned pd_size, pd_entries, pts_size;
 	int r;
 
@@ -913,7 +992,7 @@ int radeon_vm_init(struct radeon_device *rdev, struct radeon_vm *vm)
 		return -ENOMEM;
 	}
 
-	r = radeon_bo_create(rdev, pd_size, RADEON_VM_PTB_ALIGN_SIZE, false,
+	r = radeon_bo_create(rdev, pd_size, align, false,
 			     RADEON_GEM_DOMAIN_VRAM, NULL,
 			     &vm->page_directory);
 	if (r)
diff --git a/drivers/gpu/drm/radeon/rs400.c b/drivers/gpu/drm/radeon/rs400.c
index 130d5cc50d43..a0f96decece3 100644
--- a/drivers/gpu/drm/radeon/rs400.c
+++ b/drivers/gpu/drm/radeon/rs400.c
@@ -212,21 +212,16 @@ void rs400_gart_fini(struct radeon_device *rdev)
 #define RS400_PTE_WRITEABLE (1 << 2)
 #define RS400_PTE_READABLE  (1 << 3)
 
-int rs400_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr)
+void rs400_gart_set_page(struct radeon_device *rdev, unsigned i, uint64_t addr)
 {
 	uint32_t entry;
 	u32 *gtt = rdev->gart.ptr;
 
-	if (i < 0 || i > rdev->gart.num_gpu_pages) {
-		return -EINVAL;
-	}
-
 	entry = (lower_32_bits(addr) & PAGE_MASK) |
 		((upper_32_bits(addr) & 0xff) << 4) |
 		RS400_PTE_WRITEABLE | RS400_PTE_READABLE;
 	entry = cpu_to_le32(entry);
 	gtt[i] = entry;
-	return 0;
 }
 
 int rs400_mc_wait_for_idle(struct radeon_device *rdev)
diff --git a/drivers/gpu/drm/radeon/rs600.c b/drivers/gpu/drm/radeon/rs600.c
index 72d3616de08e..d1a35cb1c91d 100644
--- a/drivers/gpu/drm/radeon/rs600.c
+++ b/drivers/gpu/drm/radeon/rs600.c
@@ -109,19 +109,7 @@ void avivo_wait_for_vblank(struct radeon_device *rdev, int crtc)
 	}
 }
 
-void rs600_pre_page_flip(struct radeon_device *rdev, int crtc)
-{
-	/* enable the pflip int */
-	radeon_irq_kms_pflip_irq_get(rdev, crtc);
-}
-
-void rs600_post_page_flip(struct radeon_device *rdev, int crtc)
-{
-	/* disable the pflip int */
-	radeon_irq_kms_pflip_irq_put(rdev, crtc);
-}
-
-u32 rs600_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
+void rs600_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 {
 	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
 	u32 tmp = RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset);
@@ -148,9 +136,15 @@ u32 rs600_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 	/* Unlock the lock, so double-buffering can take place inside vblank */
 	tmp &= ~AVIVO_D1GRPH_UPDATE_LOCK;
 	WREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset, tmp);
+}
+
+bool rs600_page_flip_pending(struct radeon_device *rdev, int crtc_id)
+{
+	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
 
 	/* Return current update_pending status: */
-	return RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset) & AVIVO_D1GRPH_SURFACE_UPDATE_PENDING;
+	return !!(RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset) &
+		AVIVO_D1GRPH_SURFACE_UPDATE_PENDING);
 }
 
 void avivo_program_fmt(struct drm_encoder *encoder)
@@ -632,24 +626,16 @@ static void rs600_gart_fini(struct radeon_device *rdev)
 	radeon_gart_table_vram_free(rdev);
 }
 
-#define R600_PTE_VALID     (1 << 0)
-#define R600_PTE_SYSTEM    (1 << 1)
-#define R600_PTE_SNOOPED   (1 << 2)
-#define R600_PTE_READABLE  (1 << 5)
-#define R600_PTE_WRITEABLE (1 << 6)
-
-int rs600_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr)
+void rs600_gart_set_page(struct radeon_device *rdev, unsigned i, uint64_t addr)
 {
 	void __iomem *ptr = (void *)rdev->gart.ptr;
 
-	if (i < 0 || i > rdev->gart.num_gpu_pages) {
-		return -EINVAL;
-	}
 	addr = addr & 0xFFFFFFFFFFFFF000ULL;
-	addr |= R600_PTE_VALID | R600_PTE_SYSTEM | R600_PTE_SNOOPED;
-	addr |= R600_PTE_READABLE | R600_PTE_WRITEABLE;
+	if (addr == rdev->dummy_page.addr)
+		addr |= R600_PTE_SYSTEM | R600_PTE_SNOOPED;
+	else
+		addr |= R600_PTE_GART;
 	writeq(addr, ptr + (i * 8));
-	return 0;
 }
 
 int rs600_irq_set(struct radeon_device *rdev)
@@ -787,7 +773,7 @@ int rs600_irq_process(struct radeon_device *rdev)
 				wake_up(&rdev->irq.vblank_queue);
 			}
 			if (atomic_read(&rdev->irq.pflip[0]))
-				radeon_crtc_handle_flip(rdev, 0);
+				radeon_crtc_handle_vblank(rdev, 0);
 		}
 		if (G_007EDC_LB_D2_VBLANK_INTERRUPT(rdev->irq.stat_regs.r500.disp_int)) {
 			if (rdev->irq.crtc_vblank_int[1]) {
@@ -796,7 +782,7 @@ int rs600_irq_process(struct radeon_device *rdev)
 				wake_up(&rdev->irq.vblank_queue);
 			}
 			if (atomic_read(&rdev->irq.pflip[1]))
-				radeon_crtc_handle_flip(rdev, 1);
+				radeon_crtc_handle_vblank(rdev, 1);
 		}
 		if (G_007EDC_DC_HOT_PLUG_DETECT1_INTERRUPT(rdev->irq.stat_regs.r500.disp_int)) {
 			queue_hotplug = true;
diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c
index fef310773aad..da8703d8d455 100644
--- a/drivers/gpu/drm/radeon/rv770.c
+++ b/drivers/gpu/drm/radeon/rv770.c
@@ -801,7 +801,7 @@ u32 rv770_get_xclk(struct radeon_device *rdev)
 	return reference_clock;
 }
 
-u32 rv770_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
+void rv770_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 {
 	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
 	u32 tmp = RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset);
@@ -835,9 +835,15 @@ u32 rv770_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 	/* Unlock the lock, so double-buffering can take place inside vblank */
 	tmp &= ~AVIVO_D1GRPH_UPDATE_LOCK;
 	WREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset, tmp);
+}
+
+bool rv770_page_flip_pending(struct radeon_device *rdev, int crtc_id)
+{
+	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
 
 	/* Return current update_pending status: */
-	return RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset) & AVIVO_D1GRPH_SURFACE_UPDATE_PENDING;
+	return !!(RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset) &
+		AVIVO_D1GRPH_SURFACE_UPDATE_PENDING);
 }
 
 /* get temperature in millidegrees */
@@ -1321,6 +1327,9 @@ static void rv770_gpu_init(struct radeon_device *rdev)
 	if (tmp < rdev->config.rv770.max_simds) {
 		rdev->config.rv770.max_simds = tmp;
 	}
+	tmp = rdev->config.rv770.max_simds -
+		r600_count_pipe_bits((cc_gc_shader_pipe_config >> 16) & R7XX_MAX_SIMDS_MASK);
+	rdev->config.rv770.active_simds = tmp;
 
 	switch (rdev->config.rv770.max_tile_pipes) {
 	case 1:
diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c
index 22a63c98ba14..730cee2c34cf 100644
--- a/drivers/gpu/drm/radeon/si.c
+++ b/drivers/gpu/drm/radeon/si.c
@@ -71,6 +71,7 @@ MODULE_FIRMWARE("radeon/HAINAN_mc2.bin");
 MODULE_FIRMWARE("radeon/HAINAN_rlc.bin");
 MODULE_FIRMWARE("radeon/HAINAN_smc.bin");
 
+static u32 si_get_cu_active_bitmap(struct radeon_device *rdev, u32 se, u32 sh);
 static void si_pcie_gen3_enable(struct radeon_device *rdev);
 static void si_program_aspm(struct radeon_device *rdev);
 extern void sumo_rlc_fini(struct radeon_device *rdev);
@@ -2900,7 +2901,7 @@ static void si_gpu_init(struct radeon_device *rdev)
 	u32 sx_debug_1;
 	u32 hdp_host_path_cntl;
 	u32 tmp;
-	int i, j;
+	int i, j, k;
 
 	switch (rdev->family) {
 	case CHIP_TAHITI:
@@ -3098,6 +3099,14 @@ static void si_gpu_init(struct radeon_device *rdev)
 		     rdev->config.si.max_sh_per_se,
 		     rdev->config.si.max_cu_per_sh);
 
+	for (i = 0; i < rdev->config.si.max_shader_engines; i++) {
+		for (j = 0; j < rdev->config.si.max_sh_per_se; j++) {
+			for (k = 0; k < rdev->config.si.max_cu_per_sh; k++) {
+				rdev->config.si.active_cus +=
+					hweight32(si_get_cu_active_bitmap(rdev, i, j));
+			}
+		}
+	}
 
 	/* set HW defaults for 3D engine */
 	WREG32(CP_QUEUE_THRESHOLDS, (ROQ_IB1_START(0x16) |
@@ -3186,7 +3195,7 @@ void si_fence_ring_emit(struct radeon_device *rdev,
 	/* EVENT_WRITE_EOP - flush caches, send int */
 	radeon_ring_write(ring, PACKET3(PACKET3_EVENT_WRITE_EOP, 4));
 	radeon_ring_write(ring, EVENT_TYPE(CACHE_FLUSH_AND_INV_TS_EVENT) | EVENT_INDEX(5));
-	radeon_ring_write(ring, addr & 0xffffffff);
+	radeon_ring_write(ring, lower_32_bits(addr));
 	radeon_ring_write(ring, (upper_32_bits(addr) & 0xff) | DATA_SEL(1) | INT_SEL(2));
 	radeon_ring_write(ring, fence->seq);
 	radeon_ring_write(ring, 0);
@@ -3219,7 +3228,7 @@ void si_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib)
 			radeon_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
 			radeon_ring_write(ring, (1 << 8));
 			radeon_ring_write(ring, ring->next_rptr_gpu_addr & 0xfffffffc);
-			radeon_ring_write(ring, upper_32_bits(ring->next_rptr_gpu_addr) & 0xffffffff);
+			radeon_ring_write(ring, upper_32_bits(ring->next_rptr_gpu_addr));
 			radeon_ring_write(ring, next_rptr);
 		}
 
@@ -4044,18 +4053,21 @@ static int si_pcie_gart_enable(struct radeon_device *rdev)
 	WREG32(MC_VM_MX_L1_TLB_CNTL,
 	       (0xA << 7) |
 	       ENABLE_L1_TLB |
+	       ENABLE_L1_FRAGMENT_PROCESSING |
 	       SYSTEM_ACCESS_MODE_NOT_IN_SYS |
 	       ENABLE_ADVANCED_DRIVER_MODEL |
 	       SYSTEM_APERTURE_UNMAPPED_ACCESS_PASS_THRU);
 	/* Setup L2 cache */
 	WREG32(VM_L2_CNTL, ENABLE_L2_CACHE |
+	       ENABLE_L2_FRAGMENT_PROCESSING |
 	       ENABLE_L2_PTE_CACHE_LRU_UPDATE_BY_WRITE |
 	       ENABLE_L2_PDE0_CACHE_LRU_UPDATE_BY_WRITE |
 	       EFFECTIVE_L2_QUEUE_SIZE(7) |
 	       CONTEXT1_IDENTITY_ACCESS_MODE(1));
 	WREG32(VM_L2_CNTL2, INVALIDATE_ALL_L1_TLBS | INVALIDATE_L2_CACHE);
 	WREG32(VM_L2_CNTL3, L2_CACHE_BIGK_ASSOCIATIVITY |
-	       L2_CACHE_BIGK_FRAGMENT_SIZE(0));
+	       BANK_SELECT(4) |
+	       L2_CACHE_BIGK_FRAGMENT_SIZE(4));
 	/* setup context0 */
 	WREG32(VM_CONTEXT0_PAGE_TABLE_START_ADDR, rdev->mc.gtt_start >> 12);
 	WREG32(VM_CONTEXT0_PAGE_TABLE_END_ADDR, rdev->mc.gtt_end >> 12);
@@ -4092,6 +4104,7 @@ static int si_pcie_gart_enable(struct radeon_device *rdev)
 	       (u32)(rdev->dummy_page.addr >> 12));
 	WREG32(VM_CONTEXT1_CNTL2, 4);
 	WREG32(VM_CONTEXT1_CNTL, ENABLE_CONTEXT | PAGE_TABLE_DEPTH(1) |
+				PAGE_TABLE_BLOCK_SIZE(radeon_vm_block_size - 9) |
 				RANGE_PROTECTION_FAULT_ENABLE_INTERRUPT |
 				RANGE_PROTECTION_FAULT_ENABLE_DEFAULT |
 				DUMMY_PAGE_PROTECTION_FAULT_ENABLE_INTERRUPT |
@@ -6151,7 +6164,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[0]))
-						radeon_crtc_handle_flip(rdev, 0);
+						radeon_crtc_handle_vblank(rdev, 0);
 					rdev->irq.stat_regs.evergreen.disp_int &= ~LB_D1_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D1 vblank\n");
 				}
@@ -6177,7 +6190,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[1]))
-						radeon_crtc_handle_flip(rdev, 1);
+						radeon_crtc_handle_vblank(rdev, 1);
 					rdev->irq.stat_regs.evergreen.disp_int_cont &= ~LB_D2_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D2 vblank\n");
 				}
@@ -6203,7 +6216,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[2]))
-						radeon_crtc_handle_flip(rdev, 2);
+						radeon_crtc_handle_vblank(rdev, 2);
 					rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~LB_D3_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D3 vblank\n");
 				}
@@ -6229,7 +6242,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[3]))
-						radeon_crtc_handle_flip(rdev, 3);
+						radeon_crtc_handle_vblank(rdev, 3);
 					rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~LB_D4_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D4 vblank\n");
 				}
@@ -6255,7 +6268,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[4]))
-						radeon_crtc_handle_flip(rdev, 4);
+						radeon_crtc_handle_vblank(rdev, 4);
 					rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~LB_D5_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D5 vblank\n");
 				}
@@ -6281,7 +6294,7 @@ restart_ih:
 						wake_up(&rdev->irq.vblank_queue);
 					}
 					if (atomic_read(&rdev->irq.pflip[5]))
-						radeon_crtc_handle_flip(rdev, 5);
+						radeon_crtc_handle_vblank(rdev, 5);
 					rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~LB_D6_VBLANK_INTERRUPT;
 					DRM_DEBUG("IH: D6 vblank\n");
 				}
diff --git a/drivers/gpu/drm/radeon/si_dma.c b/drivers/gpu/drm/radeon/si_dma.c
index de0ca070122f..e24c94b6d14d 100644
--- a/drivers/gpu/drm/radeon/si_dma.c
+++ b/drivers/gpu/drm/radeon/si_dma.c
@@ -79,7 +79,25 @@ void si_dma_vm_set_page(struct radeon_device *rdev,
 
 	trace_radeon_vm_set_page(pe, addr, count, incr, flags);
 
-	if (flags & R600_PTE_SYSTEM) {
+	if (flags == R600_PTE_GART) {
+		uint64_t src = rdev->gart.table_addr + (addr >> 12) * 8;
+		while (count) {
+			unsigned bytes = count * 8;
+			if (bytes > 0xFFFF8)
+				bytes = 0xFFFF8;
+
+			ib->ptr[ib->length_dw++] = DMA_PACKET(DMA_PACKET_COPY,
+							      1, 0, 0, bytes);
+			ib->ptr[ib->length_dw++] = lower_32_bits(pe);
+			ib->ptr[ib->length_dw++] = lower_32_bits(src);
+			ib->ptr[ib->length_dw++] = upper_32_bits(pe) & 0xff;
+			ib->ptr[ib->length_dw++] = upper_32_bits(src) & 0xff;
+
+			pe += bytes;
+			src += bytes;
+			count -= bytes / 8;
+		}
+	} else if (flags & R600_PTE_SYSTEM) {
 		while (count) {
 			ndw = count * 2;
 			if (ndw > 0xFFFFE)
@@ -202,8 +220,8 @@ int si_copy_dma(struct radeon_device *rdev,
 			cur_size_in_bytes = 0xFFFFF;
 		size_in_bytes -= cur_size_in_bytes;
 		radeon_ring_write(ring, DMA_PACKET(DMA_PACKET_COPY, 1, 0, 0, cur_size_in_bytes));
-		radeon_ring_write(ring, dst_offset & 0xffffffff);
-		radeon_ring_write(ring, src_offset & 0xffffffff);
+		radeon_ring_write(ring, lower_32_bits(dst_offset));
+		radeon_ring_write(ring, lower_32_bits(src_offset));
 		radeon_ring_write(ring, upper_32_bits(dst_offset) & 0xff);
 		radeon_ring_write(ring, upper_32_bits(src_offset) & 0xff);
 		src_offset += cur_size_in_bytes;
diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c
index 9a3567bedaae..58918868f894 100644
--- a/drivers/gpu/drm/radeon/si_dpm.c
+++ b/drivers/gpu/drm/radeon/si_dpm.c
@@ -1948,6 +1948,10 @@ static void si_initialize_powertune_defaults(struct radeon_device *rdev)
 			si_pi->cac_weights = cac_weights_cape_verde_pro;
 			si_pi->dte_data = dte_data_cape_verde;
 			break;
+		case 0x682C:
+			si_pi->cac_weights = cac_weights_cape_verde_pro;
+			si_pi->dte_data = dte_data_sun_xt;
+			break;
 		case 0x6825:
 		case 0x6827:
 			si_pi->cac_weights = cac_weights_heathrow;
@@ -1971,10 +1975,9 @@ static void si_initialize_powertune_defaults(struct radeon_device *rdev)
 			si_pi->dte_data = dte_data_venus_xt;
 			break;
 		case 0x6823:
-			si_pi->cac_weights = cac_weights_chelsea_pro;
-			si_pi->dte_data = dte_data_venus_pro;
-			break;
 		case 0x682B:
+		case 0x6822:
+		case 0x682A:
 			si_pi->cac_weights = cac_weights_chelsea_pro;
 			si_pi->dte_data = dte_data_venus_pro;
 			break;
@@ -1988,6 +1991,7 @@ static void si_initialize_powertune_defaults(struct radeon_device *rdev)
 		case 0x6601:
 		case 0x6621:
 		case 0x6603:
+		case 0x6605:
 			si_pi->cac_weights = cac_weights_mars_pro;
 			si_pi->lcac_config = lcac_mars_pro;
 			si_pi->cac_override = cac_override_oland;
@@ -1998,6 +2002,7 @@ static void si_initialize_powertune_defaults(struct radeon_device *rdev)
 		case 0x6600:
 		case 0x6606:
 		case 0x6620:
+		case 0x6604:
 			si_pi->cac_weights = cac_weights_mars_xt;
 			si_pi->lcac_config = lcac_mars_pro;
 			si_pi->cac_override = cac_override_oland;
@@ -2006,6 +2011,8 @@ static void si_initialize_powertune_defaults(struct radeon_device *rdev)
 			update_dte_from_pl2 = true;
 			break;
 		case 0x6611:
+		case 0x6613:
+		case 0x6608:
 			si_pi->cac_weights = cac_weights_oland_pro;
 			si_pi->lcac_config = lcac_mars_pro;
 			si_pi->cac_override = cac_override_oland;
diff --git a/drivers/gpu/drm/radeon/sid.h b/drivers/gpu/drm/radeon/sid.h
index 7321283602ce..fd414d34d885 100644
--- a/drivers/gpu/drm/radeon/sid.h
+++ b/drivers/gpu/drm/radeon/sid.h
@@ -362,6 +362,7 @@
 #define		READ_PROTECTION_FAULT_ENABLE_DEFAULT		(1 << 16)
 #define		WRITE_PROTECTION_FAULT_ENABLE_INTERRUPT		(1 << 18)
 #define		WRITE_PROTECTION_FAULT_ENABLE_DEFAULT		(1 << 19)
+#define		PAGE_TABLE_BLOCK_SIZE(x)			(((x) & 0xF) << 24)
 #define VM_CONTEXT1_CNTL				0x1414
 #define VM_CONTEXT0_CNTL2				0x1430
 #define VM_CONTEXT1_CNTL2				0x1434
diff --git a/drivers/gpu/drm/radeon/uvd_v2_2.c b/drivers/gpu/drm/radeon/uvd_v2_2.c
index d1771004cb52..8bfdadd56598 100644
--- a/drivers/gpu/drm/radeon/uvd_v2_2.c
+++ b/drivers/gpu/drm/radeon/uvd_v2_2.c
@@ -45,7 +45,7 @@ void uvd_v2_2_fence_emit(struct radeon_device *rdev,
 	radeon_ring_write(ring, PACKET0(UVD_CONTEXT_ID, 0));
 	radeon_ring_write(ring, fence->seq);
 	radeon_ring_write(ring, PACKET0(UVD_GPCOM_VCPU_DATA0, 0));
-	radeon_ring_write(ring, addr & 0xffffffff);
+	radeon_ring_write(ring, lower_32_bits(addr));
 	radeon_ring_write(ring, PACKET0(UVD_GPCOM_VCPU_DATA1, 0));
 	radeon_ring_write(ring, upper_32_bits(addr) & 0xff);
 	radeon_ring_write(ring, PACKET0(UVD_GPCOM_VCPU_CMD, 0));
diff --git a/drivers/gpu/drm/rcar-du/Kconfig b/drivers/gpu/drm/rcar-du/Kconfig
index d8e835ac2c5e..2e3d7b5b0ad7 100644
--- a/drivers/gpu/drm/rcar-du/Kconfig
+++ b/drivers/gpu/drm/rcar-du/Kconfig
@@ -1,6 +1,7 @@
 config DRM_RCAR_DU
 	tristate "DRM Support for R-Car Display Unit"
 	depends on DRM && ARM
+	depends on ARCH_SHMOBILE || COMPILE_TEST
 	select DRM_KMS_HELPER
 	select DRM_KMS_CMA_HELPER
 	select DRM_GEM_CMA_HELPER
@@ -12,6 +13,7 @@ config DRM_RCAR_DU
 config DRM_RCAR_LVDS
 	bool "R-Car DU LVDS Encoder Support"
 	depends on DRM_RCAR_DU
+	depends on ARCH_R8A7790 || ARCH_R8A7791 || COMPILE_TEST
 	help
 	  Enable support the R-Car Display Unit embedded LVDS encoders
 	  (currently only on R8A7790).
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_lvdscon.c b/drivers/gpu/drm/rcar-du/rcar_du_lvdscon.c
index 4f3ba93cd91d..289048d1c7b2 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_lvdscon.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_lvdscon.c
@@ -57,15 +57,8 @@ static int rcar_du_lvds_connector_get_modes(struct drm_connector *connector)
 	return 1;
 }
 
-static int rcar_du_lvds_connector_mode_valid(struct drm_connector *connector,
-					    struct drm_display_mode *mode)
-{
-	return MODE_OK;
-}
-
 static const struct drm_connector_helper_funcs connector_helper_funcs = {
 	.get_modes = rcar_du_lvds_connector_get_modes,
-	.mode_valid = rcar_du_lvds_connector_mode_valid,
 	.best_encoder = rcar_du_connector_best_encoder,
 };
 
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_vgacon.c b/drivers/gpu/drm/rcar-du/rcar_du_vgacon.c
index 41d563adfeaa..ccfe64c7188f 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_vgacon.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_vgacon.c
@@ -25,15 +25,8 @@ static int rcar_du_vga_connector_get_modes(struct drm_connector *connector)
 	return 0;
 }
 
-static int rcar_du_vga_connector_mode_valid(struct drm_connector *connector,
-					    struct drm_display_mode *mode)
-{
-	return MODE_OK;
-}
-
 static const struct drm_connector_helper_funcs connector_helper_funcs = {
 	.get_modes = rcar_du_vga_connector_get_modes,
-	.mode_valid = rcar_du_vga_connector_mode_valid,
 	.best_encoder = rcar_du_connector_best_encoder,
 };
 
diff --git a/drivers/gpu/drm/savage/savage_bci.c b/drivers/gpu/drm/savage/savage_bci.c
index d2b2df9e26f3..c97cdc9ab239 100644
--- a/drivers/gpu/drm/savage/savage_bci.c
+++ b/drivers/gpu/drm/savage/savage_bci.c
@@ -1079,4 +1079,4 @@ const struct drm_ioctl_desc savage_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(SAVAGE_BCI_EVENT_WAIT, savage_bci_event_wait, DRM_AUTH),
 };
 
-int savage_max_ioctl = DRM_ARRAY_SIZE(savage_ioctls);
+int savage_max_ioctl = ARRAY_SIZE(savage_ioctls);
diff --git a/drivers/gpu/drm/shmobile/Kconfig b/drivers/gpu/drm/shmobile/Kconfig
index 2ee44ca9d67f..a50fe0eeaa0d 100644
--- a/drivers/gpu/drm/shmobile/Kconfig
+++ b/drivers/gpu/drm/shmobile/Kconfig
@@ -1,6 +1,7 @@
 config DRM_SHMOBILE
 	tristate "DRM Support for SH Mobile"
-	depends on DRM && (ARM || SUPERH)
+	depends on DRM && ARM
+	depends on ARCH_SHMOBILE || COMPILE_TEST
 	select BACKLIGHT_CLASS_DEVICE
 	select DRM_KMS_HELPER
 	select DRM_KMS_FB_HELPER
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_crtc.c b/drivers/gpu/drm/shmobile/shmob_drm_crtc.c
index e9e5e6d368cc..faf176b2daf9 100644
--- a/drivers/gpu/drm/shmobile/shmob_drm_crtc.c
+++ b/drivers/gpu/drm/shmobile/shmob_drm_crtc.c
@@ -674,12 +674,6 @@ static int shmob_drm_connector_get_modes(struct drm_connector *connector)
 	return 1;
 }
 
-static int shmob_drm_connector_mode_valid(struct drm_connector *connector,
-					  struct drm_display_mode *mode)
-{
-	return MODE_OK;
-}
-
 static struct drm_encoder *
 shmob_drm_connector_best_encoder(struct drm_connector *connector)
 {
@@ -690,7 +684,6 @@ shmob_drm_connector_best_encoder(struct drm_connector *connector)
 
 static const struct drm_connector_helper_funcs connector_helper_funcs = {
 	.get_modes = shmob_drm_connector_get_modes,
-	.mode_valid = shmob_drm_connector_mode_valid,
 	.best_encoder = shmob_drm_connector_best_encoder,
 };
 
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_drv.c b/drivers/gpu/drm/shmobile/shmob_drm_drv.c
index c839c9c89efb..82c84c7fd4f6 100644
--- a/drivers/gpu/drm/shmobile/shmob_drm_drv.c
+++ b/drivers/gpu/drm/shmobile/shmob_drm_drv.c
@@ -185,7 +185,7 @@ static int shmob_drm_load(struct drm_device *dev, unsigned long flags)
 		goto done;
 	}
 
-	ret = drm_irq_install(dev);
+	ret = drm_irq_install(dev, platform_get_irq(dev->platformdev, 0));
 	if (ret < 0) {
 		dev_err(&pdev->dev, "failed to install IRQ handler\n");
 		goto done;
diff --git a/drivers/gpu/drm/sis/sis_mm.c b/drivers/gpu/drm/sis/sis_mm.c
index 0573be0d2933..77f288e4a0a6 100644
--- a/drivers/gpu/drm/sis/sis_mm.c
+++ b/drivers/gpu/drm/sis/sis_mm.c
@@ -359,4 +359,4 @@ const struct drm_ioctl_desc sis_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(SIS_FB_INIT, sis_fb_init, DRM_AUTH | DRM_MASTER | DRM_ROOT_ONLY),
 };
 
-int sis_max_ioctl = DRM_ARRAY_SIZE(sis_ioctls);
+int sis_max_ioctl = ARRAY_SIZE(sis_ioctls);
diff --git a/drivers/gpu/drm/tegra/Makefile b/drivers/gpu/drm/tegra/Makefile
index d43f21bb4596..2c66a8db9da4 100644
--- a/drivers/gpu/drm/tegra/Makefile
+++ b/drivers/gpu/drm/tegra/Makefile
@@ -1,7 +1,6 @@
 ccflags-$(CONFIG_DRM_TEGRA_DEBUG) += -DDEBUG
 
 tegra-drm-y := \
-	bus.o \
 	drm.o \
 	gem.o \
 	fb.o \
diff --git a/drivers/gpu/drm/tegra/bus.c b/drivers/gpu/drm/tegra/bus.c
deleted file mode 100644
index 71cef5c13dc8..000000000000
--- a/drivers/gpu/drm/tegra/bus.c
+++ /dev/null
@@ -1,75 +0,0 @@
-/*
- * Copyright (C) 2013 NVIDIA Corporation
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-
-#include "drm.h"
-
-static int drm_host1x_set_busid(struct drm_device *dev,
-				struct drm_master *master)
-{
-	const char *device = dev_name(dev->dev);
-	const char *driver = dev->driver->name;
-	const char *bus = dev->dev->bus->name;
-	int length;
-
-	master->unique_len = strlen(bus) + 1 + strlen(device);
-	master->unique_size = master->unique_len;
-
-	master->unique = kmalloc(master->unique_len + 1, GFP_KERNEL);
-	if (!master->unique)
-		return -ENOMEM;
-
-	snprintf(master->unique, master->unique_len + 1, "%s:%s", bus, device);
-
-	length = strlen(driver) + 1 + master->unique_len;
-
-	dev->devname = kmalloc(length + 1, GFP_KERNEL);
-	if (!dev->devname)
-		return -ENOMEM;
-
-	snprintf(dev->devname, length + 1, "%s@%s", driver, master->unique);
-
-	return 0;
-}
-
-static struct drm_bus drm_host1x_bus = {
-	.bus_type = DRIVER_BUS_HOST1X,
-	.set_busid = drm_host1x_set_busid,
-};
-
-int drm_host1x_init(struct drm_driver *driver, struct host1x_device *device)
-{
-	struct drm_device *drm;
-	int ret;
-
-	driver->bus = &drm_host1x_bus;
-
-	drm = drm_dev_alloc(driver, &device->dev);
-	if (!drm)
-		return -ENOMEM;
-
-	ret = drm_dev_register(drm, 0);
-	if (ret)
-		goto err_free;
-
-	DRM_INFO("Initialized %s %d.%d.%d %s on minor %d\n", driver->name,
-		 driver->major, driver->minor, driver->patchlevel,
-		 driver->date, drm->primary->index);
-
-	return 0;
-
-err_free:
-	drm_dev_unref(drm);
-	return ret;
-}
-
-void drm_host1x_exit(struct drm_driver *driver, struct host1x_device *device)
-{
-	struct tegra_drm *tegra = dev_get_drvdata(&device->dev);
-
-	drm_put_dev(tegra->drm);
-}
diff --git a/drivers/gpu/drm/tegra/dc.c b/drivers/gpu/drm/tegra/dc.c
index edb871d7d395..ef40381f3909 100644
--- a/drivers/gpu/drm/tegra/dc.c
+++ b/drivers/gpu/drm/tegra/dc.c
@@ -17,6 +17,7 @@
 
 struct tegra_dc_soc_info {
 	bool supports_interlacing;
+	bool supports_cursor;
 };
 
 struct tegra_plane {
@@ -29,6 +30,254 @@ static inline struct tegra_plane *to_tegra_plane(struct drm_plane *plane)
 	return container_of(plane, struct tegra_plane, base);
 }
 
