From 849620fab413355eff48232eac5a8c53c57615c5 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Thu, 14 May 2009 17:10:52 +0200
Subject: Revert "oprofile: discover counters for op ppro too"

This reverts commit 59512900baab03c5629f2ff5efad1d5d4e682ece.

arch_perfmon_setup_counters() is actually never called for ppro, so
there is no code that changes the numbers in op_ppro_spec. The patch
as it is has no effect.

Cc: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_ppro.c | 8 +++-----
 arch/x86/oprofile/op_x86_model.h  | 2 +-
 2 files changed, 4 insertions(+), 6 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c
index 10131fbdaada..2a123990a84c 100644
--- a/arch/x86/oprofile/op_model_ppro.c
+++ b/arch/x86/oprofile/op_model_ppro.c
@@ -213,9 +213,9 @@ static void ppro_shutdown(struct op_msrs const * const msrs)
 }
 
 
-struct op_x86_model_spec op_ppro_spec = {
-	.num_counters		= 2,	/* can be overriden */
-	.num_controls		= 2,	/* dito */
+struct op_x86_model_spec const op_ppro_spec = {
+	.num_counters		= 2,
+	.num_controls		= 2,
 	.fill_in_addresses	= &ppro_fill_in_addresses,
 	.setup_ctrs		= &ppro_setup_ctrs,
 	.check_ctrs		= &ppro_check_ctrs,
@@ -251,8 +251,6 @@ void arch_perfmon_setup_counters(void)
 
 	op_arch_perfmon_spec.num_counters = num_counters;
 	op_arch_perfmon_spec.num_controls = num_counters;
-	op_ppro_spec.num_counters = num_counters;
-	op_ppro_spec.num_controls = num_counters;
 }
 
 struct op_x86_model_spec op_arch_perfmon_spec = {
diff --git a/arch/x86/oprofile/op_x86_model.h b/arch/x86/oprofile/op_x86_model.h
index 825e79064d64..2317149c94f0 100644
--- a/arch/x86/oprofile/op_x86_model.h
+++ b/arch/x86/oprofile/op_x86_model.h
@@ -45,7 +45,7 @@ struct op_x86_model_spec {
 	void (*shutdown)(struct op_msrs const * const msrs);
 };
 
-extern struct op_x86_model_spec op_ppro_spec;
+extern struct op_x86_model_spec const op_ppro_spec;
 extern struct op_x86_model_spec const op_p4_spec;
 extern struct op_x86_model_spec const op_p4_ht2_spec;
 extern struct op_x86_model_spec const op_amd_spec;
-- 
cgit v1.2.3


From e419294ed3c98cccc145202e4fe165bfd8099d63 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Sun, 12 Oct 2008 15:12:34 -0400
Subject: x86/oprofile: moving arch_perfmon counter setup to
 op_x86_model_spec.init

The function arch_perfmon_init() in nmi_int.c is model specific. This
patch moves it to op_model_ppro.c by using the init function pointer
in struct op_x86_model_spec.

Cc: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c       | 21 +++++++++------------
 arch/x86/oprofile/op_model_ppro.c |  9 ++++++++-
 arch/x86/oprofile/op_x86_model.h  |  2 --
 3 files changed, 17 insertions(+), 15 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index 3b285e656e27..dd8515301fbf 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -427,7 +427,7 @@ static int __init ppro_init(char **cpu_type)
 		*cpu_type = "i386/core_2";
 		break;
 	case 26:
-		arch_perfmon_setup_counters();
+		model = &op_arch_perfmon_spec;
 		*cpu_type = "i386/core_i7";
 		break;
 	case 28:
@@ -442,16 +442,6 @@ static int __init ppro_init(char **cpu_type)
 	return 1;
 }
 
-static int __init arch_perfmon_init(char **cpu_type)
-{
-	if (!cpu_has_arch_perfmon)
-		return 0;
-	*cpu_type = "i386/arch_perfmon";
-	model = &op_arch_perfmon_spec;
-	arch_perfmon_setup_counters();
-	return 1;
-}
-
 /* in order to get sysfs right */
 static int using_nmi;
 
@@ -509,8 +499,15 @@ int __init op_nmi_init(struct oprofile_operations *ops)
 			break;
 		}
 
-		if (!cpu_type && !arch_perfmon_init(&cpu_type))
+		if (cpu_type)
+			break;
+
+		if (!cpu_has_arch_perfmon)
 			return -ENODEV;
+
+		/* use arch perfmon as fallback */
+		cpu_type = "i386/arch_perfmon";
+		model = &op_arch_perfmon_spec;
 		break;
 
 	default:
diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c
index 2a123990a84c..ae5811966883 100644
--- a/arch/x86/oprofile/op_model_ppro.c
+++ b/arch/x86/oprofile/op_model_ppro.c
@@ -233,7 +233,7 @@ struct op_x86_model_spec const op_ppro_spec = {
  * the specific CPU.
  */
 
-void arch_perfmon_setup_counters(void)
+static void arch_perfmon_setup_counters(void)
 {
 	union cpuid10_eax eax;
 
@@ -253,7 +253,14 @@ void arch_perfmon_setup_counters(void)
 	op_arch_perfmon_spec.num_controls = num_counters;
 }
 
+static int arch_perfmon_init(struct oprofile_operations *ignore)
+{
+	arch_perfmon_setup_counters();
+	return 0;
+}
+
 struct op_x86_model_spec op_arch_perfmon_spec = {
+	.init			= &arch_perfmon_init,
 	/* num_counters/num_controls filled in at runtime */
 	.fill_in_addresses	= &ppro_fill_in_addresses,
 	/* user space does the cpuid check for available events */
diff --git a/arch/x86/oprofile/op_x86_model.h b/arch/x86/oprofile/op_x86_model.h
index 2317149c94f0..ed27783bb0d4 100644
--- a/arch/x86/oprofile/op_x86_model.h
+++ b/arch/x86/oprofile/op_x86_model.h
@@ -51,6 +51,4 @@ extern struct op_x86_model_spec const op_p4_ht2_spec;
 extern struct op_x86_model_spec const op_amd_spec;
 extern struct op_x86_model_spec op_arch_perfmon_spec;
 
-extern void arch_perfmon_setup_counters(void);
-
 #endif /* OP_X86_MODEL_H */
-- 
cgit v1.2.3


From 06552ccc36abeb12e37efc16c384dc7f30794f85 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Thu, 28 May 2009 02:12:36 +0200
Subject: x86/oprofile: minor style changes in struct op_x86_model_spec

Some vertical alignments. Variables are now located in the beginning
of the struct.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_x86_model.h | 22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_x86_model.h b/arch/x86/oprofile/op_x86_model.h
index ed27783bb0d4..bd8157d12ff0 100644
--- a/arch/x86/oprofile/op_x86_model.h
+++ b/arch/x86/oprofile/op_x86_model.h
@@ -32,17 +32,17 @@ struct pt_regs;
  * various x86 CPU models' perfctr support.
  */
 struct op_x86_model_spec {
-	int (*init)(struct oprofile_operations *ops);
-	void (*exit)(void);
-	unsigned int num_counters;
-	unsigned int num_controls;
-	void (*fill_in_addresses)(struct op_msrs * const msrs);
-	void (*setup_ctrs)(struct op_msrs const * const msrs);
-	int (*check_ctrs)(struct pt_regs * const regs,
-		struct op_msrs const * const msrs);
-	void (*start)(struct op_msrs const * const msrs);
-	void (*stop)(struct op_msrs const * const msrs);
-	void (*shutdown)(struct op_msrs const * const msrs);
+	unsigned int	num_counters;
+	unsigned int	num_controls;
+	int		(*init)(struct oprofile_operations *ops);
+	void		(*exit)(void);
+	void		(*fill_in_addresses)(struct op_msrs * const msrs);
+	void		(*setup_ctrs)(struct op_msrs const * const msrs);
+	int		(*check_ctrs)(struct pt_regs * const regs,
+				      struct op_msrs const * const msrs);
+	void		(*start)(struct op_msrs const * const msrs);
+	void		(*stop)(struct op_msrs const * const msrs);
+	void		(*shutdown)(struct op_msrs const * const msrs);
 };
 
 extern struct op_x86_model_spec const op_ppro_spec;
-- 
cgit v1.2.3


From 9063759540daac40cc1f402f83a3be6b489f8583 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Tue, 10 Mar 2009 19:15:57 +0100
Subject: x86/oprofile: remove #ifdefs in ibs functions

IBS code is moved to separate functions. This allows the removal
of #ifdefs in functions.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_amd.c | 80 +++++++++++++++++++++++-----------------
 1 file changed, 46 insertions(+), 34 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index 8fdf06e4edf9..b54c0880b7dc 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -220,6 +220,50 @@ op_amd_handle_ibs(struct pt_regs * const regs,
 	return 1;
 }
 
+static inline void op_amd_start_ibs(void)
+{
+	unsigned int low, high;
+	if (has_ibs && ibs_config.fetch_enabled) {
+		low = (ibs_config.max_cnt_fetch >> 4) & 0xFFFF;
+		high = ((ibs_config.rand_en & 0x1) << 25) /* bit 57 */
+			+ IBS_FETCH_HIGH_ENABLE;
+		wrmsr(MSR_AMD64_IBSFETCHCTL, low, high);
+	}
+
+	if (has_ibs && ibs_config.op_enabled) {
+		low = ((ibs_config.max_cnt_op >> 4) & 0xFFFF)
+			+ ((ibs_config.dispatched_ops & 0x1) << 19) /* bit 19 */
+			+ IBS_OP_LOW_ENABLE;
+		high = 0;
+		wrmsr(MSR_AMD64_IBSOPCTL, low, high);
+	}
+}
+
+static void op_amd_stop_ibs(void)
+{
+	unsigned int low, high;
+	if (has_ibs && ibs_config.fetch_enabled) {
+		/* clear max count and enable */
+		low = 0;
+		high = 0;
+		wrmsr(MSR_AMD64_IBSFETCHCTL, low, high);
+	}
+
+	if (has_ibs && ibs_config.op_enabled) {
+		/* clear max count and enable */
+		low = 0;
+		high = 0;
+		wrmsr(MSR_AMD64_IBSOPCTL, low, high);
+	}
+}
+
+#else
+
+static inline int op_amd_handle_ibs(struct pt_regs * const regs,
+				    struct op_msrs const * const msrs) { }
+static inline void op_amd_start_ibs(void) { }
+static inline void op_amd_stop_ibs(void) { }
+
 #endif
 
 static int op_amd_check_ctrs(struct pt_regs * const regs,
@@ -238,9 +282,7 @@ static int op_amd_check_ctrs(struct pt_regs * const regs,
 		}
 	}
 
-#ifdef CONFIG_OPROFILE_IBS
 	op_amd_handle_ibs(regs, msrs);
-#endif
 
 	/* See op_model_ppro.c */
 	return 1;
@@ -258,25 +300,9 @@ static void op_amd_start(struct op_msrs const * const msrs)
 		}
 	}
 
