BackMerge tag 'v4.12-rc5' into drm-next

Linux 4.12-rc5 for nouveau fixes

19 files changed, 166 insertions, 91 deletions

diff --git a/drivers/gpu/drm/i915/gvt/execlist.c b/drivers/gpu/drm/i915/gvt/execlist.c
index dca989eb2d42..24fe04d6307b 100644
--- a/drivers/gpu/drm/i915/gvt/execlist.c
+++ b/drivers/gpu/drm/i915/gvt/execlist.c
@@ -779,8 +779,26 @@ static void init_vgpu_execlist(struct intel_vgpu *vgpu, int ring_id)
 	vgpu_vreg(vgpu, ctx_status_ptr_reg) = ctx_status_ptr.dw;
 }
 
+static void clean_workloads(struct intel_vgpu *vgpu, unsigned long engine_mask)
+{
+	struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
+	struct intel_engine_cs *engine;
+	struct intel_vgpu_workload *pos, *n;
+	unsigned int tmp;
+
+	/* free the unsubmited workloads in the queues. */
+	for_each_engine_masked(engine, dev_priv, engine_mask, tmp) {
+		list_for_each_entry_safe(pos, n,
+			&vgpu->workload_q_head[engine->id], list) {
+			list_del_init(&pos->list);
+			free_workload(pos);
+		}
+	}
+}
+
 void intel_vgpu_clean_execlist(struct intel_vgpu *vgpu)
 {
+	clean_workloads(vgpu, ALL_ENGINES);
 	kmem_cache_destroy(vgpu->workloads);
 }
 
@@ -811,17 +829,9 @@ void intel_vgpu_reset_execlist(struct intel_vgpu *vgpu,
 {
 	struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
 	struct intel_engine_cs *engine;
-	struct intel_vgpu_workload *pos, *n;
 	unsigned int tmp;
 
-	for_each_engine_masked(engine, dev_priv, engine_mask, tmp) {
-		/* free the unsubmited workload in the queue */
-		list_for_each_entry_safe(pos, n,
-			&vgpu->workload_q_head[engine->id], list) {
-			list_del_init(&pos->list);
-			free_workload(pos);
-		}
-
+	clean_workloads(vgpu, engine_mask);
+	for_each_engine_masked(engine, dev_priv, engine_mask, tmp)
 		init_vgpu_execlist(vgpu, engine->id);
-	}
 }
diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c
index c995e540ff96..0ffd69654592 100644
--- a/drivers/gpu/drm/i915/gvt/handlers.c
+++ b/drivers/gpu/drm/i915/gvt/handlers.c
@@ -1366,18 +1366,28 @@ static int skl_misc_ctl_write(struct intel_vgpu *vgpu, unsigned int offset,
 		void *p_data, unsigned int bytes)
 {
 	struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
-	i915_reg_t reg = {.reg = offset};
+	u32 v = *(u32 *)p_data;
+
+	if (!IS_SKYLAKE(dev_priv) && !IS_KABYLAKE(dev_priv))
+		return intel_vgpu_default_mmio_write(vgpu,
+				offset, p_data, bytes);
 