+static unsigned int tegra_dc_format(uint32_t format, uint32_t *swap)
+{
+	/* assume no swapping of fetched data */
+	if (swap)
+		*swap = BYTE_SWAP_NOSWAP;
+
+	switch (format) {
+	case DRM_FORMAT_XBGR8888:
+		return WIN_COLOR_DEPTH_R8G8B8A8;
+
+	case DRM_FORMAT_XRGB8888:
+		return WIN_COLOR_DEPTH_B8G8R8A8;
+
+	case DRM_FORMAT_RGB565:
+		return WIN_COLOR_DEPTH_B5G6R5;
+
+	case DRM_FORMAT_UYVY:
+		return WIN_COLOR_DEPTH_YCbCr422;
+
+	case DRM_FORMAT_YUYV:
+		if (swap)
+			*swap = BYTE_SWAP_SWAP2;
+
+		return WIN_COLOR_DEPTH_YCbCr422;
+
+	case DRM_FORMAT_YUV420:
+		return WIN_COLOR_DEPTH_YCbCr420P;
+
+	case DRM_FORMAT_YUV422:
+		return WIN_COLOR_DEPTH_YCbCr422P;
+
+	default:
+		break;
+	}
+
+	WARN(1, "unsupported pixel format %u, using default\n", format);
+	return WIN_COLOR_DEPTH_B8G8R8A8;
+}
+
+static bool tegra_dc_format_is_yuv(unsigned int format, bool *planar)
+{
+	switch (format) {
+	case WIN_COLOR_DEPTH_YCbCr422:
+	case WIN_COLOR_DEPTH_YUV422:
+		if (planar)
+			*planar = false;
+
+		return true;
+
+	case WIN_COLOR_DEPTH_YCbCr420P:
+	case WIN_COLOR_DEPTH_YUV420P:
+	case WIN_COLOR_DEPTH_YCbCr422P:
+	case WIN_COLOR_DEPTH_YUV422P:
+	case WIN_COLOR_DEPTH_YCbCr422R:
+	case WIN_COLOR_DEPTH_YUV422R:
+	case WIN_COLOR_DEPTH_YCbCr422RA:
+	case WIN_COLOR_DEPTH_YUV422RA:
+		if (planar)
+			*planar = true;
+
+		return true;
+	}
+
+	return false;
+}
+
+static inline u32 compute_dda_inc(unsigned int in, unsigned int out, bool v,
+				  unsigned int bpp)
+{
+	fixed20_12 outf = dfixed_init(out);
+	fixed20_12 inf = dfixed_init(in);
+	u32 dda_inc;
+	int max;
+
+	if (v)
+		max = 15;
+	else {
+		switch (bpp) {
+		case 2:
+			max = 8;
+			break;
+
+		default:
+			WARN_ON_ONCE(1);
+			/* fallthrough */
+		case 4:
+			max = 4;
+			break;
+		}
+	}
+
+	outf.full = max_t(u32, outf.full - dfixed_const(1), dfixed_const(1));
+	inf.full -= dfixed_const(1);
+
+	dda_inc = dfixed_div(inf, outf);
+	dda_inc = min_t(u32, dda_inc, dfixed_const(max));
+
+	return dda_inc;
+}
+
+static inline u32 compute_initial_dda(unsigned int in)
+{
+	fixed20_12 inf = dfixed_init(in);
+	return dfixed_frac(inf);
+}
+
+static int tegra_dc_setup_window(struct tegra_dc *dc, unsigned int index,
+				 const struct tegra_dc_window *window)
+{
+	unsigned h_offset, v_offset, h_size, v_size, h_dda, v_dda, bpp;
+	unsigned long value;
+	bool yuv, planar;
+
+	/*
+	 * For YUV planar modes, the number of bytes per pixel takes into
+	 * account only the luma component and therefore is 1.
+	 */
+	yuv = tegra_dc_format_is_yuv(window->format, &planar);
+	if (!yuv)
+		bpp = window->bits_per_pixel / 8;
+	else
+		bpp = planar ? 1 : 2;
+
+	value = WINDOW_A_SELECT << index;
+	tegra_dc_writel(dc, value, DC_CMD_DISPLAY_WINDOW_HEADER);
+
+	tegra_dc_writel(dc, window->format, DC_WIN_COLOR_DEPTH);
+	tegra_dc_writel(dc, window->swap, DC_WIN_BYTE_SWAP);
+
+	value = V_POSITION(window->dst.y) | H_POSITION(window->dst.x);
+	tegra_dc_writel(dc, value, DC_WIN_POSITION);
+
+	value = V_SIZE(window->dst.h) | H_SIZE(window->dst.w);
+	tegra_dc_writel(dc, value, DC_WIN_SIZE);
+
+	h_offset = window->src.x * bpp;
+	v_offset = window->src.y;
+	h_size = window->src.w * bpp;
+	v_size = window->src.h;
+
+	value = V_PRESCALED_SIZE(v_size) | H_PRESCALED_SIZE(h_size);
+	tegra_dc_writel(dc, value, DC_WIN_PRESCALED_SIZE);
+
+	/*
+	 * For DDA computations the number of bytes per pixel for YUV planar
+	 * modes needs to take into account all Y, U and V components.
+	 */
+	if (yuv && planar)
+		bpp = 2;
+
+	h_dda = compute_dda_inc(window->src.w, window->dst.w, false, bpp);
+	v_dda = compute_dda_inc(window->src.h, window->dst.h, true, bpp);
+
+	value = V_DDA_INC(v_dda) | H_DDA_INC(h_dda);
+	tegra_dc_writel(dc, value, DC_WIN_DDA_INC);
+
+	h_dda = compute_initial_dda(window->src.x);
+	v_dda = compute_initial_dda(window->src.y);
+
+	tegra_dc_writel(dc, h_dda, DC_WIN_H_INITIAL_DDA);
+	tegra_dc_writel(dc, v_dda, DC_WIN_V_INITIAL_DDA);
+
+	tegra_dc_writel(dc, 0, DC_WIN_UV_BUF_STRIDE);
+	tegra_dc_writel(dc, 0, DC_WIN_BUF_STRIDE);
+
+	tegra_dc_writel(dc, window->base[0], DC_WINBUF_START_ADDR);
+
+	if (yuv && planar) {
+		tegra_dc_writel(dc, window->base[1], DC_WINBUF_START_ADDR_U);
+		tegra_dc_writel(dc, window->base[2], DC_WINBUF_START_ADDR_V);
+		value = window->stride[1] << 16 | window->stride[0];
+		tegra_dc_writel(dc, value, DC_WIN_LINE_STRIDE);
+	} else {
+		tegra_dc_writel(dc, window->stride[0], DC_WIN_LINE_STRIDE);
+	}
+
+	if (window->bottom_up)
+		v_offset += window->src.h - 1;
+
+	tegra_dc_writel(dc, h_offset, DC_WINBUF_ADDR_H_OFFSET);
+	tegra_dc_writel(dc, v_offset, DC_WINBUF_ADDR_V_OFFSET);
+
+	if (window->tiled) {
+		value = DC_WIN_BUFFER_ADDR_MODE_TILE_UV |
+			DC_WIN_BUFFER_ADDR_MODE_TILE;
+	} else {
+		value = DC_WIN_BUFFER_ADDR_MODE_LINEAR_UV |
+			DC_WIN_BUFFER_ADDR_MODE_LINEAR;
+	}
+
+	tegra_dc_writel(dc, value, DC_WIN_BUFFER_ADDR_MODE);
+
+	value = WIN_ENABLE;
+
+	if (yuv) {
+		/* setup default colorspace conversion coefficients */
+		tegra_dc_writel(dc, 0x00f0, DC_WIN_CSC_YOF);
+		tegra_dc_writel(dc, 0x012a, DC_WIN_CSC_KYRGB);
+		tegra_dc_writel(dc, 0x0000, DC_WIN_CSC_KUR);
+		tegra_dc_writel(dc, 0x0198, DC_WIN_CSC_KVR);
+		tegra_dc_writel(dc, 0x039b, DC_WIN_CSC_KUG);
+		tegra_dc_writel(dc, 0x032f, DC_WIN_CSC_KVG);
+		tegra_dc_writel(dc, 0x0204, DC_WIN_CSC_KUB);
+		tegra_dc_writel(dc, 0x0000, DC_WIN_CSC_KVB);
+
+		value |= CSC_ENABLE;
+	} else if (window->bits_per_pixel < 24) {
+		value |= COLOR_EXPAND;
+	}
+
+	if (window->bottom_up)
+		value |= V_DIRECTION;
+
+	tegra_dc_writel(dc, value, DC_WIN_WIN_OPTIONS);
+
+	/*
+	 * Disable blending and assume Window A is the bottom-most window,
+	 * Window C is the top-most window and Window B is in the middle.
+	 */
+	tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_NOKEY);
+	tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_1WIN);
+
+	switch (index) {
+	case 0:
+		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_2WIN_X);
+		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_2WIN_Y);
+		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_3WIN_XY);
+		break;
+
+	case 1:
+		tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_2WIN_X);
+		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_2WIN_Y);
+		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_3WIN_XY);
+		break;
+
+	case 2:
+		tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_2WIN_X);
+		tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_2WIN_Y);
+		tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_3WIN_XY);
+		break;
+	}
+
+	tegra_dc_writel(dc, WIN_A_UPDATE << index, DC_CMD_STATE_CONTROL);
+	tegra_dc_writel(dc, WIN_A_ACT_REQ << index, DC_CMD_STATE_CONTROL);
+
+	return 0;
+}
+
 static int tegra_plane_update(struct drm_plane *plane, struct drm_crtc *crtc,
 			      struct drm_framebuffer *fb, int crtc_x,
 			      int crtc_y, unsigned int crtc_w,
@@ -49,7 +298,7 @@ static int tegra_plane_update(struct drm_plane *plane, struct drm_crtc *crtc,
 	window.dst.y = crtc_y;
 	window.dst.w = crtc_w;
 	window.dst.h = crtc_h;
-	window.format = tegra_dc_format(fb->pixel_format);
+	window.format = tegra_dc_format(fb->pixel_format, &window.swap);
 	window.bits_per_pixel = fb->bits_per_pixel;
 	window.bottom_up = tegra_fb_is_bottom_up(fb);
 	window.tiled = tegra_fb_is_tiled(fb);
@@ -117,6 +366,7 @@ static const uint32_t plane_formats[] = {
 	DRM_FORMAT_XRGB8888,
 	DRM_FORMAT_RGB565,
 	DRM_FORMAT_UYVY,
+	DRM_FORMAT_YUYV,
 	DRM_FORMAT_YUV420,
 	DRM_FORMAT_YUV422,
 };
@@ -150,9 +400,9 @@ static int tegra_dc_add_planes(struct drm_device *drm, struct tegra_dc *dc)
 static int tegra_dc_set_base(struct tegra_dc *dc, int x, int y,
 			     struct drm_framebuffer *fb)
 {
-	unsigned int format = tegra_dc_format(fb->pixel_format);
 	struct tegra_bo *bo = tegra_fb_get_plane(fb, 0);
 	unsigned int h_offset = 0, v_offset = 0;
+	unsigned int format, swap;
 	unsigned long value;
 
 	tegra_dc_writel(dc, WINDOW_A_SELECT, DC_CMD_DISPLAY_WINDOW_HEADER);
@@ -162,7 +412,10 @@ static int tegra_dc_set_base(struct tegra_dc *dc, int x, int y,
 
 	tegra_dc_writel(dc, bo->paddr + value, DC_WINBUF_START_ADDR);
 	tegra_dc_writel(dc, fb->pitches[0], DC_WIN_LINE_STRIDE);
+
+	format = tegra_dc_format(fb->pixel_format, &swap);
 	tegra_dc_writel(dc, format, DC_WIN_COLOR_DEPTH);
+	tegra_dc_writel(dc, swap, DC_WIN_BYTE_SWAP);
 
 	if (tegra_fb_is_tiled(fb)) {
 		value = DC_WIN_BUFFER_ADDR_MODE_TILE_UV |
@@ -177,13 +430,13 @@ static int tegra_dc_set_base(struct tegra_dc *dc, int x, int y,
 	/* make sure bottom-up buffers are properly displayed */
 	if (tegra_fb_is_bottom_up(fb)) {
 		value = tegra_dc_readl(dc, DC_WIN_WIN_OPTIONS);
-		value |= INVERT_V;
+		value |= V_DIRECTION;
 		tegra_dc_writel(dc, value, DC_WIN_WIN_OPTIONS);
 
 		v_offset += fb->height - 1;
 	} else {
 		value = tegra_dc_readl(dc, DC_WIN_WIN_OPTIONS);
-		value &= ~INVERT_V;
+		value &= ~V_DIRECTION;
 		tegra_dc_writel(dc, value, DC_WIN_WIN_OPTIONS);
 	}
 
@@ -225,6 +478,109 @@ void tegra_dc_disable_vblank(struct tegra_dc *dc)
 	spin_unlock_irqrestore(&dc->lock, flags);
 }
 
+static int tegra_dc_cursor_set2(struct drm_crtc *crtc, struct drm_file *file,
+				uint32_t handle, uint32_t width,
+				uint32_t height, int32_t hot_x, int32_t hot_y)
+{
+	unsigned long value = CURSOR_CLIP_DISPLAY;
+	struct tegra_dc *dc = to_tegra_dc(crtc);
+	struct drm_gem_object *gem;
+	struct tegra_bo *bo = NULL;
+
+	if (!dc->soc->supports_cursor)
+		return -ENXIO;
+
+	if (width != height)
+		return -EINVAL;
+
+	switch (width) {
+	case 32:
+		value |= CURSOR_SIZE_32x32;
+		break;
+
+	case 64:
+		value |= CURSOR_SIZE_64x64;
+		break;
+
+	case 128:
+		value |= CURSOR_SIZE_128x128;
+
+	case 256:
+		value |= CURSOR_SIZE_256x256;
+		break;
+
+	default:
+		return -EINVAL;
+	}
+
+	if (handle) {
+		gem = drm_gem_object_lookup(crtc->dev, file, handle);
+		if (!gem)
+			return -ENOENT;
+
+		bo = to_tegra_bo(gem);
+	}
+
+	if (bo) {
+		unsigned long addr = (bo->paddr & 0xfffffc00) >> 10;
+#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
+		unsigned long high = (bo->paddr & 0xfffffffc) >> 32;
+#endif
+
+		tegra_dc_writel(dc, value | addr, DC_DISP_CURSOR_START_ADDR);
+
+#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
+		tegra_dc_writel(dc, high, DC_DISP_CURSOR_START_ADDR_HI);
+#endif
+
+		value = tegra_dc_readl(dc, DC_DISP_DISP_WIN_OPTIONS);
+		value |= CURSOR_ENABLE;
+		tegra_dc_writel(dc, value, DC_DISP_DISP_WIN_OPTIONS);
+
+		value = tegra_dc_readl(dc, DC_DISP_BLEND_CURSOR_CONTROL);
+		value &= ~CURSOR_DST_BLEND_MASK;
+		value &= ~CURSOR_SRC_BLEND_MASK;
+		value |= CURSOR_MODE_NORMAL;
+		value |= CURSOR_DST_BLEND_NEG_K1_TIMES_SRC;
+		value |= CURSOR_SRC_BLEND_K1_TIMES_SRC;
+		value |= CURSOR_ALPHA;
+		tegra_dc_writel(dc, value, DC_DISP_BLEND_CURSOR_CONTROL);
+	} else {
+		value = tegra_dc_readl(dc, DC_DISP_DISP_WIN_OPTIONS);
+		value &= ~CURSOR_ENABLE;
+		tegra_dc_writel(dc, value, DC_DISP_DISP_WIN_OPTIONS);
+	}
+
+	tegra_dc_writel(dc, CURSOR_ACT_REQ << 8, DC_CMD_STATE_CONTROL);
+	tegra_dc_writel(dc, CURSOR_ACT_REQ, DC_CMD_STATE_CONTROL);
+
+	tegra_dc_writel(dc, GENERAL_ACT_REQ << 8, DC_CMD_STATE_CONTROL);
+	tegra_dc_writel(dc, GENERAL_ACT_REQ, DC_CMD_STATE_CONTROL);
+
+	return 0;
+}
+
+static int tegra_dc_cursor_move(struct drm_crtc *crtc, int x, int y)
+{
+	struct tegra_dc *dc = to_tegra_dc(crtc);
+	unsigned long value;
+
+	if (!dc->soc->supports_cursor)
+		return -ENXIO;
+
+	value = ((y & 0x3fff) << 16) | (x & 0x3fff);
+	tegra_dc_writel(dc, value, DC_DISP_CURSOR_POSITION);
+
+	tegra_dc_writel(dc, CURSOR_ACT_REQ << 8, DC_CMD_STATE_CONTROL);
+	tegra_dc_writel(dc, CURSOR_ACT_REQ, DC_CMD_STATE_CONTROL);
+
+	/* XXX: only required on generations earlier than Tegra124? */
+	tegra_dc_writel(dc, GENERAL_ACT_REQ << 8, DC_CMD_STATE_CONTROL);
+	tegra_dc_writel(dc, GENERAL_ACT_REQ, DC_CMD_STATE_CONTROL);
+
+	return 0;
+}
+
 static void tegra_dc_finish_page_flip(struct tegra_dc *dc)
 {
 	struct drm_device *drm = dc->base.dev;
@@ -301,6 +657,8 @@ static void tegra_dc_destroy(struct drm_crtc *crtc)
 }
 
 static const struct drm_crtc_funcs tegra_crtc_funcs = {
+	.cursor_set2 = tegra_dc_cursor_set2,
+	.cursor_move = tegra_dc_cursor_move,
 	.page_flip = tegra_dc_page_flip,
 	.set_config = drm_crtc_helper_set_config,
 	.destroy = tegra_dc_destroy,
@@ -334,52 +692,11 @@ static bool tegra_crtc_mode_fixup(struct drm_crtc *crtc,
 	return true;
 }
 
-static inline u32 compute_dda_inc(unsigned int in, unsigned int out, bool v,
-				  unsigned int bpp)
-{
-	fixed20_12 outf = dfixed_init(out);
-	fixed20_12 inf = dfixed_init(in);
-	u32 dda_inc;
-	int max;
-
-	if (v)
-		max = 15;
-	else {
-		switch (bpp) {
-		case 2:
-			max = 8;
-			break;
-
-		default:
-			WARN_ON_ONCE(1);
-			/* fallthrough */
-		case 4:
-			max = 4;
-			break;
-		}
-	}
-
-	outf.full = max_t(u32, outf.full - dfixed_const(1), dfixed_const(1));
-	inf.full -= dfixed_const(1);
-
-	dda_inc = dfixed_div(inf, outf);
-	dda_inc = min_t(u32, dda_inc, dfixed_const(max));
-
-	return dda_inc;
-}
-
-static inline u32 compute_initial_dda(unsigned int in)
-{
-	fixed20_12 inf = dfixed_init(in);
-	return dfixed_frac(inf);
-}
-
 static int tegra_dc_set_timings(struct tegra_dc *dc,
 				struct drm_display_mode *mode)
 {
-	/* TODO: For HDMI compliance, h & v ref_to_sync should be set to 1 */
-	unsigned int h_ref_to_sync = 0;
-	unsigned int v_ref_to_sync = 0;
+	unsigned int h_ref_to_sync = 1;
+	unsigned int v_ref_to_sync = 1;
 	unsigned long value;
 
 	tegra_dc_writel(dc, 0x0, DC_DISP_DISP_TIMING_OPTIONS);
@@ -406,13 +723,14 @@ static int tegra_dc_set_timings(struct tegra_dc *dc,
 }
 
 static int tegra_crtc_setup_clk(struct drm_crtc *crtc,
-				struct drm_display_mode *mode,
-				unsigned long *div)
+				struct drm_display_mode *mode)
 {
-	unsigned long pclk = mode->clock * 1000, rate;
+	unsigned long pclk = mode->clock * 1000;
 	struct tegra_dc *dc = to_tegra_dc(crtc);
 	struct tegra_output *output = NULL;
 	struct drm_encoder *encoder;
+	unsigned int div;
+	u32 value;
 	long err;
 
 	list_for_each_entry(encoder, &crtc->dev->mode_config.encoder_list, head)
@@ -425,221 +743,23 @@ static int tegra_crtc_setup_clk(struct drm_crtc *crtc,
 		return -ENODEV;
 
 	/*
-	 * This assumes that the display controller will divide its parent
-	 * clock by 2 to generate the pixel clock.
+	 * This assumes that the parent clock is pll_d_out0 or pll_d2_out
+	 * respectively, each of which divides the base pll_d by 2.
 	 */
-	err = tegra_output_setup_clock(output, dc->clk, pclk * 2);
+	err = tegra_output_setup_clock(output, dc->clk, pclk, &div);
 	if (err < 0) {
 		dev_err(dc->dev, "failed to setup clock: %ld\n", err);
 		return err;
 	}
 
-	rate = clk_get_rate(dc->clk);
-	*div = (rate * 2 / pclk) - 2;
-
-	DRM_DEBUG_KMS("rate: %lu, div: %lu\n", rate, *div);
-
-	return 0;
-}
-
-static bool tegra_dc_format_is_yuv(unsigned int format, bool *planar)
-{
-	switch (format) {
-	case WIN_COLOR_DEPTH_YCbCr422:
-	case WIN_COLOR_DEPTH_YUV422:
-		if (planar)
-			*planar = false;
-
-		return true;
-
-	case WIN_COLOR_DEPTH_YCbCr420P:
-	case WIN_COLOR_DEPTH_YUV420P:
-	case WIN_COLOR_DEPTH_YCbCr422P:
-	case WIN_COLOR_DEPTH_YUV422P:
-	case WIN_COLOR_DEPTH_YCbCr422R:
-	case WIN_COLOR_DEPTH_YUV422R:
-	case WIN_COLOR_DEPTH_YCbCr422RA:
-	case WIN_COLOR_DEPTH_YUV422RA:
-		if (planar)
-			*planar = true;
-
-		return true;
-	}
-
-	return false;
-}
-
-int tegra_dc_setup_window(struct tegra_dc *dc, unsigned int index,
-			  const struct tegra_dc_window *window)
-{
-	unsigned h_offset, v_offset, h_size, v_size, h_dda, v_dda, bpp;
-	unsigned long value;
-	bool yuv, planar;
-
-	/*
-	 * For YUV planar modes, the number of bytes per pixel takes into
-	 * account only the luma component and therefore is 1.
-	 */
-	yuv = tegra_dc_format_is_yuv(window->format, &planar);
-	if (!yuv)
-		bpp = window->bits_per_pixel / 8;
-	else
-		bpp = planar ? 1 : 2;
-
-	value = WINDOW_A_SELECT << index;
-	tegra_dc_writel(dc, value, DC_CMD_DISPLAY_WINDOW_HEADER);
-
-	tegra_dc_writel(dc, window->format, DC_WIN_COLOR_DEPTH);
-	tegra_dc_writel(dc, 0, DC_WIN_BYTE_SWAP);
-
-	value = V_POSITION(window->dst.y) | H_POSITION(window->dst.x);
-	tegra_dc_writel(dc, value, DC_WIN_POSITION);
-
-	value = V_SIZE(window->dst.h) | H_SIZE(window->dst.w);
-	tegra_dc_writel(dc, value, DC_WIN_SIZE);
-
-	h_offset = window->src.x * bpp;
-	v_offset = window->src.y;
-	h_size = window->src.w * bpp;
-	v_size = window->src.h;
-
-	value = V_PRESCALED_SIZE(v_size) | H_PRESCALED_SIZE(h_size);
-	tegra_dc_writel(dc, value, DC_WIN_PRESCALED_SIZE);
-
-	/*
-	 * For DDA computations the number of bytes per pixel for YUV planar
-	 * modes needs to take into account all Y, U and V components.
-	 */
-	if (yuv && planar)
-		bpp = 2;
-
-	h_dda = compute_dda_inc(window->src.w, window->dst.w, false, bpp);
-	v_dda = compute_dda_inc(window->src.h, window->dst.h, true, bpp);
-
-	value = V_DDA_INC(v_dda) | H_DDA_INC(h_dda);
-	tegra_dc_writel(dc, value, DC_WIN_DDA_INC);
-
-	h_dda = compute_initial_dda(window->src.x);
-	v_dda = compute_initial_dda(window->src.y);
-
-	tegra_dc_writel(dc, h_dda, DC_WIN_H_INITIAL_DDA);
-	tegra_dc_writel(dc, v_dda, DC_WIN_V_INITIAL_DDA);
-
-	tegra_dc_writel(dc, 0, DC_WIN_UV_BUF_STRIDE);
-	tegra_dc_writel(dc, 0, DC_WIN_BUF_STRIDE);
-
-	tegra_dc_writel(dc, window->base[0], DC_WINBUF_START_ADDR);
-
-	if (yuv && planar) {
-		tegra_dc_writel(dc, window->base[1], DC_WINBUF_START_ADDR_U);
-		tegra_dc_writel(dc, window->base[2], DC_WINBUF_START_ADDR_V);
-		value = window->stride[1] << 16 | window->stride[0];
-		tegra_dc_writel(dc, value, DC_WIN_LINE_STRIDE);
-	} else {
-		tegra_dc_writel(dc, window->stride[0], DC_WIN_LINE_STRIDE);
-	}
-
-	if (window->bottom_up)
-		v_offset += window->src.h - 1;
+	DRM_DEBUG_KMS("rate: %lu, div: %u\n", clk_get_rate(dc->clk), div);
 
-	tegra_dc_writel(dc, h_offset, DC_WINBUF_ADDR_H_OFFSET);
-	tegra_dc_writel(dc, v_offset, DC_WINBUF_ADDR_V_OFFSET);
-
-	if (window->tiled) {
-		value = DC_WIN_BUFFER_ADDR_MODE_TILE_UV |
-			DC_WIN_BUFFER_ADDR_MODE_TILE;
-	} else {
-		value = DC_WIN_BUFFER_ADDR_MODE_LINEAR_UV |
-			DC_WIN_BUFFER_ADDR_MODE_LINEAR;
-	}
-
-	tegra_dc_writel(dc, value, DC_WIN_BUFFER_ADDR_MODE);
-
-	value = WIN_ENABLE;
-
-	if (yuv) {
-		/* setup default colorspace conversion coefficients */
-		tegra_dc_writel(dc, 0x00f0, DC_WIN_CSC_YOF);
-		tegra_dc_writel(dc, 0x012a, DC_WIN_CSC_KYRGB);
-		tegra_dc_writel(dc, 0x0000, DC_WIN_CSC_KUR);
-		tegra_dc_writel(dc, 0x0198, DC_WIN_CSC_KVR);
-		tegra_dc_writel(dc, 0x039b, DC_WIN_CSC_KUG);
-		tegra_dc_writel(dc, 0x032f, DC_WIN_CSC_KVG);
-		tegra_dc_writel(dc, 0x0204, DC_WIN_CSC_KUB);
-		tegra_dc_writel(dc, 0x0000, DC_WIN_CSC_KVB);
-
-		value |= CSC_ENABLE;
-	} else if (window->bits_per_pixel < 24) {
-		value |= COLOR_EXPAND;
-	}
-
-	if (window->bottom_up)
-		value |= INVERT_V;
-
-	tegra_dc_writel(dc, value, DC_WIN_WIN_OPTIONS);
-
-	/*
-	 * Disable blending and assume Window A is the bottom-most window,
-	 * Window C is the top-most window and Window B is in the middle.
-	 */
-	tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_NOKEY);
-	tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_1WIN);
-
-	switch (index) {
-	case 0:
-		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_2WIN_X);
-		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_2WIN_Y);
-		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_3WIN_XY);
-		break;
-
-	case 1:
-		tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_2WIN_X);
-		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_2WIN_Y);
-		tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_3WIN_XY);
-		break;
-
-	case 2:
-		tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_2WIN_X);
-		tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_2WIN_Y);
-		tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_3WIN_XY);
-		break;
-	}
-
-	tegra_dc_writel(dc, WIN_A_UPDATE << index, DC_CMD_STATE_CONTROL);
-	tegra_dc_writel(dc, WIN_A_ACT_REQ << index, DC_CMD_STATE_CONTROL);
+	value = SHIFT_CLK_DIVIDER(div) | PIXEL_CLK_DIVIDER_PCD1;
+	tegra_dc_writel(dc, value, DC_DISP_DISP_CLOCK_CONTROL);
 
 	return 0;
 }
 
-unsigned int tegra_dc_format(uint32_t format)
-{
-	switch (format) {
-	case DRM_FORMAT_XBGR8888:
-		return WIN_COLOR_DEPTH_R8G8B8A8;
-
-	case DRM_FORMAT_XRGB8888:
-		return WIN_COLOR_DEPTH_B8G8R8A8;
-
-	case DRM_FORMAT_RGB565:
-		return WIN_COLOR_DEPTH_B5G6R5;
-
-	case DRM_FORMAT_UYVY:
-		return WIN_COLOR_DEPTH_YCbCr422;
-
-	case DRM_FORMAT_YUV420:
-		return WIN_COLOR_DEPTH_YCbCr420P;
-
-	case DRM_FORMAT_YUV422:
-		return WIN_COLOR_DEPTH_YCbCr422P;
-
-	default:
-		break;
-	}
-
-	WARN(1, "unsupported pixel format %u, using default\n", format);
-	return WIN_COLOR_DEPTH_B8G8R8A8;
-}
-
 static int tegra_crtc_mode_set(struct drm_crtc *crtc,
 			       struct drm_display_mode *mode,
 			       struct drm_display_mode *adjusted,
@@ -648,12 +768,12 @@ static int tegra_crtc_mode_set(struct drm_crtc *crtc,
 	struct tegra_bo *bo = tegra_fb_get_plane(crtc->primary->fb, 0);
 	struct tegra_dc *dc = to_tegra_dc(crtc);
 	struct tegra_dc_window window;
-	unsigned long div, value;
+	u32 value;
 	int err;
 
 	drm_vblank_pre_modeset(crtc->dev, dc->pipe);
 
-	err = tegra_crtc_setup_clk(crtc, mode, &div);
+	err = tegra_crtc_setup_clk(crtc, mode);
 	if (err) {
 		dev_err(dc->dev, "failed to setup clock for CRTC: %d\n", err);
 		return err;
@@ -669,9 +789,6 @@ static int tegra_crtc_mode_set(struct drm_crtc *crtc,
 		tegra_dc_writel(dc, value, DC_DISP_INTERLACE_CONTROL);
 	}
 