-#ifdef CONFIG_OPROFILE_IBS
-	if (has_ibs && ibs_config.fetch_enabled) {
-		low = (ibs_config.max_cnt_fetch >> 4) & 0xFFFF;
-		high = ((ibs_config.rand_en & 0x1) << 25) /* bit 57 */
-			+ IBS_FETCH_HIGH_ENABLE;
-		wrmsr(MSR_AMD64_IBSFETCHCTL, low, high);
-	}
-
-	if (has_ibs && ibs_config.op_enabled) {
-		low = ((ibs_config.max_cnt_op >> 4) & 0xFFFF)
-			+ ((ibs_config.dispatched_ops & 0x1) << 19) /* bit 19 */
-			+ IBS_OP_LOW_ENABLE;
-		high = 0;
-		wrmsr(MSR_AMD64_IBSOPCTL, low, high);
-	}
-#endif
+	op_amd_start_ibs();
 }
 
-
 static void op_amd_stop(struct op_msrs const * const msrs)
 {
 	unsigned int low, high;
@@ -294,21 +320,7 @@ static void op_amd_stop(struct op_msrs const * const msrs)
 		CTRL_WRITE(low, high, msrs, i);
 	}
 
-#ifdef CONFIG_OPROFILE_IBS
-	if (has_ibs && ibs_config.fetch_enabled) {
-		/* clear max count and enable */
-		low = 0;
-		high = 0;
-		wrmsr(MSR_AMD64_IBSFETCHCTL, low, high);
-	}
-
-	if (has_ibs && ibs_config.op_enabled) {
-		/* clear max count and enable */
-		low = 0;
-		high = 0;
-		wrmsr(MSR_AMD64_IBSOPCTL, low, high);
-	}
-#endif
+	op_amd_stop_ibs();
 }
 
 static void op_amd_shutdown(struct op_msrs const * const msrs)
-- 
cgit v1.2.3


From d20f24c66011f8a397bca6c5d1a6a7c7e612d2d7 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Sun, 11 Jan 2009 13:01:16 +0100
Subject: x86/oprofile: simplify AMD cpu init code

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c | 15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index dd8515301fbf..ae0ab03959b4 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -460,27 +460,26 @@ int __init op_nmi_init(struct oprofile_operations *ops)
 		/* Needs to be at least an Athlon (or hammer in 32bit mode) */
 
 		switch (family) {
-		default:
-			return -ENODEV;
 		case 6:
-			model = &op_amd_spec;
 			cpu_type = "i386/athlon";
 			break;
 		case 0xf:
-			model = &op_amd_spec;
-			/* Actually it could be i386/hammer too, but give
-			 user space an consistent name. */
+			/*
+			 * Actually it could be i386/hammer too, but
+			 * give user space an consistent name.
+			 */
 			cpu_type = "x86-64/hammer";
 			break;
 		case 0x10:
-			model = &op_amd_spec;
 			cpu_type = "x86-64/family10";
 			break;
 		case 0x11:
-			model = &op_amd_spec;
 			cpu_type = "x86-64/family11h";
 			break;
+		default:
+			return -ENODEV;
 		}
+		model = &op_amd_spec;
 		break;
 
 	case X86_VENDOR_INTEL:
-- 
cgit v1.2.3


From ff9faa8b676e195476b86f03fe58db0f01bda8f3 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Fri, 22 May 2009 15:36:29 +0200
Subject: x86/oprofile: move common macros to op_x86_model.h

There are duplicate macro implementations in model specific code. This
patch moves all common macros to op_x86_model.h.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_amd.c  | 8 --------
 arch/x86/oprofile/op_model_p4.c   | 2 --
 arch/x86/oprofile/op_model_ppro.c | 8 --------
 arch/x86/oprofile/op_x86_model.h  | 9 +++++++++
 4 files changed, 9 insertions(+), 18 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index b54c0880b7dc..4b9254a67e68 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -26,22 +26,14 @@
 #define NUM_COUNTERS 4
 #define NUM_CONTROLS 4
 
-#define CTR_IS_RESERVED(msrs, c) (msrs->counters[(c)].addr ? 1 : 0)
 #define CTR_READ(l, h, msrs, c) do {rdmsr(msrs->counters[(c)].addr, (l), (h)); } while (0)
 #define CTR_WRITE(l, msrs, c) do {wrmsr(msrs->counters[(c)].addr, -(unsigned int)(l), -1); } while (0)
 #define CTR_OVERFLOWED(n) (!((n) & (1U<<31)))
 
-#define CTRL_IS_RESERVED(msrs, c) (msrs->controls[(c)].addr ? 1 : 0)
 #define CTRL_READ(l, h, msrs, c) do {rdmsr(msrs->controls[(c)].addr, (l), (h)); } while (0)
 #define CTRL_WRITE(l, h, msrs, c) do {wrmsr(msrs->controls[(c)].addr, (l), (h)); } while (0)
-#define CTRL_SET_ACTIVE(n) (n |= (1<<22))
-#define CTRL_SET_INACTIVE(n) (n &= ~(1<<22))
 #define CTRL_CLEAR_LO(x) (x &= (1<<21))
 #define CTRL_CLEAR_HI(x) (x &= 0xfffffcf0)
-#define CTRL_SET_ENABLE(val) (val |= 1<<20)
-#define CTRL_SET_USR(val, u) (val |= ((u & 1) << 16))
-#define CTRL_SET_KERN(val, k) (val |= ((k & 1) << 17))
-#define CTRL_SET_UM(val, m) (val |= (m << 8))
 #define CTRL_SET_EVENT_LOW(val, e) (val |= (e & 0xff))
 #define CTRL_SET_EVENT_HIGH(val, e) (val |= ((e >> 8) & 0xf))
 #define CTRL_SET_HOST_ONLY(val, h) (val |= ((h & 1) << 9))
diff --git a/arch/x86/oprofile/op_model_p4.c b/arch/x86/oprofile/op_model_p4.c
index 819b131fd752..420c15e71237 100644
--- a/arch/x86/oprofile/op_model_p4.c
+++ b/arch/x86/oprofile/op_model_p4.c
@@ -366,8 +366,6 @@ static struct p4_event_binding p4_events[NUM_EVENTS] = {
 #define CCCR_OVF_P(cccr) ((cccr) & (1U<<31))
 #define CCCR_CLEAR_OVF(cccr) ((cccr) &= (~(1U<<31)))
 
-#define CTRL_IS_RESERVED(msrs, c) (msrs->controls[(c)].addr ? 1 : 0)
-#define CTR_IS_RESERVED(msrs, c) (msrs->counters[(c)].addr ? 1 : 0)
 #define CTR_READ(l, h, i) do {rdmsr(p4_counters[(i)].counter_address, (l), (h)); } while (0)
 #define CTR_WRITE(l, i) do {wrmsr(p4_counters[(i)].counter_address, -(u32)(l), -1); } while (0)
 #define CTR_OVERFLOW_P(ctr) (!((ctr) & 0x80000000))
diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c
index ae5811966883..a922a1a815c3 100644
--- a/arch/x86/oprofile/op_model_ppro.c
+++ b/arch/x86/oprofile/op_model_ppro.c
@@ -26,19 +26,11 @@
 static int num_counters = 2;
 static int counter_width = 32;
 
-#define CTR_IS_RESERVED(msrs, c) (msrs->counters[(c)].addr ? 1 : 0)
 #define CTR_OVERFLOWED(n) (!((n) & (1ULL<<(counter_width-1))))
 
-#define CTRL_IS_RESERVED(msrs, c) (msrs->controls[(c)].addr ? 1 : 0)
 #define CTRL_READ(l, h, msrs, c) do {rdmsr((msrs->controls[(c)].addr), (l), (h)); } while (0)
 #define CTRL_WRITE(l, h, msrs, c) do {wrmsr((msrs->controls[(c)].addr), (l), (h)); } while (0)
-#define CTRL_SET_ACTIVE(n) (n |= (1<<22))
-#define CTRL_SET_INACTIVE(n) (n &= ~(1<<22))
 #define CTRL_CLEAR(x) (x &= (1<<21))
-#define CTRL_SET_ENABLE(val) (val |= 1<<20)
-#define CTRL_SET_USR(val, u) (val |= ((u & 1) << 16))
-#define CTRL_SET_KERN(val, k) (val |= ((k & 1) << 17))
-#define CTRL_SET_UM(val, m) (val |= (m << 8))
 #define CTRL_SET_EVENT(val, e) (val |= e)
 
 static u64 *reset_value;
diff --git a/arch/x86/oprofile/op_x86_model.h b/arch/x86/oprofile/op_x86_model.h
index bd8157d12ff0..c80ec7d09993 100644
--- a/arch/x86/oprofile/op_x86_model.h
+++ b/arch/x86/oprofile/op_x86_model.h
@@ -11,6 +11,15 @@
 #ifndef OP_X86_MODEL_H
 #define OP_X86_MODEL_H
 
+#define CTR_IS_RESERVED(msrs, c) (msrs->counters[(c)].addr ? 1 : 0)
+#define CTRL_IS_RESERVED(msrs, c) (msrs->controls[(c)].addr ? 1 : 0)
+#define CTRL_SET_ACTIVE(n) (n |= (1<<22))
+#define CTRL_SET_ENABLE(val) (val |= 1<<20)
+#define CTRL_SET_INACTIVE(n) (n &= ~(1<<22))
+#define CTRL_SET_KERN(val, k) (val |= ((k & 1) << 17))
+#define CTRL_SET_UM(val, m) (val |= (m << 8))
+#define CTRL_SET_USR(val, u) (val |= ((u & 1) << 16))
+
 struct op_saved_msr {
 	unsigned int high;
 	unsigned int low;
-- 
cgit v1.2.3


From d2731a4387ad6c6bca07abfe9ed41d450fb6d665 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Fri, 22 May 2009 19:47:38 +0200
Subject: x86/oprofile: remove MSR macros for AMD cpus

The macros CTRL_READ() and CTRL_WRITE() make the code hard to read and
maintain. This patch replaces them by rdmsr()/wrmsr() functions and
simplifies the code.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_amd.c | 29 ++++++++++++-----------------
 1 file changed, 12 insertions(+), 17 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index 4b9254a67e68..c6181c265ae0 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -26,12 +26,7 @@
 #define NUM_COUNTERS 4
 #define NUM_CONTROLS 4
 
-#define CTR_READ(l, h, msrs, c) do {rdmsr(msrs->counters[(c)].addr, (l), (h)); } while (0)
-#define CTR_WRITE(l, msrs, c) do {wrmsr(msrs->counters[(c)].addr, -(unsigned int)(l), -1); } while (0)
 #define CTR_OVERFLOWED(n) (!((n) & (1U<<31)))
-
-#define CTRL_READ(l, h, msrs, c) do {rdmsr(msrs->controls[(c)].addr, (l), (h)); } while (0)
-#define CTRL_WRITE(l, h, msrs, c) do {wrmsr(msrs->controls[(c)].addr, (l), (h)); } while (0)
 #define CTRL_CLEAR_LO(x) (x &= (1<<21))
 #define CTRL_CLEAR_HI(x) (x &= 0xfffffcf0)
 #define CTRL_SET_EVENT_LOW(val, e) (val |= (e & 0xff))
@@ -101,17 +96,17 @@ static void op_amd_setup_ctrs(struct op_msrs const * const msrs)
 	for (i = 0 ; i < NUM_CONTROLS; ++i) {
 		if (unlikely(!CTRL_IS_RESERVED(msrs, i)))
 			continue;
-		CTRL_READ(low, high, msrs, i);
+		rdmsr(msrs->controls[i].addr, low, high);
 		CTRL_CLEAR_LO(low);
 		CTRL_CLEAR_HI(high);
-		CTRL_WRITE(low, high, msrs, i);
+		wrmsr(msrs->controls[i].addr, low, high);
 	}
 