 	switch (offset) {
 	case 0x4ddc:
-		vgpu_vreg(vgpu, offset) = 0x8000003c;
-		/* WaCompressedResourceSamplerPbeMediaNewHashMode:skl */
-		I915_WRITE(reg, vgpu_vreg(vgpu, offset));
+		/* bypass WaCompressedResourceSamplerPbeMediaNewHashMode */
+		vgpu_vreg(vgpu, offset) = v & ~(1 << 31);
 		break;
 	case 0x42080:
-		vgpu_vreg(vgpu, offset) = 0x8000;
-		/* WaCompressedResourceDisplayNewHashMode:skl */
-		I915_WRITE(reg, vgpu_vreg(vgpu, offset));
+		/* bypass WaCompressedResourceDisplayNewHashMode */
+		vgpu_vreg(vgpu, offset) = v & ~(1 << 15);
+		break;
+	case 0xe194:
+		/* bypass WaCompressedResourceSamplerPbeMediaNewHashMode */
+		vgpu_vreg(vgpu, offset) = v & ~(1 << 8);
+		break;
+	case 0x7014:
+		/* bypass WaCompressedResourceSamplerPbeMediaNewHashMode */
+		vgpu_vreg(vgpu, offset) = v & ~(1 << 13);
 		break;
 	default:
 		return -EINVAL;
@@ -1634,7 +1644,8 @@ static int init_generic_mmio_info(struct intel_gvt *gvt)
 	MMIO_DFH(GAM_ECOCHK, D_ALL, F_CMD_ACCESS, NULL, NULL);
 	MMIO_DFH(GEN7_COMMON_SLICE_CHICKEN1, D_ALL, F_MODE_MASK | F_CMD_ACCESS,
 		NULL, NULL);
-	MMIO_DFH(COMMON_SLICE_CHICKEN2, D_ALL, F_MODE_MASK | F_CMD_ACCESS, NULL, NULL);
+	MMIO_DFH(COMMON_SLICE_CHICKEN2, D_ALL, F_MODE_MASK | F_CMD_ACCESS, NULL,
+		 skl_misc_ctl_write);
 	MMIO_DFH(0x9030, D_ALL, F_CMD_ACCESS, NULL, NULL);
 	MMIO_DFH(0x20a0, D_ALL, F_CMD_ACCESS, NULL, NULL);
 	MMIO_DFH(0x2420, D_ALL, F_CMD_ACCESS, NULL, NULL);
@@ -2568,7 +2579,8 @@ static int init_broadwell_mmio_info(struct intel_gvt *gvt)
 	MMIO_D(0x6e570, D_BDW_PLUS);
 	MMIO_D(0x65f10, D_BDW_PLUS);
 
-	MMIO_DFH(0xe194, D_BDW_PLUS, F_MODE_MASK | F_CMD_ACCESS, NULL, NULL);
+	MMIO_DFH(0xe194, D_BDW_PLUS, F_MODE_MASK | F_CMD_ACCESS, NULL,
+		 skl_misc_ctl_write);
 	MMIO_DFH(0xe188, D_BDW_PLUS, F_MODE_MASK | F_CMD_ACCESS, NULL, NULL);
 	MMIO_DFH(HALF_SLICE_CHICKEN2, D_BDW_PLUS, F_MODE_MASK | F_CMD_ACCESS, NULL, NULL);
 	MMIO_DFH(0x2580, D_BDW_PLUS, F_MODE_MASK | F_CMD_ACCESS, NULL, NULL);
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 7b8c72776f46..d703897786e9 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -1240,6 +1240,15 @@ int i915_driver_load(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto out_fini;
 
 	pci_set_drvdata(pdev, &dev_priv->drm);
+	/*
+	 * Disable the system suspend direct complete optimization, which can
+	 * leave the device suspended skipping the driver's suspend handlers
+	 * if the device was already runtime suspended. This is needed due to
+	 * the difference in our runtime and system suspend sequence and
+	 * becaue the HDA driver may require us to enable the audio power
+	 * domain during system suspend.
+	 */
+	pdev->dev_flags |= PCI_DEV_FLAGS_NEEDS_RESUME;
 
 	ret = i915_driver_init_early(dev_priv, ent);
 	if (ret < 0)
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 35e161b5b90e..35f71b559808 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -611,7 +611,8 @@ struct intel_link_m_n {
 
 void intel_link_compute_m_n(int bpp, int nlanes,
 			    int pixel_clock, int link_clock,
-			    struct intel_link_m_n *m_n);
+			    struct intel_link_m_n *m_n,
+			    bool reduce_m_n);
 
 /* Interface history:
  *
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 7ab47a84671f..7b676fd1f075 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -3348,6 +3348,10 @@ int i915_gem_wait_for_idle(struct drm_i915_private *i915, unsigned int flags)
 {
 	int ret;
 
+	/* If the device is asleep, we have no requests outstanding */
+	if (!READ_ONCE(i915->gt.awake))
+		return 0;
+
 	if (flags & I915_WAIT_LOCKED) {
 		struct i915_gem_timeline *tl;
 