-	value = SHIFT_CLK_DIVIDER(div) | PIXEL_CLK_DIVIDER_PCD1;
-	tegra_dc_writel(dc, value, DC_DISP_DISP_CLOCK_CONTROL);
-
 	/* setup window parameters */
 	memset(&window, 0, sizeof(window));
 	window.src.x = 0;
@@ -682,7 +799,8 @@ static int tegra_crtc_mode_set(struct drm_crtc *crtc,
 	window.dst.y = 0;
 	window.dst.w = mode->hdisplay;
 	window.dst.h = mode->vdisplay;
-	window.format = tegra_dc_format(crtc->primary->fb->pixel_format);
+	window.format = tegra_dc_format(crtc->primary->fb->pixel_format,
+					&window.swap);
 	window.bits_per_pixel = crtc->primary->fb->bits_per_pixel;
 	window.stride[0] = crtc->primary->fb->pitches[0];
 	window.base[0] = bo->paddr;
@@ -728,10 +846,6 @@ static void tegra_crtc_prepare(struct drm_crtc *crtc)
 		WIN_A_OF_INT | WIN_B_OF_INT | WIN_C_OF_INT;
 	tegra_dc_writel(dc, value, DC_CMD_INT_POLARITY);
 
-	value = PW0_ENABLE | PW1_ENABLE | PW2_ENABLE | PW3_ENABLE |
-		PW4_ENABLE | PM0_ENABLE | PM1_ENABLE;
-	tegra_dc_writel(dc, value, DC_CMD_DISPLAY_POWER_CONTROL);
-
 	/* initialize timer */
 	value = CURSOR_THRESHOLD(0) | WINDOW_A_THRESHOLD(0x20) |
 		WINDOW_B_THRESHOLD(0x20) | WINDOW_C_THRESHOLD(0x20);
@@ -991,6 +1105,8 @@ static int tegra_dc_show_regs(struct seq_file *s, void *data)
 	DUMP_REG(DC_DISP_SD_BL_CONTROL);
 	DUMP_REG(DC_DISP_SD_HW_K_VALUES);
 	DUMP_REG(DC_DISP_SD_MAN_K_VALUES);
+	DUMP_REG(DC_DISP_CURSOR_START_ADDR_HI);
+	DUMP_REG(DC_DISP_BLEND_CURSOR_CONTROL);
 	DUMP_REG(DC_WIN_WIN_OPTIONS);
 	DUMP_REG(DC_WIN_BYTE_SWAP);
 	DUMP_REG(DC_WIN_BUFFER_CONTROL);
@@ -1096,26 +1212,26 @@ static int tegra_dc_debugfs_exit(struct tegra_dc *dc)
 
 static int tegra_dc_init(struct host1x_client *client)
 {
-	struct tegra_drm *tegra = dev_get_drvdata(client->parent);
+	struct drm_device *drm = dev_get_drvdata(client->parent);
 	struct tegra_dc *dc = host1x_client_to_dc(client);
 	int err;
 
-	drm_crtc_init(tegra->drm, &dc->base, &tegra_crtc_funcs);
+	drm_crtc_init(drm, &dc->base, &tegra_crtc_funcs);
 	drm_mode_crtc_set_gamma_size(&dc->base, 256);
 	drm_crtc_helper_add(&dc->base, &tegra_crtc_helper_funcs);
 
-	err = tegra_dc_rgb_init(tegra->drm, dc);
+	err = tegra_dc_rgb_init(drm, dc);
 	if (err < 0 && err != -ENODEV) {
 		dev_err(dc->dev, "failed to initialize RGB output: %d\n", err);
 		return err;
 	}
 
-	err = tegra_dc_add_planes(tegra->drm, dc);
+	err = tegra_dc_add_planes(drm, dc);
 	if (err < 0)
 		return err;
 
 	if (IS_ENABLED(CONFIG_DEBUG_FS)) {
-		err = tegra_dc_debugfs_init(dc, tegra->drm->primary);
+		err = tegra_dc_debugfs_init(dc, drm->primary);
 		if (err < 0)
 			dev_err(dc->dev, "debugfs setup failed: %d\n", err);
 	}
@@ -1160,14 +1276,17 @@ static const struct host1x_client_ops dc_client_ops = {
 
 static const struct tegra_dc_soc_info tegra20_dc_soc_info = {
 	.supports_interlacing = false,
+	.supports_cursor = false,
 };
 
 static const struct tegra_dc_soc_info tegra30_dc_soc_info = {
 	.supports_interlacing = false,
+	.supports_cursor = false,
 };
 
 static const struct tegra_dc_soc_info tegra124_dc_soc_info = {
 	.supports_interlacing = true,
+	.supports_cursor = true,
 };
 
 static const struct of_device_id tegra_dc_of_match[] = {
diff --git a/drivers/gpu/drm/tegra/dc.h b/drivers/gpu/drm/tegra/dc.h
index c94101494826..78c5feff95d2 100644
--- a/drivers/gpu/drm/tegra/dc.h
+++ b/drivers/gpu/drm/tegra/dc.h
@@ -67,10 +67,12 @@
 #define WIN_A_ACT_REQ   (1 <<  1)
 #define WIN_B_ACT_REQ   (1 <<  2)
 #define WIN_C_ACT_REQ   (1 <<  3)
+#define CURSOR_ACT_REQ  (1 <<  7)
 #define GENERAL_UPDATE  (1 <<  8)
 #define WIN_A_UPDATE    (1 <<  9)
 #define WIN_B_UPDATE    (1 << 10)
 #define WIN_C_UPDATE    (1 << 11)
+#define CURSOR_UPDATE   (1 << 15)
 #define NC_HOST_TRIG    (1 << 24)
 
 #define DC_CMD_DISPLAY_WINDOW_HEADER		0x042
@@ -116,9 +118,10 @@
 #define DC_DISP_DISP_SIGNAL_OPTIONS1		0x401
 
 #define DC_DISP_DISP_WIN_OPTIONS		0x402
-#define HDMI_ENABLE (1 << 30)
-#define DSI_ENABLE  (1 << 29)
-#define SOR_ENABLE  (1 << 25)
+#define HDMI_ENABLE	(1 << 30)
+#define DSI_ENABLE	(1 << 29)
+#define SOR_ENABLE	(1 << 25)
+#define CURSOR_ENABLE	(1 << 16)
 
 #define DC_DISP_DISP_MEM_HIGH_PRIORITY		0x403
 #define CURSOR_THRESHOLD(x)   (((x) & 0x03) << 24)
@@ -266,6 +269,14 @@
 #define DC_DISP_CURSOR_BACKGROUND		0x43d
 
 #define DC_DISP_CURSOR_START_ADDR		0x43e
+#define CURSOR_CLIP_DISPLAY	(0 << 28)
+#define CURSOR_CLIP_WIN_A	(1 << 28)
+#define CURSOR_CLIP_WIN_B	(2 << 28)
+#define CURSOR_CLIP_WIN_C	(3 << 28)
+#define CURSOR_SIZE_32x32	(0 << 24)
+#define CURSOR_SIZE_64x64	(1 << 24)
+#define CURSOR_SIZE_128x128	(2 << 24)
+#define CURSOR_SIZE_256x256	(3 << 24)
 #define DC_DISP_CURSOR_START_ADDR_NS		0x43f
 
 #define DC_DISP_CURSOR_POSITION			0x440
@@ -302,6 +313,19 @@
 #define  INTERLACE_START  (1 << 1)
 #define  INTERLACE_ENABLE (1 << 0)
 
+#define DC_DISP_CURSOR_START_ADDR_HI		0x4ec
+#define DC_DISP_BLEND_CURSOR_CONTROL		0x4f1
+#define CURSOR_MODE_LEGACY			(0 << 24)
+#define CURSOR_MODE_NORMAL			(1 << 24)
+#define CURSOR_DST_BLEND_ZERO			(0 << 16)
+#define CURSOR_DST_BLEND_K1			(1 << 16)
+#define CURSOR_DST_BLEND_NEG_K1_TIMES_SRC	(2 << 16)
+#define CURSOR_DST_BLEND_MASK			(3 << 16)
+#define CURSOR_SRC_BLEND_K1			(0 << 8)
+#define CURSOR_SRC_BLEND_K1_TIMES_SRC		(1 << 8)
+#define CURSOR_SRC_BLEND_MASK			(3 << 8)
+#define CURSOR_ALPHA				0xff
+
 #define DC_WIN_CSC_YOF				0x611
 #define DC_WIN_CSC_KYRGB			0x612
 #define DC_WIN_CSC_KUR				0x613
@@ -312,7 +336,8 @@
 #define DC_WIN_CSC_KVB				0x618
 
 #define DC_WIN_WIN_OPTIONS			0x700
-#define INVERT_V     (1 <<  2)
+#define H_DIRECTION  (1 <<  0)
+#define V_DIRECTION  (1 <<  2)
 #define COLOR_EXPAND (1 <<  6)
 #define CSC_ENABLE   (1 << 18)
 #define WIN_ENABLE   (1 << 30)
diff --git a/drivers/gpu/drm/tegra/dpaux.c b/drivers/gpu/drm/tegra/dpaux.c
index 005c19bd92df..3f132e356e9c 100644
--- a/drivers/gpu/drm/tegra/dpaux.c
+++ b/drivers/gpu/drm/tegra/dpaux.c
@@ -15,6 +15,7 @@
 #include <linux/platform_device.h>
 #include <linux/reset.h>
 #include <linux/regulator/consumer.h>
+#include <linux/workqueue.h>
 
 #include <drm/drm_dp_helper.h>
 #include <drm/drm_panel.h>
@@ -41,6 +42,7 @@ struct tegra_dpaux {
 	struct regulator *vdd;
 
 	struct completion complete;
+	struct work_struct work;
 	struct list_head list;
 };
 
@@ -49,6 +51,11 @@ static inline struct tegra_dpaux *to_dpaux(struct drm_dp_aux *aux)
 	return container_of(aux, struct tegra_dpaux, aux);
 }
 
+static inline struct tegra_dpaux *work_to_dpaux(struct work_struct *work)
+{
+	return container_of(work, struct tegra_dpaux, work);
+}
+
 static inline unsigned long tegra_dpaux_readl(struct tegra_dpaux *dpaux,
 					      unsigned long offset)
 {
@@ -231,6 +238,14 @@ static ssize_t tegra_dpaux_transfer(struct drm_dp_aux *aux,
 	return ret;
 }
 
+static void tegra_dpaux_hotplug(struct work_struct *work)
+{
+	struct tegra_dpaux *dpaux = work_to_dpaux(work);
+
+	if (dpaux->output)
+		drm_helper_hpd_irq_event(dpaux->output->connector.dev);
+}
+
 static irqreturn_t tegra_dpaux_irq(int irq, void *data)
 {
 	struct tegra_dpaux *dpaux = data;
@@ -241,16 +256,8 @@ static irqreturn_t tegra_dpaux_irq(int irq, void *data)
 	value = tegra_dpaux_readl(dpaux, DPAUX_INTR_AUX);
 	tegra_dpaux_writel(dpaux, value, DPAUX_INTR_AUX);
 
-	if (value & DPAUX_INTR_PLUG_EVENT) {
-		if (dpaux->output) {
-			drm_helper_hpd_irq_event(dpaux->output->connector.dev);
-		}
-	}
-
-	if (value & DPAUX_INTR_UNPLUG_EVENT) {
-		if (dpaux->output)
-			drm_helper_hpd_irq_event(dpaux->output->connector.dev);
-	}
+	if (value & (DPAUX_INTR_PLUG_EVENT | DPAUX_INTR_UNPLUG_EVENT))
+		schedule_work(&dpaux->work);
 
 	if (value & DPAUX_INTR_IRQ_EVENT) {
 		/* TODO: handle this */
@@ -273,6 +280,7 @@ static int tegra_dpaux_probe(struct platform_device *pdev)
 	if (!dpaux)
 		return -ENOMEM;
 
+	INIT_WORK(&dpaux->work, tegra_dpaux_hotplug);
 	init_completion(&dpaux->complete);
 	INIT_LIST_HEAD(&dpaux->list);
 	dpaux->dev = &pdev->dev;
@@ -332,7 +340,7 @@ static int tegra_dpaux_probe(struct platform_device *pdev)
 	dpaux->aux.transfer = tegra_dpaux_transfer;
 	dpaux->aux.dev = &pdev->dev;
 
-	err = drm_dp_aux_register_i2c_bus(&dpaux->aux);
+	err = drm_dp_aux_register(&dpaux->aux);
 	if (err < 0)
 		return err;
 
@@ -355,12 +363,14 @@ static int tegra_dpaux_remove(struct platform_device *pdev)
 {
 	struct tegra_dpaux *dpaux = platform_get_drvdata(pdev);
 
-	drm_dp_aux_unregister_i2c_bus(&dpaux->aux);
+	drm_dp_aux_unregister(&dpaux->aux);
 
 	mutex_lock(&dpaux_lock);
 	list_del(&dpaux->list);
 	mutex_unlock(&dpaux_lock);
 
+	cancel_work_sync(&dpaux->work);
+
 	clk_disable_unprepare(dpaux->clk_parent);
 	reset_control_assert(dpaux->rst);
 	clk_disable_unprepare(dpaux->clk);
@@ -404,6 +414,7 @@ int tegra_dpaux_attach(struct tegra_dpaux *dpaux, struct tegra_output *output)
 	unsigned long timeout;
 	int err;
 
+	output->connector.polled = DRM_CONNECTOR_POLL_HPD;
 	dpaux->output = output;
 
 	err = regulator_enable(dpaux->vdd);
diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c
index 6f5b6e2f552e..3396f9f6a9f7 100644
--- a/drivers/gpu/drm/tegra/drm.c
+++ b/drivers/gpu/drm/tegra/drm.c
@@ -33,7 +33,6 @@ static int tegra_drm_load(struct drm_device *drm, unsigned long flags)
 	if (!tegra)
 		return -ENOMEM;
 
-	dev_set_drvdata(drm->dev, tegra);
 	mutex_init(&tegra->clients_lock);
 	INIT_LIST_HEAD(&tegra->clients);
 	drm->dev_private = tegra;
@@ -640,14 +639,40 @@ int tegra_drm_unregister_client(struct tegra_drm *tegra,
 	return 0;
 }
 
-static int host1x_drm_probe(struct host1x_device *device)
+static int host1x_drm_probe(struct host1x_device *dev)
 {
-	return drm_host1x_init(&tegra_drm_driver, device);
+	struct drm_driver *driver = &tegra_drm_driver;
+	struct drm_device *drm;
+	int err;
+
+	drm = drm_dev_alloc(driver, &dev->dev);
+	if (!drm)
+		return -ENOMEM;
+
+	drm_dev_set_unique(drm, dev_name(&dev->dev));
+	dev_set_drvdata(&dev->dev, drm);
+
+	err = drm_dev_register(drm, 0);
+	if (err < 0)
+		goto unref;
+
+	DRM_INFO("Initialized %s %d.%d.%d %s on minor %d\n", driver->name,
+		 driver->major, driver->minor, driver->patchlevel,
+		 driver->date, drm->primary->index);
+
+	return 0;
+
+unref:
+	drm_dev_unref(drm);
+	return err;
 }
 
-static int host1x_drm_remove(struct host1x_device *device)
+static int host1x_drm_remove(struct host1x_device *dev)
 {
-	drm_host1x_exit(&tegra_drm_driver, device);
+	struct drm_device *drm = dev_get_drvdata(&dev->dev);
+
+	drm_dev_unregister(drm);
+	drm_dev_unref(drm);
 
 	return 0;
 }
@@ -666,6 +691,7 @@ static const struct of_device_id host1x_drm_subdevs[] = {
 	{ .compatible = "nvidia,tegra114-gr3d", },
 	{ .compatible = "nvidia,tegra124-dc", },
 	{ .compatible = "nvidia,tegra124-sor", },
+	{ .compatible = "nvidia,tegra124-hdmi", },
 	{ /* sentinel */ }
 };
 
diff --git a/drivers/gpu/drm/tegra/drm.h b/drivers/gpu/drm/tegra/drm.h
index 126332c3ecbb..6b8fe9d86ed4 100644
--- a/drivers/gpu/drm/tegra/drm.h
+++ b/drivers/gpu/drm/tegra/drm.h
@@ -80,13 +80,13 @@ host1x_to_drm_client(struct host1x_client *client)
 	return container_of(client, struct tegra_drm_client, base);
 }
 
-extern int tegra_drm_register_client(struct tegra_drm *tegra,
-				     struct tegra_drm_client *client);
-extern int tegra_drm_unregister_client(struct tegra_drm *tegra,
-				       struct tegra_drm_client *client);
+int tegra_drm_register_client(struct tegra_drm *tegra,
+			      struct tegra_drm_client *client);
+int tegra_drm_unregister_client(struct tegra_drm *tegra,
+				struct tegra_drm_client *client);
 
-extern int tegra_drm_init(struct tegra_drm *tegra, struct drm_device *drm);
-extern int tegra_drm_exit(struct tegra_drm *tegra);
+int tegra_drm_init(struct tegra_drm *tegra, struct drm_device *drm);
+int tegra_drm_exit(struct tegra_drm *tegra);
 
 struct tegra_dc_soc_info;
 struct tegra_output;
@@ -156,6 +156,7 @@ struct tegra_dc_window {
 	} dst;
 	unsigned int bits_per_pixel;
 	unsigned int format;
+	unsigned int swap;
 	unsigned int stride[2];
 	unsigned long base[3];
 	bool bottom_up;
@@ -163,19 +164,15 @@ struct tegra_dc_window {
 };
 
 /* from dc.c */
-extern unsigned int tegra_dc_format(uint32_t format);
-extern int tegra_dc_setup_window(struct tegra_dc *dc, unsigned int index,
-				 const struct tegra_dc_window *window);
-extern void tegra_dc_enable_vblank(struct tegra_dc *dc);
-extern void tegra_dc_disable_vblank(struct tegra_dc *dc);
-extern void tegra_dc_cancel_page_flip(struct drm_crtc *crtc,
-				      struct drm_file *file);
+void tegra_dc_enable_vblank(struct tegra_dc *dc);
+void tegra_dc_disable_vblank(struct tegra_dc *dc);
+void tegra_dc_cancel_page_flip(struct drm_crtc *crtc, struct drm_file *file);
 
 struct tegra_output_ops {
 	int (*enable)(struct tegra_output *output);
 	int (*disable)(struct tegra_output *output);
 	int (*setup_clock)(struct tegra_output *output, struct clk *clk,
-			   unsigned long pclk);
+			   unsigned long pclk, unsigned int *div);
 	int (*check_mode)(struct tegra_output *output,
 			  struct drm_display_mode *mode,
 			  enum drm_mode_status *status);
@@ -233,10 +230,11 @@ static inline int tegra_output_disable(struct tegra_output *output)
 }
 
 static inline int tegra_output_setup_clock(struct tegra_output *output,
-					   struct clk *clk, unsigned long pclk)
+					   struct clk *clk, unsigned long pclk,
+					   unsigned int *div)
 {
 	if (output && output->ops && output->ops->setup_clock)
-		return output->ops->setup_clock(output, clk, pclk);
+		return output->ops->setup_clock(output, clk, pclk, div);
 
 	return output ? -ENOSYS : -EINVAL;
 }
@@ -251,27 +249,21 @@ static inline int tegra_output_check_mode(struct tegra_output *output,
 	return output ? -ENOSYS : -EINVAL;
 }
 
-/* from bus.c */
-int drm_host1x_init(struct drm_driver *driver, struct host1x_device *device);
-void drm_host1x_exit(struct drm_driver *driver, struct host1x_device *device);
-
 /* from rgb.c */
-extern int tegra_dc_rgb_probe(struct tegra_dc *dc);
-extern int tegra_dc_rgb_remove(struct tegra_dc *dc);
-extern int tegra_dc_rgb_init(struct drm_device *drm, struct tegra_dc *dc);
-extern int tegra_dc_rgb_exit(struct tegra_dc *dc);
+int tegra_dc_rgb_probe(struct tegra_dc *dc);
+int tegra_dc_rgb_remove(struct tegra_dc *dc);
+int tegra_dc_rgb_init(struct drm_device *drm, struct tegra_dc *dc);
+int tegra_dc_rgb_exit(struct tegra_dc *dc);
 
 /* from output.c */
-extern int tegra_output_probe(struct tegra_output *output);
-extern int tegra_output_remove(struct tegra_output *output);
-extern int tegra_output_init(struct drm_device *drm, struct tegra_output *output);
-extern int tegra_output_exit(struct tegra_output *output);
+int tegra_output_probe(struct tegra_output *output);
+int tegra_output_remove(struct tegra_output *output);
+int tegra_output_init(struct drm_device *drm, struct tegra_output *output);
+int tegra_output_exit(struct tegra_output *output);
 
 /* from dpaux.c */
-
 struct tegra_dpaux;
 struct drm_dp_link;
-struct drm_dp_aux;
 
 struct tegra_dpaux *tegra_dpaux_find_by_of_node(struct device_node *np);
 enum drm_connector_status tegra_dpaux_detect(struct tegra_dpaux *dpaux);
@@ -288,10 +280,10 @@ struct tegra_bo *tegra_fb_get_plane(struct drm_framebuffer *framebuffer,
 				    unsigned int index);
 bool tegra_fb_is_bottom_up(struct drm_framebuffer *framebuffer);
 bool tegra_fb_is_tiled(struct drm_framebuffer *framebuffer);
-extern int tegra_drm_fb_init(struct drm_device *drm);
-extern void tegra_drm_fb_exit(struct drm_device *drm);
+int tegra_drm_fb_init(struct drm_device *drm);
+void tegra_drm_fb_exit(struct drm_device *drm);
 #ifdef CONFIG_DRM_TEGRA_FBDEV
-extern void tegra_fbdev_restore_mode(struct tegra_fbdev *fbdev);
+void tegra_fbdev_restore_mode(struct tegra_fbdev *fbdev);
 #endif
 
 extern struct platform_driver tegra_dc_driver;
diff --git a/drivers/gpu/drm/tegra/dsi.c b/drivers/gpu/drm/tegra/dsi.c
index 0e599f0417c0..bd56f2affa78 100644
--- a/drivers/gpu/drm/tegra/dsi.c
+++ b/drivers/gpu/drm/tegra/dsi.c
@@ -14,6 +14,8 @@
 #include <linux/platform_device.h>
 #include <linux/reset.h>
 
+#include <linux/regulator/consumer.h>
+
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_panel.h>
 
@@ -43,11 +45,15 @@ struct tegra_dsi {
 	struct drm_minor *minor;
 	struct dentry *debugfs;
 
+	unsigned long flags;
 	enum mipi_dsi_pixel_format format;
 	unsigned int lanes;
 
 	struct tegra_mipi_device *mipi;
 	struct mipi_dsi_host host;
+
+	struct regulator *vdd;
+	bool enabled;
 };
 
 static inline struct tegra_dsi *
@@ -244,8 +250,10 @@ static int tegra_dsi_debugfs_exit(struct tegra_dsi *dsi)
 #define PKT_LP		(1 << 30)
 #define NUM_PKT_SEQ	12
 
-/* non-burst mode with sync-end */
-static const u32 pkt_seq_vnb_syne[NUM_PKT_SEQ] = {
+/*
+ * non-burst mode with sync pulses
+ */
+static const u32 pkt_seq_video_non_burst_sync_pulses[NUM_PKT_SEQ] = {
 	[ 0] = PKT_ID0(MIPI_DSI_V_SYNC_START) | PKT_LEN0(0) |
 	       PKT_ID1(MIPI_DSI_BLANKING_PACKET) | PKT_LEN1(1) |
 	       PKT_ID2(MIPI_DSI_H_SYNC_END) | PKT_LEN2(0) |
@@ -280,6 +288,36 @@ static const u32 pkt_seq_vnb_syne[NUM_PKT_SEQ] = {
 	       PKT_ID2(MIPI_DSI_BLANKING_PACKET) | PKT_LEN2(4),
 };
 
+/*
+ * non-burst mode with sync events
+ */
+static const u32 pkt_seq_video_non_burst_sync_events[NUM_PKT_SEQ] = {
+	[ 0] = PKT_ID0(MIPI_DSI_V_SYNC_START) | PKT_LEN0(0) |
+	       PKT_ID1(MIPI_DSI_END_OF_TRANSMISSION) | PKT_LEN1(7) |
+	       PKT_LP,
+	[ 1] = 0,
+	[ 2] = PKT_ID0(MIPI_DSI_H_SYNC_START) | PKT_LEN0(0) |
+	       PKT_ID1(MIPI_DSI_END_OF_TRANSMISSION) | PKT_LEN1(7) |
+	       PKT_LP,
+	[ 3] = 0,
+	[ 4] = PKT_ID0(MIPI_DSI_H_SYNC_START) | PKT_LEN0(0) |
+	       PKT_ID1(MIPI_DSI_END_OF_TRANSMISSION) | PKT_LEN1(7) |
+	       PKT_LP,
+	[ 5] = 0,
+	[ 6] = PKT_ID0(MIPI_DSI_H_SYNC_START) | PKT_LEN0(0) |
+	       PKT_ID1(MIPI_DSI_BLANKING_PACKET) | PKT_LEN1(2) |
+	       PKT_ID2(MIPI_DSI_PACKED_PIXEL_STREAM_24) | PKT_LEN2(3),
+	[ 7] = PKT_ID0(MIPI_DSI_BLANKING_PACKET) | PKT_LEN0(4),
+	[ 8] = PKT_ID0(MIPI_DSI_H_SYNC_START) | PKT_LEN0(0) |
+	       PKT_ID1(MIPI_DSI_END_OF_TRANSMISSION) | PKT_LEN1(7) |
+	       PKT_LP,
+	[ 9] = 0,
+	[10] = PKT_ID0(MIPI_DSI_H_SYNC_START) | PKT_LEN0(0) |
+	       PKT_ID1(MIPI_DSI_BLANKING_PACKET) | PKT_LEN1(2) |
+	       PKT_ID2(MIPI_DSI_PACKED_PIXEL_STREAM_24) | PKT_LEN2(3),
+	[11] = PKT_ID0(MIPI_DSI_BLANKING_PACKET) | PKT_LEN0(4),
+};
+
 static int tegra_dsi_set_phy_timing(struct tegra_dsi *dsi)
 {
 	struct mipi_dphy_timing timing;
@@ -361,28 +399,70 @@ static int tegra_dsi_get_muldiv(enum mipi_dsi_pixel_format format,
 	return 0;
 }
 
+static int tegra_dsi_get_format(enum mipi_dsi_pixel_format format,
+				enum tegra_dsi_format *fmt)
+{
+	switch (format) {
+	case MIPI_DSI_FMT_RGB888:
+		*fmt = TEGRA_DSI_FORMAT_24P;
+		break;
+
+	case MIPI_DSI_FMT_RGB666:
+		*fmt = TEGRA_DSI_FORMAT_18NP;
+		break;
+
+	case MIPI_DSI_FMT_RGB666_PACKED:
+		*fmt = TEGRA_DSI_FORMAT_18P;
+		break;
+
+	case MIPI_DSI_FMT_RGB565:
+		*fmt = TEGRA_DSI_FORMAT_16P;
+		break;
+
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 static int tegra_output_dsi_enable(struct tegra_output *output)
 {
 	struct tegra_dc *dc = to_tegra_dc(output->encoder.crtc);
 	struct drm_display_mode *mode = &dc->base.mode;
 	unsigned int hact, hsw, hbp, hfp, i, mul, div;
 	struct tegra_dsi *dsi = to_dsi(output);
-	/* FIXME: don't hardcode this */
-	const u32 *pkt_seq = pkt_seq_vnb_syne;
+	enum tegra_dsi_format format;
 	unsigned long value;
+	const u32 *pkt_seq;
 	int err;
 
+	if (dsi->enabled)
+		return 0;
+
+	if (dsi->flags & MIPI_DSI_MODE_VIDEO_SYNC_PULSE) {
+		DRM_DEBUG_KMS("Non-burst video mode with sync pulses\n");
+		pkt_seq = pkt_seq_video_non_burst_sync_pulses;
+	} else {
+		DRM_DEBUG_KMS("Non-burst video mode with sync events\n");
+		pkt_seq = pkt_seq_video_non_burst_sync_events;
+	}
+
 	err = tegra_dsi_get_muldiv(dsi->format, &mul, &div);
 	if (err < 0)
 		return err;
 
+	err = tegra_dsi_get_format(dsi->format, &format);
+	if (err < 0)
+		return err;
+
 	err = clk_enable(dsi->clk);
 	if (err < 0)
 		return err;
 
 	reset_control_deassert(dsi->rst);
 
-	value = DSI_CONTROL_CHANNEL(0) | DSI_CONTROL_FORMAT(dsi->format) |
+	value = DSI_CONTROL_CHANNEL(0) | DSI_CONTROL_FORMAT(format) |
 		DSI_CONTROL_LANES(dsi->lanes - 1) |
 		DSI_CONTROL_SOURCE(dc->pipe);
 	tegra_dsi_writel(dsi, value, DSI_CONTROL);
@@ -454,6 +534,8 @@ static int tegra_output_dsi_enable(struct tegra_output *output)
 	value |= DSI_POWER_CONTROL_ENABLE;
 	tegra_dsi_writel(dsi, value, DSI_POWER_CONTROL);
 
+	dsi->enabled = true;
+
 	return 0;
 }
 
@@ -463,9 +545,12 @@ static int tegra_output_dsi_disable(struct tegra_output *output)
 	struct tegra_dsi *dsi = to_dsi(output);
 	unsigned long value;
 
+	if (!dsi->enabled)
+		return 0;
+
 	/* disable DSI controller */
 	value = tegra_dsi_readl(dsi, DSI_POWER_CONTROL);
-	value &= DSI_POWER_CONTROL_ENABLE;
+	value &= ~DSI_POWER_CONTROL_ENABLE;
 	tegra_dsi_writel(dsi, value, DSI_POWER_CONTROL);
 
 	/*
@@ -492,30 +577,44 @@ static int tegra_output_dsi_disable(struct tegra_output *output)
 
 	clk_disable(dsi->clk);
 
+	dsi->enabled = false;
+
 	return 0;
 }
 
 static int tegra_output_dsi_setup_clock(struct tegra_output *output,
-					struct clk *clk, unsigned long pclk)
+					struct clk *clk, unsigned long pclk,
+					unsigned int *divp)
 {
 	struct tegra_dc *dc = to_tegra_dc(output->encoder.crtc);
 	struct drm_display_mode *mode = &dc->base.mode;
 	unsigned int timeout, mul, div, vrefresh;
 	struct tegra_dsi *dsi = to_dsi(output);
 	unsigned long bclk, plld, value;
-	struct clk *base;
 	int err;
 
 	err = tegra_dsi_get_muldiv(dsi->format, &mul, &div);
 	if (err < 0)
 		return err;
 
+	DRM_DEBUG_KMS("mul: %u, div: %u, lanes: %u\n", mul, div, dsi->lanes);
 	vrefresh = drm_mode_vrefresh(mode);
+	DRM_DEBUG_KMS("vrefresh: %u\n", vrefresh);
 
-	pclk = mode->htotal * mode->vtotal * vrefresh;
+	/* compute byte clock */
 	bclk = (pclk * mul) / (div * dsi->lanes);
-	plld = DIV_ROUND_UP(bclk * 8, 1000000);
-	pclk = (plld * 1000000) / 2;
+
+	/*
+	 * Compute bit clock and round up to the next MHz.
+	 */
+	plld = DIV_ROUND_UP(bclk * 8, 1000000) * 1000000;
+
+	/*
+	 * We divide the frequency by two here, but we make up for that by
+	 * setting the shift clock divider (further below) to half of the
+	 * correct value.
+	 */
+	plld /= 2;
 
 	err = clk_set_parent(clk, dsi->clk_parent);
 	if (err < 0) {
@@ -523,20 +622,26 @@ static int tegra_output_dsi_setup_clock(struct tegra_output *output,
 		return err;
 	}
 