 	/* avoid a false detection of ctr overflows in NMI handler */
 	for (i = 0; i < NUM_COUNTERS; ++i) {
 		if (unlikely(!CTR_IS_RESERVED(msrs, i)))
 			continue;
-		CTR_WRITE(1, msrs, i);
+		wrmsr(msrs->counters[i].addr, -1, -1);
 	}
 
 	/* enable active counters */
@@ -119,9 +114,9 @@ static void op_amd_setup_ctrs(struct op_msrs const * const msrs)
 		if ((counter_config[i].enabled) && (CTR_IS_RESERVED(msrs, i))) {
 			reset_value[i] = counter_config[i].count;
 
-			CTR_WRITE(counter_config[i].count, msrs, i);
+			wrmsr(msrs->counters[i].addr, -(unsigned int)counter_config[i].count, -1);
 
-			CTRL_READ(low, high, msrs, i);
+			rdmsr(msrs->controls[i].addr, low, high);
 			CTRL_CLEAR_LO(low);
 			CTRL_CLEAR_HI(high);
 			CTRL_SET_ENABLE(low);
@@ -133,7 +128,7 @@ static void op_amd_setup_ctrs(struct op_msrs const * const msrs)
 			CTRL_SET_HOST_ONLY(high, 0);
 			CTRL_SET_GUEST_ONLY(high, 0);
 
-			CTRL_WRITE(low, high, msrs, i);
+			wrmsr(msrs->controls[i].addr, low, high);
 		} else {
 			reset_value[i] = 0;
 		}
@@ -267,10 +262,10 @@ static int op_amd_check_ctrs(struct pt_regs * const regs,
 	for (i = 0 ; i < NUM_COUNTERS; ++i) {
 		if (!reset_value[i])
 			continue;
-		CTR_READ(low, high, msrs, i);
+		rdmsr(msrs->counters[i].addr, low, high);
 		if (CTR_OVERFLOWED(low)) {
 			oprofile_add_sample(regs, i);
-			CTR_WRITE(reset_value[i], msrs, i);
+			wrmsr(msrs->counters[i].addr, -(unsigned int)reset_value[i], -1);
 		}
 	}
 
@@ -286,9 +281,9 @@ static void op_amd_start(struct op_msrs const * const msrs)
 	int i;
 	for (i = 0 ; i < NUM_COUNTERS ; ++i) {
 		if (reset_value[i]) {
-			CTRL_READ(low, high, msrs, i);
+			rdmsr(msrs->controls[i].addr, low, high);
 			CTRL_SET_ACTIVE(low);
-			CTRL_WRITE(low, high, msrs, i);
+			wrmsr(msrs->controls[i].addr, low, high);
 		}
 	}
 
@@ -307,9 +302,9 @@ static void op_amd_stop(struct op_msrs const * const msrs)
 	for (i = 0 ; i < NUM_COUNTERS ; ++i) {
 		if (!reset_value[i])
 			continue;
-		CTRL_READ(low, high, msrs, i);
+		rdmsr(msrs->controls[i].addr, low, high);
 		CTRL_SET_INACTIVE(low);
-		CTRL_WRITE(low, high, msrs, i);
+		wrmsr(msrs->controls[i].addr, low, high);
 	}
 
 	op_amd_stop_ibs();
-- 
cgit v1.2.3


From 74c9a5c341bb1f6cbb5095b07c77230f19682ce8 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Fri, 22 May 2009 19:47:38 +0200
Subject: x86/oprofile: remove MSR macros for ppro cpus

The macros CTRL_READ() and CTRL_WRITE() make the code hard to read and
maintain. This patch replaces them by rdmsr()/wrmsr() functions and
simplifies the code.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_ppro.c | 19 ++++++++-----------
 1 file changed, 8 insertions(+), 11 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c
index a922a1a815c3..6c5d288c566e 100644
--- a/arch/x86/oprofile/op_model_ppro.c
+++ b/arch/x86/oprofile/op_model_ppro.c
@@ -27,9 +27,6 @@ static int num_counters = 2;
 static int counter_width = 32;
 
 #define CTR_OVERFLOWED(n) (!((n) & (1ULL<<(counter_width-1))))
-
-#define CTRL_READ(l, h, msrs, c) do {rdmsr((msrs->controls[(c)].addr), (l), (h)); } while (0)
-#define CTRL_WRITE(l, h, msrs, c) do {wrmsr((msrs->controls[(c)].addr), (l), (h)); } while (0)
 #define CTRL_CLEAR(x) (x &= (1<<21))
 #define CTRL_SET_EVENT(val, e) (val |= e)
 
@@ -88,9 +85,9 @@ static void ppro_setup_ctrs(struct op_msrs const * const msrs)
 	for (i = 0 ; i < num_counters; ++i) {
 		if (unlikely(!CTRL_IS_RESERVED(msrs, i)))
 			continue;
-		CTRL_READ(low, high, msrs, i);
+		rdmsr(msrs->controls[i].addr, low, high);
 		CTRL_CLEAR(low);
-		CTRL_WRITE(low, high, msrs, i);
+		wrmsr(msrs->controls[i].addr, low, high);
 	}
 
 	/* avoid a false detection of ctr overflows in NMI handler */
@@ -107,14 +104,14 @@ static void ppro_setup_ctrs(struct op_msrs const * const msrs)
 
 			wrmsrl(msrs->counters[i].addr, -reset_value[i]);
 
-			CTRL_READ(low, high, msrs, i);
+			rdmsr(msrs->controls[i].addr, low, high);
 			CTRL_CLEAR(low);
 			CTRL_SET_ENABLE(low);
 			CTRL_SET_USR(low, counter_config[i].user);
 			CTRL_SET_KERN(low, counter_config[i].kernel);
 			CTRL_SET_UM(low, counter_config[i].unit_mask);
 			CTRL_SET_EVENT(low, counter_config[i].event);
-			CTRL_WRITE(low, high, msrs, i);
+			wrmsr(msrs->controls[i].addr, low, high);
 		} else {
 			reset_value[i] = 0;
 		}
@@ -162,9 +159,9 @@ static void ppro_start(struct op_msrs const * const msrs)
 		return;
 	for (i = 0; i < num_counters; ++i) {
 		if (reset_value[i]) {
-			CTRL_READ(low, high, msrs, i);
+			rdmsr(msrs->controls[i].addr, low, high);
 			CTRL_SET_ACTIVE(low);
-			CTRL_WRITE(low, high, msrs, i);
+			wrmsr(msrs->controls[i].addr, low, high);
 		}
 	}
 }
@@ -180,9 +177,9 @@ static void ppro_stop(struct op_msrs const * const msrs)
 	for (i = 0; i < num_counters; ++i) {
 		if (!reset_value[i])
 			continue;
-		CTRL_READ(low, high, msrs, i);
+		rdmsr(msrs->controls[i].addr, low, high);
 		CTRL_SET_INACTIVE(low);
-		CTRL_WRITE(low, high, msrs, i);
+		wrmsr(msrs->controls[i].addr, low, high);
 	}
 }
 
-- 
cgit v1.2.3


From 1131a478245b00664ae2dbc0f68db987b51fa806 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Mon, 25 May 2009 20:23:23 +0200
Subject: x86/oprofile: remove MSR macros for p4 cpus

The macros CTRL_READ() and CTRL_WRITE() make the code hard to read and
maintain. This patch replaces them by rdmsr()/wrmsr() functions and
simplifies the code.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_p4.c | 39 +++++++++++++++++++--------------------
 1 file changed, 19 insertions(+), 20 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_model_p4.c b/arch/x86/oprofile/op_model_p4.c
index 420c15e71237..365d8a9c03d3 100644
--- a/arch/x86/oprofile/op_model_p4.c
+++ b/arch/x86/oprofile/op_model_p4.c
@@ -350,8 +350,6 @@ static struct p4_event_binding p4_events[NUM_EVENTS] = {
 #define ESCR_SET_OS_1(escr, os) ((escr) |= (((os) & 1) << 1))
 #define ESCR_SET_EVENT_SELECT(escr, sel) ((escr) |= (((sel) & 0x3f) << 25))
 #define ESCR_SET_EVENT_MASK(escr, mask) ((escr) |= (((mask) & 0xffff) << 9))
-#define ESCR_READ(escr, high, ev, i) do {rdmsr(ev->bindings[(i)].escr_address, (escr), (high)); } while (0)
-#define ESCR_WRITE(escr, high, ev, i) do {wrmsr(ev->bindings[(i)].escr_address, (escr), (high)); } while (0)
 
 #define CCCR_RESERVED_BITS 0x38030FFF
 #define CCCR_CLEAR(cccr) ((cccr) &= CCCR_RESERVED_BITS)
@@ -361,13 +359,9 @@ static struct p4_event_binding p4_events[NUM_EVENTS] = {
 #define CCCR_SET_PMI_OVF_1(cccr) ((cccr) |= (1<<27))
 #define CCCR_SET_ENABLE(cccr) ((cccr) |= (1<<12))
 #define CCCR_SET_DISABLE(cccr) ((cccr) &= ~(1<<12))
-#define CCCR_READ(low, high, i) do {rdmsr(p4_counters[(i)].cccr_address, (low), (high)); } while (0)
-#define CCCR_WRITE(low, high, i) do {wrmsr(p4_counters[(i)].cccr_address, (low), (high)); } while (0)
 #define CCCR_OVF_P(cccr) ((cccr) & (1U<<31))
 #define CCCR_CLEAR_OVF(cccr) ((cccr) &= (~(1U<<31)))
 
-#define CTR_READ(l, h, i) do {rdmsr(p4_counters[(i)].counter_address, (l), (h)); } while (0)
-#define CTR_WRITE(l, i) do {wrmsr(p4_counters[(i)].counter_address, -(u32)(l), -1); } while (0)
 #define CTR_OVERFLOW_P(ctr) (!((ctr) & 0x80000000))
 
 
@@ -513,7 +507,7 @@ static void pmc_setup_one_p4_counter(unsigned int ctr)
 		if (ev->bindings[i].virt_counter & counter_bit) {
 
 			/* modify ESCR */
-			ESCR_READ(escr, high, ev, i);
+			rdmsr(ev->bindings[i].escr_address, escr, high);
 			ESCR_CLEAR(escr);
 			if (stag == 0) {
 				ESCR_SET_USR_0(escr, counter_config[ctr].user);
@@ -524,10 +518,11 @@ static void pmc_setup_one_p4_counter(unsigned int ctr)
 			}
 			ESCR_SET_EVENT_SELECT(escr, ev->event_select);
 			ESCR_SET_EVENT_MASK(escr, counter_config[ctr].unit_mask);
-			ESCR_WRITE(escr, high, ev, i);
+			wrmsr(ev->bindings[i].escr_address, escr, high);
 