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 0c1008a2bbda..1489c3af7145 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -3100,7 +3100,8 @@ void i915_ggtt_enable_guc(struct drm_i915_private *i915)
 
 void i915_ggtt_disable_guc(struct drm_i915_private *i915)
 {
-	i915->ggtt.invalidate = gen6_ggtt_invalidate;
+	if (i915->ggtt.invalidate == guc_ggtt_invalidate)
+		i915->ggtt.invalidate = gen6_ggtt_invalidate;
 }
 
 void i915_gem_restore_gtt_mappings(struct drm_i915_private *dev_priv)
diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c b/drivers/gpu/drm/i915/i915_gem_tiling.c
index a0d6d4317a49..fb5231f98c0d 100644
--- a/drivers/gpu/drm/i915/i915_gem_tiling.c
+++ b/drivers/gpu/drm/i915/i915_gem_tiling.c
@@ -278,7 +278,7 @@ i915_gem_object_set_tiling(struct drm_i915_gem_object *obj,
 			obj->mm.quirked = false;
 		}
 		if (!i915_gem_object_is_tiled(obj)) {
-			GEM_BUG_ON(!obj->mm.quirked);
+			GEM_BUG_ON(obj->mm.quirked);
 			__i915_gem_object_pin_pages(obj);
 			obj->mm.quirked = true;
 		}
diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c
index f80db2ccd92f..f940e486a62a 100644
--- a/drivers/gpu/drm/i915/i915_pci.c
+++ b/drivers/gpu/drm/i915/i915_pci.c
@@ -208,7 +208,7 @@ static const struct intel_device_info intel_ironlake_d_info = {
 static const struct intel_device_info intel_ironlake_m_info = {
 	GEN5_FEATURES,
 	.platform = INTEL_IRONLAKE,
-	.is_mobile = 1,
+	.is_mobile = 1, .has_fbc = 1,
 };
 
 #define GEN6_FEATURES \
@@ -385,7 +385,6 @@ static const struct intel_device_info intel_skylake_gt3_info = {
 	.has_gmbus_irq = 1, \
 	.has_logical_ring_contexts = 1, \
 	.has_guc = 1, \
-	.has_decoupled_mmio = 1, \
 	.has_aliasing_ppgtt = 1, \
 	.has_full_ppgtt = 1, \
 	.has_full_48bit_ppgtt = 1, \
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 92cfcae7b3d8..e59ad8e20256 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -6116,7 +6116,7 @@ retry:
 	pipe_config->fdi_lanes = lane;
 
 	intel_link_compute_m_n(pipe_config->pipe_bpp, lane, fdi_dotclock,
-			       link_bw, &pipe_config->fdi_m_n);
+			       link_bw, &pipe_config->fdi_m_n, false);
 
 	ret = ironlake_check_fdi_lanes(dev, intel_crtc->pipe, pipe_config);
 	if (ret == -EINVAL && pipe_config->pipe_bpp > 6*3) {
@@ -6292,7 +6292,8 @@ intel_reduce_m_n_ratio(uint32_t *num, uint32_t *den)
 }
 
 static void compute_m_n(unsigned int m, unsigned int n,
-			uint32_t *ret_m, uint32_t *ret_n)
+			uint32_t *ret_m, uint32_t *ret_n,
+			bool reduce_m_n)
 {
 	/*
 	 * Reduce M/N as much as possible without loss in precision. Several DP
@@ -6300,9 +6301,11 @@ static void compute_m_n(unsigned int m, unsigned int n,
 	 * values. The passed in values are more likely to have the least
 	 * significant bits zero than M after rounding below, so do this first.
 	 */
-	while ((m & 1) == 0 && (n & 1) == 0) {
-		m >>= 1;
-		n >>= 1;
+	if (reduce_m_n) {
+		while ((m & 1) == 0 && (n & 1) == 0) {
+			m >>= 1;
+			n >>= 1;
+		}
 	}
 
 	*ret_n = min_t(unsigned int, roundup_pow_of_two(n), DATA_LINK_N_MAX);
@@ -6313,16 +6316,19 @@ static void compute_m_n(unsigned int m, unsigned int n,
 void
 intel_link_compute_m_n(int bits_per_pixel, int nlanes,
 		       int pixel_clock, int link_clock,
-		       struct intel_link_m_n *m_n)
+		       struct intel_link_m_n *m_n,
+		       bool reduce_m_n)
 {
 	m_n->tu = 64;
 
 	compute_m_n(bits_per_pixel * pixel_clock,
 		    link_clock * nlanes * 8,
-		    &m_n->gmch_m, &m_n->gmch_n);
+		    &m_n->gmch_m, &m_n->gmch_n,
+		    reduce_m_n);
 