-	base = clk_get_parent(dsi->clk_parent);
-
-	/*
-	 * This assumes that the parent clock is pll_d_out0 or pll_d2_out
-	 * respectively, each of which divides the base pll_d by 2.
-	 */
-	err = clk_set_rate(base, pclk * 2);
+	err = clk_set_rate(dsi->clk_parent, plld);
 	if (err < 0) {
 		dev_err(dsi->dev, "failed to set base clock rate to %lu Hz\n",
-			pclk * 2);
+			plld);
 		return err;
 	}
 
 	/*
+	 * Derive pixel clock from bit clock using the shift clock divider.
+	 * Note that this is only half of what we would expect, but we need
+	 * that to make up for the fact that we divided the bit clock by a
+	 * factor of two above.
+	 *
+	 * It's not clear exactly why this is necessary, but the display is
+	 * not working properly otherwise. Perhaps the PLLs cannot generate
+	 * frequencies sufficiently high.
+	 */
+	*divp = ((8 * mul) / (div * dsi->lanes)) - 2;
+
+	/*
 	 * XXX: Move the below somewhere else so that we don't need to have
 	 * access to the vrefresh in this function?
 	 */
@@ -610,61 +715,32 @@ static int tegra_dsi_pad_calibrate(struct tegra_dsi *dsi)
 
 static int tegra_dsi_init(struct host1x_client *client)
 {
-	struct tegra_drm *tegra = dev_get_drvdata(client->parent);
+	struct drm_device *drm = dev_get_drvdata(client->parent);
 	struct tegra_dsi *dsi = host1x_client_to_dsi(client);
-	unsigned long value, i;
 	int err;
 
 	dsi->output.type = TEGRA_OUTPUT_DSI;
 	dsi->output.dev = client->dev;
 	dsi->output.ops = &dsi_ops;
 
-	err = tegra_output_init(tegra->drm, &dsi->output);
+	err = tegra_output_init(drm, &dsi->output);
 	if (err < 0) {
 		dev_err(client->dev, "output setup failed: %d\n", err);
 		return err;
 	}
 
 	if (IS_ENABLED(CONFIG_DEBUG_FS)) {
-		err = tegra_dsi_debugfs_init(dsi, tegra->drm->primary);
+		err = tegra_dsi_debugfs_init(dsi, drm->primary);
 		if (err < 0)
 			dev_err(dsi->dev, "debugfs setup failed: %d\n", err);
 	}
 
-	/*
-	 * enable high-speed mode, checksum generation, ECC generation and
-	 * disable raw mode
-	 */
-	value = tegra_dsi_readl(dsi, DSI_HOST_CONTROL);
-	value |= DSI_HOST_CONTROL_ECC | DSI_HOST_CONTROL_CS |
-		 DSI_HOST_CONTROL_HS;
-	value &= ~DSI_HOST_CONTROL_RAW;
-	tegra_dsi_writel(dsi, value, DSI_HOST_CONTROL);
-
-	tegra_dsi_writel(dsi, 0, DSI_SOL_DELAY);
-	tegra_dsi_writel(dsi, 0, DSI_MAX_THRESHOLD);
-
-	tegra_dsi_writel(dsi, 0, DSI_INIT_SEQ_CONTROL);
-
-	for (i = 0; i < 8; i++) {
-		tegra_dsi_writel(dsi, 0, DSI_INIT_SEQ_DATA_0 + i);
-		tegra_dsi_writel(dsi, 0, DSI_INIT_SEQ_DATA_8 + i);
-	}
-
-	for (i = 0; i < 12; i++)
-		tegra_dsi_writel(dsi, 0, DSI_PKT_SEQ_0_LO + i);
-
-	tegra_dsi_writel(dsi, 0, DSI_DCS_CMDS);
-
 	err = tegra_dsi_pad_calibrate(dsi);
 	if (err < 0) {
 		dev_err(dsi->dev, "MIPI calibration failed: %d\n", err);
 		return err;
 	}
 
-	tegra_dsi_writel(dsi, DSI_POWER_CONTROL_ENABLE, DSI_POWER_CONTROL);
-	usleep_range(300, 1000);
-
 	return 0;
 }
 
@@ -715,66 +791,13 @@ static int tegra_dsi_setup_clocks(struct tegra_dsi *dsi)
 	return 0;
 }
 
-static void tegra_dsi_initialize(struct tegra_dsi *dsi)
-{
-	unsigned int i;
-
-	tegra_dsi_writel(dsi, 0, DSI_POWER_CONTROL);
-
-	tegra_dsi_writel(dsi, 0, DSI_INT_ENABLE);
-	tegra_dsi_writel(dsi, 0, DSI_INT_STATUS);
-	tegra_dsi_writel(dsi, 0, DSI_INT_MASK);
-
-	tegra_dsi_writel(dsi, 0, DSI_HOST_CONTROL);
-	tegra_dsi_writel(dsi, 0, DSI_CONTROL);
-
-	tegra_dsi_writel(dsi, 0, DSI_SOL_DELAY);
-	tegra_dsi_writel(dsi, 0, DSI_MAX_THRESHOLD);
-
-	tegra_dsi_writel(dsi, 0, DSI_INIT_SEQ_CONTROL);
-
-	for (i = 0; i < 8; i++) {
-		tegra_dsi_writel(dsi, 0, DSI_INIT_SEQ_DATA_0 + i);
-		tegra_dsi_writel(dsi, 0, DSI_INIT_SEQ_DATA_8 + i);
-	}
-
-	for (i = 0; i < 12; i++)
-		tegra_dsi_writel(dsi, 0, DSI_PKT_SEQ_0_LO + i);
-
-	tegra_dsi_writel(dsi, 0, DSI_DCS_CMDS);
-
-	for (i = 0; i < 4; i++)
-		tegra_dsi_writel(dsi, 0, DSI_PKT_LEN_0_1 + i);
-
-	tegra_dsi_writel(dsi, 0x00000000, DSI_PHY_TIMING_0);
-	tegra_dsi_writel(dsi, 0x00000000, DSI_PHY_TIMING_1);
-	tegra_dsi_writel(dsi, 0x000000ff, DSI_PHY_TIMING_2);
-	tegra_dsi_writel(dsi, 0x00000000, DSI_BTA_TIMING);
-
-	tegra_dsi_writel(dsi, 0, DSI_TIMEOUT_0);
-	tegra_dsi_writel(dsi, 0, DSI_TIMEOUT_1);
-	tegra_dsi_writel(dsi, 0, DSI_TO_TALLY);
-
-	tegra_dsi_writel(dsi, 0, DSI_PAD_CONTROL_0);
-	tegra_dsi_writel(dsi, 0, DSI_PAD_CONTROL_CD);
-	tegra_dsi_writel(dsi, 0, DSI_PAD_CD_STATUS);
-	tegra_dsi_writel(dsi, 0, DSI_VIDEO_MODE_CONTROL);
-	tegra_dsi_writel(dsi, 0, DSI_PAD_CONTROL_1);
-	tegra_dsi_writel(dsi, 0, DSI_PAD_CONTROL_2);
-	tegra_dsi_writel(dsi, 0, DSI_PAD_CONTROL_3);
-	tegra_dsi_writel(dsi, 0, DSI_PAD_CONTROL_4);
-
-	tegra_dsi_writel(dsi, 0, DSI_GANGED_MODE_CONTROL);
-	tegra_dsi_writel(dsi, 0, DSI_GANGED_MODE_START);
-	tegra_dsi_writel(dsi, 0, DSI_GANGED_MODE_SIZE);
-}
-
 static int tegra_dsi_host_attach(struct mipi_dsi_host *host,
 				 struct mipi_dsi_device *device)
 {
 	struct tegra_dsi *dsi = host_to_tegra(host);
 	struct tegra_output *output = &dsi->output;
 
+	dsi->flags = device->mode_flags;
 	dsi->format = device->format;
 	dsi->lanes = device->lanes;
 
@@ -829,6 +852,7 @@ static int tegra_dsi_probe(struct platform_device *pdev)
 	 * attaches to the DSI host, the parameters will be taken from
 	 * the attached device.
 	 */
+	dsi->flags = MIPI_DSI_MODE_VIDEO;
 	dsi->format = MIPI_DSI_FMT_RGB888;
 	dsi->lanes = 4;
 
@@ -872,6 +896,18 @@ static int tegra_dsi_probe(struct platform_device *pdev)
 		return err;
 	}
 
+	dsi->vdd = devm_regulator_get(&pdev->dev, "avdd-dsi-csi");
+	if (IS_ERR(dsi->vdd)) {
+		dev_err(&pdev->dev, "cannot get VDD supply\n");
+		return PTR_ERR(dsi->vdd);
+	}
+
+	err = regulator_enable(dsi->vdd);
+	if (err < 0) {
+		dev_err(&pdev->dev, "cannot enable VDD supply\n");
+		return err;
+	}
+
 	err = tegra_dsi_setup_clocks(dsi);
 	if (err < 0) {
 		dev_err(&pdev->dev, "cannot setup clocks\n");
@@ -883,8 +919,6 @@ static int tegra_dsi_probe(struct platform_device *pdev)
 	if (IS_ERR(dsi->regs))
 		return PTR_ERR(dsi->regs);
 
-	tegra_dsi_initialize(dsi);
-
 	dsi->mipi = tegra_mipi_request(&pdev->dev);
 	if (IS_ERR(dsi->mipi))
 		return PTR_ERR(dsi->mipi);
@@ -929,9 +963,11 @@ static int tegra_dsi_remove(struct platform_device *pdev)
 	mipi_dsi_host_unregister(&dsi->host);
 	tegra_mipi_free(dsi->mipi);
 
+	regulator_disable(dsi->vdd);
 	clk_disable_unprepare(dsi->clk_parent);
 	clk_disable_unprepare(dsi->clk_lp);
 	clk_disable_unprepare(dsi->clk);
+	reset_control_assert(dsi->rst);
 
 	err = tegra_output_remove(&dsi->output);
 	if (err < 0) {
diff --git a/drivers/gpu/drm/tegra/dsi.h b/drivers/gpu/drm/tegra/dsi.h
index 1db5cc24ea91..5ce610d08d77 100644
--- a/drivers/gpu/drm/tegra/dsi.h
+++ b/drivers/gpu/drm/tegra/dsi.h
@@ -117,4 +117,14 @@
 #define DSI_INIT_SEQ_DATA_14		0x5e
 #define DSI_INIT_SEQ_DATA_15		0x5f
 
+/*
+ * pixel format as used in the DSI_CONTROL_FORMAT field
+ */
+enum tegra_dsi_format {
+	TEGRA_DSI_FORMAT_16P,
+	TEGRA_DSI_FORMAT_18NP,
+	TEGRA_DSI_FORMAT_18P,
+	TEGRA_DSI_FORMAT_24P,
+};
+
 #endif
diff --git a/drivers/gpu/drm/tegra/fb.c b/drivers/gpu/drm/tegra/fb.c
index f7fca09d4921..9798a7080322 100644
--- a/drivers/gpu/drm/tegra/fb.c
+++ b/drivers/gpu/drm/tegra/fb.c
@@ -346,11 +346,8 @@ static void tegra_fbdev_free(struct tegra_fbdev *fbdev)
 
 void tegra_fbdev_restore_mode(struct tegra_fbdev *fbdev)
 {
-	if (fbdev) {
-		drm_modeset_lock_all(fbdev->base.dev);
-		drm_fb_helper_restore_fbdev_mode(&fbdev->base);
-		drm_modeset_unlock_all(fbdev->base.dev);
-	}
+	if (fbdev)
+		drm_fb_helper_restore_fbdev_mode_unlocked(&fbdev->base);
 }
 
 static void tegra_fb_output_poll_changed(struct drm_device *drm)
diff --git a/drivers/gpu/drm/tegra/gem.c b/drivers/gpu/drm/tegra/gem.c
index bcf9895cef9f..aa85b7b26f10 100644
--- a/drivers/gpu/drm/tegra/gem.c
+++ b/drivers/gpu/drm/tegra/gem.c
@@ -169,7 +169,8 @@ err:
 	return ERR_PTR(ret);
 }
 
-struct tegra_bo *tegra_bo_import(struct drm_device *drm, struct dma_buf *buf)
+static struct tegra_bo *tegra_bo_import(struct drm_device *drm,
+					struct dma_buf *buf)
 {
 	struct dma_buf_attachment *attach;
 	struct tegra_bo *bo;
diff --git a/drivers/gpu/drm/tegra/gr2d.c b/drivers/gpu/drm/tegra/gr2d.c
index 2c7ca748edf5..7c53941f2a9e 100644
--- a/drivers/gpu/drm/tegra/gr2d.c
+++ b/drivers/gpu/drm/tegra/gr2d.c
@@ -28,7 +28,7 @@ static inline struct gr2d *to_gr2d(struct tegra_drm_client *client)
 static int gr2d_init(struct host1x_client *client)
 {
 	struct tegra_drm_client *drm = host1x_to_drm_client(client);
-	struct tegra_drm *tegra = dev_get_drvdata(client->parent);
+	struct drm_device *dev = dev_get_drvdata(client->parent);
 	unsigned long flags = HOST1X_SYNCPT_HAS_BASE;
 	struct gr2d *gr2d = to_gr2d(drm);
 
@@ -42,17 +42,17 @@ static int gr2d_init(struct host1x_client *client)
 		return -ENOMEM;
 	}
 
-	return tegra_drm_register_client(tegra, drm);
+	return tegra_drm_register_client(dev->dev_private, drm);
 }
 
 static int gr2d_exit(struct host1x_client *client)
 {
 	struct tegra_drm_client *drm = host1x_to_drm_client(client);
-	struct tegra_drm *tegra = dev_get_drvdata(client->parent);
+	struct drm_device *dev = dev_get_drvdata(client->parent);
 	struct gr2d *gr2d = to_gr2d(drm);
 	int err;
 
-	err = tegra_drm_unregister_client(tegra, drm);
+	err = tegra_drm_unregister_client(dev->dev_private, drm);
 	if (err < 0)
 		return err;
 
diff --git a/drivers/gpu/drm/tegra/gr3d.c b/drivers/gpu/drm/tegra/gr3d.c
index 0cbb24b1ae04..30f5ba9bd6d0 100644
--- a/drivers/gpu/drm/tegra/gr3d.c
+++ b/drivers/gpu/drm/tegra/gr3d.c
@@ -37,7 +37,7 @@ static inline struct gr3d *to_gr3d(struct tegra_drm_client *client)
 static int gr3d_init(struct host1x_client *client)
 {
 	struct tegra_drm_client *drm = host1x_to_drm_client(client);
-	struct tegra_drm *tegra = dev_get_drvdata(client->parent);
+	struct drm_device *dev = dev_get_drvdata(client->parent);
 	unsigned long flags = HOST1X_SYNCPT_HAS_BASE;
 	struct gr3d *gr3d = to_gr3d(drm);
 
@@ -51,17 +51,17 @@ static int gr3d_init(struct host1x_client *client)
 		return -ENOMEM;
 	}
 
-	return tegra_drm_register_client(tegra, drm);
+	return tegra_drm_register_client(dev->dev_private, drm);
 }
 
 static int gr3d_exit(struct host1x_client *client)
 {
 	struct tegra_drm_client *drm = host1x_to_drm_client(client);
-	struct tegra_drm *tegra = dev_get_drvdata(client->parent);
+	struct drm_device *dev = dev_get_drvdata(client->parent);
 	struct gr3d *gr3d = to_gr3d(drm);
 	int err;
 
-	err = tegra_drm_unregister_client(tegra, drm);
+	err = tegra_drm_unregister_client(dev->dev_private, drm);
 	if (err < 0)
 		return err;
 
diff --git a/drivers/gpu/drm/tegra/hdmi.c b/drivers/gpu/drm/tegra/hdmi.c
index 6928015d11a4..ba067bb767e3 100644
--- a/drivers/gpu/drm/tegra/hdmi.c
+++ b/drivers/gpu/drm/tegra/hdmi.c
@@ -42,8 +42,9 @@ struct tegra_hdmi {
 	struct device *dev;
 	bool enabled;
 
-	struct regulator *vdd;
+	struct regulator *hdmi;
 	struct regulator *pll;
+	struct regulator *vdd;
 
 	void __iomem *regs;
 	unsigned int irq;
@@ -317,6 +318,85 @@ static const struct tmds_config tegra114_tmds_config[] = {
 	},
 };
 
+static const struct tmds_config tegra124_tmds_config[] = {
+	{ /* 480p/576p / 25.2MHz/27MHz modes */
+		.pclk = 27000000,
+		.pll0 = SOR_PLL_ICHPMP(1) | SOR_PLL_BG_V17_S(3) |
+			SOR_PLL_VCOCAP(0) | SOR_PLL_RESISTORSEL,
+		.pll1 = SOR_PLL_LOADADJ(3) | SOR_PLL_TMDS_TERMADJ(0),
+		.pe_current = PE_CURRENT0(PE_CURRENT_0_mA_T114) |
+			PE_CURRENT1(PE_CURRENT_0_mA_T114) |
+			PE_CURRENT2(PE_CURRENT_0_mA_T114) |
+			PE_CURRENT3(PE_CURRENT_0_mA_T114),
+		.drive_current =
+			DRIVE_CURRENT_LANE0_T114(DRIVE_CURRENT_10_400_mA_T114) |
+			DRIVE_CURRENT_LANE1_T114(DRIVE_CURRENT_10_400_mA_T114) |
+			DRIVE_CURRENT_LANE2_T114(DRIVE_CURRENT_10_400_mA_T114) |
+			DRIVE_CURRENT_LANE3_T114(DRIVE_CURRENT_10_400_mA_T114),
+		.peak_current = PEAK_CURRENT_LANE0(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE1(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE2(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE3(PEAK_CURRENT_0_000_mA),
+	}, { /* 720p / 74.25MHz modes */
+		.pclk = 74250000,
+		.pll0 = SOR_PLL_ICHPMP(1) | SOR_PLL_BG_V17_S(3) |
+			SOR_PLL_VCOCAP(1) | SOR_PLL_RESISTORSEL,
+		.pll1 = SOR_PLL_PE_EN | SOR_PLL_LOADADJ(3) |
+			SOR_PLL_TMDS_TERMADJ(0),
+		.pe_current = PE_CURRENT0(PE_CURRENT_15_mA_T114) |
+			PE_CURRENT1(PE_CURRENT_15_mA_T114) |
+			PE_CURRENT2(PE_CURRENT_15_mA_T114) |
+			PE_CURRENT3(PE_CURRENT_15_mA_T114),
+		.drive_current =
+			DRIVE_CURRENT_LANE0_T114(DRIVE_CURRENT_10_400_mA_T114) |
+			DRIVE_CURRENT_LANE1_T114(DRIVE_CURRENT_10_400_mA_T114) |
+			DRIVE_CURRENT_LANE2_T114(DRIVE_CURRENT_10_400_mA_T114) |
+			DRIVE_CURRENT_LANE3_T114(DRIVE_CURRENT_10_400_mA_T114),
+		.peak_current = PEAK_CURRENT_LANE0(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE1(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE2(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE3(PEAK_CURRENT_0_000_mA),
+	}, { /* 1080p / 148.5MHz modes */
+		.pclk = 148500000,
+		.pll0 = SOR_PLL_ICHPMP(1) | SOR_PLL_BG_V17_S(3) |
+			SOR_PLL_VCOCAP(3) | SOR_PLL_RESISTORSEL,
+		.pll1 = SOR_PLL_PE_EN | SOR_PLL_LOADADJ(3) |
+			SOR_PLL_TMDS_TERMADJ(0),
+		.pe_current = PE_CURRENT0(PE_CURRENT_10_mA_T114) |
+			PE_CURRENT1(PE_CURRENT_10_mA_T114) |
+			PE_CURRENT2(PE_CURRENT_10_mA_T114) |
+			PE_CURRENT3(PE_CURRENT_10_mA_T114),
+		.drive_current =
+			DRIVE_CURRENT_LANE0_T114(DRIVE_CURRENT_12_400_mA_T114) |
+			DRIVE_CURRENT_LANE1_T114(DRIVE_CURRENT_12_400_mA_T114) |
+			DRIVE_CURRENT_LANE2_T114(DRIVE_CURRENT_12_400_mA_T114) |
+			DRIVE_CURRENT_LANE3_T114(DRIVE_CURRENT_12_400_mA_T114),
+		.peak_current = PEAK_CURRENT_LANE0(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE1(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE2(PEAK_CURRENT_0_000_mA) |
+			PEAK_CURRENT_LANE3(PEAK_CURRENT_0_000_mA),
+	}, { /* 225/297MHz modes */
+		.pclk = UINT_MAX,
+		.pll0 = SOR_PLL_ICHPMP(1) | SOR_PLL_BG_V17_S(3) |
+			SOR_PLL_VCOCAP(0xf) | SOR_PLL_RESISTORSEL,
+		.pll1 = SOR_PLL_LOADADJ(3) | SOR_PLL_TMDS_TERMADJ(7)
+			| SOR_PLL_TMDS_TERM_ENABLE,
+		.pe_current = PE_CURRENT0(PE_CURRENT_0_mA_T114) |
+			PE_CURRENT1(PE_CURRENT_0_mA_T114) |
+			PE_CURRENT2(PE_CURRENT_0_mA_T114) |
+			PE_CURRENT3(PE_CURRENT_0_mA_T114),
+		.drive_current =
+			DRIVE_CURRENT_LANE0_T114(DRIVE_CURRENT_25_200_mA_T114) |
+			DRIVE_CURRENT_LANE1_T114(DRIVE_CURRENT_25_200_mA_T114) |
+			DRIVE_CURRENT_LANE2_T114(DRIVE_CURRENT_25_200_mA_T114) |
+			DRIVE_CURRENT_LANE3_T114(DRIVE_CURRENT_19_200_mA_T114),
+		.peak_current = PEAK_CURRENT_LANE0(PEAK_CURRENT_3_000_mA) |
+			PEAK_CURRENT_LANE1(PEAK_CURRENT_3_000_mA) |
+			PEAK_CURRENT_LANE2(PEAK_CURRENT_3_000_mA) |
+			PEAK_CURRENT_LANE3(PEAK_CURRENT_0_800_mA),
+	},
+};
+
 static const struct tegra_hdmi_audio_config *
 tegra_hdmi_get_audio_config(unsigned int audio_freq, unsigned int pclk)
 {
@@ -716,13 +796,9 @@ static int tegra_output_hdmi_enable(struct tegra_output *output)
 		return err;
 	}
 
-	/*
-	 * This assumes that the display controller will divide its parent
-	 * clock by 2 to generate the pixel clock.
-	 */
-	err = tegra_output_setup_clock(output, hdmi->clk, pclk * 2);
+	err = regulator_enable(hdmi->vdd);
 	if (err < 0) {
-		dev_err(hdmi->dev, "failed to setup clock: %d\n", err);
+		dev_err(hdmi->dev, "failed to enable VDD regulator: %d\n", err);
 		return err;
 	}
 
@@ -730,7 +806,7 @@ static int tegra_output_hdmi_enable(struct tegra_output *output)
 	if (err < 0)
 		return err;
 
-	err = clk_enable(hdmi->clk);
+	err = clk_prepare_enable(hdmi->clk);
 	if (err < 0) {
 		dev_err(hdmi->dev, "failed to enable clock: %d\n", err);
 		return err;
@@ -740,6 +816,17 @@ static int tegra_output_hdmi_enable(struct tegra_output *output)
 	usleep_range(1000, 2000);
 	reset_control_deassert(hdmi->rst);
 
+	/* power up sequence */
+	value = tegra_hdmi_readl(hdmi, HDMI_NV_PDISP_SOR_PLL0);
+	value &= ~SOR_PLL_PDBG;
+	tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_SOR_PLL0);
+
+	usleep_range(10, 20);
+
+	value = tegra_hdmi_readl(hdmi, HDMI_NV_PDISP_SOR_PLL0);
+	value &= ~SOR_PLL_PWR;
+	tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_SOR_PLL0);
+
 	tegra_dc_writel(dc, VSYNC_H_POSITION(1),
 			DC_DISP_DISP_TIMING_OPTIONS);
 	tegra_dc_writel(dc, DITHER_CONTROL_DISABLE | BASE_COLOR_SIZE888,
@@ -838,9 +925,13 @@ static int tegra_output_hdmi_enable(struct tegra_output *output)
 	tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_SOR_SEQ_INST(0));
 	tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_SOR_SEQ_INST(8));
 
-	value = 0x1c800;
+	value = tegra_hdmi_readl(hdmi, HDMI_NV_PDISP_SOR_CSTM);
 	value &= ~SOR_CSTM_ROTCLK(~0);
 	value |= SOR_CSTM_ROTCLK(2);
+	value |= SOR_CSTM_PLLDIV;
+	value &= ~SOR_CSTM_LVDS_ENABLE;
+	value &= ~SOR_CSTM_MODE_MASK;
+	value |= SOR_CSTM_MODE_TMDS;
 	tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_SOR_CSTM);
 
 	/* start SOR */
@@ -930,10 +1021,18 @@ static int tegra_output_hdmi_disable(struct tegra_output *output)
 	 * sure it's only executed when the output is attached to one.
 	 */
 	if (dc) {
+		/*
+		 * XXX: We can't do this here because it causes HDMI to go
+		 * into an erroneous state with the result that HDMI won't
+		 * properly work once disabled. See also a similar symptom
+		 * for the SOR output.
+		 */
+		/*
 		value = tegra_dc_readl(dc, DC_CMD_DISPLAY_POWER_CONTROL);
 		value &= ~(PW0_ENABLE | PW1_ENABLE | PW2_ENABLE | PW3_ENABLE |
 			   PW4_ENABLE | PM0_ENABLE | PM1_ENABLE);
 		tegra_dc_writel(dc, value, DC_CMD_DISPLAY_POWER_CONTROL);
+		*/
 
 		value = tegra_dc_readl(dc, DC_CMD_DISPLAY_COMMAND);
 		value &= ~DISP_CTRL_MODE_MASK;
@@ -947,8 +1046,9 @@ static int tegra_output_hdmi_disable(struct tegra_output *output)
 		tegra_dc_writel(dc, GENERAL_ACT_REQ, DC_CMD_STATE_CONTROL);
 	}
 
+	clk_disable_unprepare(hdmi->clk);
 	reset_control_assert(hdmi->rst);
-	clk_disable(hdmi->clk);
+	regulator_disable(hdmi->vdd);
 	regulator_disable(hdmi->pll);
 
 	hdmi->enabled = false;
@@ -957,10 +1057,10 @@ static int tegra_output_hdmi_disable(struct tegra_output *output)
 }
 
 static int tegra_output_hdmi_setup_clock(struct tegra_output *output,
-					 struct clk *clk, unsigned long pclk)
+					 struct clk *clk, unsigned long pclk,
+					 unsigned int *div)
 {
 	struct tegra_hdmi *hdmi = to_hdmi(output);
-	struct clk *base;
 	int err;
 
 	err = clk_set_parent(clk, hdmi->clk_parent);
@@ -969,17 +1069,12 @@ static int tegra_output_hdmi_setup_clock(struct tegra_output *output,
 		return err;
 	}
 
-	base = clk_get_parent(hdmi->clk_parent);
-
-	/*
-	 * This assumes that the parent clock is pll_d_out0 or pll_d2_out
-	 * respectively, each of which divides the base pll_d by 2.
-	 */
-	err = clk_set_rate(base, pclk * 2);
+	err = clk_set_rate(hdmi->clk_parent, pclk);
 	if (err < 0)
-		dev_err(output->dev,
-			"failed to set base clock rate to %lu Hz\n",
-			pclk * 2);
+		dev_err(output->dev, "failed to set clock rate to %lu Hz\n",
+			pclk);
+
+	*div = 0;
 
 	return 0;
 }
@@ -1017,7 +1112,7 @@ static int tegra_hdmi_show_regs(struct seq_file *s, void *data)
 	struct tegra_hdmi *hdmi = node->info_ent->data;
 	int err;
 
-	err = clk_enable(hdmi->clk);
+	err = clk_prepare_enable(hdmi->clk);
 	if (err)
 		return err;
 
@@ -1186,7 +1281,7 @@ static int tegra_hdmi_show_regs(struct seq_file *s, void *data)
 
 #undef DUMP_REG
 
-	clk_disable(hdmi->clk);
+	clk_disable_unprepare(hdmi->clk);
 
 	return 0;
 }
@@ -1252,33 +1347,33 @@ static int tegra_hdmi_debugfs_exit(struct tegra_hdmi *hdmi)
 
 static int tegra_hdmi_init(struct host1x_client *client)
 {
-	struct tegra_drm *tegra = dev_get_drvdata(client->parent);
+	struct drm_device *drm = dev_get_drvdata(client->parent);
 	struct tegra_hdmi *hdmi = host1x_client_to_hdmi(client);
 	int err;
 
-	err = regulator_enable(hdmi->vdd);
-	if (err < 0) {
-		dev_err(client->dev, "failed to enable VDD regulator: %d\n",
-			err);
-		return err;
-	}
-
 	hdmi->output.type = TEGRA_OUTPUT_HDMI;
 	hdmi->output.dev = client->dev;
 	hdmi->output.ops = &hdmi_ops;
 
-	err = tegra_output_init(tegra->drm, &hdmi->output);
+	err = tegra_output_init(drm, &hdmi->output);
 	if (err < 0) {
 		dev_err(client->dev, "output setup failed: %d\n", err);
 		return err;
 	}
 
 	if (IS_ENABLED(CONFIG_DEBUG_FS)) {
-		err = tegra_hdmi_debugfs_init(hdmi, tegra->drm->primary);
+		err = tegra_hdmi_debugfs_init(hdmi, drm->primary);
 		if (err < 0)
 			dev_err(client->dev, "debugfs setup failed: %d\n", err);
 	}
 
+	err = regulator_enable(hdmi->hdmi);
+	if (err < 0) {
+		dev_err(client->dev, "failed to enable HDMI regulator: %d\n",
+			err);
+		return err;
+	}
+
 	return 0;
 }
 
@@ -1287,6 +1382,8 @@ static int tegra_hdmi_exit(struct host1x_client *client)
 	struct tegra_hdmi *hdmi = host1x_client_to_hdmi(client);
 	int err;
 
+	regulator_disable(hdmi->hdmi);
+
 	if (IS_ENABLED(CONFIG_DEBUG_FS)) {
 		err = tegra_hdmi_debugfs_exit(hdmi);
 		if (err < 0)
@@ -1306,8 +1403,6 @@ static int tegra_hdmi_exit(struct host1x_client *client)
 		return err;
 	}
 
-	regulator_disable(hdmi->vdd);
-
 	return 0;
 }
 
@@ -1340,7 +1435,16 @@ static const struct tegra_hdmi_config tegra114_hdmi_config = {
 	.has_sor_io_peak_current = true,
 };
 