 			/* modify CCCR */
-			CCCR_READ(cccr, high, VIRT_CTR(stag, ctr));
+			rdmsr(p4_counters[VIRT_CTR(stag, ctr)].cccr_address,
+			      cccr, high);
 			CCCR_CLEAR(cccr);
 			CCCR_SET_REQUIRED_BITS(cccr);
 			CCCR_SET_ESCR_SELECT(cccr, ev->escr_select);
@@ -535,7 +530,8 @@ static void pmc_setup_one_p4_counter(unsigned int ctr)
 				CCCR_SET_PMI_OVF_0(cccr);
 			else
 				CCCR_SET_PMI_OVF_1(cccr);
-			CCCR_WRITE(cccr, high, VIRT_CTR(stag, ctr));
+			wrmsr(p4_counters[VIRT_CTR(stag, ctr)].cccr_address,
+			      cccr, high);
 			return;
 		}
 	}
@@ -582,7 +578,8 @@ static void p4_setup_ctrs(struct op_msrs const * const msrs)
 		if ((counter_config[i].enabled) && (CTRL_IS_RESERVED(msrs, i))) {
 			reset_value[i] = counter_config[i].count;
 			pmc_setup_one_p4_counter(i);
-			CTR_WRITE(counter_config[i].count, VIRT_CTR(stag, i));
+			wrmsr(p4_counters[VIRT_CTR(stag, i)].counter_address,
+			      -(u32)counter_config[i].count, -1);
 		} else {
 			reset_value[i] = 0;
 		}
@@ -622,14 +619,16 @@ static int p4_check_ctrs(struct pt_regs * const regs,
 
 		real = VIRT_CTR(stag, i);
 
-		CCCR_READ(low, high, real);
-		CTR_READ(ctr, high, real);
+		rdmsr(p4_counters[real].cccr_address, low, high);
+		rdmsr(p4_counters[real].counter_address, ctr, high);
 		if (CCCR_OVF_P(low) || CTR_OVERFLOW_P(ctr)) {
 			oprofile_add_sample(regs, i);
-			CTR_WRITE(reset_value[i], real);
+			wrmsr(p4_counters[real].counter_address,
+			      -(u32)reset_value[i], -1);
 			CCCR_CLEAR_OVF(low);
-			CCCR_WRITE(low, high, real);
-			CTR_WRITE(reset_value[i], real);
+			wrmsr(p4_counters[real].cccr_address, low, high);
+			wrmsr(p4_counters[real].counter_address,
+			      -(u32)reset_value[i], -1);
 		}
 	}
 
@@ -651,9 +650,9 @@ static void p4_start(struct op_msrs const * const msrs)
 	for (i = 0; i < num_counters; ++i) {
 		if (!reset_value[i])
 			continue;
-		CCCR_READ(low, high, VIRT_CTR(stag, i));
+		rdmsr(p4_counters[VIRT_CTR(stag, i)].cccr_address, low, high);
 		CCCR_SET_ENABLE(low);
-		CCCR_WRITE(low, high, VIRT_CTR(stag, i));
+		wrmsr(p4_counters[VIRT_CTR(stag, i)].cccr_address, low, high);
 	}
 }
 
@@ -668,9 +667,9 @@ static void p4_stop(struct op_msrs const * const msrs)
 	for (i = 0; i < num_counters; ++i) {
 		if (!reset_value[i])
 			continue;
-		CCCR_READ(low, high, VIRT_CTR(stag, i));
+		rdmsr(p4_counters[VIRT_CTR(stag, i)].cccr_address, low, high);
 		CCCR_SET_DISABLE(low);
-		CCCR_WRITE(low, high, VIRT_CTR(stag, i));
+		wrmsr(p4_counters[VIRT_CTR(stag, i)].cccr_address, low, high);
 	}
 }
 
-- 
cgit v1.2.3


From ec064c093e254f4433afb17dcef7f964c76436af Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Mon, 25 May 2009 15:05:50 +0200
Subject: x86/oprofile: fix and cleanup CTRL_SET_* macros

This patch fixes missing braces around macro parameters. Macro
definitions from intel_arch_perfmon.h are used where possible.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_ppro.c |  1 -
 arch/x86/oprofile/op_x86_model.h  | 18 ++++++++++--------
 2 files changed, 10 insertions(+), 9 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c
index 6c5d288c566e..61ee8f640535 100644
--- a/arch/x86/oprofile/op_model_ppro.c
+++ b/arch/x86/oprofile/op_model_ppro.c
@@ -18,7 +18,6 @@
 #include <asm/msr.h>
 #include <asm/apic.h>
 #include <asm/nmi.h>
-#include <asm/intel_arch_perfmon.h>
 
 #include "op_x86_model.h"
 #include "op_counter.h"
diff --git a/arch/x86/oprofile/op_x86_model.h b/arch/x86/oprofile/op_x86_model.h
index c80ec7d09993..a207b1c46e26 100644
--- a/arch/x86/oprofile/op_x86_model.h
+++ b/arch/x86/oprofile/op_x86_model.h
@@ -11,14 +11,16 @@
 #ifndef OP_X86_MODEL_H
 #define OP_X86_MODEL_H
 
-#define CTR_IS_RESERVED(msrs, c) (msrs->counters[(c)].addr ? 1 : 0)
-#define CTRL_IS_RESERVED(msrs, c) (msrs->controls[(c)].addr ? 1 : 0)
-#define CTRL_SET_ACTIVE(n) (n |= (1<<22))
-#define CTRL_SET_ENABLE(val) (val |= 1<<20)
-#define CTRL_SET_INACTIVE(n) (n &= ~(1<<22))
-#define CTRL_SET_KERN(val, k) (val |= ((k & 1) << 17))
-#define CTRL_SET_UM(val, m) (val |= (m << 8))
-#define CTRL_SET_USR(val, u) (val |= ((u & 1) << 16))
+#include <asm/intel_arch_perfmon.h>
+
+#define CTR_IS_RESERVED(msrs, c)	((msrs)->counters[(c)].addr ? 1 : 0)
+#define CTRL_IS_RESERVED(msrs, c)	((msrs)->controls[(c)].addr ? 1 : 0)
+#define CTRL_SET_ACTIVE(val)		((val) |= ARCH_PERFMON_EVENTSEL0_ENABLE)
+#define CTRL_SET_ENABLE(val)		((val) |= ARCH_PERFMON_EVENTSEL_INT)
+#define CTRL_SET_INACTIVE(val)		((val) &= ~ARCH_PERFMON_EVENTSEL0_ENABLE)
+#define CTRL_SET_KERN(val, k)		((val) |= ((k) ? ARCH_PERFMON_EVENTSEL_OS : 0))
+#define CTRL_SET_USR(val, u)		((val) |= ((u) ? ARCH_PERFMON_EVENTSEL_USR : 0))
+#define CTRL_SET_UM(val, m)		((val) |= ((m) << 8))
 
 struct op_saved_msr {
 	unsigned int high;
-- 
cgit v1.2.3


From 9c59354b48ce9cf28048b02fea73dd0236f876ea Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Mon, 25 May 2009 18:16:43 +0200
Subject: x86/oprofile: remove unused macros for AMD virtualization profiling

The use of the macros has no effect. The oprofilefs has to be extended
first to support these features.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_amd.c | 5 -----
 1 file changed, 5 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index c6181c265ae0..aaa7ffaed6b9 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -31,8 +31,6 @@
 #define CTRL_CLEAR_HI(x) (x &= 0xfffffcf0)
 #define CTRL_SET_EVENT_LOW(val, e) (val |= (e & 0xff))
 #define CTRL_SET_EVENT_HIGH(val, e) (val |= ((e >> 8) & 0xf))
-#define CTRL_SET_HOST_ONLY(val, h) (val |= ((h & 1) << 9))
-#define CTRL_SET_GUEST_ONLY(val, h) (val |= ((h & 1) << 8))
 
 static unsigned long reset_value[NUM_COUNTERS];
 
@@ -125,9 +123,6 @@ static void op_amd_setup_ctrs(struct op_msrs const * const msrs)
 			CTRL_SET_UM(low, counter_config[i].unit_mask);
 			CTRL_SET_EVENT_LOW(low, counter_config[i].event);
 			CTRL_SET_EVENT_HIGH(high, counter_config[i].event);
-			CTRL_SET_HOST_ONLY(high, 0);
-			CTRL_SET_GUEST_ONLY(high, 0);
-
 			wrmsr(msrs->controls[i].addr, low, high);
 		} else {
 			reset_value[i] = 0;
-- 
cgit v1.2.3


From ef8828ddf828174785421af67c281144d4b8e796 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Mon, 25 May 2009 19:31:44 +0200
Subject: x86/oprofile: pass the model to setup_ctrs() functions

In follow-on patches the setup_ctrs() functions will need data that
describes the model. This patch extends the function argument list to
pass a pointer of the model to these function.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c       | 2 +-
 arch/x86/oprofile/op_model_amd.c  | 3 ++-
 arch/x86/oprofile/op_model_p4.c   | 3 ++-
 arch/x86/oprofile/op_model_ppro.c | 3 ++-
 arch/x86/oprofile/op_x86_model.h  | 3 ++-
 5 files changed, 9 insertions(+), 5 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index ae0ab03959b4..c31f87bbf436 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -125,7 +125,7 @@ static void nmi_cpu_setup(void *dummy)
 	int cpu = smp_processor_id();
 	struct op_msrs *msrs = &per_cpu(cpu_msrs, cpu);
 	spin_lock(&oprofilefs_lock);
-	model->setup_ctrs(msrs);
+	model->setup_ctrs(model, msrs);
 	spin_unlock(&oprofilefs_lock);
 	per_cpu(saved_lvtpc, cpu) = apic_read(APIC_LVTPC);
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index aaa7ffaed6b9..86e0a01ba125 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -85,7 +85,8 @@ static void op_amd_fill_in_addresses(struct op_msrs * const msrs)
 }
 
 
-static void op_amd_setup_ctrs(struct op_msrs const * const msrs)
+static void op_amd_setup_ctrs(struct op_x86_model_spec const *model,
+			      struct op_msrs const * const msrs)
 {
 	unsigned int low, high;
 	int i;
diff --git a/arch/x86/oprofile/op_model_p4.c b/arch/x86/oprofile/op_model_p4.c
index 365d8a9c03d3..05ba0287b1f7 100644
--- a/arch/x86/oprofile/op_model_p4.c
+++ b/arch/x86/oprofile/op_model_p4.c
@@ -542,7 +542,8 @@ static void pmc_setup_one_p4_counter(unsigned int ctr)
 }
 
 
-static void p4_setup_ctrs(struct op_msrs const * const msrs)
+static void p4_setup_ctrs(struct op_x86_model_spec const *model,
+			  struct op_msrs const * const msrs)
 {
 	unsigned int i;
 	unsigned int low, high;
diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c
index 61ee8f640535..40b44ee521d5 100644
--- a/arch/x86/oprofile/op_model_ppro.c
+++ b/arch/x86/oprofile/op_model_ppro.c
@@ -51,7 +51,8 @@ static void ppro_fill_in_addresses(struct op_msrs * const msrs)
 }
 