 	compute_m_n(pixel_clock, link_clock,
-		    &m_n->link_m, &m_n->link_n);
+		    &m_n->link_m, &m_n->link_n,
+		    reduce_m_n);
 }
 
 static inline bool intel_panel_use_ssc(struct drm_i915_private *dev_priv)
@@ -12384,6 +12390,15 @@ static void update_scanline_offset(struct intel_crtc *crtc)
 	 * 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.
+	 *
+	 * On VLV/CHV DSI the scanline counter would appear to increment
+	 * approx. 1/3 of a scanline before start of vblank. Unfortunately
+	 * that means we can't tell whether we're in vblank or not while
+	 * we're on that particular line. We must still set scanline_offset
+	 * to 1 so that the vblank timestamps come out correct when we query
+	 * the scanline counter from within the vblank interrupt handler.
+	 * However if queried just before the start of vblank we'll get an
+	 * answer that's slightly in the future.
 	 */
 	if (IS_GEN2(dev_priv)) {
 		const struct drm_display_mode *adjusted_mode = &crtc->config->base.adjusted_mode;
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 4a6feb6a69bd..de4b1f2d367d 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -1548,35 +1548,15 @@ static void intel_dp_print_rates(struct intel_dp *intel_dp)
 	DRM_DEBUG_KMS("common rates: %s\n", str);
 }
 
-bool
-__intel_dp_read_desc(struct intel_dp *intel_dp, struct intel_dp_desc *desc)
-{
-	u32 base = drm_dp_is_branch(intel_dp->dpcd) ? DP_BRANCH_OUI :
-						      DP_SINK_OUI;
-
-	return drm_dp_dpcd_read(&intel_dp->aux, base, desc, sizeof(*desc)) ==
-	       sizeof(*desc);
-}
-
-bool intel_dp_read_desc(struct intel_dp *intel_dp)
+static int rate_to_index(int find, const int *rates)
 {
-	struct intel_dp_desc *desc = &intel_dp->desc;
-	bool oui_sup = intel_dp->dpcd[DP_DOWN_STREAM_PORT_COUNT] &
-		       DP_OUI_SUPPORT;
-	int dev_id_len;
+	int i = 0;
 
-	if (!__intel_dp_read_desc(intel_dp, desc))
-		return false;
-
-	dev_id_len = strnlen(desc->device_id, sizeof(desc->device_id));
-	DRM_DEBUG_KMS("DP %s: OUI %*phD%s dev-ID %*pE HW-rev %d.%d SW-rev %d.%d\n",
-		      drm_dp_is_branch(intel_dp->dpcd) ? "branch" : "sink",
-		      (int)sizeof(desc->oui), desc->oui, oui_sup ? "" : "(NS)",
-		      dev_id_len, desc->device_id,
-		      desc->hw_rev >> 4, desc->hw_rev & 0xf,
-		      desc->sw_major_rev, desc->sw_minor_rev);
+	for (i = 0; i < DP_MAX_SUPPORTED_RATES; ++i)
+		if (find == rates[i])
+			break;
 
-	return true;
+	return i;
 }
 
 int
@@ -1658,6 +1638,8 @@ intel_dp_compute_config(struct intel_encoder *encoder,
 	int link_avail, link_clock;
 	int common_len;
 	uint8_t link_bw, rate_select;
+	bool reduce_m_n = drm_dp_has_quirk(&intel_dp->desc,
+					   DP_DPCD_QUIRK_LIMITED_M_N);
 
 	common_len = intel_dp_common_len_rate_limit(intel_dp,
 						    intel_dp->max_link_rate);
@@ -1790,7 +1772,8 @@ found:
 	intel_link_compute_m_n(bpp, lane_count,
 			       adjusted_mode->crtc_clock,
 			       pipe_config->port_clock,
-			       &pipe_config->dp_m_n);
+			       &pipe_config->dp_m_n,
+			       reduce_m_n);
 
 	if (intel_connector->panel.downclock_mode != NULL &&
 		dev_priv->drrs.type == SEAMLESS_DRRS_SUPPORT) {
@@ -1798,7 +1781,8 @@ found:
 			intel_link_compute_m_n(bpp, lane_count,
 				intel_connector->panel.downclock_mode->clock,
 				pipe_config->port_clock,
-				&pipe_config->dp_m2_n2);
+				&pipe_config->dp_m2_n2,
+				reduce_m_n);
 	}
 