+static const struct tegra_hdmi_config tegra124_hdmi_config = {
+	.tmds = tegra124_tmds_config,
+	.num_tmds = ARRAY_SIZE(tegra124_tmds_config),
+	.fuse_override_offset = HDMI_NV_PDISP_SOR_PAD_CTLS0,
+	.fuse_override_value = 1 << 31,
+	.has_sor_io_peak_current = true,
+};
+
 static const struct of_device_id tegra_hdmi_of_match[] = {
+	{ .compatible = "nvidia,tegra124-hdmi", .data = &tegra124_hdmi_config },
 	{ .compatible = "nvidia,tegra114-hdmi", .data = &tegra114_hdmi_config },
 	{ .compatible = "nvidia,tegra30-hdmi", .data = &tegra30_hdmi_config },
 	{ .compatible = "nvidia,tegra20-hdmi", .data = &tegra20_hdmi_config },
@@ -1381,28 +1485,20 @@ static int tegra_hdmi_probe(struct platform_device *pdev)
 		return PTR_ERR(hdmi->rst);
 	}
 
-	err = clk_prepare(hdmi->clk);
-	if (err < 0)
-		return err;
-
 	hdmi->clk_parent = devm_clk_get(&pdev->dev, "parent");
 	if (IS_ERR(hdmi->clk_parent))
 		return PTR_ERR(hdmi->clk_parent);
 
-	err = clk_prepare(hdmi->clk_parent);
-	if (err < 0)
-		return err;
-
 	err = clk_set_parent(hdmi->clk, hdmi->clk_parent);
 	if (err < 0) {
 		dev_err(&pdev->dev, "failed to setup clocks: %d\n", err);
 		return err;
 	}
 
-	hdmi->vdd = devm_regulator_get(&pdev->dev, "vdd");
-	if (IS_ERR(hdmi->vdd)) {
-		dev_err(&pdev->dev, "failed to get VDD regulator\n");
-		return PTR_ERR(hdmi->vdd);
+	hdmi->hdmi = devm_regulator_get(&pdev->dev, "hdmi");
+	if (IS_ERR(hdmi->hdmi)) {
+		dev_err(&pdev->dev, "failed to get HDMI regulator\n");
+		return PTR_ERR(hdmi->hdmi);
 	}
 
 	hdmi->pll = devm_regulator_get(&pdev->dev, "pll");
@@ -1411,6 +1507,12 @@ static int tegra_hdmi_probe(struct platform_device *pdev)
 		return PTR_ERR(hdmi->pll);
 	}
 
+	hdmi->vdd = devm_regulator_get(&pdev->dev, "vdd");
+	if (IS_ERR(hdmi->vdd)) {
+		dev_err(&pdev->dev, "failed to get VDD regulator\n");
+		return PTR_ERR(hdmi->vdd);
+	}
+
 	hdmi->output.dev = &pdev->dev;
 
 	err = tegra_output_probe(&hdmi->output);
@@ -1462,8 +1564,8 @@ static int tegra_hdmi_remove(struct platform_device *pdev)
 		return err;
 	}
 
-	clk_unprepare(hdmi->clk_parent);
-	clk_unprepare(hdmi->clk);
+	clk_disable_unprepare(hdmi->clk_parent);
+	clk_disable_unprepare(hdmi->clk);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/tegra/hdmi.h b/drivers/gpu/drm/tegra/hdmi.h
index 0aebc485f7fa..919a19df4e1b 100644
--- a/drivers/gpu/drm/tegra/hdmi.h
+++ b/drivers/gpu/drm/tegra/hdmi.h
@@ -190,6 +190,11 @@
 
 #define HDMI_NV_PDISP_SOR_CSTM					0x5a
 #define SOR_CSTM_ROTCLK(x) (((x) & 0xf) << 24)
+#define SOR_CSTM_PLLDIV (1 << 21)
+#define SOR_CSTM_LVDS_ENABLE (1 << 16)
+#define SOR_CSTM_MODE_LVDS (0 << 12)
+#define SOR_CSTM_MODE_TMDS (1 << 12)
+#define SOR_CSTM_MODE_MASK (3 << 12)
 
 #define HDMI_NV_PDISP_SOR_LVDS					0x5b
 #define HDMI_NV_PDISP_SOR_CRCA					0x5c
diff --git a/drivers/gpu/drm/tegra/rgb.c b/drivers/gpu/drm/tegra/rgb.c
index 0266fb40479e..d6af9be48f42 100644
--- a/drivers/gpu/drm/tegra/rgb.c
+++ b/drivers/gpu/drm/tegra/rgb.c
@@ -159,11 +159,38 @@ static int tegra_output_rgb_disable(struct tegra_output *output)
 }
 
 static int tegra_output_rgb_setup_clock(struct tegra_output *output,
-					struct clk *clk, unsigned long pclk)
+					struct clk *clk, unsigned long pclk,
+					unsigned int *div)
 {
 	struct tegra_rgb *rgb = to_rgb(output);
+	int err;
+
+	err = clk_set_parent(clk, rgb->clk_parent);
+	if (err < 0) {
+		dev_err(output->dev, "failed to set parent: %d\n", err);
+		return err;
+	}
 
-	return clk_set_parent(clk, rgb->clk_parent);
+	/*
+	 * We may not want to change the frequency of the parent clock, since
+	 * it may be a parent for other peripherals. This is due to the fact
+	 * that on Tegra20 there's only a single clock dedicated to display
+	 * (pll_d_out0), whereas later generations have a second one that can
+	 * be used to independently drive a second output (pll_d2_out0).
+	 *
+	 * As a way to support multiple outputs on Tegra20 as well, pll_p is
+	 * typically used as the parent clock for the display controllers.
+	 * But this comes at a cost: pll_p is the parent of several other
+	 * peripherals, so its frequency shouldn't change out of the blue.
+	 *
+	 * The best we can do at this point is to use the shift clock divider
+	 * and hope that the desired frequency can be matched (or at least
+	 * matched sufficiently close that the panel will still work).
+	 */
+
+	*div = ((clk_get_rate(clk) * 2) / pclk) - 2;
+
+	return 0;
 }
 
 static int tegra_output_rgb_check_mode(struct tegra_output *output,
diff --git a/drivers/gpu/drm/tegra/sor.c b/drivers/gpu/drm/tegra/sor.c
index 49ef5729f435..27c979b50111 100644
--- a/drivers/gpu/drm/tegra/sor.c
+++ b/drivers/gpu/drm/tegra/sor.c
@@ -7,6 +7,7 @@
  */
 
 #include <linux/clk.h>
+#include <linux/debugfs.h>
 #include <linux/io.h>
 #include <linux/platform_device.h>
 #include <linux/reset.h>
@@ -33,7 +34,23 @@ struct tegra_sor {
 
 	struct tegra_dpaux *dpaux;
 
+	struct mutex lock;
 	bool enabled;
+
+	struct dentry *debugfs;
+};
+
+struct tegra_sor_config {
+	u32 bits_per_pixel;
+
+	u32 active_polarity;
+	u32 active_count;
+	u32 tu_size;
+	u32 active_frac;
+	u32 watermark;
+
+	u32 hblank_symbols;
+	u32 vblank_symbols;
 };
 
 static inline struct tegra_sor *
@@ -289,34 +306,232 @@ static int tegra_sor_power_up(struct tegra_sor *sor, unsigned long timeout)
 	return -ETIMEDOUT;
 }
 
+struct tegra_sor_params {
+	/* number of link clocks per line */
+	unsigned int num_clocks;
+	/* ratio between input and output */
+	u64 ratio;
+	/* precision factor */
+	u64 precision;
+
+	unsigned int active_polarity;
+	unsigned int active_count;
+	unsigned int active_frac;
+	unsigned int tu_size;
+	unsigned int error;
+};
+
+static int tegra_sor_compute_params(struct tegra_sor *sor,
+				    struct tegra_sor_params *params,
+				    unsigned int tu_size)
+{
+	u64 active_sym, active_count, frac, approx;
+	u32 active_polarity, active_frac = 0;
+	const u64 f = params->precision;
+	s64 error;
+
+	active_sym = params->ratio * tu_size;
+	active_count = div_u64(active_sym, f) * f;
+	frac = active_sym - active_count;
+
+	/* fraction < 0.5 */
+	if (frac >= (f / 2)) {
+		active_polarity = 1;
+		frac = f - frac;
+	} else {
+		active_polarity = 0;
+	}
+
+	if (frac != 0) {
+		frac = div_u64(f * f,  frac); /* 1/fraction */
+		if (frac <= (15 * f)) {
+			active_frac = div_u64(frac, f);
+
+			/* round up */
+			if (active_polarity)
+				active_frac++;
+		} else {
+			active_frac = active_polarity ? 1 : 15;
+		}
+	}
+
+	if (active_frac == 1)
+		active_polarity = 0;
+
+	if (active_polarity == 1) {
+		if (active_frac) {
+			approx = active_count + (active_frac * (f - 1)) * f;
+			approx = div_u64(approx, active_frac * f);
+		} else {
+			approx = active_count + f;
+		}
+	} else {
+		if (active_frac)
+			approx = active_count + div_u64(f, active_frac);
+		else
+			approx = active_count;
+	}
+
+	error = div_s64(active_sym - approx, tu_size);
+	error *= params->num_clocks;
+
+	if (error <= 0 && abs64(error) < params->error) {
+		params->active_count = div_u64(active_count, f);
+		params->active_polarity = active_polarity;
+		params->active_frac = active_frac;
+		params->error = abs64(error);
+		params->tu_size = tu_size;
+
+		if (error == 0)
+			return true;
+	}
+
+	return false;
+}
+
+static int tegra_sor_calc_config(struct tegra_sor *sor,
+				 struct drm_display_mode *mode,
+				 struct tegra_sor_config *config,
+				 struct drm_dp_link *link)
+{
+	const u64 f = 100000, link_rate = link->rate * 1000;
+	const u64 pclk = mode->clock * 1000;
+	u64 input, output, watermark, num;
+	struct tegra_sor_params params;
+	u32 num_syms_per_line;
+	unsigned int i;
+
+	if (!link_rate || !link->num_lanes || !pclk || !config->bits_per_pixel)
+		return -EINVAL;
+
+	output = link_rate * 8 * link->num_lanes;
+	input = pclk * config->bits_per_pixel;
+
+	if (input >= output)
+		return -ERANGE;
+
+	memset(&params, 0, sizeof(params));
+	params.ratio = div64_u64(input * f, output);
+	params.num_clocks = div_u64(link_rate * mode->hdisplay, pclk);
+	params.precision = f;
+	params.error = 64 * f;
+	params.tu_size = 64;
+
+	for (i = params.tu_size; i >= 32; i--)
+		if (tegra_sor_compute_params(sor, &params, i))
+			break;
+
+	if (params.active_frac == 0) {
+		config->active_polarity = 0;
+		config->active_count = params.active_count;
+
+		if (!params.active_polarity)
+			config->active_count--;
+
+		config->tu_size = params.tu_size;
+		config->active_frac = 1;
+	} else {
+		config->active_polarity = params.active_polarity;
+		config->active_count = params.active_count;
+		config->active_frac = params.active_frac;
+		config->tu_size = params.tu_size;
+	}
+
+	dev_dbg(sor->dev,
+		"polarity: %d active count: %d tu size: %d active frac: %d\n",
+		config->active_polarity, config->active_count,
+		config->tu_size, config->active_frac);
+
+	watermark = params.ratio * config->tu_size * (f - params.ratio);
+	watermark = div_u64(watermark, f);
+
+	watermark = div_u64(watermark + params.error, f);
+	config->watermark = watermark + (config->bits_per_pixel / 8) + 2;
+	num_syms_per_line = (mode->hdisplay * config->bits_per_pixel) *
+			    (link->num_lanes * 8);
+
+	if (config->watermark > 30) {
+		config->watermark = 30;
+		dev_err(sor->dev,
+			"unable to compute TU size, forcing watermark to %u\n",
+			config->watermark);
+	} else if (config->watermark > num_syms_per_line) {
+		config->watermark = num_syms_per_line;
+		dev_err(sor->dev, "watermark too high, forcing to %u\n",
+			config->watermark);
+	}
+
+	/* compute the number of symbols per horizontal blanking interval */
+	num = ((mode->htotal - mode->hdisplay) - 7) * link_rate;
+	config->hblank_symbols = div_u64(num, pclk);
+
+	if (link->capabilities & DP_LINK_CAP_ENHANCED_FRAMING)
+		config->hblank_symbols -= 3;
+
+	config->hblank_symbols -= 12 / link->num_lanes;
+
+	/* compute the number of symbols per vertical blanking interval */
+	num = (mode->hdisplay - 25) * link_rate;
+	config->vblank_symbols = div_u64(num, pclk);
+	config->vblank_symbols -= 36 / link->num_lanes + 4;
+
+	dev_dbg(sor->dev, "blank symbols: H:%u V:%u\n", config->hblank_symbols,
+		config->vblank_symbols);
+
+	return 0;
+}
+
 static int tegra_output_sor_enable(struct tegra_output *output)
 {
 	struct tegra_dc *dc = to_tegra_dc(output->encoder.crtc);
 	struct drm_display_mode *mode = &dc->base.mode;
 	unsigned int vbe, vse, hbe, hse, vbs, hbs, i;
 	struct tegra_sor *sor = to_sor(output);
+	struct tegra_sor_config config;
+	struct drm_dp_link link;
+	struct drm_dp_aux *aux;
 	unsigned long value;
-	int err;
+	int err = 0;
+
+	mutex_lock(&sor->lock);
 
 	if (sor->enabled)
-		return 0;
+		goto unlock;
 
 	err = clk_prepare_enable(sor->clk);
 	if (err < 0)
-		return err;
+		goto unlock;
 
 	reset_control_deassert(sor->rst);
 
+	/* FIXME: properly convert to struct drm_dp_aux */
+	aux = (struct drm_dp_aux *)sor->dpaux;
+
 	if (sor->dpaux) {
 		err = tegra_dpaux_enable(sor->dpaux);
 		if (err < 0)
 			dev_err(sor->dev, "failed to enable DP: %d\n", err);
+
+		err = drm_dp_link_probe(aux, &link);
+		if (err < 0) {
+			dev_err(sor->dev, "failed to probe eDP link: %d\n",
+				err);
+			return err;
+		}
 	}
 
 	err = clk_set_parent(sor->clk, sor->clk_safe);
 	if (err < 0)
 		dev_err(sor->dev, "failed to set safe parent clock: %d\n", err);
 
+	memset(&config, 0, sizeof(config));
+	config.bits_per_pixel = 24; /* XXX: don't hardcode? */
+
+	err = tegra_sor_calc_config(sor, mode, &config, &link);
+	if (err < 0)
+		dev_err(sor->dev, "failed to compute link configuration: %d\n",
+			err);
+
 	value = tegra_sor_readl(sor, SOR_CLK_CNTRL);
 	value &= ~SOR_CLK_CNTRL_DP_CLK_SEL_MASK;
 	value |= SOR_CLK_CNTRL_DP_CLK_SEL_SINGLE_DPCLK;
@@ -385,7 +600,7 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 	err = tegra_io_rail_power_on(TEGRA_IO_RAIL_LVDS);
 	if (err < 0) {
 		dev_err(sor->dev, "failed to power on I/O rail: %d\n", err);
-		return err;
+		goto unlock;
 	}
 
 	usleep_range(5, 100);
@@ -419,15 +634,29 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 	if (err < 0)
 		dev_err(sor->dev, "failed to set DP parent clock: %d\n", err);
 
-	/* power dplanes (XXX parameterize based on link?) */
+	/* power DP lanes */
 	value = tegra_sor_readl(sor, SOR_DP_PADCTL_0);
-	value |= SOR_DP_PADCTL_PD_TXD_3 | SOR_DP_PADCTL_PD_TXD_0 |
-		 SOR_DP_PADCTL_PD_TXD_1 | SOR_DP_PADCTL_PD_TXD_2;
+
+	if (link.num_lanes <= 2)
+		value &= ~(SOR_DP_PADCTL_PD_TXD_3 | SOR_DP_PADCTL_PD_TXD_2);
+	else
+		value |= SOR_DP_PADCTL_PD_TXD_3 | SOR_DP_PADCTL_PD_TXD_2;
+
+	if (link.num_lanes <= 1)
+		value &= ~SOR_DP_PADCTL_PD_TXD_1;
+	else
+		value |= SOR_DP_PADCTL_PD_TXD_1;
+
+	if (link.num_lanes == 0)
+		value &= ~SOR_DP_PADCTL_PD_TXD_0;
+	else
+		value |= SOR_DP_PADCTL_PD_TXD_0;
+
 	tegra_sor_writel(sor, value, SOR_DP_PADCTL_0);
 
 	value = tegra_sor_readl(sor, SOR_DP_LINKCTL_0);
 	value &= ~SOR_DP_LINKCTL_LANE_COUNT_MASK;
-	value |= SOR_DP_LINKCTL_LANE_COUNT(4);
+	value |= SOR_DP_LINKCTL_LANE_COUNT(link.num_lanes);
 	tegra_sor_writel(sor, value, SOR_DP_LINKCTL_0);
 
 	/* start lane sequencer */
@@ -443,10 +672,10 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 		usleep_range(250, 1000);
 	}
 
-	/* set link bandwidth (2.7 GHz, XXX: parameterize based on link?) */
+	/* set link bandwidth */
 	value = tegra_sor_readl(sor, SOR_CLK_CNTRL);
 	value &= ~SOR_CLK_CNTRL_DP_LINK_SPEED_MASK;
-	value |= SOR_CLK_CNTRL_DP_LINK_SPEED_G2_70;
+	value |= drm_dp_link_rate_to_bw_code(link.rate) << 2;
 	tegra_sor_writel(sor, value, SOR_CLK_CNTRL);
 
 	/* set linkctl */
@@ -454,7 +683,7 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 	value |= SOR_DP_LINKCTL_ENABLE;
 
 	value &= ~SOR_DP_LINKCTL_TU_SIZE_MASK;
-	value |= SOR_DP_LINKCTL_TU_SIZE(59); /* XXX: don't hardcode? */
+	value |= SOR_DP_LINKCTL_TU_SIZE(config.tu_size);
 
 	value |= SOR_DP_LINKCTL_ENHANCED_FRAME;
 	tegra_sor_writel(sor, value, SOR_DP_LINKCTL_0);
@@ -470,28 +699,31 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 
 	value = tegra_sor_readl(sor, SOR_DP_CONFIG_0);
 	value &= ~SOR_DP_CONFIG_WATERMARK_MASK;
-	value |= SOR_DP_CONFIG_WATERMARK(14); /* XXX: don't hardcode? */
+	value |= SOR_DP_CONFIG_WATERMARK(config.watermark);
 
 	value &= ~SOR_DP_CONFIG_ACTIVE_SYM_COUNT_MASK;
-	value |= SOR_DP_CONFIG_ACTIVE_SYM_COUNT(47); /* XXX: don't hardcode? */
+	value |= SOR_DP_CONFIG_ACTIVE_SYM_COUNT(config.active_count);
 
 	value &= ~SOR_DP_CONFIG_ACTIVE_SYM_FRAC_MASK;
-	value |= SOR_DP_CONFIG_ACTIVE_SYM_FRAC(9); /* XXX: don't hardcode? */
+	value |= SOR_DP_CONFIG_ACTIVE_SYM_FRAC(config.active_frac);
 
-	value &= ~SOR_DP_CONFIG_ACTIVE_SYM_POLARITY; /* XXX: don't hardcode? */
+	if (config.active_polarity)
+		value |= SOR_DP_CONFIG_ACTIVE_SYM_POLARITY;
+	else
+		value &= ~SOR_DP_CONFIG_ACTIVE_SYM_POLARITY;
 
 	value |= SOR_DP_CONFIG_ACTIVE_SYM_ENABLE;
-	value |= SOR_DP_CONFIG_DISPARITY_NEGATIVE; /* XXX: don't hardcode? */
+	value |= SOR_DP_CONFIG_DISPARITY_NEGATIVE;
 	tegra_sor_writel(sor, value, SOR_DP_CONFIG_0);
 
 	value = tegra_sor_readl(sor, SOR_DP_AUDIO_HBLANK_SYMBOLS);
 	value &= ~SOR_DP_AUDIO_HBLANK_SYMBOLS_MASK;
-	value |= 137; /* XXX: don't hardcode? */
+	value |= config.hblank_symbols & 0xffff;
 	tegra_sor_writel(sor, value, SOR_DP_AUDIO_HBLANK_SYMBOLS);
 
 	value = tegra_sor_readl(sor, SOR_DP_AUDIO_VBLANK_SYMBOLS);
 	value &= ~SOR_DP_AUDIO_VBLANK_SYMBOLS_MASK;
-	value |= 2368; /* XXX: don't hardcode? */
+	value |= config.vblank_symbols & 0xffff;
 	tegra_sor_writel(sor, value, SOR_DP_AUDIO_VBLANK_SYMBOLS);
 
 	/* enable pad calibration logic */
@@ -500,30 +732,27 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 	tegra_sor_writel(sor, value, SOR_DP_PADCTL_0);
 
 	if (sor->dpaux) {
-		/* FIXME: properly convert to struct drm_dp_aux */
-		struct drm_dp_aux *aux = (struct drm_dp_aux *)sor->dpaux;
-		struct drm_dp_link link;
 		u8 rate, lanes;
 
 		err = drm_dp_link_probe(aux, &link);
 		if (err < 0) {
 			dev_err(sor->dev, "failed to probe eDP link: %d\n",
 				err);
-			return err;
+			goto unlock;
 		}
 
 		err = drm_dp_link_power_up(aux, &link);
 		if (err < 0) {
 			dev_err(sor->dev, "failed to power up eDP link: %d\n",
 				err);
-			return err;
+			goto unlock;
 		}
 
 		err = drm_dp_link_configure(aux, &link);
 		if (err < 0) {
 			dev_err(sor->dev, "failed to configure eDP link: %d\n",
 				err);
-			return err;
+			goto unlock;
 		}
 
 		rate = drm_dp_link_rate_to_bw_code(link.rate);
@@ -558,7 +787,7 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 		if (err < 0) {
 			dev_err(sor->dev, "DP fast link training failed: %d\n",
 				err);
-			return err;
+			goto unlock;
 		}
 
 		dev_dbg(sor->dev, "fast link training succeeded\n");
@@ -567,7 +796,7 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 	err = tegra_sor_power_up(sor, 250);
 	if (err < 0) {
 		dev_err(sor->dev, "failed to power up SOR: %d\n", err);
-		return err;
+		goto unlock;
 	}
 
 	/* start display controller in continuous mode */
@@ -586,12 +815,26 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 	 * configure panel (24bpp, vsync-, hsync-, DP-A protocol, complete
 	 * raster, associate with display controller)
 	 */
-	value = SOR_STATE_ASY_PIXELDEPTH_BPP_24_444 |
-		SOR_STATE_ASY_VSYNCPOL |
+	value = SOR_STATE_ASY_VSYNCPOL |
 		SOR_STATE_ASY_HSYNCPOL |
 		SOR_STATE_ASY_PROTOCOL_DP_A |
 		SOR_STATE_ASY_CRC_MODE_COMPLETE |
 		SOR_STATE_ASY_OWNER(dc->pipe + 1);
+
+	switch (config.bits_per_pixel) {
+	case 24:
+		value |= SOR_STATE_ASY_PIXELDEPTH_BPP_24_444;
+		break;
+
+	case 18:
+		value |= SOR_STATE_ASY_PIXELDEPTH_BPP_18_444;
+		break;
+
+	default:
+		BUG();
+		break;
+	}
+
 	tegra_sor_writel(sor, value, SOR_STATE_1);
 
 	/*
@@ -620,11 +863,8 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 	value = ((vbs & 0x7fff) << 16) | (hbs & 0x7fff);
 	tegra_sor_writel(sor, value, SOR_HEAD_STATE_4(0));
 
-	/* XXX interlaced mode */
-	tegra_sor_writel(sor, 0x00000001, SOR_HEAD_STATE_5(0));
-
 	/* CSTM (LVDS, link A/B, upper) */
-	value = SOR_CSTM_LVDS | SOR_CSTM_LINK_ACT_B | SOR_CSTM_LINK_ACT_B |
+	value = SOR_CSTM_LVDS | SOR_CSTM_LINK_ACT_A | SOR_CSTM_LINK_ACT_B |
 		SOR_CSTM_UPPER;
 	tegra_sor_writel(sor, value, SOR_CSTM);
 
@@ -632,7 +872,7 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 	err = tegra_sor_setup_pwm(sor, 250);
 	if (err < 0) {
 		dev_err(sor->dev, "failed to setup PWM: %d\n", err);
-		return err;
+		goto unlock;
 	}
 
 	value = tegra_dc_readl(dc, DC_DISP_DISP_WIN_OPTIONS);
@@ -644,18 +884,20 @@ static int tegra_output_sor_enable(struct tegra_output *output)
 	err = tegra_sor_attach(sor);
 	if (err < 0) {
 		dev_err(sor->dev, "failed to attach SOR: %d\n", err);
-		return err;
+		goto unlock;
 	}
 
 	err = tegra_sor_wakeup(sor);
 	if (err < 0) {
 		dev_err(sor->dev, "failed to enable DC: %d\n", err);
-		return err;
+		goto unlock;
 	}
 
 	sor->enabled = true;
 
-	return 0;
+unlock:
+	mutex_unlock(&sor->lock);
+	return err;
 }
 
 static int tegra_sor_detach(struct tegra_sor *sor)
@@ -740,7 +982,7 @@ static int tegra_sor_power_down(struct tegra_sor *sor)
 	tegra_sor_writel(sor, value, SOR_DP_PADCTL_0);
 
 	/* stop lane sequencer */
-	value = SOR_LANE_SEQ_CTL_TRIGGER | SOR_LANE_SEQ_CTL_SEQUENCE_DOWN |
+	value = SOR_LANE_SEQ_CTL_TRIGGER | SOR_LANE_SEQ_CTL_SEQUENCE_UP |
 		SOR_LANE_SEQ_CTL_POWER_STATE_DOWN;
 	tegra_sor_writel(sor, value, SOR_LANE_SEQ_CTL);
 
@@ -783,15 +1025,17 @@ static int tegra_output_sor_disable(struct tegra_output *output)
 	struct tegra_dc *dc = to_tegra_dc(output->encoder.crtc);
 	struct tegra_sor *sor = to_sor(output);
 	unsigned long value;
-	int err;
+	int err = 0;
+
+	mutex_lock(&sor->lock);
 
 	if (!sor->enabled)
-		return 0;
+		goto unlock;
 
 	err = tegra_sor_detach(sor);
 	if (err < 0) {
 		dev_err(sor->dev, "failed to detach SOR: %d\n", err);
-		return err;
+		goto unlock;
 	}
 
 	tegra_sor_writel(sor, 0, SOR_STATE_1);
@@ -832,21 +1076,21 @@ static int tegra_output_sor_disable(struct tegra_output *output)
 	err = tegra_sor_power_down(sor);
 	if (err < 0) {
 		dev_err(sor->dev, "failed to power down SOR: %d\n", err);
-		return err;
+		goto unlock;
 	}
 
 	if (sor->dpaux) {
 		err = tegra_dpaux_disable(sor->dpaux);
 		if (err < 0) {
 			dev_err(sor->dev, "failed to disable DP: %d\n", err);
-			return err;
+			goto unlock;
 		}
 	}
 
 	err = tegra_io_rail_power_off(TEGRA_IO_RAIL_LVDS);
 	if (err < 0) {
 		dev_err(sor->dev, "failed to power off I/O rail: %d\n", err);
-		return err;
+		goto unlock;
 	}
 
 	reset_control_assert(sor->rst);
@@ -854,18 +1098,18 @@ static int tegra_output_sor_disable(struct tegra_output *output)
 
 	sor->enabled = false;
 
-	return 0;
+unlock:
+	mutex_unlock(&sor->lock);
+	return err;
 }
 
 static int tegra_output_sor_setup_clock(struct tegra_output *output,
-					struct clk *clk, unsigned long pclk)
+					struct clk *clk, unsigned long pclk,
+					unsigned int *div)
 {
 	struct tegra_sor *sor = to_sor(output);
 	int err;
 
-	/* round to next MHz */
-	pclk = DIV_ROUND_UP(pclk / 2, 1000000) * 1000000;
-
 	err = clk_set_parent(clk, sor->clk_parent);
 	if (err < 0) {
 		dev_err(sor->dev, "failed to set parent clock: %d\n", err);
@@ -874,11 +1118,12 @@ static int tegra_output_sor_setup_clock(struct tegra_output *output,
 
 	err = clk_set_rate(sor->clk_parent, pclk);
 	if (err < 0) {
-		dev_err(sor->dev, "failed to set base clock rate to %lu Hz\n",
-			pclk * 2);
+		dev_err(sor->dev, "failed to set clock rate to %lu Hz\n", pclk);
 		return err;
 	}
 
+	*div = 0;
+
 	return 0;
 }
 
@@ -914,9 +1159,124 @@ static const struct tegra_output_ops sor_ops = {
 	.detect = tegra_output_sor_detect,
 };
 
+static int tegra_sor_crc_open(struct inode *inode, struct file *file)
+{
+	file->private_data = inode->i_private;
+
+	return 0;
+}
+
+static int tegra_sor_crc_release(struct inode *inode, struct file *file)
+{
+	return 0;
+}
+
+static int tegra_sor_crc_wait(struct tegra_sor *sor, unsigned long timeout)
+{
+	u32 value;
+
+	timeout = jiffies + msecs_to_jiffies(timeout);
+
+	while (time_before(jiffies, timeout)) {
+		value = tegra_sor_readl(sor, SOR_CRC_A);
+		if (value & SOR_CRC_A_VALID)
+			return 0;
+
+		usleep_range(100, 200);
+	}
+
+	return -ETIMEDOUT;
+}
+
+static ssize_t tegra_sor_crc_read(struct file *file, char __user *buffer,
+				  size_t size, loff_t *ppos)
+{
+	struct tegra_sor *sor = file->private_data;
+	ssize_t num, err;
+	char buf[10];
+	u32 value;
+
+	mutex_lock(&sor->lock);
+
+	if (!sor->enabled) {
+		err = -EAGAIN;
+		goto unlock;
+	}
+
+	value = tegra_sor_readl(sor, SOR_STATE_1);
+	value &= ~SOR_STATE_ASY_CRC_MODE_MASK;
+	tegra_sor_writel(sor, value, SOR_STATE_1);
+
+	value = tegra_sor_readl(sor, SOR_CRC_CNTRL);
+	value |= SOR_CRC_CNTRL_ENABLE;
+	tegra_sor_writel(sor, value, SOR_CRC_CNTRL);
+
+	value = tegra_sor_readl(sor, SOR_TEST);
+	value &= ~SOR_TEST_CRC_POST_SERIALIZE;
+	tegra_sor_writel(sor, value, SOR_TEST);
+
+	err = tegra_sor_crc_wait(sor, 100);
+	if (err < 0)
+		goto unlock;
+
+	tegra_sor_writel(sor, SOR_CRC_A_RESET, SOR_CRC_A);
+	value = tegra_sor_readl(sor, SOR_CRC_B);
+
+	num = scnprintf(buf, sizeof(buf), "%08x\n", value);
+
+	err = simple_read_from_buffer(buffer, size, ppos, buf, num);
+
+unlock:
+	mutex_unlock(&sor->lock);
+	return err;
+}
+
+static const struct file_operations tegra_sor_crc_fops = {
+	.owner = THIS_MODULE,
+	.open = tegra_sor_crc_open,
+	.read = tegra_sor_crc_read,
+	.release = tegra_sor_crc_release,
+};
+
+static int tegra_sor_debugfs_init(struct tegra_sor *sor,
+				  struct drm_minor *minor)
+{
+	struct dentry *entry;
+	int err = 0;
+
+	sor->debugfs = debugfs_create_dir("sor", minor->debugfs_root);
+	if (!sor->debugfs)
+		return -ENOMEM;
+
+	entry = debugfs_create_file("crc", 0644, sor->debugfs, sor,
+				    &tegra_sor_crc_fops);
+	if (!entry) {
+		dev_err(sor->dev,
+			"cannot create /sys/kernel/debug/dri/%s/sor/crc\n",
+			minor->debugfs_root->d_name.name);
+		err = -ENOMEM;
+		goto remove;
+	}
+
+	return err;
+
+remove:
+	debugfs_remove(sor->debugfs);
+	sor->debugfs = NULL;
+	return err;
+}
+
+static int tegra_sor_debugfs_exit(struct tegra_sor *sor)
+{
+	debugfs_remove_recursive(sor->debugfs);
+	sor->debugfs = NULL;
+
+	return 0;
+}
+
 static int tegra_sor_init(struct host1x_client *client)
 {
-	struct tegra_drm *tegra = dev_get_drvdata(client->parent);
+	struct drm_device *drm = dev_get_drvdata(client->parent);
 	struct tegra_sor *sor = host1x_client_to_sor(client);
 	int err;
 