 
-static void ppro_setup_ctrs(struct op_msrs const * const msrs)
+static void ppro_setup_ctrs(struct op_x86_model_spec const *model,
+			    struct op_msrs const * const msrs)
 {
 	unsigned int low, high;
 	int i;
diff --git a/arch/x86/oprofile/op_x86_model.h b/arch/x86/oprofile/op_x86_model.h
index a207b1c46e26..6161c7f0e7fb 100644
--- a/arch/x86/oprofile/op_x86_model.h
+++ b/arch/x86/oprofile/op_x86_model.h
@@ -48,7 +48,8 @@ struct op_x86_model_spec {
 	int		(*init)(struct oprofile_operations *ops);
 	void		(*exit)(void);
 	void		(*fill_in_addresses)(struct op_msrs * const msrs);
-	void		(*setup_ctrs)(struct op_msrs const * const msrs);
+	void		(*setup_ctrs)(struct op_x86_model_spec const *model,
+				      struct op_msrs const * const msrs);
 	int		(*check_ctrs)(struct pt_regs * const regs,
 				      struct op_msrs const * const msrs);
 	void		(*start)(struct op_msrs const * const msrs);
-- 
cgit v1.2.3


From 3370d358569755625aba4d9a846a040ce691d9ed Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Mon, 25 May 2009 15:10:32 +0200
Subject: x86/oprofile: replace macros to calculate control register

This patch introduces op_x86_get_ctrl() to calculate the value of the
performance control register. This is generic code usable for all
models. The event and reserved masks are model specific and stored in
struct op_x86_model_spec. 64 bit MSR functions are used now. The patch
removes many hard to read macros used for ctrl calculation.

The function op_x86_get_ctrl() is common code and the first step to
further merge performance counter implementations for x86 models.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c       | 20 +++++++++++++++++++
 arch/x86/oprofile/op_model_amd.c  | 41 +++++++++++++++------------------------
 arch/x86/oprofile/op_model_ppro.c | 29 +++++++++++++--------------
 arch/x86/oprofile/op_x86_model.h  | 15 ++++++++++----
 4 files changed, 60 insertions(+), 45 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index c31f87bbf436..388ee15e0e42 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -31,6 +31,26 @@ static DEFINE_PER_CPU(unsigned long, saved_lvtpc);
 /* 0 == registered but off, 1 == registered and on */
 static int nmi_enabled = 0;
 
+/* common functions */
+
+u64 op_x86_get_ctrl(struct op_x86_model_spec const *model,
+		    struct op_counter_config *counter_config)
+{
+	u64 val = 0;
+	u16 event = (u16)counter_config->event;
+
+	val |= ARCH_PERFMON_EVENTSEL_INT;
+	val |= counter_config->user ? ARCH_PERFMON_EVENTSEL_USR : 0;
+	val |= counter_config->kernel ? ARCH_PERFMON_EVENTSEL_OS : 0;
+	val |= (counter_config->unit_mask & 0xFF) << 8;
+	event &= model->event_mask ? model->event_mask : 0xFF;
+	val |= event & 0xFF;
+	val |= (event & 0x0F00) << 24;
+
+	return val;
+}
+
+
 static int profile_exceptions_notify(struct notifier_block *self,
 				     unsigned long val, void *data)
 {
diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index 86e0a01ba125..2406ab863605 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -25,12 +25,11 @@
 
 #define NUM_COUNTERS 4
 #define NUM_CONTROLS 4
+#define OP_EVENT_MASK			0x0FFF
+
+#define MSR_AMD_EVENTSEL_RESERVED	((0xFFFFFCF0ULL<<32)|(1ULL<<21))
 
 #define CTR_OVERFLOWED(n) (!((n) & (1U<<31)))
-#define CTRL_CLEAR_LO(x) (x &= (1<<21))
-#define CTRL_CLEAR_HI(x) (x &= 0xfffffcf0)
-#define CTRL_SET_EVENT_LOW(val, e) (val |= (e & 0xff))
-#define CTRL_SET_EVENT_HIGH(val, e) (val |= ((e >> 8) & 0xf))
 
 static unsigned long reset_value[NUM_COUNTERS];
 
@@ -84,21 +83,19 @@ static void op_amd_fill_in_addresses(struct op_msrs * const msrs)
 	}
 }
 
-
 static void op_amd_setup_ctrs(struct op_x86_model_spec const *model,
 			      struct op_msrs const * const msrs)
 {
-	unsigned int low, high;
+	u64 val;
 	int i;
 
 	/* clear all counters */
 	for (i = 0 ; i < NUM_CONTROLS; ++i) {
 		if (unlikely(!CTRL_IS_RESERVED(msrs, i)))
 			continue;
-		rdmsr(msrs->controls[i].addr, low, high);
-		CTRL_CLEAR_LO(low);
-		CTRL_CLEAR_HI(high);
-		wrmsr(msrs->controls[i].addr, low, high);
+		rdmsrl(msrs->controls[i].addr, val);
+		val &= model->reserved;
+		wrmsrl(msrs->controls[i].addr, val);
 	}
 
 	/* avoid a false detection of ctr overflows in NMI handler */
@@ -112,19 +109,11 @@ static void op_amd_setup_ctrs(struct op_x86_model_spec const *model,
 	for (i = 0; i < NUM_COUNTERS; ++i) {
 		if ((counter_config[i].enabled) && (CTR_IS_RESERVED(msrs, i))) {
 			reset_value[i] = counter_config[i].count;
-
 			wrmsr(msrs->counters[i].addr, -(unsigned int)counter_config[i].count, -1);
-
-			rdmsr(msrs->controls[i].addr, low, high);
-			CTRL_CLEAR_LO(low);
-			CTRL_CLEAR_HI(high);
-			CTRL_SET_ENABLE(low);
-			CTRL_SET_USR(low, counter_config[i].user);
-			CTRL_SET_KERN(low, counter_config[i].kernel);
-			CTRL_SET_UM(low, counter_config[i].unit_mask);
-			CTRL_SET_EVENT_LOW(low, counter_config[i].event);
-			CTRL_SET_EVENT_HIGH(high, counter_config[i].event);
-			wrmsr(msrs->controls[i].addr, low, high);
+			rdmsrl(msrs->controls[i].addr, val);
+			val &= model->reserved;
+			val |= op_x86_get_ctrl(model, &counter_config[i]);
+			wrmsrl(msrs->controls[i].addr, val);
 		} else {
 			reset_value[i] = 0;
 		}
@@ -486,14 +475,16 @@ static void op_amd_exit(void) {}
 #endif /* CONFIG_OPROFILE_IBS */
 
 struct op_x86_model_spec const op_amd_spec = {
-	.init			= op_amd_init,
-	.exit			= op_amd_exit,
 	.num_counters		= NUM_COUNTERS,
 	.num_controls		= NUM_CONTROLS,
+	.reserved		= MSR_AMD_EVENTSEL_RESERVED,
+	.event_mask		= OP_EVENT_MASK,
+	.init			= op_amd_init,
+	.exit			= op_amd_exit,
 	.fill_in_addresses	= &op_amd_fill_in_addresses,
 	.setup_ctrs		= &op_amd_setup_ctrs,
 	.check_ctrs		= &op_amd_check_ctrs,
 	.start			= &op_amd_start,
 	.stop			= &op_amd_stop,
-	.shutdown		= &op_amd_shutdown
+	.shutdown		= &op_amd_shutdown,
 };
diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c
index 40b44ee521d5..3092f998baf2 100644
--- a/arch/x86/oprofile/op_model_ppro.c
+++ b/arch/x86/oprofile/op_model_ppro.c
@@ -10,6 +10,7 @@
  * @author Philippe Elie
  * @author Graydon Hoare
  * @author Andi Kleen
+ * @author Robert Richter <robert.richter@amd.com>
  */
 
 #include <linux/oprofile.h>
@@ -26,8 +27,8 @@ static int num_counters = 2;
 static int counter_width = 32;
 
 #define CTR_OVERFLOWED(n) (!((n) & (1ULL<<(counter_width-1))))
-#define CTRL_CLEAR(x) (x &= (1<<21))
-#define CTRL_SET_EVENT(val, e) (val |= e)
+
+#define MSR_PPRO_EVENTSEL_RESERVED	((0xFFFFFFFFULL<<32)|(1ULL<<21))
 
 static u64 *reset_value;
 
@@ -54,7 +55,7 @@ static void ppro_fill_in_addresses(struct op_msrs * const msrs)
 static void ppro_setup_ctrs(struct op_x86_model_spec const *model,
 			    struct op_msrs const * const msrs)
 {
-	unsigned int low, high;
+	u64 val;
 	int i;
 
 	if (!reset_value) {
@@ -85,9 +86,9 @@ static void ppro_setup_ctrs(struct op_x86_model_spec const *model,
 	for (i = 0 ; i < num_counters; ++i) {
 		if (unlikely(!CTRL_IS_RESERVED(msrs, i)))
 			continue;
-		rdmsr(msrs->controls[i].addr, low, high);
-		CTRL_CLEAR(low);
-		wrmsr(msrs->controls[i].addr, low, high);
+		rdmsrl(msrs->controls[i].addr, val);
+		val &= model->reserved;
+		wrmsrl(msrs->controls[i].addr, val);
 	}
 
 	/* avoid a false detection of ctr overflows in NMI handler */
@@ -101,17 +102,11 @@ static void ppro_setup_ctrs(struct op_x86_model_spec const *model,
 	for (i = 0; i < num_counters; ++i) {
 		if ((counter_config[i].enabled) && (CTR_IS_RESERVED(msrs, i))) {
 			reset_value[i] = counter_config[i].count;
-
 			wrmsrl(msrs->counters[i].addr, -reset_value[i]);
-
-			rdmsr(msrs->controls[i].addr, low, high);
-			CTRL_CLEAR(low);
-			CTRL_SET_ENABLE(low);
-			CTRL_SET_USR(low, counter_config[i].user);
-			CTRL_SET_KERN(low, counter_config[i].kernel);
-			CTRL_SET_UM(low, counter_config[i].unit_mask);
-			CTRL_SET_EVENT(low, counter_config[i].event);
-			wrmsr(msrs->controls[i].addr, low, high);
+			rdmsrl(msrs->controls[i].addr, val);
+			val &= model->reserved;
+			val |= op_x86_get_ctrl(model, &counter_config[i]);
+			wrmsrl(msrs->controls[i].addr, val);
 		} else {
 			reset_value[i] = 0;
 		}
@@ -205,6 +200,7 @@ static void ppro_shutdown(struct op_msrs const * const msrs)
 struct op_x86_model_spec const op_ppro_spec = {
 	.num_counters		= 2,
 	.num_controls		= 2,
+	.reserved		= MSR_PPRO_EVENTSEL_RESERVED,
 	.fill_in_addresses	= &ppro_fill_in_addresses,
 	.setup_ctrs		= &ppro_setup_ctrs,
 	.check_ctrs		= &ppro_check_ctrs,
@@ -249,6 +245,7 @@ static int arch_perfmon_init(struct oprofile_operations *ignore)
 }
 
 struct op_x86_model_spec op_arch_perfmon_spec = {
+	.reserved		= MSR_PPRO_EVENTSEL_RESERVED,
 	.init			= &arch_perfmon_init,
 	/* num_counters/num_controls filled in at runtime */
 	.fill_in_addresses	= &ppro_fill_in_addresses,
diff --git a/arch/x86/oprofile/op_x86_model.h b/arch/x86/oprofile/op_x86_model.h
index 6161c7f0e7fb..3220d4ce6322 100644
--- a/arch/x86/oprofile/op_x86_model.h
+++ b/arch/x86/oprofile/op_x86_model.h
@@ -6,21 +6,19 @@
  * @remark Read the file COPYING
  *
  * @author Graydon Hoare
+ * @author Robert Richter <robert.richter@amd.com>
  */
 