 	/*
@@ -3662,7 +3646,8 @@ intel_edp_init_dpcd(struct intel_dp *intel_dp)
 	if (!intel_dp_read_dpcd(intel_dp))
 		return false;
 
-	intel_dp_read_desc(intel_dp);
+	drm_dp_read_desc(&intel_dp->aux, &intel_dp->desc,
+			 drm_dp_is_branch(intel_dp->dpcd));
 
 	if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11)
 		dev_priv->no_aux_handshake = intel_dp->dpcd[DP_MAX_DOWNSPREAD] &
@@ -4677,7 +4662,8 @@ intel_dp_long_pulse(struct intel_connector *intel_connector)
 
 	intel_dp_print_rates(intel_dp);
 
-	intel_dp_read_desc(intel_dp);
+	drm_dp_read_desc(&intel_dp->aux, &intel_dp->desc,
+			 drm_dp_is_branch(intel_dp->dpcd));
 
 	intel_dp_configure_mst(intel_dp);
 
diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c
index 3715386e4272..2cf046beae0f 100644
--- a/drivers/gpu/drm/i915/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/intel_dp_mst.c
@@ -44,6 +44,8 @@ static bool intel_dp_mst_compute_config(struct intel_encoder *encoder,
 	int lane_count, slots;
 	const struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode;
 	int mst_pbn;
+	bool reduce_m_n = drm_dp_has_quirk(&intel_dp->desc,
+					   DP_DPCD_QUIRK_LIMITED_M_N);
 
 	pipe_config->has_pch_encoder = false;
 	bpp = 24;
@@ -80,7 +82,8 @@ static bool intel_dp_mst_compute_config(struct intel_encoder *encoder,
 	intel_link_compute_m_n(bpp, lane_count,
 			       adjusted_mode->crtc_clock,
 			       pipe_config->port_clock,
-			       &pipe_config->dp_m_n);
+			       &pipe_config->dp_m_n,
+			       reduce_m_n);
 
 	pipe_config->dp_m_n.tu = slots;
 
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index bd500977b3fc..71f94a01aedd 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -938,14 +938,6 @@ enum link_m_n_set {
 	M2_N2
 };
 
-struct intel_dp_desc {
-	u8 oui[3];
-	u8 device_id[6];
-	u8 hw_rev;
-	u8 sw_major_rev;
-	u8 sw_minor_rev;
-} __packed;
-
 struct intel_dp_compliance_data {
 	unsigned long edid;
 	uint8_t video_pattern;
@@ -996,7 +988,7 @@ struct intel_dp {
 	/* Max rate for the current link */
 	int max_link_rate;
 	/* sink or branch descriptor */
-	struct intel_dp_desc desc;
+	struct drm_dp_desc desc;
 	struct drm_dp_aux aux;
 	enum intel_display_power_domain aux_power_domain;
 	uint8_t train_set[4];
@@ -1571,9 +1563,6 @@ static inline unsigned int intel_dp_unused_lane_mask(int lane_count)
 }
 
 bool intel_dp_read_dpcd(struct intel_dp *intel_dp);
-bool __intel_dp_read_desc(struct intel_dp *intel_dp,
-			  struct intel_dp_desc *desc);
-bool intel_dp_read_desc(struct intel_dp *intel_dp);
 int intel_dp_link_required(int pixel_clock, int bpp);
 int intel_dp_max_data_rate(int max_link_clock, int max_lanes);
 bool intel_digital_port_connected(struct drm_i915_private *dev_priv,
diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c b/drivers/gpu/drm/i915/intel_engine_cs.c
index 413bfd8d4bf4..699f2d3861c7 100644
--- a/drivers/gpu/drm/i915/intel_engine_cs.c
+++ b/drivers/gpu/drm/i915/intel_engine_cs.c
@@ -1205,6 +1205,22 @@ int intel_ring_workarounds_emit(struct drm_i915_gem_request *req)
 	return 0;
 }
 