@@ -928,12 +1288,18 @@ static int tegra_sor_init(struct host1x_client *client)
 	sor->output.dev = sor->dev;
 	sor->output.ops = &sor_ops;
 
-	err = tegra_output_init(tegra->drm, &sor->output);
+	err = tegra_output_init(drm, &sor->output);
 	if (err < 0) {
 		dev_err(sor->dev, "output setup failed: %d\n", err);
 		return err;
 	}
 
+	if (IS_ENABLED(CONFIG_DEBUG_FS)) {
+		err = tegra_sor_debugfs_init(sor, drm->primary);
+		if (err < 0)
+			dev_err(sor->dev, "debugfs setup failed: %d\n", err);
+	}
+
 	if (sor->dpaux) {
 		err = tegra_dpaux_attach(sor->dpaux, &sor->output);
 		if (err < 0) {
@@ -964,6 +1330,12 @@ static int tegra_sor_exit(struct host1x_client *client)
 		}
 	}
 
+	if (IS_ENABLED(CONFIG_DEBUG_FS)) {
+		err = tegra_sor_debugfs_exit(sor);
+		if (err < 0)
+			dev_err(sor->dev, "debugfs cleanup failed: %d\n", err);
+	}
+
 	err = tegra_output_exit(&sor->output);
 	if (err < 0) {
 		dev_err(sor->dev, "output cleanup failed: %d\n", err);
@@ -1045,6 +1417,8 @@ static int tegra_sor_probe(struct platform_device *pdev)
 	sor->client.ops = &sor_client_ops;
 	sor->client.dev = &pdev->dev;
 
+	mutex_init(&sor->lock);
+
 	err = host1x_client_register(&sor->client);
 	if (err < 0) {
 		dev_err(&pdev->dev, "failed to register host1x client: %d\n",
diff --git a/drivers/gpu/drm/tegra/sor.h b/drivers/gpu/drm/tegra/sor.h
index f4156d54cd05..a5f8853fedb5 100644
--- a/drivers/gpu/drm/tegra/sor.h
+++ b/drivers/gpu/drm/tegra/sor.h
@@ -47,6 +47,7 @@
 #define SOR_HEAD_STATE_4(x) (0x0d + (x))
 #define SOR_HEAD_STATE_5(x) (0x0f + (x))
 #define SOR_CRC_CNTRL 0x11
+#define  SOR_CRC_CNTRL_ENABLE			(1 << 0)
 #define SOR_DP_DEBUG_MVID 0x12
 
 #define SOR_CLK_CNTRL 0x13
@@ -69,6 +70,7 @@
 #define  SOR_PWR_NORMAL_STATE_PU		(1 << 0)
 
 #define SOR_TEST 0x16
+#define  SOR_TEST_CRC_POST_SERIALIZE		(1 << 23)
 #define  SOR_TEST_ATTACHED			(1 << 10)
 #define  SOR_TEST_HEAD_MODE_MASK		(3 << 8)
 #define  SOR_TEST_HEAD_MODE_AWAKE		(2 << 8)
@@ -115,6 +117,8 @@
 
 #define SOR_LVDS 0x1c
 #define SOR_CRC_A 0x1d
+#define  SOR_CRC_A_VALID			(1 << 0)
+#define  SOR_CRC_A_RESET			(1 << 0)
 #define SOR_CRC_B 0x1e
 #define SOR_BLANK 0x1f
 #define SOR_SEQ_CTL 0x20
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_drv.c b/drivers/gpu/drm/tilcdc/tilcdc_drv.c
index 171a8203892c..b20b69488dc9 100644
--- a/drivers/gpu/drm/tilcdc/tilcdc_drv.c
+++ b/drivers/gpu/drm/tilcdc/tilcdc_drv.c
@@ -268,7 +268,7 @@ static int tilcdc_load(struct drm_device *dev, unsigned long flags)
 	}
 
 	pm_runtime_get_sync(dev->dev);
-	ret = drm_irq_install(dev);
+	ret = drm_irq_install(dev, platform_get_irq(dev->platformdev, 0));
 	pm_runtime_put_sync(dev->dev);
 	if (ret < 0) {
 		dev_err(dev->dev, "failed to install IRQ handler\n");
diff --git a/drivers/gpu/drm/udl/udl_main.c b/drivers/gpu/drm/udl/udl_main.c
index afdf383f630a..7094b92d1ec7 100644
--- a/drivers/gpu/drm/udl/udl_main.c
+++ b/drivers/gpu/drm/udl/udl_main.c
@@ -294,6 +294,7 @@ int udl_driver_load(struct drm_device *dev, unsigned long flags)
 	dev->dev_private = udl;
 
 	if (!udl_parse_vendor_descriptor(dev, dev->usbdev)) {
+		ret = -ENODEV;
 		DRM_ERROR("firmware not recognized. Assume incompatible device\n");
 		goto err;
 	}
diff --git a/drivers/gpu/drm/via/via_dma.c b/drivers/gpu/drm/via/via_dma.c
index a18479c6b6da..6fc0648dd37f 100644
--- a/drivers/gpu/drm/via/via_dma.c
+++ b/drivers/gpu/drm/via/via_dma.c
@@ -737,4 +737,4 @@ const struct drm_ioctl_desc via_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(VIA_BLIT_SYNC, via_dma_blit_sync, DRM_AUTH)
 };
 
-int via_max_ioctl = DRM_ARRAY_SIZE(via_ioctls);
+int via_max_ioctl = ARRAY_SIZE(via_ioctls);
diff --git a/drivers/gpu/drm/via/via_mm.c b/drivers/gpu/drm/via/via_mm.c
index 927889105483..d70b1e1544bf 100644
--- a/drivers/gpu/drm/via/via_mm.c
+++ b/drivers/gpu/drm/via/via_mm.c
@@ -79,7 +79,7 @@ int via_final_context(struct drm_device *dev, int context)
 
 	/* Linux specific until context tracking code gets ported to BSD */
 	/* Last context, perform cleanup */
-	if (list_is_singular(&dev->ctxlist) && dev->dev_private) {
+	if (list_is_singular(&dev->ctxlist)) {
 		DRM_DEBUG("Last Context\n");
 		drm_irq_uninstall(dev);
 		via_cleanup_futex(dev_priv);
diff --git a/drivers/gpu/drm/vmwgfx/Kconfig b/drivers/gpu/drm/vmwgfx/Kconfig
index b71bcd0bfbbf..67720f70fe29 100644
--- a/drivers/gpu/drm/vmwgfx/Kconfig
+++ b/drivers/gpu/drm/vmwgfx/Kconfig
@@ -1,11 +1,14 @@
 config DRM_VMWGFX
 	tristate "DRM driver for VMware Virtual GPU"
-	depends on DRM && PCI && FB
+	depends on DRM && PCI
 	select FB_DEFERRED_IO
 	select FB_CFB_FILLRECT
 	select FB_CFB_COPYAREA
 	select FB_CFB_IMAGEBLIT
 	select DRM_TTM
+	# Only needed for the transitional use of drm_crtc_init - can be removed
+	# again once vmwgfx sets up the primary plane itself.
+	select DRM_KMS_HELPER
 	help
 	  Choose this option if you would like to run 3D acceleration
 	  in a VMware virtual machine.
@@ -14,7 +17,7 @@ config DRM_VMWGFX
 	  The compiled module will be called "vmwgfx.ko".
 
 config DRM_VMWGFX_FBCON
-	depends on DRM_VMWGFX
+	depends on DRM_VMWGFX && FB
 	bool "Enable framebuffer console under vmwgfx by default"
 	help
 	   Choose this option if you are shipping a new vmwgfx
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index 4a223bbea3b3..246a62bab378 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -806,7 +806,7 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
 	}
 
 	if (dev_priv->capabilities & SVGA_CAP_IRQMASK) {
-		ret = drm_irq_install(dev);
+		ret = drm_irq_install(dev, dev->pdev->irq);
 		if (ret != 0) {
 			DRM_ERROR("Failed installing irq: %d\n", ret);
 			goto out_no_irq;
@@ -1417,7 +1417,7 @@ static struct drm_driver driver = {
 	.enable_vblank = vmw_enable_vblank,
 	.disable_vblank = vmw_disable_vblank,
 	.ioctls = vmw_ioctls,
-	.num_ioctls = DRM_ARRAY_SIZE(vmw_ioctls),
+	.num_ioctls = ARRAY_SIZE(vmw_ioctls),
 	.master_create = vmw_master_create,
 	.master_destroy = vmw_master_destroy,
 	.master_set = vmw_master_set,
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
index a2dde5ad8138..8f3edc4710f2 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
@@ -187,7 +187,7 @@ int vmw_du_crtc_cursor_set(struct drm_crtc *crtc, struct drm_file *file_priv,
 	 * can do this since the caller in the drm core doesn't check anything
 	 * which is protected by any looks.
 	 */
-	mutex_unlock(&crtc->mutex);
+	drm_modeset_unlock(&crtc->mutex);
 	drm_modeset_lock_all(dev_priv->dev);
 
 	/* A lot of the code assumes this */
@@ -252,7 +252,7 @@ int vmw_du_crtc_cursor_set(struct drm_crtc *crtc, struct drm_file *file_priv,
 	ret = 0;
 out:
 	drm_modeset_unlock_all(dev_priv->dev);
-	mutex_lock(&crtc->mutex);
+	drm_modeset_lock(&crtc->mutex, NULL);
 
 	return ret;
 }
@@ -273,7 +273,7 @@ int vmw_du_crtc_cursor_move(struct drm_crtc *crtc, int x, int y)
 	 * can do this since the caller in the drm core doesn't check anything
 	 * which is protected by any looks.
 	 */
-	mutex_unlock(&crtc->mutex);
+	drm_modeset_unlock(&crtc->mutex);
 	drm_modeset_lock_all(dev_priv->dev);
 
 	vmw_cursor_update_position(dev_priv, shown,
@@ -281,7 +281,7 @@ int vmw_du_crtc_cursor_move(struct drm_crtc *crtc, int x, int y)
 				   du->cursor_y + du->hotspot_y);
 
 	drm_modeset_unlock_all(dev_priv->dev);
-	mutex_lock(&crtc->mutex);
+	drm_modeset_lock(&crtc->mutex, NULL);
 
 	return 0;
 }
@@ -2001,7 +2001,7 @@ int vmw_du_connector_fill_modes(struct drm_connector *connector,
 	if (du->pref_mode)
 		list_move(&du->pref_mode->head, &connector->probed_modes);
 
-	drm_mode_connector_list_update(connector);
+	drm_mode_connector_list_update(connector, true);
 
 	return 1;
 }
diff --git a/drivers/gpu/host1x/bus.c b/drivers/gpu/host1x/bus.c
index ccdd2e6da5e3..aaf54859adb0 100644
--- a/drivers/gpu/host1x/bus.c
+++ b/drivers/gpu/host1x/bus.c
@@ -216,8 +216,8 @@ int host1x_device_exit(struct host1x_device *device)
 }
 EXPORT_SYMBOL(host1x_device_exit);
 
-static int host1x_register_client(struct host1x *host1x,
-				  struct host1x_client *client)
+static int host1x_add_client(struct host1x *host1x,
+			     struct host1x_client *client)
 {
 	struct host1x_device *device;
 	struct host1x_subdev *subdev;
@@ -238,8 +238,8 @@ static int host1x_register_client(struct host1x *host1x,
 	return -ENODEV;
 }
 
-static int host1x_unregister_client(struct host1x *host1x,
-				    struct host1x_client *client)
+static int host1x_del_client(struct host1x *host1x,
+			     struct host1x_client *client)
 {
 	struct host1x_device *device, *dt;
 	struct host1x_subdev *subdev;
@@ -503,7 +503,7 @@ int host1x_client_register(struct host1x_client *client)
 	mutex_lock(&devices_lock);
 
 	list_for_each_entry(host1x, &devices, list) {
-		err = host1x_register_client(host1x, client);
+		err = host1x_add_client(host1x, client);
 		if (!err) {
 			mutex_unlock(&devices_lock);
 			return 0;
@@ -529,7 +529,7 @@ int host1x_client_unregister(struct host1x_client *client)
 	mutex_lock(&devices_lock);
 
 	list_for_each_entry(host1x, &devices, list) {
-		err = host1x_unregister_client(host1x, client);
+		err = host1x_del_client(host1x, client);
 		if (!err) {
 			mutex_unlock(&devices_lock);
 			return 0;
diff --git a/drivers/gpu/ipu-v3/Kconfig b/drivers/gpu/ipu-v3/Kconfig
new file mode 100644
index 000000000000..2f228a2f2a48
--- /dev/null
+++ b/drivers/gpu/ipu-v3/Kconfig
@@ -0,0 +1,7 @@
+config IMX_IPUV3_CORE
+	tristate "IPUv3 core support"
+	depends on SOC_IMX5 || SOC_IMX6Q || SOC_IMX6SL || ARCH_MULTIPLATFORM
+	depends on RESET_CONTROLLER
+	help
+	  Choose this if you have a i.MX5/6 system and want to use the Image
+	  Processing Unit. This option only enables IPU base support.
diff --git a/drivers/staging/imx-drm/ipu-v3/Makefile b/drivers/gpu/ipu-v3/Makefile
index 28ed72e98a96..1887972b4ac2 100644
--- a/drivers/staging/imx-drm/ipu-v3/Makefile
+++ b/drivers/gpu/ipu-v3/Makefile
@@ -1,3 +1,3 @@
-obj-$(CONFIG_DRM_IMX_IPUV3_CORE) += imx-ipu-v3.o
+obj-$(CONFIG_IMX_IPUV3_CORE) += imx-ipu-v3.o
 
-imx-ipu-v3-objs := ipu-common.o ipu-dc.o ipu-di.o ipu-dp.o ipu-dmfc.o
+imx-ipu-v3-objs := ipu-common.o ipu-dc.o ipu-di.o ipu-dp.o ipu-dmfc.o ipu-smfc.o
diff --git a/drivers/staging/imx-drm/ipu-v3/ipu-common.c b/drivers/gpu/ipu-v3/ipu-common.c
index a1f7b2001c8a..04e7b2eafbdd 100644
--- a/drivers/staging/imx-drm/ipu-v3/ipu-common.c
+++ b/drivers/gpu/ipu-v3/ipu-common.c
@@ -31,7 +31,7 @@
 
 #include <drm/drm_fourcc.h>
 
-#include "imx-ipu-v3.h"
+#include <video/imx-ipu-v3.h>
 #include "ipu-prv.h"
 
 static inline u32 ipu_cm_read(struct ipu_soc *ipu, unsigned offset)
@@ -661,6 +661,39 @@ int ipu_module_disable(struct ipu_soc *ipu, u32 mask)
 }
 EXPORT_SYMBOL_GPL(ipu_module_disable);
 
+int ipu_csi_enable(struct ipu_soc *ipu, int csi)
+{
+	return ipu_module_enable(ipu, csi ? IPU_CONF_CSI1_EN : IPU_CONF_CSI0_EN);
+}
+EXPORT_SYMBOL_GPL(ipu_csi_enable);
+
+int ipu_csi_disable(struct ipu_soc *ipu, int csi)
+{
+	return ipu_module_disable(ipu, csi ? IPU_CONF_CSI1_EN : IPU_CONF_CSI0_EN);
+}
+EXPORT_SYMBOL_GPL(ipu_csi_disable);
+
+int ipu_smfc_enable(struct ipu_soc *ipu)
+{
+	return ipu_module_enable(ipu, IPU_CONF_SMFC_EN);
+}
+EXPORT_SYMBOL_GPL(ipu_smfc_enable);
+
+int ipu_smfc_disable(struct ipu_soc *ipu)
+{
+	return ipu_module_disable(ipu, IPU_CONF_SMFC_EN);
+}
+EXPORT_SYMBOL_GPL(ipu_smfc_disable);
+
+int ipu_idmac_get_current_buffer(struct ipuv3_channel *channel)
+{
+	struct ipu_soc *ipu = channel->ipu;
+	unsigned int chno = channel->num;
+
+	return (ipu_cm_read(ipu, IPU_CHA_CUR_BUF(chno)) & idma_mask(chno)) ? 1 : 0;
+}
+EXPORT_SYMBOL_GPL(ipu_idmac_get_current_buffer);
+
 void ipu_idmac_select_buffer(struct ipuv3_channel *channel, u32 buf_num)
 {
 	struct ipu_soc *ipu = channel->ipu;
@@ -896,8 +929,17 @@ static int ipu_submodules_init(struct ipu_soc *ipu,
 		goto err_dp;
 	}
 
+	ret = ipu_smfc_init(ipu, dev, ipu_base +
+			devtype->cm_ofs + IPU_CM_SMFC_REG_OFS);
+	if (ret) {
+		unit = "smfc";
+		goto err_smfc;
+	}
+
 	return 0;
 
+err_smfc:
+	ipu_dp_exit(ipu);
 err_dp:
 	ipu_dmfc_exit(ipu);
 err_dmfc:
@@ -977,6 +1019,7 @@ EXPORT_SYMBOL_GPL(ipu_idmac_channel_irq);
 
 static void ipu_submodules_exit(struct ipu_soc *ipu)
 {
+	ipu_smfc_exit(ipu);
 	ipu_dp_exit(ipu);
 	ipu_dmfc_exit(ipu);
 	ipu_dc_exit(ipu);
@@ -1001,6 +1044,7 @@ static void platform_device_unregister_children(struct platform_device *pdev)
 struct ipu_platform_reg {
 	struct ipu_client_platformdata pdata;
 	const char *name;
+	int reg_offset;
 };
 
 static const struct ipu_platform_reg client_reg[] = {
@@ -1022,13 +1066,29 @@ static const struct ipu_platform_reg client_reg[] = {
 			.dma[1] = -EINVAL,
 		},
 		.name = "imx-ipuv3-crtc",
+	}, {
+		.pdata = {
+			.csi = 0,
+			.dma[0] = IPUV3_CHANNEL_CSI0,
+			.dma[1] = -EINVAL,
+		},
+		.reg_offset = IPU_CM_CSI0_REG_OFS,
+		.name = "imx-ipuv3-camera",
+	}, {
+		.pdata = {
+			.csi = 1,
+			.dma[0] = IPUV3_CHANNEL_CSI1,
+			.dma[1] = -EINVAL,
+		},
+		.reg_offset = IPU_CM_CSI1_REG_OFS,
+		.name = "imx-ipuv3-camera",
 	},
 };
 
 static DEFINE_MUTEX(ipu_client_id_mutex);
 static int ipu_client_id;
 
-static int ipu_add_client_devices(struct ipu_soc *ipu)
+static int ipu_add_client_devices(struct ipu_soc *ipu, unsigned long ipu_base)
 {
 	struct device *dev = ipu->dev;
 	unsigned i;
@@ -1042,9 +1102,19 @@ static int ipu_add_client_devices(struct ipu_soc *ipu)
 	for (i = 0; i < ARRAY_SIZE(client_reg); i++) {
 		const struct ipu_platform_reg *reg = &client_reg[i];
 		struct platform_device *pdev;
-
-		pdev = platform_device_register_data(dev, reg->name,
-			id++, &reg->pdata, sizeof(reg->pdata));
+		struct resource res;
+
+		if (reg->reg_offset) {
+			memset(&res, 0, sizeof(res));
+			res.flags = IORESOURCE_MEM;
+			res.start = ipu_base + ipu->devtype->cm_ofs + reg->reg_offset;
+			res.end = res.start + PAGE_SIZE - 1;
+			pdev = platform_device_register_resndata(dev, reg->name,
+				id++, &res, 1, &reg->pdata, sizeof(reg->pdata));
+		} else {
+			pdev = platform_device_register_data(dev, reg->name,
+				id++, &reg->pdata, sizeof(reg->pdata));
+		}
 
 		if (IS_ERR(pdev))
 			goto err_register;
@@ -1241,7 +1311,7 @@ static int ipu_probe(struct platform_device *pdev)
 	if (ret)
 		goto failed_submodules_init;
 
-	ret = ipu_add_client_devices(ipu);
+	ret = ipu_add_client_devices(ipu, ipu_base);
 	if (ret) {
 		dev_err(&pdev->dev, "adding client devices failed with %d\n",
 				ret);
diff --git a/drivers/staging/imx-drm/ipu-v3/ipu-dc.c b/drivers/gpu/ipu-v3/ipu-dc.c
index 784a4a13eac3..2326c752d89b 100644
--- a/drivers/staging/imx-drm/ipu-v3/ipu-dc.c
+++ b/drivers/gpu/ipu-v3/ipu-dc.c
@@ -21,8 +21,7 @@
 #include <linux/interrupt.h>
 #include <linux/io.h>
 
-#include "../imx-drm.h"
-#include "imx-ipu-v3.h"
+#include <video/imx-ipu-v3.h>
 #include "ipu-prv.h"
 
 #define DC_MAP_CONF_PTR(n)	(0x108 + ((n) & ~0x1) * 2)
diff --git a/drivers/staging/imx-drm/ipu-v3/ipu-di.c b/drivers/gpu/ipu-v3/ipu-di.c
index 849b3e120ef0..c490ba4384fc 100644
--- a/drivers/staging/imx-drm/ipu-v3/ipu-di.c
+++ b/drivers/gpu/ipu-v3/ipu-di.c
@@ -20,7 +20,7 @@
 #include <linux/err.h>
 #include <linux/platform_device.h>
 
-#include "imx-ipu-v3.h"
+#include <video/imx-ipu-v3.h>
 #include "ipu-prv.h"
 
 struct ipu_di {
diff --git a/drivers/staging/imx-drm/ipu-v3/ipu-dmfc.c b/drivers/gpu/ipu-v3/ipu-dmfc.c
index 59f182b28fc1..042c3958e2a0 100644
--- a/drivers/staging/imx-drm/ipu-v3/ipu-dmfc.c
+++ b/drivers/gpu/ipu-v3/ipu-dmfc.c
@@ -17,7 +17,7 @@
 #include <linux/errno.h>
 #include <linux/io.h>
 
-#include "imx-ipu-v3.h"
+#include <video/imx-ipu-v3.h>
 #include "ipu-prv.h"
 
 #define DMFC_RD_CHAN		0x0000
diff --git a/drivers/staging/imx-drm/ipu-v3/ipu-dp.c b/drivers/gpu/ipu-v3/ipu-dp.c
index d90f82a87d19..98686edbcdbb 100644
--- a/drivers/staging/imx-drm/ipu-v3/ipu-dp.c
+++ b/drivers/gpu/ipu-v3/ipu-dp.c
@@ -19,7 +19,7 @@
 #include <linux/io.h>
 #include <linux/err.h>
 
-#include "imx-ipu-v3.h"
+#include <video/imx-ipu-v3.h>
 #include "ipu-prv.h"
 
 #define DP_SYNC 0
diff --git a/drivers/staging/imx-drm/ipu-v3/ipu-prv.h b/drivers/gpu/ipu-v3/ipu-prv.h
index bfc1b3366488..c93f50ec04f7 100644
--- a/drivers/staging/imx-drm/ipu-v3/ipu-prv.h
+++ b/drivers/gpu/ipu-v3/ipu-prv.h
@@ -22,7 +22,7 @@ struct ipu_soc;
 #include <linux/clk.h>
 #include <linux/platform_device.h>
 
-#include "imx-ipu-v3.h"
+#include <video/imx-ipu-v3.h>
 
 #define IPUV3_CHANNEL_CSI0			 0
 #define IPUV3_CHANNEL_CSI1			 1
@@ -151,6 +151,8 @@ struct ipuv3_channel {
 struct ipu_dc_priv;
 struct ipu_dmfc_priv;
 struct ipu_di;
+struct ipu_smfc_priv;
+
 struct ipu_devtype;
 
 struct ipu_soc {
@@ -178,6 +180,7 @@ struct ipu_soc {
 	struct ipu_dp_priv	*dp_priv;
 	struct ipu_dmfc_priv	*dmfc_priv;
 	struct ipu_di		*di_priv[2];
+	struct ipu_smfc_priv	*smfc_priv;
 };
 
 void ipu_srm_dp_sync_update(struct ipu_soc *ipu);
@@ -206,4 +209,7 @@ void ipu_dc_exit(struct ipu_soc *ipu);
 int ipu_cpmem_init(struct ipu_soc *ipu, struct device *dev, unsigned long base);
 void ipu_cpmem_exit(struct ipu_soc *ipu);
 
+int ipu_smfc_init(struct ipu_soc *ipu, struct device *dev, unsigned long base);
+void ipu_smfc_exit(struct ipu_soc *ipu);
+
 #endif				/* __IPU_PRV_H__ */
diff --git a/drivers/gpu/ipu-v3/ipu-smfc.c b/drivers/gpu/ipu-v3/ipu-smfc.c
new file mode 100644
index 000000000000..e4f85ad286fc
--- /dev/null
+++ b/drivers/gpu/ipu-v3/ipu-smfc.c
@@ -0,0 +1,97 @@
+/*
+ * Copyright 2008-2010 Freescale Semiconductor, Inc. All Rights Reserved.
+ *
+ * The code contained herein is licensed under the GNU General Public
+ * License. You may obtain a copy of the GNU General Public License
+ * Version 2 or later at the following locations:
+ *
+ * http://www.opensource.org/licenses/gpl-license.html
+ * http://www.gnu.org/copyleft/gpl.html
+ */
+#define DEBUG
+#include <linux/export.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/io.h>
+#include <linux/errno.h>
+#include <linux/spinlock.h>
+#include <linux/delay.h>
+#include <linux/clk.h>
+#include <video/imx-ipu-v3.h>
+
+#include "ipu-prv.h"
+
+struct ipu_smfc_priv {
+	void __iomem *base;
+	spinlock_t lock;
+};
+
+/*SMFC Registers */
+#define SMFC_MAP	0x0000
+#define SMFC_WMC	0x0004
+#define SMFC_BS		0x0008
+
+int ipu_smfc_set_burstsize(struct ipu_soc *ipu, int channel, int burstsize)
+{
+	struct ipu_smfc_priv *smfc = ipu->smfc_priv;
+	unsigned long flags;
+	u32 val, shift;
+
+	spin_lock_irqsave(&smfc->lock, flags);
+
+	shift = channel * 4;
+	val = readl(smfc->base + SMFC_BS);
+	val &= ~(0xf << shift);
+	val |= burstsize << shift;
+	writel(val, smfc->base + SMFC_BS);
+
+	spin_unlock_irqrestore(&smfc->lock, flags);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ipu_smfc_set_burstsize);
+
+int ipu_smfc_map_channel(struct ipu_soc *ipu, int channel, int csi_id, int mipi_id)
+{
+	struct ipu_smfc_priv *smfc = ipu->smfc_priv;
+	unsigned long flags;
+	u32 val, shift;
+
+	spin_lock_irqsave(&smfc->lock, flags);
+
+	shift = channel * 3;
+	val = readl(smfc->base + SMFC_MAP);
+	val &= ~(0x7 << shift);
+	val |= ((csi_id << 2) | mipi_id) << shift;
+	writel(val, smfc->base + SMFC_MAP);
+
+	spin_unlock_irqrestore(&smfc->lock, flags);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ipu_smfc_map_channel);
+
+int ipu_smfc_init(struct ipu_soc *ipu, struct device *dev,
+		  unsigned long base)
+{
+	struct ipu_smfc_priv *smfc;
+
+	smfc = devm_kzalloc(dev, sizeof(*smfc), GFP_KERNEL);
+	if (!smfc)
+		return -ENOMEM;
+
+	ipu->smfc_priv = smfc;
+	spin_lock_init(&smfc->lock);
+
+	smfc->base = devm_ioremap(dev, base, PAGE_SIZE);
+	if (!smfc->base)
+		return -ENOMEM;
+
+	pr_debug("%s: ioremap 0x%08lx -> %p\n", __func__, base, smfc->base);
+
+	return 0;
+}
+
+void ipu_smfc_exit(struct ipu_soc *ipu)
+{
+}
diff --git a/drivers/gpu/vga/vga_switcheroo.c b/drivers/gpu/vga/vga_switcheroo.c
index ec0ae2d1686a..6866448083b2 100644
--- a/drivers/gpu/vga/vga_switcheroo.c
+++ b/drivers/gpu/vga/vga_switcheroo.c
@@ -623,7 +623,8 @@ static int vga_switcheroo_runtime_suspend(struct device *dev)
 	ret = dev->bus->pm->runtime_suspend(dev);
 	if (ret)
 		return ret;
-
+	if (vgasr_priv.handler->switchto)
+		vgasr_priv.handler->switchto(VGA_SWITCHEROO_IGD);
 	vga_switcheroo_power_switch(pdev, VGA_SWITCHEROO_OFF);
 	return 0;
 }
diff --git a/drivers/staging/imx-drm/Kconfig b/drivers/staging/imx-drm/Kconfig
index c6e8ba7b3e4e..82fb758a29bc 100644
--- a/drivers/staging/imx-drm/Kconfig
+++ b/drivers/staging/imx-drm/Kconfig
@@ -39,19 +39,10 @@ config DRM_IMX_LDB
 	  Choose this to enable the internal LVDS Display Bridge (LDB)
 	  found on i.MX53 and i.MX6 processors.
 