 #ifndef OP_X86_MODEL_H
 #define OP_X86_MODEL_H
 
+#include <asm/types.h>
 #include <asm/intel_arch_perfmon.h>
 
 #define CTR_IS_RESERVED(msrs, c)	((msrs)->counters[(c)].addr ? 1 : 0)
 #define CTRL_IS_RESERVED(msrs, c)	((msrs)->controls[(c)].addr ? 1 : 0)
 #define CTRL_SET_ACTIVE(val)		((val) |= ARCH_PERFMON_EVENTSEL0_ENABLE)
-#define CTRL_SET_ENABLE(val)		((val) |= ARCH_PERFMON_EVENTSEL_INT)
 #define CTRL_SET_INACTIVE(val)		((val) &= ~ARCH_PERFMON_EVENTSEL0_ENABLE)
-#define CTRL_SET_KERN(val, k)		((val) |= ((k) ? ARCH_PERFMON_EVENTSEL_OS : 0))
-#define CTRL_SET_USR(val, u)		((val) |= ((u) ? ARCH_PERFMON_EVENTSEL_USR : 0))
-#define CTRL_SET_UM(val, m)		((val) |= ((m) << 8))
 
 struct op_saved_msr {
 	unsigned int high;
@@ -39,12 +37,16 @@ struct op_msrs {
 
 struct pt_regs;
 
+struct oprofile_operations;
+
 /* The model vtable abstracts the differences between
  * various x86 CPU models' perfctr support.
  */
 struct op_x86_model_spec {
 	unsigned int	num_counters;
 	unsigned int	num_controls;
+	u64		reserved;
+	u16		event_mask;
 	int		(*init)(struct oprofile_operations *ops);
 	void		(*exit)(void);
 	void		(*fill_in_addresses)(struct op_msrs * const msrs);
@@ -57,6 +59,11 @@ struct op_x86_model_spec {
 	void		(*shutdown)(struct op_msrs const * const msrs);
 };
 
+struct op_counter_config;
+
+extern u64 op_x86_get_ctrl(struct op_x86_model_spec const *model,
+			   struct op_counter_config *counter_config);
+
 extern struct op_x86_model_spec const op_ppro_spec;
 extern struct op_x86_model_spec const op_p4_spec;
 extern struct op_x86_model_spec const op_p4_ht2_spec;
-- 
cgit v1.2.3


From 42399adb239d4f1413899cc618ecf640779e79df Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Mon, 25 May 2009 17:59:06 +0200
Subject: x86/oprofile: replace CTR_OVERFLOWED macros

The patch replaces all CTR_OVERFLOWED macros. 64 bit MSR functions and
64 bit counter values are used now. Thus, it will be easier to later
extend the models to use more than 32 bit width counters.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_amd.c  | 16 ++++++++--------
 arch/x86/oprofile/op_model_p4.c   |  6 +++---
 arch/x86/oprofile/op_model_ppro.c | 10 ++++------
 3 files changed, 15 insertions(+), 17 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index 2406ab863605..b5d678fbf038 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -26,11 +26,10 @@
 #define NUM_COUNTERS 4
 #define NUM_CONTROLS 4
 #define OP_EVENT_MASK			0x0FFF
+#define OP_CTR_OVERFLOW			(1ULL<<31)
 
 #define MSR_AMD_EVENTSEL_RESERVED	((0xFFFFFCF0ULL<<32)|(1ULL<<21))
 
-#define CTR_OVERFLOWED(n) (!((n) & (1U<<31)))
-
 static unsigned long reset_value[NUM_COUNTERS];
 
 #ifdef CONFIG_OPROFILE_IBS
@@ -241,17 +240,18 @@ static inline void op_amd_stop_ibs(void) { }
 static int op_amd_check_ctrs(struct pt_regs * const regs,
 			     struct op_msrs const * const msrs)
 {
-	unsigned int low, high;
+	u64 val;
 	int i;
 
 	for (i = 0 ; i < NUM_COUNTERS; ++i) {
 		if (!reset_value[i])
 			continue;
-		rdmsr(msrs->counters[i].addr, low, high);
-		if (CTR_OVERFLOWED(low)) {
-			oprofile_add_sample(regs, i);
-			wrmsr(msrs->counters[i].addr, -(unsigned int)reset_value[i], -1);
-		}
+		rdmsrl(msrs->counters[i].addr, val);
+		/* bit is clear if overflowed: */
+		if (val & OP_CTR_OVERFLOW)
+			continue;
+		oprofile_add_sample(regs, i);
+		wrmsr(msrs->counters[i].addr, -(unsigned int)reset_value[i], -1);
 	}
 
 	op_amd_handle_ibs(regs, msrs);
diff --git a/arch/x86/oprofile/op_model_p4.c b/arch/x86/oprofile/op_model_p4.c
index 05ba0287b1f7..ac4ca28b9ed5 100644
--- a/arch/x86/oprofile/op_model_p4.c
+++ b/arch/x86/oprofile/op_model_p4.c
@@ -32,6 +32,8 @@
 #define NUM_CCCRS_HT2 9
 #define NUM_CONTROLS_HT2 (NUM_ESCRS_HT2 + NUM_CCCRS_HT2)
 
+#define OP_CTR_OVERFLOW			(1ULL<<31)
+
 static unsigned int num_counters = NUM_COUNTERS_NON_HT;
 static unsigned int num_controls = NUM_CONTROLS_NON_HT;
 
@@ -362,8 +364,6 @@ static struct p4_event_binding p4_events[NUM_EVENTS] = {
 #define CCCR_OVF_P(cccr) ((cccr) & (1U<<31))
 #define CCCR_CLEAR_OVF(cccr) ((cccr) &= (~(1U<<31)))
 
-#define CTR_OVERFLOW_P(ctr) (!((ctr) & 0x80000000))
-
 
 /* this assigns a "stagger" to the current CPU, which is used throughout
    the code in this module as an extra array offset, to select the "even"
@@ -622,7 +622,7 @@ static int p4_check_ctrs(struct pt_regs * const regs,
 
 		rdmsr(p4_counters[real].cccr_address, low, high);
 		rdmsr(p4_counters[real].counter_address, ctr, high);
-		if (CCCR_OVF_P(low) || CTR_OVERFLOW_P(ctr)) {
+		if (CCCR_OVF_P(low) || !(ctr & OP_CTR_OVERFLOW)) {
 			oprofile_add_sample(regs, i);
 			wrmsr(p4_counters[real].counter_address,
 			      -(u32)reset_value[i], -1);
diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c
index 3092f998baf2..82db396dc3ef 100644
--- a/arch/x86/oprofile/op_model_ppro.c
+++ b/arch/x86/oprofile/op_model_ppro.c
@@ -26,8 +26,6 @@
 static int num_counters = 2;
 static int counter_width = 32;
 
-#define CTR_OVERFLOWED(n) (!((n) & (1ULL<<(counter_width-1))))
-
 #define MSR_PPRO_EVENTSEL_RESERVED	((0xFFFFFFFFULL<<32)|(1ULL<<21))
 
 static u64 *reset_value;
@@ -124,10 +122,10 @@ static int ppro_check_ctrs(struct pt_regs * const regs,
 		if (!reset_value[i])
 			continue;
 		rdmsrl(msrs->counters[i].addr, val);
-		if (CTR_OVERFLOWED(val)) {
-			oprofile_add_sample(regs, i);
-			wrmsrl(msrs->counters[i].addr, -reset_value[i]);
-		}
+		if (val & (1ULL << (counter_width - 1)))
+			continue;
+		oprofile_add_sample(regs, i);
+		wrmsrl(msrs->counters[i].addr, -reset_value[i]);
 	}
 
 	/* Only P6 based Pentium M need to re-unmask the apic vector but it
-- 
cgit v1.2.3


From dea3766ca052a4f572b16a23a322553c064d75af Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Mon, 25 May 2009 18:11:52 +0200
Subject: x86/oprofile: replace CTRL_SET_*ACTIVE macros

The patch replaces all CTRL_SET_*ACTIVE macros. 64 bit MSR functions
and 64 bit counter values are used now. The code uses bit masks from
<asm/intel_arch_perfmon.h>.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_amd.c  | 16 ++++++++--------
 arch/x86/oprofile/op_model_ppro.c | 16 ++++++++--------
 arch/x86/oprofile/op_x86_model.h  |  2 --
 3 files changed, 16 insertions(+), 18 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index b5d678fbf038..4ac9d283e8d2 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -262,13 +262,13 @@ static int op_amd_check_ctrs(struct pt_regs * const regs,
 
 static void op_amd_start(struct op_msrs const * const msrs)
 {
-	unsigned int low, high;
+	u64 val;
 	int i;
 	for (i = 0 ; i < NUM_COUNTERS ; ++i) {
 		if (reset_value[i]) {
-			rdmsr(msrs->controls[i].addr, low, high);
-			CTRL_SET_ACTIVE(low);
-			wrmsr(msrs->controls[i].addr, low, high);
+			rdmsrl(msrs->controls[i].addr, val);
+			val |= ARCH_PERFMON_EVENTSEL0_ENABLE;
+			wrmsrl(msrs->controls[i].addr, val);
 		}
 	}
 
@@ -277,7 +277,7 @@ static void op_amd_start(struct op_msrs const * const msrs)
 
 static void op_amd_stop(struct op_msrs const * const msrs)
 {
-	unsigned int low, high;
+	u64 val;
 	int i;
 