+static bool ring_is_idle(struct intel_engine_cs *engine)
+{
+	struct drm_i915_private *dev_priv = engine->i915;
+	bool idle = true;
+
+	intel_runtime_pm_get(dev_priv);
+
+	/* No bit for gen2, so assume the CS parser is idle */
+	if (INTEL_GEN(dev_priv) > 2 && !(I915_READ_MODE(engine) & MODE_IDLE))
+		idle = false;
+
+	intel_runtime_pm_put(dev_priv);
+
+	return idle;
+}
+
 /**
  * intel_engine_is_idle() - Report if the engine has finished process all work
  * @engine: the intel_engine_cs
@@ -1237,7 +1253,7 @@ bool intel_engine_is_idle(struct intel_engine_cs *engine)
 		return false;
 
 	/* Ring stopped? */
-	if (INTEL_GEN(dev_priv) > 2 && !(I915_READ_MODE(engine) & MODE_IDLE))
+	if (!ring_is_idle(engine))
 		return false;
 
 	return true;
diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c
index ff2fc5bc4af4..860b8c26d29b 100644
--- a/drivers/gpu/drm/i915/intel_fbc.c
+++ b/drivers/gpu/drm/i915/intel_fbc.c
@@ -82,20 +82,10 @@ static unsigned int get_crtc_fence_y_offset(struct intel_crtc *crtc)
 static void intel_fbc_get_plane_source_size(struct intel_fbc_state_cache *cache,
 					    int *width, int *height)
 {
-	int w, h;
-
-	if (drm_rotation_90_or_270(cache->plane.rotation)) {
-		w = cache->plane.src_h;
-		h = cache->plane.src_w;
-	} else {
-		w = cache->plane.src_w;
-		h = cache->plane.src_h;
-	}
-
 	if (width)
-		*width = w;
+		*width = cache->plane.src_w;
 	if (height)
-		*height = h;
+		*height = cache->plane.src_h;
 }
 
 static int intel_fbc_calculate_cfb_size(struct drm_i915_private *dev_priv,
@@ -746,6 +736,11 @@ static void intel_fbc_update_state_cache(struct intel_crtc *crtc,
 		cache->crtc.hsw_bdw_pixel_rate = crtc_state->pixel_rate;
 
 	cache->plane.rotation = plane_state->base.rotation;
+	/*
+	 * Src coordinates are already rotated by 270 degrees for
+	 * the 90/270 degree plane rotation cases (to match the
+	 * GTT mapping), hence no need to account for rotation here.
+	 */
 	cache->plane.src_w = drm_rect_width(&plane_state->base.src) >> 16;
 	cache->plane.src_h = drm_rect_height(&plane_state->base.src) >> 16;
 	cache->plane.visible = plane_state->base.visible;
diff --git a/drivers/gpu/drm/i915/intel_lspcon.c b/drivers/gpu/drm/i915/intel_lspcon.c
index 71cbe9c08932..5abef482eacf 100644
--- a/drivers/gpu/drm/i915/intel_lspcon.c
+++ b/drivers/gpu/drm/i915/intel_lspcon.c
@@ -240,7 +240,7 @@ bool lspcon_init(struct intel_digital_port *intel_dig_port)
 		return false;
 	}
 
-	intel_dp_read_desc(dp);
+	drm_dp_read_desc(&dp->aux, &dp->desc, drm_dp_is_branch(dp->dpcd));
 
 	DRM_DEBUG_KMS("Success: LSPCON init\n");
 	return true;
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 936eef1634c7..fce4bc5ccc99 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -4819,11 +4819,19 @@ skl_compute_wm(struct drm_atomic_state *state)
 	struct drm_crtc_state *cstate;
 	struct intel_atomic_state *intel_state = to_intel_atomic_state(state);
 	struct skl_wm_values *results = &intel_state->wm_results;
+	struct drm_device *dev = state->dev;
 	struct skl_pipe_wm *pipe_wm;
 	bool changed = false;
 	int ret, i;
 