-config DRM_IMX_IPUV3_CORE
-	tristate "IPUv3 core support"
-	depends on DRM_IMX
-	depends on RESET_CONTROLLER
-	help
-	  Choose this if you have a i.MX5/6 system and want
-	  to use the IPU. This option only enables IPU base
-	  support.
-
 config DRM_IMX_IPUV3
 	tristate "DRM Support for i.MX IPUv3"
 	depends on DRM_IMX
-	depends on DRM_IMX_IPUV3_CORE
+	depends on IMX_IPUV3_CORE
 	help
 	  Choose this if you have a i.MX5 or i.MX6 processor.
 
diff --git a/drivers/staging/imx-drm/Makefile b/drivers/staging/imx-drm/Makefile
index 129e3a3f59f1..582c438d8cbd 100644
--- a/drivers/staging/imx-drm/Makefile
+++ b/drivers/staging/imx-drm/Makefile
@@ -6,7 +6,6 @@ obj-$(CONFIG_DRM_IMX) += imxdrm.o
 obj-$(CONFIG_DRM_IMX_PARALLEL_DISPLAY) += parallel-display.o
 obj-$(CONFIG_DRM_IMX_TVE) += imx-tve.o
 obj-$(CONFIG_DRM_IMX_LDB) += imx-ldb.o
-obj-$(CONFIG_DRM_IMX_IPUV3_CORE) += ipu-v3/
 
 imx-ipuv3-crtc-objs  := ipuv3-crtc.o ipuv3-plane.o
 obj-$(CONFIG_DRM_IMX_IPUV3)	+= imx-ipuv3-crtc.o
diff --git a/drivers/staging/imx-drm/imx-drm-core.c b/drivers/staging/imx-drm/imx-drm-core.c
index c270c9ae6d27..def8280d7ee6 100644
--- a/drivers/staging/imx-drm/imx-drm-core.c
+++ b/drivers/staging/imx-drm/imx-drm-core.c
@@ -200,13 +200,6 @@ static const struct file_operations imx_drm_driver_fops = {
 	.llseek = noop_llseek,
 };
 
-int imx_drm_connector_mode_valid(struct drm_connector *connector,
-	struct drm_display_mode *mode)
-{
-	return MODE_OK;
-}
-EXPORT_SYMBOL(imx_drm_connector_mode_valid);
-
 void imx_drm_connector_destroy(struct drm_connector *connector)
 {
 	drm_sysfs_connector_remove(connector);
@@ -305,7 +298,7 @@ static int imx_drm_driver_load(struct drm_device *drm, unsigned long flags)
 			dev_err(drm->dev,
 				"[CONNECTOR:%d:%s] drm_sysfs_connector_add failed: %d\n",
 				connector->base.id,
-				drm_get_connector_name(connector), ret);
+				connector->name, ret);
 			goto err_unbind;
 		}
 	}
diff --git a/drivers/staging/imx-drm/imx-drm.h b/drivers/staging/imx-drm/imx-drm.h
index a322bac55414..7453ae00c412 100644
--- a/drivers/staging/imx-drm/imx-drm.h
+++ b/drivers/staging/imx-drm/imx-drm.h
@@ -50,8 +50,6 @@ int imx_drm_encoder_get_mux_id(struct device_node *node,
 int imx_drm_encoder_parse_of(struct drm_device *drm,
 	struct drm_encoder *encoder, struct device_node *np);
 
-int imx_drm_connector_mode_valid(struct drm_connector *connector,
-	struct drm_display_mode *mode);
 void imx_drm_connector_destroy(struct drm_connector *connector);
 void imx_drm_encoder_destroy(struct drm_encoder *encoder);
 
diff --git a/drivers/staging/imx-drm/imx-hdmi.c b/drivers/staging/imx-drm/imx-hdmi.c
index 1b440483f28f..18c9ccd460b7 100644
--- a/drivers/staging/imx-drm/imx-hdmi.c
+++ b/drivers/staging/imx-drm/imx-hdmi.c
@@ -27,8 +27,8 @@
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_encoder_slave.h>
+#include <video/imx-ipu-v3.h>
 
-#include "ipu-v3/imx-ipu-v3.h"
 #include "imx-hdmi.h"
 #include "imx-drm.h"
 
@@ -1490,7 +1490,6 @@ static struct drm_connector_funcs imx_hdmi_connector_funcs = {
 
 static struct drm_connector_helper_funcs imx_hdmi_connector_helper_funcs = {
 	.get_modes = imx_hdmi_connector_get_modes,
-	.mode_valid = imx_drm_connector_mode_valid,
 	.best_encoder = imx_hdmi_connector_best_encoder,
 };
 
diff --git a/drivers/staging/imx-drm/imx-ldb.c b/drivers/staging/imx-drm/imx-ldb.c
index fe4c1ef4e7a5..7e3f019d7e72 100644
--- a/drivers/staging/imx-drm/imx-ldb.c
+++ b/drivers/staging/imx-drm/imx-ldb.c
@@ -317,7 +317,6 @@ static struct drm_connector_funcs imx_ldb_connector_funcs = {
 static struct drm_connector_helper_funcs imx_ldb_connector_helper_funcs = {
 	.get_modes = imx_ldb_connector_get_modes,
 	.best_encoder = imx_ldb_connector_best_encoder,
-	.mode_valid = imx_drm_connector_mode_valid,
 };
 
 static struct drm_encoder_funcs imx_ldb_encoder_funcs = {
diff --git a/drivers/staging/imx-drm/imx-tve.c b/drivers/staging/imx-drm/imx-tve.c
index a23f4f773146..c628fcdc22ae 100644
--- a/drivers/staging/imx-drm/imx-tve.c
+++ b/drivers/staging/imx-drm/imx-tve.c
@@ -30,8 +30,8 @@
 #include <drm/drmP.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_crtc_helper.h>
+#include <video/imx-ipu-v3.h>
 
-#include "ipu-v3/imx-ipu-v3.h"
 #include "imx-drm.h"
 
 #define TVE_COM_CONF_REG	0x00
@@ -249,11 +249,6 @@ static int imx_tve_connector_mode_valid(struct drm_connector *connector,
 {
 	struct imx_tve *tve = con_to_tve(connector);
 	unsigned long rate;
-	int ret;
-
-	ret = imx_drm_connector_mode_valid(connector, mode);
-	if (ret != MODE_OK)
-		return ret;
 
 	/* pixel clock with 2x oversampling */
 	rate = clk_round_rate(tve->clk, 2000UL * mode->clock) / 2000;
diff --git a/drivers/staging/imx-drm/ipuv3-crtc.c b/drivers/staging/imx-drm/ipuv3-crtc.c
index 47bec5e17358..720868bff35b 100644
--- a/drivers/staging/imx-drm/ipuv3-crtc.c
+++ b/drivers/staging/imx-drm/ipuv3-crtc.c
@@ -30,7 +30,7 @@
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_fb_cma_helper.h>
 
-#include "ipu-v3/imx-ipu-v3.h"
+#include <video/imx-ipu-v3.h>
 #include "imx-drm.h"
 #include "ipuv3-plane.h"
 
diff --git a/drivers/staging/imx-drm/ipuv3-plane.c b/drivers/staging/imx-drm/ipuv3-plane.c
index 5697e59ddf1d..6f393a11f44d 100644
--- a/drivers/staging/imx-drm/ipuv3-plane.c
+++ b/drivers/staging/imx-drm/ipuv3-plane.c
@@ -17,7 +17,7 @@
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 
-#include "ipu-v3/imx-ipu-v3.h"
+#include "video/imx-ipu-v3.h"
 #include "ipuv3-plane.h"
 
 #define to_ipu_plane(x)	container_of(x, struct ipu_plane, base)
diff --git a/drivers/staging/imx-drm/parallel-display.c b/drivers/staging/imx-drm/parallel-display.c
index eaf4dda1a0c4..b5678328fc40 100644
--- a/drivers/staging/imx-drm/parallel-display.c
+++ b/drivers/staging/imx-drm/parallel-display.c
@@ -148,7 +148,6 @@ static struct drm_connector_funcs imx_pd_connector_funcs = {
 static struct drm_connector_helper_funcs imx_pd_connector_helper_funcs = {
 	.get_modes = imx_pd_connector_get_modes,
 	.best_encoder = imx_pd_connector_best_encoder,
-	.mode_valid = imx_drm_connector_mode_valid,
 };
 
 static struct drm_encoder_funcs imx_pd_encoder_funcs = {
diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig
index c7b4f0f927b1..8bf495ffb020 100644
--- a/drivers/video/Kconfig
+++ b/drivers/video/Kconfig
@@ -20,6 +20,7 @@ source "drivers/char/agp/Kconfig"
 source "drivers/gpu/vga/Kconfig"
 
 source "drivers/gpu/host1x/Kconfig"
+source "drivers/gpu/ipu-v3/Kconfig"
 
 menu "Direct Rendering Manager"
 source "drivers/gpu/drm/Kconfig"
diff --git a/include/drm/drmP.h b/include/drm/drmP.h
index a7c2a862b4f4..8af71a8e2c00 100644
--- a/include/drm/drmP.h
+++ b/include/drm/drmP.h
@@ -143,11 +143,6 @@ int drm_err(const char *func, const char *format, ...);
 #define DRIVER_PRIME       0x4000
 #define DRIVER_RENDER      0x8000
 
-#define DRIVER_BUS_PCI 0x1
-#define DRIVER_BUS_PLATFORM 0x2
-#define DRIVER_BUS_USB 0x3
-#define DRIVER_BUS_HOST1X 0x4
-
 /***********************************************************************/
 /** \name Begin the DRM... */
 /*@{*/
@@ -239,8 +234,6 @@ int drm_err(const char *func, const char *format, ...);
 /** \name Internal types and structures */
 /*@{*/
 
-#define DRM_ARRAY_SIZE(x) ARRAY_SIZE(x)
-
 #define DRM_IF_VERSION(maj, min) (maj << 16 | min)
 
 /**
@@ -731,13 +724,7 @@ struct drm_master {
 #define DRM_SCANOUTPOS_ACCURATE     (1 << 2)
 
 struct drm_bus {
-	int bus_type;
-	int (*get_irq)(struct drm_device *dev);
-	const char *(*get_name)(struct drm_device *dev);
 	int (*set_busid)(struct drm_device *dev, struct drm_master *master);
-	int (*set_unique)(struct drm_device *dev, struct drm_master *master,
-			  struct drm_unique *unique);
-	int (*irq_by_busid)(struct drm_device *dev, struct drm_irq_busid *p);
 };
 
 /**
@@ -974,11 +961,6 @@ struct drm_driver {
 	const struct drm_ioctl_desc *ioctls;
 	int num_ioctls;
 	const struct file_operations *fops;
-	union {
-		struct pci_driver *pci;
-		struct platform_device *platform_device;
-		struct usb_driver *usb;
-	} kdriver;
 	struct drm_bus *bus;
 
 	/* List of devices hanging off this driver with stealth attach. */
@@ -1040,14 +1022,17 @@ struct drm_pending_vblank_event {
 };
 
 struct drm_vblank_crtc {
+	struct drm_device *dev;		/* pointer to the drm_device */
 	wait_queue_head_t queue;	/**< VBLANK wait queue */
 	struct timeval time[DRM_VBLANKTIME_RBSIZE];	/**< timestamp of current count */
+	struct timer_list disable_timer;		/* delayed disable timer */
 	atomic_t count;			/**< number of VBLANK interrupts */
 	atomic_t refcount;		/* number of users of vblank interruptsper crtc */
 	u32 last;			/* protected by dev->vbl_lock, used */
 					/* for wraparound handling */
 	u32 last_wait;			/* Last vblank seqno waited per CRTC */
 	unsigned int inmodeset;		/* Display driver is setting mode */
+	int crtc;			/* crtc index */
 	bool enabled;			/* so we don't call enable more than
 					   once per disable */
 };
@@ -1058,7 +1043,6 @@ struct drm_vblank_crtc {
  */
 struct drm_device {
 	struct list_head legacy_dev_list;/**< list of devices per driver for stealth attach cleanup */
-	char *devname;			/**< For /proc/interrupts */
 	int if_version;			/**< Highest interface version set */
 
 	/** \name Lifetime Management */
@@ -1072,18 +1056,19 @@ struct drm_device {
 	struct drm_minor *render;		/**< Render node */
 	atomic_t unplugged;			/**< Flag whether dev is dead */
 	struct inode *anon_inode;		/**< inode for private address-space */
+	char *unique;				/**< unique name of the device */
 	/*@} */
 
 	/** \name Locks */
 	/*@{ */
-	spinlock_t count_lock;		/**< For inuse, drm_device::open_count, drm_device::buf_use */
 	struct mutex struct_mutex;	/**< For others */
 	struct mutex master_mutex;      /**< For drm_minor::master and drm_file::is_master */
 	/*@} */
 
 	/** \name Usage Counters */
 	/*@{ */
-	int open_count;			/**< Outstanding files open */
+	int open_count;			/**< Outstanding files open, protected by drm_global_mutex. */
+	spinlock_t buf_lock;		/**< For drm_device::buf_use and a few other things. */
 	int buf_use;			/**< Buffers in use -- cannot alloc */
 	atomic_t buf_alloc;		/**< Buffer allocation in progress */
 	/*@} */
@@ -1114,6 +1099,8 @@ struct drm_device {
 	/** \name Context support */
 	/*@{ */
 	bool irq_enabled;		/**< True if irq handler is enabled */
+	int irq;
+
 	__volatile__ long context_flag;	/**< Context swapping flag */
 	int last_context;		/**< Last current context */
 	/*@} */
@@ -1134,7 +1121,6 @@ struct drm_device {
 
 	spinlock_t vblank_time_lock;    /**< Protects vblank count and time updates during vblank enable/disable */
 	spinlock_t vbl_lock;
-	struct timer_list vblank_disable_timer;
 
 	u32 max_vblank_count;           /**< size of vblank counter register */
 
@@ -1186,11 +1172,6 @@ static __inline__ int drm_core_check_feature(struct drm_device *dev,
 	return ((dev->driver->driver_features & feature) ? 1 : 0);
 }
 
-static inline int drm_dev_to_irq(struct drm_device *dev)
-{
-	return dev->driver->bus->get_irq(dev);
-}
-
 static inline void drm_device_set_unplugged(struct drm_device *dev)
 {
 	smp_wmb();
@@ -1204,11 +1185,6 @@ static inline int drm_device_is_unplugged(struct drm_device *dev)
 	return ret;
 }
 
-static inline bool drm_modeset_is_locked(struct drm_device *dev)
-{
-	return mutex_is_locked(&dev->mode_config.mutex);
-}
-
 static inline bool drm_is_render_client(const struct drm_file *file_priv)
 {
 	return file_priv->minor->type == DRM_MINOR_RENDER;
@@ -1310,7 +1286,7 @@ extern int drm_remove_magic(struct drm_master *master, drm_magic_t magic);
 /* Cache management (drm_cache.c) */
 void drm_clflush_pages(struct page *pages[], unsigned long num_pages);
 void drm_clflush_sg(struct sg_table *st);
-void drm_clflush_virt_range(char *addr, unsigned long length);
+void drm_clflush_virt_range(void *addr, unsigned long length);
 
 				/* Locking IOCTL support (drm_lock.h) */
 extern int drm_lock(struct drm_device *dev, void *data,
@@ -1363,7 +1339,7 @@ extern void drm_core_reclaim_buffers(struct drm_device *dev,
 				/* IRQ support (drm_irq.h) */
 extern int drm_control(struct drm_device *dev, void *data,
 		       struct drm_file *file_priv);
-extern int drm_irq_install(struct drm_device *dev);
+extern int drm_irq_install(struct drm_device *dev, int irq);
 extern int drm_irq_uninstall(struct drm_device *dev);
 
 extern int drm_vblank_init(struct drm_device *dev, int num_crtcs);
@@ -1377,8 +1353,14 @@ extern void drm_send_vblank_event(struct drm_device *dev, int crtc,
 extern bool drm_handle_vblank(struct drm_device *dev, int crtc);
 extern int drm_vblank_get(struct drm_device *dev, int crtc);
 extern void drm_vblank_put(struct drm_device *dev, int crtc);
+extern int drm_crtc_vblank_get(struct drm_crtc *crtc);
+extern void drm_crtc_vblank_put(struct drm_crtc *crtc);
 extern void drm_vblank_off(struct drm_device *dev, int crtc);
+extern void drm_vblank_on(struct drm_device *dev, int crtc);
+extern void drm_crtc_vblank_off(struct drm_crtc *crtc);
+extern void drm_crtc_vblank_on(struct drm_crtc *crtc);
 extern void drm_vblank_cleanup(struct drm_device *dev);
+
 extern u32 drm_get_last_vbltimestamp(struct drm_device *dev, int crtc,
 				     struct timeval *tvblank, unsigned flags);
 extern int drm_calc_vbltimestamp_from_scanoutpos(struct drm_device *dev,
@@ -1522,6 +1504,9 @@ extern drm_dma_handle_t *drm_pci_alloc(struct drm_device *dev, size_t size,
 				       size_t align);
 extern void __drm_pci_free(struct drm_device *dev, drm_dma_handle_t * dmah);
 extern void drm_pci_free(struct drm_device *dev, drm_dma_handle_t * dmah);
+extern int drm_pci_set_unique(struct drm_device *dev,
+			      struct drm_master *master,
+			      struct drm_unique *u);
 
 			       /* sysfs support (drm_sysfs.c) */
 struct drm_sysfs_class;
@@ -1631,6 +1616,7 @@ void drm_dev_ref(struct drm_device *dev);
 void drm_dev_unref(struct drm_device *dev);
 int drm_dev_register(struct drm_device *dev, unsigned long flags);
 void drm_dev_unregister(struct drm_device *dev);
+int drm_dev_set_unique(struct drm_device *dev, const char *fmt, ...);
 
 struct drm_minor *drm_minor_acquire(unsigned int minor_id);
 void drm_minor_release(struct drm_minor *minor);
diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h
index e55fccbe7c42..251b75e6bf7a 100644
--- a/include/drm/drm_crtc.h
+++ b/include/drm/drm_crtc.h
@@ -33,6 +33,7 @@
 #include <linux/hdmi.h>
 #include <drm/drm_mode.h>
 #include <drm/drm_fourcc.h>
+#include <drm/drm_modeset_lock.h>
 
 struct drm_device;
 struct drm_mode_set;
@@ -50,6 +51,7 @@ struct drm_clip_rect;
 #define DRM_MODE_OBJECT_BLOB 0xbbbbbbbb
 #define DRM_MODE_OBJECT_PLANE 0xeeeeeeee
 #define DRM_MODE_OBJECT_BRIDGE 0xbdbdbdbd
+#define DRM_MODE_OBJECT_ANY 0
 
 struct drm_mode_object {
 	uint32_t id;
@@ -64,6 +66,15 @@ struct drm_object_properties {
 	uint64_t values[DRM_OBJECT_MAX_PROPERTY];
 };
 
+static inline int64_t U642I64(uint64_t val)
+{
+	return (int64_t)*((int64_t *)&val);
+}
+static inline uint64_t I642U64(int64_t val)
+{
+	return (uint64_t)*((uint64_t *)&val);
+}
+
 enum drm_connector_force {
 	DRM_FORCE_UNSPECIFIED,
 	DRM_FORCE_OFF,
@@ -110,6 +121,9 @@ struct drm_display_info {
 	enum subpixel_order subpixel_order;
 	u32 color_formats;
 
+	/* Mask of supported hdmi deep color modes */
+	u8 edid_hdmi_dc_modes;
+
 	u8 cea_rev;
 };
 
@@ -190,10 +204,15 @@ struct drm_property {
 	char name[DRM_PROP_NAME_LEN];
 	uint32_t num_values;
 	uint64_t *values;
+	struct drm_device *dev;
 
 	struct list_head enum_blob_list;
 };
 
+void drm_modeset_lock_all(struct drm_device *dev);
+void drm_modeset_unlock_all(struct drm_device *dev);
+void drm_warn_on_modeset_not_all_locked(struct drm_device *dev);
+
 struct drm_crtc;
 struct drm_connector;
 struct drm_encoder;
@@ -269,6 +288,7 @@ struct drm_crtc_funcs {
  * drm_crtc - central CRTC control structure
  * @dev: parent DRM device
  * @head: list management
+ * @mutex: per-CRTC locking
  * @base: base KMS object for ID tracking etc.
  * @primary: primary plane for this CRTC
  * @cursor: cursor plane for this CRTC
@@ -303,7 +323,7 @@ struct drm_crtc {
 	 * state, ...) and a write lock for everything which can be update
 	 * without a full modeset (fb, cursor data, ...)
 	 */
-	struct mutex mutex;
+	struct drm_modeset_lock mutex;
 
 	struct drm_mode_object base;
 
@@ -400,6 +420,7 @@ struct drm_encoder_funcs {
  * @dev: parent DRM device
  * @head: list management
  * @base: base KMS object
+ * @name: encoder name
  * @encoder_type: one of the %DRM_MODE_ENCODER_<foo> types in drm_mode.h
  * @possible_crtcs: bitmask of potential CRTC bindings
  * @possible_clones: bitmask of potential sibling encoders for cloning
@@ -416,6 +437,7 @@ struct drm_encoder {
 	struct list_head head;
 
 	struct drm_mode_object base;
+	char *name;
 	int encoder_type;
 	uint32_t possible_crtcs;
 	uint32_t possible_clones;
@@ -444,6 +466,7 @@ struct drm_encoder {
  * @attr: sysfs attributes
  * @head: list management
  * @base: base KMS object
+ * @name: connector name
  * @connector_type: one of the %DRM_MODE_CONNECTOR_<foo> types from drm_mode.h
  * @connector_type_id: index into connector type enum
  * @interlace_allowed: can this connector handle interlaced modes?
@@ -482,6 +505,7 @@ struct drm_connector {
 
 	struct drm_mode_object base;
 
+	char *name;
 	int connector_type;
 	int connector_type_id;
 	bool interlace_allowed;
@@ -723,6 +747,8 @@ struct drm_mode_group {
  */
 struct drm_mode_config {
 	struct mutex mutex; /* protects configuration (mode lists etc.) */
+	struct drm_modeset_lock connection_mutex; /* protects connector->encoder and encoder->crtc links */
+	struct drm_modeset_acquire_ctx *acquire_ctx; /* for legacy _lock_all() / _unlock_all() */
 	struct mutex idr_mutex; /* for IDR management */
 	struct idr crtc_idr; /* use this idr for all IDs, fb, crtc, connector, modes - just makes life easier */
 	/* this is limited to one for now */
@@ -823,10 +849,6 @@ struct drm_prop_enum_list {
 	char *name;
 };
 
-extern void drm_modeset_lock_all(struct drm_device *dev);
-extern void drm_modeset_unlock_all(struct drm_device *dev);
-extern void drm_warn_on_modeset_not_all_locked(struct drm_device *dev);
-
 extern int drm_crtc_init_with_planes(struct drm_device *dev,
 				     struct drm_crtc *crtc,
 				     struct drm_plane *primary,
@@ -905,7 +927,6 @@ extern int drm_crtc_check_viewport(const struct drm_crtc *crtc,
 
 extern void drm_encoder_cleanup(struct drm_encoder *encoder);
 
-extern const char *drm_get_connector_name(const struct drm_connector *connector);
 extern const char *drm_get_connector_status_name(enum drm_connector_status status);
 extern const char *drm_get_subpixel_order_name(enum subpixel_order order);
 extern const char *drm_get_dpms_name(int val);
@@ -915,6 +936,7 @@ extern const char *drm_get_tv_subconnector_name(int val);
 extern const char *drm_get_tv_select_name(int val);
 extern void drm_fb_release(struct drm_file *file_priv);
 extern int drm_mode_group_init_legacy_group(struct drm_device *dev, struct drm_mode_group *group);
+extern void drm_mode_group_destroy(struct drm_mode_group *group);
 extern bool drm_probe_ddc(struct i2c_adapter *adapter);
 extern struct edid *drm_get_edid(struct drm_connector *connector,
 				 struct i2c_adapter *adapter);
@@ -926,6 +948,23 @@ extern void drm_mode_config_cleanup(struct drm_device *dev);
 
 extern int drm_mode_connector_update_edid_property(struct drm_connector *connector,
 						struct edid *edid);
+
+static inline bool drm_property_type_is(struct drm_property *property,
+		uint32_t type)
+{
+	/* instanceof for props.. handles extended type vs original types: */
+	if (property->flags & DRM_MODE_PROP_EXTENDED_TYPE)
+		return (property->flags & DRM_MODE_PROP_EXTENDED_TYPE) == type;
+	return property->flags & type;
+}
+
+static inline bool drm_property_type_valid(struct drm_property *property)
+{
+	if (property->flags & DRM_MODE_PROP_EXTENDED_TYPE)
+		return !(property->flags & DRM_MODE_PROP_LEGACY_TYPE);
+	return !!(property->flags & DRM_MODE_PROP_LEGACY_TYPE);
+}
+
 extern int drm_object_property_set_value(struct drm_mode_object *obj,
 					 struct drm_property *property,
 					 uint64_t val);
@@ -959,6 +998,11 @@ struct drm_property *drm_property_create_bitmask(struct drm_device *dev,
 struct drm_property *drm_property_create_range(struct drm_device *dev, int flags,
 					 const char *name,
 					 uint64_t min, uint64_t max);
+struct drm_property *drm_property_create_signed_range(struct drm_device *dev,
+					 int flags, const char *name,
+					 int64_t min, int64_t max);
+struct drm_property *drm_property_create_object(struct drm_device *dev,
+					 int flags, const char *name, uint32_t type);
 extern void drm_property_destroy(struct drm_device *dev, struct drm_property *property);
 extern int drm_property_add_enum(struct drm_property *property, int index,
 				 uint64_t value, const char *name);
@@ -967,7 +1011,6 @@ extern int drm_mode_create_tv_properties(struct drm_device *dev, int num_formats
 				     char *formats[]);
 extern int drm_mode_create_scaling_mode_property(struct drm_device *dev);
 extern int drm_mode_create_dirty_info_property(struct drm_device *dev);
-extern const char *drm_get_encoder_name(const struct drm_encoder *encoder);
 
 extern int drm_mode_connector_attach_encoder(struct drm_connector *connector,
 					     struct drm_encoder *encoder);
@@ -975,6 +1018,7 @@ extern int drm_mode_crtc_set_gamma_size(struct drm_crtc *crtc,
 					 int gamma_size);
 extern struct drm_mode_object *drm_mode_object_find(struct drm_device *dev,
 		uint32_t id, uint32_t type);
+
 /* IOCTLs */
 extern int drm_mode_getresources(struct drm_device *dev,
 				 void *data, struct drm_file *file_priv);
@@ -1020,6 +1064,7 @@ extern int drm_mode_gamma_get_ioctl(struct drm_device *dev,
 extern int drm_mode_gamma_set_ioctl(struct drm_device *dev,
 				    void *data, struct drm_file *file_priv);
 extern u8 drm_match_cea_mode(const struct drm_display_mode *to_match);
+extern enum hdmi_picture_aspect drm_get_cea_aspect_ratio(const u8 video_code);
 extern bool drm_detect_hdmi_monitor(struct edid *edid);
 extern bool drm_detect_monitor_audio(struct edid *edid);
 extern bool drm_rgb_quant_range_selectable(struct edid *edid);
@@ -1057,6 +1102,15 @@ extern int drm_format_vert_chroma_subsampling(uint32_t format);
 extern const char *drm_get_format_name(uint32_t format);
 
 /* Helpers */
+
+static inline struct drm_plane *drm_plane_find(struct drm_device *dev,
+		uint32_t id)
+{
+	struct drm_mode_object *mo;
+	mo = drm_mode_object_find(dev, id, DRM_MODE_OBJECT_PLANE);
+	return mo ? obj_to_plane(mo) : NULL;
+}
+
 static inline struct drm_crtc *drm_crtc_find(struct drm_device *dev,
 	uint32_t id)
 {
@@ -1073,6 +1127,30 @@ static inline struct drm_encoder *drm_encoder_find(struct drm_device *dev,
 	return mo ? obj_to_encoder(mo) : NULL;
 }
 
+static inline struct drm_connector *drm_connector_find(struct drm_device *dev,
+		uint32_t id)
+{
+	struct drm_mode_object *mo;
+	mo = drm_mode_object_find(dev, id, DRM_MODE_OBJECT_CONNECTOR);
+	return mo ? obj_to_connector(mo) : NULL;
+}
+
+static inline struct drm_property *drm_property_find(struct drm_device *dev,
+		uint32_t id)
+{
+	struct drm_mode_object *mo;
+	mo = drm_mode_object_find(dev, id, DRM_MODE_OBJECT_PROPERTY);
+	return mo ? obj_to_property(mo) : NULL;
+}
+
+static inline struct drm_property_blob *
+drm_property_blob_find(struct drm_device *dev, uint32_t id)
+{
+	struct drm_mode_object *mo;
+	mo = drm_mode_object_find(dev, id, DRM_MODE_OBJECT_BLOB);
+	return mo ? obj_to_blob(mo) : NULL;
+}
+
 /* Plane list iterator for legacy (overlay only) planes. */
 #define drm_for_each_legacy_plane(plane, planelist) \
 	list_for_each_entry(plane, planelist, head) \
diff --git a/include/drm/drm_crtc_helper.h b/include/drm/drm_crtc_helper.h
index 36a5febac2a6..a3d75fefd010 100644
--- a/include/drm/drm_crtc_helper.h
+++ b/include/drm/drm_crtc_helper.h
@@ -114,7 +114,7 @@ struct drm_encoder_helper_funcs {
 /**
  * drm_connector_helper_funcs - helper operations for connectors
  * @get_modes: get mode list for this connector
- * @mode_valid: is this mode valid on the given connector?
+ * @mode_valid (optional): is this mode valid on the given connector?
  *
  * The helper operations are called by the mid-layer CRTC helper.
  */
@@ -165,6 +165,10 @@ extern void drm_helper_resume_force_mode(struct drm_device *dev);
 extern int drm_helper_probe_single_connector_modes(struct drm_connector
 						   *connector, uint32_t maxX,
 						   uint32_t maxY);
+extern int drm_helper_probe_single_connector_modes_nomerge(struct drm_connector
+							   *connector,
+							   uint32_t maxX,
+							   uint32_t maxY);
 extern void drm_kms_helper_poll_init(struct drm_device *dev);
 extern void drm_kms_helper_poll_fini(struct drm_device *dev);
 extern bool drm_helper_hpd_irq_event(struct drm_device *dev);
diff --git a/include/drm/drm_dp_helper.h b/include/drm/drm_dp_helper.h
index cfcacec5b89d..a21568bf1514 100644
--- a/include/drm/drm_dp_helper.h
+++ b/include/drm/drm_dp_helper.h
@@ -37,6 +37,7 @@
  * eDP: Embedded DisplayPort version 1
  * DPI: DisplayPort Interoperability Guideline v1.1a
  * 1.2: DisplayPort 1.2
+ * MST: Multistream Transport - part of DP 1.2a
  *
  * 1.2 formally includes both eDP and DPI definitions.
  */
@@ -103,9 +104,14 @@
 #define DP_TRAINING_AUX_RD_INTERVAL         0x00e   /* XXX 1.2? */
 
 /* Multiple stream transport */
+#define DP_FAUX_CAP			    0x020   /* 1.2 */
+# define DP_FAUX_CAP_1			    (1 << 0)
+
 #define DP_MSTM_CAP			    0x021   /* 1.2 */
 # define DP_MST_CAP			    (1 << 0)
 
+#define DP_GUID				    0x030   /* 1.2 */
+
 #define DP_PSR_SUPPORT                      0x070   /* XXX 1.2? */
 # define DP_PSR_IS_SUPPORTED                1
 #define DP_PSR_CAPS                         0x071   /* XXX 1.2? */
@@ -221,6 +227,16 @@
 # define DP_PSR_CRC_VERIFICATION	    (1 << 2)
 # define DP_PSR_FRAME_CAPTURE		    (1 << 3)
 