 	/*
@@ -287,9 +287,9 @@ static void op_amd_stop(struct op_msrs const * const msrs)
 	for (i = 0 ; i < NUM_COUNTERS ; ++i) {
 		if (!reset_value[i])
 			continue;
-		rdmsr(msrs->controls[i].addr, low, high);
-		CTRL_SET_INACTIVE(low);
-		wrmsr(msrs->controls[i].addr, low, high);
+		rdmsrl(msrs->controls[i].addr, val);
+		val &= ~ARCH_PERFMON_EVENTSEL0_ENABLE;
+		wrmsrl(msrs->controls[i].addr, val);
 	}
 
 	op_amd_stop_ibs();
diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c
index 82db396dc3ef..566b43f0b6c6 100644
--- a/arch/x86/oprofile/op_model_ppro.c
+++ b/arch/x86/oprofile/op_model_ppro.c
@@ -145,16 +145,16 @@ static int ppro_check_ctrs(struct pt_regs * const regs,
 
 static void ppro_start(struct op_msrs const * const msrs)
 {
-	unsigned int low, high;
+	u64 val;
 	int i;
 
 	if (!reset_value)
 		return;
 	for (i = 0; i < num_counters; ++i) {
 		if (reset_value[i]) {
-			rdmsr(msrs->controls[i].addr, low, high);
-			CTRL_SET_ACTIVE(low);
-			wrmsr(msrs->controls[i].addr, low, high);
+			rdmsrl(msrs->controls[i].addr, val);
+			val |= ARCH_PERFMON_EVENTSEL0_ENABLE;
+			wrmsrl(msrs->controls[i].addr, val);
 		}
 	}
 }
@@ -162,7 +162,7 @@ static void ppro_start(struct op_msrs const * const msrs)
 
 static void ppro_stop(struct op_msrs const * const msrs)
 {
-	unsigned int low, high;
+	u64 val;
 	int i;
 
 	if (!reset_value)
@@ -170,9 +170,9 @@ static void ppro_stop(struct op_msrs const * const msrs)
 	for (i = 0; i < num_counters; ++i) {
 		if (!reset_value[i])
 			continue;
-		rdmsr(msrs->controls[i].addr, low, high);
-		CTRL_SET_INACTIVE(low);
-		wrmsr(msrs->controls[i].addr, low, high);
+		rdmsrl(msrs->controls[i].addr, val);
+		val &= ~ARCH_PERFMON_EVENTSEL0_ENABLE;
+		wrmsrl(msrs->controls[i].addr, val);
 	}
 }
 
diff --git a/arch/x86/oprofile/op_x86_model.h b/arch/x86/oprofile/op_x86_model.h
index 3220d4ce6322..1c4577795a92 100644
--- a/arch/x86/oprofile/op_x86_model.h
+++ b/arch/x86/oprofile/op_x86_model.h
@@ -17,8 +17,6 @@
 
 #define CTR_IS_RESERVED(msrs, c)	((msrs)->counters[(c)].addr ? 1 : 0)
 #define CTRL_IS_RESERVED(msrs, c)	((msrs)->controls[(c)].addr ? 1 : 0)
-#define CTRL_SET_ACTIVE(val)		((val) |= ARCH_PERFMON_EVENTSEL0_ENABLE)
-#define CTRL_SET_INACTIVE(val)		((val) &= ~ARCH_PERFMON_EVENTSEL0_ENABLE)
 
 struct op_saved_msr {
 	unsigned int high;
-- 
cgit v1.2.3


From 217d3cfb959756cb493fc03106c0253baa420ce8 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Thu, 4 Jun 2009 02:36:44 +0200
Subject: x86/oprofile: replace CTR*_IS_RESERVED macros

The patch replaces all CTR*_IS_RESERVED macros.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_amd.c  | 10 +++++-----
 arch/x86/oprofile/op_model_p4.c   | 10 +++++-----
 arch/x86/oprofile/op_model_ppro.c | 10 +++++-----
 arch/x86/oprofile/op_x86_model.h  |  3 ---
 4 files changed, 15 insertions(+), 18 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index 4ac9d283e8d2..c5c5eec2fa74 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -90,7 +90,7 @@ static void op_amd_setup_ctrs(struct op_x86_model_spec const *model,
 
 	/* clear all counters */
 	for (i = 0 ; i < NUM_CONTROLS; ++i) {
-		if (unlikely(!CTRL_IS_RESERVED(msrs, i)))
+		if (unlikely(!msrs->controls[i].addr))
 			continue;
 		rdmsrl(msrs->controls[i].addr, val);
 		val &= model->reserved;
@@ -99,14 +99,14 @@ static void op_amd_setup_ctrs(struct op_x86_model_spec const *model,
 
 	/* avoid a false detection of ctr overflows in NMI handler */
 	for (i = 0; i < NUM_COUNTERS; ++i) {
-		if (unlikely(!CTR_IS_RESERVED(msrs, i)))
+		if (unlikely(!msrs->counters[i].addr))
 			continue;
 		wrmsr(msrs->counters[i].addr, -1, -1);
 	}
 
 	/* enable active counters */
 	for (i = 0; i < NUM_COUNTERS; ++i) {
-		if ((counter_config[i].enabled) && (CTR_IS_RESERVED(msrs, i))) {
+		if (counter_config[i].enabled && msrs->counters[i].addr) {
 			reset_value[i] = counter_config[i].count;
 			wrmsr(msrs->counters[i].addr, -(unsigned int)counter_config[i].count, -1);
 			rdmsrl(msrs->controls[i].addr, val);
@@ -300,11 +300,11 @@ static void op_amd_shutdown(struct op_msrs const * const msrs)
 	int i;
 
 	for (i = 0 ; i < NUM_COUNTERS ; ++i) {
-		if (CTR_IS_RESERVED(msrs, i))
+		if (msrs->counters[i].addr)
 			release_perfctr_nmi(MSR_K7_PERFCTR0 + i);
 	}
 	for (i = 0 ; i < NUM_CONTROLS ; ++i) {
-		if (CTRL_IS_RESERVED(msrs, i))
+		if (msrs->controls[i].addr)
 			release_evntsel_nmi(MSR_K7_EVNTSEL0 + i);
 	}
 }
diff --git a/arch/x86/oprofile/op_model_p4.c b/arch/x86/oprofile/op_model_p4.c
index ac4ca28b9ed5..9db0ca9af764 100644
--- a/arch/x86/oprofile/op_model_p4.c
+++ b/arch/x86/oprofile/op_model_p4.c
@@ -559,7 +559,7 @@ static void p4_setup_ctrs(struct op_x86_model_spec const *model,
 
 	/* clear the cccrs we will use */
 	for (i = 0 ; i < num_counters ; i++) {
-		if (unlikely(!CTRL_IS_RESERVED(msrs, i)))
+		if (unlikely(!msrs->controls[i].addr))
 			continue;
 		rdmsr(p4_counters[VIRT_CTR(stag, i)].cccr_address, low, high);
 		CCCR_CLEAR(low);
@@ -569,14 +569,14 @@ static void p4_setup_ctrs(struct op_x86_model_spec const *model,
 
 	/* clear all escrs (including those outside our concern) */
 	for (i = num_counters; i < num_controls; i++) {
-		if (unlikely(!CTRL_IS_RESERVED(msrs, i)))
+		if (unlikely(!msrs->controls[i].addr))
 			continue;
 		wrmsr(msrs->controls[i].addr, 0, 0);
 	}
 
 	/* setup all counters */
 	for (i = 0 ; i < num_counters ; ++i) {
-		if ((counter_config[i].enabled) && (CTRL_IS_RESERVED(msrs, i))) {
+		if (counter_config[i].enabled && msrs->controls[i].addr) {
 			reset_value[i] = counter_config[i].count;
 			pmc_setup_one_p4_counter(i);
 			wrmsr(p4_counters[VIRT_CTR(stag, i)].counter_address,
@@ -679,7 +679,7 @@ static void p4_shutdown(struct op_msrs const * const msrs)
 	int i;
 
 	for (i = 0 ; i < num_counters ; ++i) {
-		if (CTR_IS_RESERVED(msrs, i))
+		if (msrs->counters[i].addr)
 			release_perfctr_nmi(msrs->counters[i].addr);
 	}
 	/*
@@ -688,7 +688,7 @@ static void p4_shutdown(struct op_msrs const * const msrs)
 	 * This saves a few bits.
 	 */
 	for (i = num_counters ; i < num_controls ; ++i) {
-		if (CTRL_IS_RESERVED(msrs, i))
+		if (msrs->controls[i].addr)
 			release_evntsel_nmi(msrs->controls[i].addr);
 	}
 }
diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c
index 566b43f0b6c6..0a261a5c696e 100644
--- a/arch/x86/oprofile/op_model_ppro.c
+++ b/arch/x86/oprofile/op_model_ppro.c
@@ -82,7 +82,7 @@ static void ppro_setup_ctrs(struct op_x86_model_spec const *model,
 
 	/* clear all counters */
 	for (i = 0 ; i < num_counters; ++i) {
-		if (unlikely(!CTRL_IS_RESERVED(msrs, i)))
+		if (unlikely(!msrs->controls[i].addr))
 			continue;
 		rdmsrl(msrs->controls[i].addr, val);
 		val &= model->reserved;
@@ -91,14 +91,14 @@ static void ppro_setup_ctrs(struct op_x86_model_spec const *model,
 
 	/* avoid a false detection of ctr overflows in NMI handler */
 	for (i = 0; i < num_counters; ++i) {
-		if (unlikely(!CTR_IS_RESERVED(msrs, i)))
+		if (unlikely(!msrs->counters[i].addr))
 			continue;
 		wrmsrl(msrs->counters[i].addr, -1LL);
 	}
 
 	/* enable active counters */
 	for (i = 0; i < num_counters; ++i) {
-		if ((counter_config[i].enabled) && (CTR_IS_RESERVED(msrs, i))) {
+		if (counter_config[i].enabled && msrs->counters[i].addr) {
 			reset_value[i] = counter_config[i].count;
 			wrmsrl(msrs->counters[i].addr, -reset_value[i]);
 			rdmsrl(msrs->controls[i].addr, val);
@@ -181,11 +181,11 @@ static void ppro_shutdown(struct op_msrs const * const msrs)
 	int i;
 
 	for (i = 0 ; i < num_counters ; ++i) {
-		if (CTR_IS_RESERVED(msrs, i))
+		if (msrs->counters[i].addr)
 			release_perfctr_nmi(MSR_P6_PERFCTR0 + i);
 	}
 	for (i = 0 ; i < num_counters ; ++i) {
-		if (CTRL_IS_RESERVED(msrs, i))
+		if (msrs->controls[i].addr)
 			release_evntsel_nmi(MSR_P6_EVNTSEL0 + i);
 	}
 	if (reset_value) {
diff --git a/arch/x86/oprofile/op_x86_model.h b/arch/x86/oprofile/op_x86_model.h
index 1c4577795a92..69f1eb46e1b3 100644
--- a/arch/x86/oprofile/op_x86_model.h
+++ b/arch/x86/oprofile/op_x86_model.h
@@ -15,9 +15,6 @@
 #include <asm/types.h>
 #include <asm/intel_arch_perfmon.h>
 
-#define CTR_IS_RESERVED(msrs, c)	((msrs)->counters[(c)].addr ? 1 : 0)
-#define CTRL_IS_RESERVED(msrs, c)	((msrs)->controls[(c)].addr ? 1 : 0)
-
 struct op_saved_msr {
 	unsigned int high;
 	unsigned int low;
-- 
cgit v1.2.3


From bbc5986d2db427fdd61b6116ff8b9ed988e663a8 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Mon, 25 May 2009 17:38:19 +0200
Subject: x86/oprofile: use 64 bit wrmsr functions

This patch replaces some wrmsr() functions with wrmsrl().