 	/*
+	 * When we distrust bios wm we always need to recompute to set the
+	 * expected DDB allocations for each CRTC.
+	 */
+	if (to_i915(dev)->wm.distrust_bios_wm)
+		changed = true;
+
+	/*
 	 * If this transaction isn't actually touching any CRTC's, don't
 	 * bother with watermark calculation.  Note that if we pass this
 	 * test, we're guaranteed to hold at least one CRTC state mutex,
@@ -4833,6 +4841,7 @@ skl_compute_wm(struct drm_atomic_state *state)
 	 */
 	for_each_new_crtc_in_state(state, crtc, cstate, i)
 		changed = true;
+
 	if (!changed)
 		return 0;
 
diff --git a/drivers/gpu/drm/i915/intel_psr.c b/drivers/gpu/drm/i915/intel_psr.c
index c3780d0d2baf..559f1ab42bfc 100644
--- a/drivers/gpu/drm/i915/intel_psr.c
+++ b/drivers/gpu/drm/i915/intel_psr.c
@@ -435,8 +435,9 @@ static bool intel_psr_match_conditions(struct intel_dp *intel_dp)
 	}
 
 	/* PSR2 is restricted to work with panel resolutions upto 3200x2000 */
-	if (intel_crtc->config->pipe_src_w > 3200 ||
-				intel_crtc->config->pipe_src_h > 2000) {
+	if (dev_priv->psr.psr2_support &&
+	    (intel_crtc->config->pipe_src_w > 3200 ||
+	     intel_crtc->config->pipe_src_h > 2000)) {
 		dev_priv->psr.psr2_support = false;
 		return false;
 	}
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index c4bf19364e49..0c650c2cbca8 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -83,10 +83,13 @@ int intel_usecs_to_scanlines(const struct drm_display_mode *adjusted_mode,
  */
 void intel_pipe_update_start(struct intel_crtc *crtc)
 {
+	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
 	const struct drm_display_mode *adjusted_mode = &crtc->config->base.adjusted_mode;
 	long timeout = msecs_to_jiffies_timeout(1);
 	int scanline, min, max, vblank_start;
 	wait_queue_head_t *wq = drm_crtc_vblank_waitqueue(&crtc->base);
+	bool need_vlv_dsi_wa = (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) &&
+		intel_crtc_has_type(crtc->config, INTEL_OUTPUT_DSI);
 	DEFINE_WAIT(wait);
 
 	vblank_start = adjusted_mode->crtc_vblank_start;
@@ -139,6 +142,24 @@ void intel_pipe_update_start(struct intel_crtc *crtc)
 
 	drm_crtc_vblank_put(&crtc->base);
 
+	/*
+	 * On VLV/CHV DSI the scanline counter would appear to
+	 * increment approx. 1/3 of a scanline before start of vblank.
+	 * The registers still get latched at start of vblank however.
+	 * This means we must not write any registers on the first
+	 * line of vblank (since not the whole line is actually in
+	 * vblank). And unfortunately we can't use the interrupt to
+	 * wait here since it will fire too soon. We could use the
+	 * frame start interrupt instead since it will fire after the
+	 * critical scanline, but that would require more changes
+	 * in the interrupt code. So for now we'll just do the nasty
+	 * thing and poll for the bad scanline to pass us by.
+	 *
+	 * FIXME figure out if BXT+ DSI suffers from this as well
+	 */
+	while (need_vlv_dsi_wa && scanline == vblank_start)
+		scanline = intel_get_crtc_scanline(crtc);
+
 	crtc->debug.scanline_start = scanline;
 	crtc->debug.start_vbl_time = ktime_get();
 	crtc->debug.start_vbl_count = intel_crtc_get_vblank_counter(crtc);
diff --git a/drivers/gpu/drm/i915/intel_uc.h b/drivers/gpu/drm/i915/intel_uc.h
index 69daf4c01cd0..22ae52b17b0f 100644
--- a/drivers/gpu/drm/i915/intel_uc.h
+++ b/drivers/gpu/drm/i915/intel_uc.h
@@ -59,6 +59,10 @@ struct drm_i915_gem_request;
  *                available in the work queue (note, the queue is shared,
  *                not per-engine). It is OK for this to be nonzero, but
  *                it should not be huge!
+ *   b_fail: failed to ring the doorbell. This should never happen, unless
+ *           somehow the hardware misbehaves, or maybe if the GuC firmware
+ *           crashes? We probably need to reset the GPU to recover.
+ *   retcode: errno from last guc_submit()
  */
 struct i915_guc_client {
 	struct i915_vma *vma;