+#define DP_ADAPTER_CTRL			    0x1a0
+# define DP_ADAPTER_CTRL_FORCE_LOAD_SENSE   (1 << 0)
+
+#define DP_BRANCH_DEVICE_CTRL		    0x1a1
+# define DP_BRANCH_DEVICE_IRQ_HPD	    (1 << 0)
+
+#define DP_PAYLOAD_ALLOCATE_SET		    0x1c0
+#define DP_PAYLOAD_ALLOCATE_START_TIME_SLOT 0x1c1
+#define DP_PAYLOAD_ALLOCATE_TIME_SLOT_COUNT 0x1c2
+
 #define DP_SINK_COUNT			    0x200
 /* prior to 1.2 bit 7 was reserved mbz */
 # define DP_GET_SINK_COUNT(x)		    ((((x) & 0x80) >> 1) | ((x) & 0x3f))
@@ -230,6 +246,9 @@
 # define DP_REMOTE_CONTROL_COMMAND_PENDING  (1 << 0)
 # define DP_AUTOMATED_TEST_REQUEST	    (1 << 1)
 # define DP_CP_IRQ			    (1 << 2)
+# define DP_MCCS_IRQ			    (1 << 3)
+# define DP_DOWN_REP_MSG_RDY		    (1 << 4) /* 1.2 MST */
+# define DP_UP_REQ_MSG_RDY		    (1 << 5) /* 1.2 MST */
 # define DP_SINK_SPECIFIC_IRQ		    (1 << 6)
 
 #define DP_LANE0_1_STATUS		    0x202
@@ -291,9 +310,18 @@
 # define DP_TEST_NAK			    (1 << 1)
 # define DP_TEST_EDID_CHECKSUM_WRITE	    (1 << 2)
 
+#define DP_TEST_EDID_CHECKSUM		    0x261
+
 #define DP_TEST_SINK			    0x270
 #define DP_TEST_SINK_START	    (1 << 0)
 
+#define DP_PAYLOAD_TABLE_UPDATE_STATUS      0x2c0   /* 1.2 MST */
+# define DP_PAYLOAD_TABLE_UPDATED           (1 << 0)
+# define DP_PAYLOAD_ACT_HANDLED             (1 << 1)
+
+#define DP_VC_PAYLOAD_ID_SLOT_1             0x2c1   /* 1.2 MST */
+/* up to ID_SLOT_63 at 0x2ff */
+
 #define DP_SOURCE_OUI			    0x300
 #define DP_SINK_OUI			    0x400
 #define DP_BRANCH_OUI			    0x500
@@ -303,6 +331,21 @@
 # define DP_SET_POWER_D3                    0x2
 # define DP_SET_POWER_MASK                  0x3
 
+#define DP_SIDEBAND_MSG_DOWN_REQ_BASE	    0x1000   /* 1.2 MST */
+#define DP_SIDEBAND_MSG_UP_REP_BASE	    0x1200   /* 1.2 MST */
+#define DP_SIDEBAND_MSG_DOWN_REP_BASE	    0x1400   /* 1.2 MST */
+#define DP_SIDEBAND_MSG_UP_REQ_BASE	    0x1600   /* 1.2 MST */
+
+#define DP_SINK_COUNT_ESI		    0x2002   /* 1.2 */
+/* 0-5 sink count */
+# define DP_SINK_COUNT_CP_READY             (1 << 6)
+
+#define DP_DEVICE_SERVICE_IRQ_VECTOR_ESI0   0x2003   /* 1.2 */
+
+#define DP_DEVICE_SERVICE_IRQ_VECTOR_ESI1   0x2004   /* 1.2 */
+
+#define DP_LINK_SERVICE_IRQ_VECTOR_ESI0     0x2005   /* 1.2 */
+
 #define DP_PSR_ERROR_STATUS                 0x2006  /* XXX 1.2? */
 # define DP_PSR_LINK_CRC_ERROR              (1 << 0)
 # define DP_PSR_RFB_STORAGE_ERROR           (1 << 1)
@@ -319,6 +362,43 @@
 # define DP_PSR_SINK_INTERNAL_ERROR         7
 # define DP_PSR_SINK_STATE_MASK             0x07
 
+/* DP 1.2 Sideband message defines */
+/* peer device type - DP 1.2a Table 2-92 */
+#define DP_PEER_DEVICE_NONE		0x0
+#define DP_PEER_DEVICE_SOURCE_OR_SST	0x1
+#define DP_PEER_DEVICE_MST_BRANCHING	0x2
+#define DP_PEER_DEVICE_SST_SINK		0x3
+#define DP_PEER_DEVICE_DP_LEGACY_CONV	0x4
+
+/* DP 1.2 MST sideband request names DP 1.2a Table 2-80 */
+#define DP_LINK_ADDRESS			0x01
+#define DP_CONNECTION_STATUS_NOTIFY	0x02
+#define DP_ENUM_PATH_RESOURCES		0x10
+#define DP_ALLOCATE_PAYLOAD		0x11
+#define DP_QUERY_PAYLOAD		0x12
+#define DP_RESOURCE_STATUS_NOTIFY	0x13
+#define DP_CLEAR_PAYLOAD_ID_TABLE	0x14
+#define DP_REMOTE_DPCD_READ		0x20
+#define DP_REMOTE_DPCD_WRITE		0x21
+#define DP_REMOTE_I2C_READ		0x22
+#define DP_REMOTE_I2C_WRITE		0x23
+#define DP_POWER_UP_PHY			0x24
+#define DP_POWER_DOWN_PHY		0x25
+#define DP_SINK_EVENT_NOTIFY		0x30
+#define DP_QUERY_STREAM_ENC_STATUS	0x38
+
+/* DP 1.2 MST sideband nak reasons - table 2.84 */
+#define DP_NAK_WRITE_FAILURE		0x01
+#define DP_NAK_INVALID_READ		0x02
+#define DP_NAK_CRC_FAILURE		0x03
+#define DP_NAK_BAD_PARAM		0x04
+#define DP_NAK_DEFER			0x05
+#define DP_NAK_LINK_FAILURE		0x06
+#define DP_NAK_NO_RESOURCES		0x07
+#define DP_NAK_DPCD_FAIL		0x08
+#define DP_NAK_I2C_NAK			0x09
+#define DP_NAK_ALLOCATE_FAIL		0x0a
+
 #define MODE_I2C_START	1
 #define MODE_I2C_WRITE	2
 #define MODE_I2C_READ	4
@@ -431,8 +511,10 @@ struct drm_dp_aux_msg {
 
 /**
  * struct drm_dp_aux - DisplayPort AUX channel
+ * @name: user-visible name of this AUX channel and the I2C-over-AUX adapter
  * @ddc: I2C adapter that can be used for I2C-over-AUX communication
  * @dev: pointer to struct device that is the parent for this AUX channel
+ * @hw_mutex: internal mutex used for locking transfers
  * @transfer: transfers a message representing a single AUX transaction
  *
  * The .dev field should be set to a pointer to the device that implements
@@ -465,7 +547,7 @@ struct drm_dp_aux {
 	const char *name;
 	struct i2c_adapter ddc;
 	struct device *dev;
-
+	struct mutex hw_mutex;
 	ssize_t (*transfer)(struct drm_dp_aux *aux,
 			    struct drm_dp_aux_msg *msg);
 };
@@ -524,7 +606,7 @@ int drm_dp_link_probe(struct drm_dp_aux *aux, struct drm_dp_link *link);
 int drm_dp_link_power_up(struct drm_dp_aux *aux, struct drm_dp_link *link);
 int drm_dp_link_configure(struct drm_dp_aux *aux, struct drm_dp_link *link);
 
-int drm_dp_aux_register_i2c_bus(struct drm_dp_aux *aux);
-void drm_dp_aux_unregister_i2c_bus(struct drm_dp_aux *aux);
+int drm_dp_aux_register(struct drm_dp_aux *aux);
+void drm_dp_aux_unregister(struct drm_dp_aux *aux);
 
 #endif /* _DRM_DP_HELPER_H_ */
diff --git a/include/drm/drm_edid.h b/include/drm/drm_edid.h
index a1441c5ac63d..b96031d947a0 100644
--- a/include/drm/drm_edid.h
+++ b/include/drm/drm_edid.h
@@ -202,6 +202,11 @@ struct detailed_timing {
 #define DRM_EDID_FEATURE_PM_SUSPEND       (1 << 6)
 #define DRM_EDID_FEATURE_PM_STANDBY       (1 << 7)
 
+#define DRM_EDID_HDMI_DC_48               (1 << 6)
+#define DRM_EDID_HDMI_DC_36               (1 << 5)
+#define DRM_EDID_HDMI_DC_30               (1 << 4)
+#define DRM_EDID_HDMI_DC_Y444             (1 << 3)
+
 struct edid {
 	u8 header[8];
 	/* Vendor & product info */
diff --git a/include/drm/drm_fb_helper.h b/include/drm/drm_fb_helper.h
index 6e622f7d481d..7997246d4039 100644
--- a/include/drm/drm_fb_helper.h
+++ b/include/drm/drm_fb_helper.h
@@ -108,7 +108,7 @@ int drm_fb_helper_set_par(struct fb_info *info);
 int drm_fb_helper_check_var(struct fb_var_screeninfo *var,
 			    struct fb_info *info);
 
-bool drm_fb_helper_restore_fbdev_mode(struct drm_fb_helper *fb_helper);
+bool drm_fb_helper_restore_fbdev_mode_unlocked(struct drm_fb_helper *fb_helper);
 void drm_fb_helper_fill_var(struct fb_info *info, struct drm_fb_helper *fb_helper,
 			    uint32_t fb_width, uint32_t fb_height);
 void drm_fb_helper_fill_fix(struct fb_info *info, uint32_t pitch,
diff --git a/include/drm/drm_flip_work.h b/include/drm/drm_flip_work.h
index 35c776ae7d3b..9eed34dcd6af 100644
--- a/include/drm/drm_flip_work.h
+++ b/include/drm/drm_flip_work.h
@@ -57,6 +57,7 @@ typedef void (*drm_flip_func_t)(struct drm_flip_work *work, void *val);
  * @count: number of committed items
  * @func: callback fxn called for each committed item
  * @worker: worker which calls @func
+ * @fifo: queue of committed items
  */
 struct drm_flip_work {
 	const char *name;
diff --git a/include/drm/drm_mipi_dsi.h b/include/drm/drm_mipi_dsi.h
index 7209df15a3cd..944f33f8ba38 100644
--- a/include/drm/drm_mipi_dsi.h
+++ b/include/drm/drm_mipi_dsi.h
@@ -135,11 +135,13 @@ ssize_t mipi_dsi_dcs_read(struct mipi_dsi_device *dsi, unsigned int channel,
  * @driver: device driver model driver
  * @probe: callback for device binding
  * @remove: callback for device unbinding
+ * @shutdown: called at shutdown time to quiesce the device
  */
 struct mipi_dsi_driver {
 	struct device_driver driver;
 	int(*probe)(struct mipi_dsi_device *dsi);
 	int(*remove)(struct mipi_dsi_device *dsi);
+	void (*shutdown)(struct mipi_dsi_device *dsi);
 };
 
 #define to_mipi_dsi_driver(d) container_of(d, struct mipi_dsi_driver, driver)
diff --git a/include/drm/drm_modes.h b/include/drm/drm_modes.h
index 2dbbf9976669..91d0582f924e 100644
--- a/include/drm/drm_modes.h
+++ b/include/drm/drm_modes.h
@@ -223,7 +223,7 @@ void drm_mode_validate_size(struct drm_device *dev,
 void drm_mode_prune_invalid(struct drm_device *dev,
 			    struct list_head *mode_list, bool verbose);
 void drm_mode_sort(struct list_head *mode_list);
-void drm_mode_connector_list_update(struct drm_connector *connector);
+void drm_mode_connector_list_update(struct drm_connector *connector, bool merge_type_bits);
 
 /* parsing cmdline modes */
 bool
diff --git a/include/drm/drm_modeset_lock.h b/include/drm/drm_modeset_lock.h
new file mode 100644
index 000000000000..402aa7a6a058
--- /dev/null
+++ b/include/drm/drm_modeset_lock.h
@@ -0,0 +1,126 @@
+/*
+ * Copyright (C) 2014 Red Hat
+ * Author: Rob Clark <robdclark@gmail.com>
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef DRM_MODESET_LOCK_H_
+#define DRM_MODESET_LOCK_H_
+
+#include <linux/ww_mutex.h>
+
+struct drm_modeset_lock;
+
+/**
+ * drm_modeset_acquire_ctx - locking context (see ww_acquire_ctx)
+ * @ww_ctx: base acquire ctx
+ * @contended: used internally for -EDEADLK handling
+ * @locked: list of held locks
+ *
+ * Each thread competing for a set of locks must use one acquire
+ * ctx.  And if any lock fxn returns -EDEADLK, it must backoff and
+ * retry.
+ */
+struct drm_modeset_acquire_ctx {
+
+	struct ww_acquire_ctx ww_ctx;
+
+	/**
+	 * Contended lock: if a lock is contended you should only call
+	 * drm_modeset_backoff() which drops locks and slow-locks the
+	 * contended lock.
+	 */
+	struct drm_modeset_lock *contended;
+
+	/**
+	 * list of held locks (drm_modeset_lock)
+	 */
+	struct list_head locked;
+};
+
+/**
+ * drm_modeset_lock - used for locking modeset resources.
+ * @mutex: resource locking
+ * @head: used to hold it's place on state->locked list when
+ *    part of an atomic update
+ *
+ * Used for locking CRTCs and other modeset resources.
+ */
+struct drm_modeset_lock {
+	/**
+	 * modeset lock
+	 */
+	struct ww_mutex mutex;
+
+	/**
+	 * Resources that are locked as part of an atomic update are added
+	 * to a list (so we know what to unlock at the end).
+	 */
+	struct list_head head;
+};
+
+extern struct ww_class crtc_ww_class;
+
+void drm_modeset_acquire_init(struct drm_modeset_acquire_ctx *ctx,
+		uint32_t flags);
+void drm_modeset_acquire_fini(struct drm_modeset_acquire_ctx *ctx);
+void drm_modeset_drop_locks(struct drm_modeset_acquire_ctx *ctx);
+void drm_modeset_backoff(struct drm_modeset_acquire_ctx *ctx);
+int drm_modeset_backoff_interruptible(struct drm_modeset_acquire_ctx *ctx);
+
+/**
+ * drm_modeset_lock_init - initialize lock
+ * @lock: lock to init
+ */
+static inline void drm_modeset_lock_init(struct drm_modeset_lock *lock)
+{
+	ww_mutex_init(&lock->mutex, &crtc_ww_class);
+	INIT_LIST_HEAD(&lock->head);
+}
+
+/**
+ * drm_modeset_lock_fini - cleanup lock
+ * @lock: lock to cleanup
+ */
+static inline void drm_modeset_lock_fini(struct drm_modeset_lock *lock)
+{
+	WARN_ON(!list_empty(&lock->head));
+}
+
+/**
+ * drm_modeset_is_locked - equivalent to mutex_is_locked()
+ * @lock: lock to check
+ */
+static inline bool drm_modeset_is_locked(struct drm_modeset_lock *lock)
+{
+	return ww_mutex_is_locked(&lock->mutex);
+}
+
+int drm_modeset_lock(struct drm_modeset_lock *lock,
+		struct drm_modeset_acquire_ctx *ctx);
+int drm_modeset_lock_interruptible(struct drm_modeset_lock *lock,
+		struct drm_modeset_acquire_ctx *ctx);
+void drm_modeset_unlock(struct drm_modeset_lock *lock);
+
+struct drm_device;
+int drm_modeset_lock_all_crtcs(struct drm_device *dev,
+		struct drm_modeset_acquire_ctx *ctx);
+
+#endif /* DRM_MODESET_LOCK_H_ */
diff --git a/include/drm/drm_plane_helper.h b/include/drm/drm_plane_helper.h
index 09824becee3e..52e6870534b2 100644
--- a/include/drm/drm_plane_helper.h
+++ b/include/drm/drm_plane_helper.h
@@ -24,6 +24,17 @@
 #ifndef DRM_PLANE_HELPER_H
 #define DRM_PLANE_HELPER_H
 
+#include <drm/drm_rect.h>
+
+/*
+ * Drivers that don't allow primary plane scaling may pass this macro in place
+ * of the min/max scale parameters of the update checker function.
+ *
+ * Due to src being in 16.16 fixed point and dest being in integer pixels,
+ * 1<<16 represents no scaling.
+ */
+#define DRM_PLANE_HELPER_NO_SCALING (1<<16)
+
 /**
  * DOC: plane helpers
  *
@@ -31,6 +42,17 @@
  * planes.
  */
 
+extern int drm_plane_helper_check_update(struct drm_plane *plane,
+					 struct drm_crtc *crtc,
+					 struct drm_framebuffer *fb,
+					 struct drm_rect *src,
+					 struct drm_rect *dest,
+					 const struct drm_rect *clip,
+					 int min_scale,
+					 int max_scale,
+					 bool can_position,
+					 bool can_update_disabled,
+					 bool *visible);
 extern int drm_primary_helper_update(struct drm_plane *plane,
 				     struct drm_crtc *crtc,
 				     struct drm_framebuffer *fb,
@@ -42,7 +64,7 @@ extern int drm_primary_helper_disable(struct drm_plane *plane);
 extern void drm_primary_helper_destroy(struct drm_plane *plane);
 extern const struct drm_plane_funcs drm_primary_helper_funcs;
 extern struct drm_plane *drm_primary_helper_create_plane(struct drm_device *dev,
-							 uint32_t *formats,
+							 const uint32_t *formats,
 							 int num_formats);
 
 
diff --git a/include/drm/i915_pciids.h b/include/drm/i915_pciids.h
index 012d58fa8ff0..0572035673f3 100644
--- a/include/drm/i915_pciids.h
+++ b/include/drm/i915_pciids.h
@@ -223,14 +223,32 @@
 	_INTEL_BDW_D(gt, 0x160A, info), /* Server */ \
 	_INTEL_BDW_D(gt, 0x160D, info) /* Workstation */
 
-#define INTEL_BDW_M_IDS(info) \
+#define INTEL_BDW_GT12M_IDS(info) \
 	_INTEL_BDW_M_IDS(1, info), \
-	_INTEL_BDW_M_IDS(2, info), \
-	_INTEL_BDW_M_IDS(3, info)
+	_INTEL_BDW_M_IDS(2, info)
 
-#define INTEL_BDW_D_IDS(info) \
+#define INTEL_BDW_GT12D_IDS(info) \
 	_INTEL_BDW_D_IDS(1, info), \
-	_INTEL_BDW_D_IDS(2, info), \
+	_INTEL_BDW_D_IDS(2, info)
+
+#define INTEL_BDW_GT3M_IDS(info) \
+	_INTEL_BDW_M_IDS(3, info)
+
+#define INTEL_BDW_GT3D_IDS(info) \
 	_INTEL_BDW_D_IDS(3, info)
 
+#define INTEL_BDW_M_IDS(info) \
+	INTEL_BDW_GT12M_IDS(info), \
+	INTEL_BDW_GT3M_IDS(info)
+
+#define INTEL_BDW_D_IDS(info) \
+	INTEL_BDW_GT12D_IDS(info), \
+	INTEL_BDW_GT3D_IDS(info)
+
+#define INTEL_CHV_IDS(info) \
+	INTEL_VGA_DEVICE(0x22b0, info), \
+	INTEL_VGA_DEVICE(0x22b1, info), \
+	INTEL_VGA_DEVICE(0x22b2, info), \
+	INTEL_VGA_DEVICE(0x22b3, info)
+
 #endif /* _I915_PCIIDS_H */
diff --git a/include/drm/ttm/ttm_bo_api.h b/include/drm/ttm/ttm_bo_api.h
index ee127ec33c60..7526c5bf5610 100644
--- a/include/drm/ttm/ttm_bo_api.h
+++ b/include/drm/ttm/ttm_bo_api.h
@@ -485,13 +485,12 @@ extern int ttm_bo_init(struct ttm_bo_device *bdev,
 			void (*destroy) (struct ttm_buffer_object *));
 
 /**
- * ttm_bo_synccpu_object_init
+ * ttm_bo_create
  *
  * @bdev: Pointer to a ttm_bo_device struct.
- * @bo: Pointer to a ttm_buffer_object to be initialized.
  * @size: Requested size of buffer object.
  * @type: Requested type of buffer object.
- * @flags: Initial placement flags.
+ * @placement: Initial placement.
  * @page_alignment: Data alignment in pages.
  * @interruptible: If needing to sleep while waiting for GPU resources,
  * sleep interruptible.
diff --git a/include/uapi/drm/drm_mode.h b/include/uapi/drm/drm_mode.h
index f104c2603ebe..def54f9e07ca 100644
--- a/include/uapi/drm/drm_mode.h
+++ b/include/uapi/drm/drm_mode.h
@@ -181,6 +181,7 @@ struct drm_mode_get_plane_res {
 #define DRM_MODE_ENCODER_TVDAC	4
 #define DRM_MODE_ENCODER_VIRTUAL 5
 #define DRM_MODE_ENCODER_DSI	6
+#define DRM_MODE_ENCODER_DPMST	7
 
 struct drm_mode_get_encoder {
 	__u32 encoder_id;
@@ -251,6 +252,21 @@ struct drm_mode_get_connector {
 #define DRM_MODE_PROP_BLOB	(1<<4)
 #define DRM_MODE_PROP_BITMASK	(1<<5) /* bitmask of enumerated types */
 
+/* non-extended types: legacy bitmask, one bit per type: */
+#define DRM_MODE_PROP_LEGACY_TYPE  ( \
+		DRM_MODE_PROP_RANGE | \
+		DRM_MODE_PROP_ENUM | \
+		DRM_MODE_PROP_BLOB | \
+		DRM_MODE_PROP_BITMASK)
+
+/* extended-types: rather than continue to consume a bit per type,
+ * grab a chunk of the bits to use as integer type id.
+ */
+#define DRM_MODE_PROP_EXTENDED_TYPE	0x0000ffc0
+#define DRM_MODE_PROP_TYPE(n)		((n) << 6)
+#define DRM_MODE_PROP_OBJECT		DRM_MODE_PROP_TYPE(1)
+#define DRM_MODE_PROP_SIGNED_RANGE	DRM_MODE_PROP_TYPE(2)
+
 struct drm_mode_property_enum {
 	__u64 value;
 	char name[DRM_PROP_NAME_LEN];
diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h
index 126bfaa8bb6b..ff57f07c3249 100644
--- a/include/uapi/drm/i915_drm.h
+++ b/include/uapi/drm/i915_drm.h
@@ -223,6 +223,7 @@ typedef struct _drm_i915_sarea {
 #define DRM_I915_GEM_GET_CACHING	0x30
 #define DRM_I915_REG_READ		0x31
 #define DRM_I915_GET_RESET_STATS	0x32
+#define DRM_I915_GEM_USERPTR		0x33
 
 #define DRM_IOCTL_I915_INIT		DRM_IOW( DRM_COMMAND_BASE + DRM_I915_INIT, drm_i915_init_t)
 #define DRM_IOCTL_I915_FLUSH		DRM_IO ( DRM_COMMAND_BASE + DRM_I915_FLUSH)
@@ -273,6 +274,7 @@ typedef struct _drm_i915_sarea {
 #define DRM_IOCTL_I915_GEM_CONTEXT_DESTROY	DRM_IOW (DRM_COMMAND_BASE + DRM_I915_GEM_CONTEXT_DESTROY, struct drm_i915_gem_context_destroy)
 #define DRM_IOCTL_I915_REG_READ			DRM_IOWR (DRM_COMMAND_BASE + DRM_I915_REG_READ, struct drm_i915_reg_read)
 #define DRM_IOCTL_I915_GET_RESET_STATS		DRM_IOWR (DRM_COMMAND_BASE + DRM_I915_GET_RESET_STATS, struct drm_i915_reset_stats)
+#define DRM_IOCTL_I915_GEM_USERPTR			DRM_IOWR (DRM_COMMAND_BASE + DRM_I915_GEM_USERPTR, struct drm_i915_gem_userptr)
 
 /* Allow drivers to submit batchbuffers directly to hardware, relying
  * on the security mechanisms provided by hardware.
@@ -337,6 +339,7 @@ typedef struct drm_i915_irq_wait {
 #define I915_PARAM_HAS_EXEC_NO_RELOC	 25
 #define I915_PARAM_HAS_EXEC_HANDLE_LUT   26
 #define I915_PARAM_HAS_WT     	 	 27
+#define I915_PARAM_CMD_PARSER_VERSION	 28
 
 typedef struct drm_i915_getparam {
 	int param;
@@ -1049,4 +1052,18 @@ struct drm_i915_reset_stats {
 	__u32 pad;
 };
 
+struct drm_i915_gem_userptr {
+	__u64 user_ptr;
+	__u64 user_size;
+	__u32 flags;
+#define I915_USERPTR_READ_ONLY 0x1
+#define I915_USERPTR_UNSYNCHRONIZED 0x80000000
+	/**
+	 * Returned handle for the object.
+	 *
+	 * Object handles are nonzero.
+	 */
+	__u32 handle;
+};
+
 #endif /* _UAPI_I915_DRM_H_ */
diff --git a/include/uapi/drm/radeon_drm.h b/include/uapi/drm/radeon_drm.h
index aefa2f6afa3b..1cc0b610f162 100644
--- a/include/uapi/drm/radeon_drm.h
+++ b/include/uapi/drm/radeon_drm.h
@@ -1007,7 +1007,7 @@ struct drm_radeon_cs {
 #define RADEON_INFO_NUM_BYTES_MOVED	0x1d
 #define RADEON_INFO_VRAM_USAGE		0x1e
 #define RADEON_INFO_GTT_USAGE		0x1f
-
+#define RADEON_INFO_ACTIVE_CU_COUNT	0x20
 
 struct drm_radeon_info {
 	uint32_t		request;
diff --git a/drivers/staging/imx-drm/ipu-v3/imx-ipu-v3.h b/include/video/imx-ipu-v3.h
index c2c6fab05eaa..3e43e22cdff9 100644
--- a/drivers/staging/imx-drm/ipu-v3/imx-ipu-v3.h
+++ b/include/video/imx-ipu-v3.h
@@ -104,6 +104,7 @@ int ipu_idmac_wait_busy(struct ipuv3_channel *channel, int ms);
 
 void ipu_idmac_set_double_buffer(struct ipuv3_channel *channel,
 		bool doublebuffer);
+int ipu_idmac_get_current_buffer(struct ipuv3_channel *channel);
 void ipu_idmac_select_buffer(struct ipuv3_channel *channel, u32 buf_num);
 
 /*
@@ -165,6 +166,20 @@ int ipu_dp_set_window_pos(struct ipu_dp *, u16 x_pos, u16 y_pos);
 int ipu_dp_set_global_alpha(struct ipu_dp *dp, bool enable, u8 alpha,
 		bool bg_chan);
 
+/*
+ * IPU CMOS Sensor Interface (csi) functions
+ */
+int ipu_csi_enable(struct ipu_soc *ipu, int csi);
+int ipu_csi_disable(struct ipu_soc *ipu, int csi);
+
+/*
+ * IPU Sensor Multiple FIFO Controller (SMFC) functions
+ */
+int ipu_smfc_enable(struct ipu_soc *ipu);
+int ipu_smfc_disable(struct ipu_soc *ipu);
+int ipu_smfc_map_channel(struct ipu_soc *ipu, int channel, int csi_id, int mipi_id);
+int ipu_smfc_set_burstsize(struct ipu_soc *ipu, int channel, int burstsize);
+
 #define IPU_CPMEM_WORD(word, ofs, size) ((((word) * 160 + (ofs)) << 8) | (size))
 
 #define IPU_FIELD_UBO		IPU_CPMEM_WORD(0, 46, 22)
@@ -321,6 +336,7 @@ static inline void ipu_cpmem_set_burstsize(struct ipu_ch_param __iomem *p,
 };
 
 struct ipu_client_platformdata {
+	int csi;
 	int di;
 	int dc;
 	int dp;
diff --git a/lib/Kconfig b/lib/Kconfig
index 4771fb3f4da4..334f7722a999 100644
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -331,6 +331,20 @@ config TEXTSEARCH_FSM
 config BTREE
 	boolean
 
+config INTERVAL_TREE
+	boolean
+	help
+	  Simple, embeddable, interval-tree. Can find the start of an
+	  overlapping range in log(n) time and then iterate over all
+	  overlapping nodes. The algorithm is implemented as an
+	  augmented rbtree.
+
+	  See:
+
+		Documentation/rbtree.txt
+
+	  for more information.
+
 config ASSOCIATIVE_ARRAY
 	bool
 	help
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index ccca32264748..e34d11d70bbf 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -1511,6 +1511,7 @@ config RBTREE_TEST
 config INTERVAL_TREE_TEST
 	tristate "Interval tree test"
 	depends on m && DEBUG_KERNEL
+	select INTERVAL_TREE
 	help
 	  A benchmark measuring the performance of the interval tree library
 
diff --git a/lib/Makefile b/lib/Makefile
index 74a32dc49a93..4a4078987a4c 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -50,6 +50,7 @@ CFLAGS_hweight.o = $(subst $(quote),,$(CONFIG_ARCH_HWEIGHT_CFLAGS))
 obj-$(CONFIG_GENERIC_HWEIGHT) += hweight.o
 
 obj-$(CONFIG_BTREE) += btree.o
+obj-$(CONFIG_INTERVAL_TREE) += interval_tree.o
 obj-$(CONFIG_ASSOCIATIVE_ARRAY) += assoc_array.o
 obj-$(CONFIG_DEBUG_PREEMPT) += smp_processor_id.o
 obj-$(CONFIG_DEBUG_LIST) += list_debug.o
@@ -157,8 +158,6 @@ lib-$(CONFIG_LIBFDT) += $(libfdt_files)
 obj-$(CONFIG_RBTREE_TEST) += rbtree_test.o
 obj-$(CONFIG_INTERVAL_TREE_TEST) += interval_tree_test.o
 
-interval_tree_test-objs := interval_tree_test_main.o interval_tree.o
-
 obj-$(CONFIG_PERCPU_TEST) += percpu_test.o
 
 obj-$(CONFIG_ASN1) += asn1_decoder.o
diff --git a/lib/interval_tree.c b/lib/interval_tree.c
index e6eb406f2d65..f367f9ad544c 100644
--- a/lib/interval_tree.c
+++ b/lib/interval_tree.c
@@ -1,6 +1,7 @@
 #include <linux/init.h>
 #include <linux/interval_tree.h>
 #include <linux/interval_tree_generic.h>
+#include <linux/module.h>
 
 #define START(node) ((node)->start)
 #define LAST(node)  ((node)->last)
@@ -8,3 +9,8 @@
 INTERVAL_TREE_DEFINE(struct interval_tree_node, rb,
 		     unsigned long, __subtree_last,
 		     START, LAST,, interval_tree)
+
+EXPORT_SYMBOL_GPL(interval_tree_insert);
+EXPORT_SYMBOL_GPL(interval_tree_remove);
+EXPORT_SYMBOL_GPL(interval_tree_iter_first);
+EXPORT_SYMBOL_GPL(interval_tree_iter_next);
diff --git a/lib/interval_tree_test_main.c b/lib/interval_tree_test.c
index 245900b98c8e..245900b98c8e 100644
--- a/lib/interval_tree_test_main.c
+++ b/lib/interval_tree_test.c