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_amd.c |  7 ++++---
 arch/x86/oprofile/op_model_p4.c  | 12 ++++++------
 2 files changed, 10 insertions(+), 9 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index c5c5eec2fa74..9bf901762411 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -101,14 +101,15 @@ static void op_amd_setup_ctrs(struct op_x86_model_spec const *model,
 	for (i = 0; i < NUM_COUNTERS; ++i) {
 		if (unlikely(!msrs->counters[i].addr))
 			continue;
-		wrmsr(msrs->counters[i].addr, -1, -1);
+		wrmsrl(msrs->counters[i].addr, -1LL);
 	}
 
 	/* enable active counters */
 	for (i = 0; i < NUM_COUNTERS; ++i) {
 		if (counter_config[i].enabled && msrs->counters[i].addr) {
 			reset_value[i] = counter_config[i].count;
-			wrmsr(msrs->counters[i].addr, -(unsigned int)counter_config[i].count, -1);
+			wrmsrl(msrs->counters[i].addr,
+			       -(s64)counter_config[i].count);
 			rdmsrl(msrs->controls[i].addr, val);
 			val &= model->reserved;
 			val |= op_x86_get_ctrl(model, &counter_config[i]);
@@ -251,7 +252,7 @@ static int op_amd_check_ctrs(struct pt_regs * const regs,
 		if (val & OP_CTR_OVERFLOW)
 			continue;
 		oprofile_add_sample(regs, i);
-		wrmsr(msrs->counters[i].addr, -(unsigned int)reset_value[i], -1);
+		wrmsrl(msrs->counters[i].addr, -(s64)reset_value[i]);
 	}
 
 	op_amd_handle_ibs(regs, msrs);
diff --git a/arch/x86/oprofile/op_model_p4.c b/arch/x86/oprofile/op_model_p4.c
index 9db0ca9af764..f01e53b118fa 100644
--- a/arch/x86/oprofile/op_model_p4.c
+++ b/arch/x86/oprofile/op_model_p4.c
@@ -579,8 +579,8 @@ static void p4_setup_ctrs(struct op_x86_model_spec const *model,
 		if (counter_config[i].enabled && msrs->controls[i].addr) {
 			reset_value[i] = counter_config[i].count;
 			pmc_setup_one_p4_counter(i);
-			wrmsr(p4_counters[VIRT_CTR(stag, i)].counter_address,
-			      -(u32)counter_config[i].count, -1);
+			wrmsrl(p4_counters[VIRT_CTR(stag, i)].counter_address,
+			       -(s64)counter_config[i].count);
 		} else {
 			reset_value[i] = 0;
 		}
@@ -624,12 +624,12 @@ static int p4_check_ctrs(struct pt_regs * const regs,
 		rdmsr(p4_counters[real].counter_address, ctr, high);
 		if (CCCR_OVF_P(low) || !(ctr & OP_CTR_OVERFLOW)) {
 			oprofile_add_sample(regs, i);
-			wrmsr(p4_counters[real].counter_address,
-			      -(u32)reset_value[i], -1);
+			wrmsrl(p4_counters[real].counter_address,
+			       -(s64)reset_value[i]);
 			CCCR_CLEAR_OVF(low);
 			wrmsr(p4_counters[real].cccr_address, low, high);
-			wrmsr(p4_counters[real].counter_address,
-			      -(u32)reset_value[i], -1);
+			wrmsrl(p4_counters[real].counter_address,
+			       -(s64)reset_value[i]);
 		}
 	}
 
-- 
cgit v1.2.3


From 95e74e62c1540b1115fe8cec5b592f22960f2bb2 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Wed, 3 Jun 2009 19:09:27 +0200
Subject: x86/oprofile: use 64 bit values to save MSR states

This patch removes struct op_saved_msr and replaces it by an u64
variable. This makes code easier and it is possible to use 64 bit MSR
functions.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c      | 28 ++++++++--------------------
 arch/x86/oprofile/op_x86_model.h |  9 ++-------
 2 files changed, 10 insertions(+), 27 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index 388ee15e0e42..3b84b789de0b 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -78,19 +78,13 @@ static void nmi_cpu_save_registers(struct op_msrs *msrs)
 	unsigned int i;
 
 	for (i = 0; i < nr_ctrs; ++i) {
-		if (counters[i].addr) {
-			rdmsr(counters[i].addr,
-				counters[i].saved.low,
-				counters[i].saved.high);
-		}
+		if (counters[i].addr)
+			rdmsrl(counters[i].addr, counters[i].saved);
 	}
 
 	for (i = 0; i < nr_ctrls; ++i) {
-		if (controls[i].addr) {
-			rdmsr(controls[i].addr,
-				controls[i].saved.low,
-				controls[i].saved.high);
-		}
+		if (controls[i].addr)
+			rdmsrl(controls[i].addr, controls[i].saved);
 	}
 }
 
@@ -204,19 +198,13 @@ static void nmi_restore_registers(struct op_msrs *msrs)
 	unsigned int i;
 
 	for (i = 0; i < nr_ctrls; ++i) {
-		if (controls[i].addr) {
-			wrmsr(controls[i].addr,
-				controls[i].saved.low,
-				controls[i].saved.high);
-		}
+		if (controls[i].addr)
+			wrmsrl(controls[i].addr, controls[i].saved);
 	}
 
 	for (i = 0; i < nr_ctrs; ++i) {
-		if (counters[i].addr) {
-			wrmsr(counters[i].addr,
-				counters[i].saved.low,
-				counters[i].saved.high);
-		}
+		if (counters[i].addr)
+			wrmsrl(counters[i].addr, counters[i].saved);
 	}
 }
 
diff --git a/arch/x86/oprofile/op_x86_model.h b/arch/x86/oprofile/op_x86_model.h
index 69f1eb46e1b3..fda52b4c1b95 100644
--- a/arch/x86/oprofile/op_x86_model.h
+++ b/arch/x86/oprofile/op_x86_model.h
@@ -15,14 +15,9 @@
 #include <asm/types.h>
 #include <asm/intel_arch_perfmon.h>
 
-struct op_saved_msr {
-	unsigned int high;
-	unsigned int low;
-};
-
 struct op_msr {
-	unsigned long addr;
-	struct op_saved_msr saved;
+	unsigned long	addr;
+	u64		saved;
 };
 
 struct op_msrs {
-- 
cgit v1.2.3


From 1a245c45343651a87ff63afc5ddeb8e24d731835 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Fri, 5 Jun 2009 15:54:24 +0200
Subject: x86/oprofile: remove some local variables in MSR save/restore
 functions

The patch removes some local variables in these functions.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c | 12 ++++--------
 1 file changed, 4 insertions(+), 8 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index 3b84b789de0b..80b63d5db509 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -71,18 +71,16 @@ static int profile_exceptions_notify(struct notifier_block *self,
 
 static void nmi_cpu_save_registers(struct op_msrs *msrs)
 {
-	unsigned int const nr_ctrs = model->num_counters;
-	unsigned int const nr_ctrls = model->num_controls;
 	struct op_msr *counters = msrs->counters;
 	struct op_msr *controls = msrs->controls;
 	unsigned int i;
 
-	for (i = 0; i < nr_ctrs; ++i) {
+	for (i = 0; i < model->num_counters; ++i) {
 		if (counters[i].addr)
 			rdmsrl(counters[i].addr, counters[i].saved);
 	}
 
-	for (i = 0; i < nr_ctrls; ++i) {
+	for (i = 0; i < model->num_controls; ++i) {
 		if (controls[i].addr)
 			rdmsrl(controls[i].addr, controls[i].saved);
 	}
@@ -191,18 +189,16 @@ static int nmi_setup(void)
 
 static void nmi_restore_registers(struct op_msrs *msrs)
 {
-	unsigned int const nr_ctrs = model->num_counters;
-	unsigned int const nr_ctrls = model->num_controls;
 	struct op_msr *counters = msrs->counters;
 	struct op_msr *controls = msrs->controls;
 	unsigned int i;
 
-	for (i = 0; i < nr_ctrls; ++i) {
+	for (i = 0; i < model->num_controls; ++i) {
 		if (controls[i].addr)
 			wrmsrl(controls[i].addr, controls[i].saved);
 	}
 
-	for (i = 0; i < nr_ctrs; ++i) {
+	for (i = 0; i < model->num_counters; ++i) {
 		if (counters[i].addr)
 			wrmsrl(counters[i].addr, counters[i].saved);
 	}
-- 
cgit v1.2.3


From c572ae4efd1b0a5cc76c5e6aae05c1b182b6a69c Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Wed, 3 Jun 2009 20:10:39 +0200
Subject: x86/oprofile: use 64 bit values in IBS functions

The IBS code internally uses 32 bit values (a low and a high value) to
represent a 64 bit value. This patch changes this and now 64 bit
values are used instead. 64 bit MSR functions can be used now.

No functional changes.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_amd.c | 131 ++++++++++++++++++---------------------
 1 file changed, 61 insertions(+), 70 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index 9bf901762411..6493ef7ae9ad 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -35,16 +35,18 @@ static unsigned long reset_value[NUM_COUNTERS];
 #ifdef CONFIG_OPROFILE_IBS
 
 /* IbsFetchCtl bits/masks */
-#define IBS_FETCH_HIGH_VALID_BIT	(1UL << 17)	/* bit 49 */
-#define IBS_FETCH_HIGH_ENABLE		(1UL << 16)	/* bit 48 */
-#define IBS_FETCH_LOW_MAX_CNT_MASK	0x0000FFFFUL	/* MaxCnt mask */
+#define IBS_FETCH_RAND_EN		(1ULL<<57)
+#define IBS_FETCH_VAL			(1ULL<<49)
+#define IBS_FETCH_ENABLE		(1ULL<<48)
+#define IBS_FETCH_CNT_MASK		0xFFFF0000ULL
 
 /*IbsOpCtl bits */
-#define IBS_OP_LOW_VALID_BIT		(1ULL<<18)	/* bit 18 */
-#define IBS_OP_LOW_ENABLE		(1ULL<<17)	/* bit 17 */
+#define IBS_OP_CNT_CTL			(1ULL<<19)
+#define IBS_OP_VAL			(1ULL<<18)
+#define IBS_OP_ENABLE			(1ULL<<17)
 
-#define IBS_FETCH_SIZE	6
-#define IBS_OP_SIZE	12
+#define IBS_FETCH_SIZE			6
+#define IBS_OP_SIZE			12
 
 static int has_ibs;	/* AMD Family10h and later */
 
@@ -126,66 +128,63 @@ static inline int
 op_amd_handle_ibs(struct pt_regs * const regs,
 		  struct op_msrs const * const msrs)
 {
-	u32 low, high;
-	u64 msr;
+	u64 val, ctl;
 	struct op_entry entry;
 
 	if (!has_ibs)
 		return 1;
 
 	if (ibs_config.fetch_enabled) {
-		rdmsr(MSR_AMD64_IBSFETCHCTL, low, high);
-		if (high & IBS_FETCH_HIGH_VALID_BIT) {
-			rdmsrl(MSR_AMD64_IBSFETCHLINAD, msr);
-			oprofile_write_reserve(&entry, regs, msr,
+		rdmsrl(MSR_AMD64_IBSFETCHCTL, ctl);
+		if (ctl & IBS_FETCH_VAL) {
+			rdmsrl(MSR_AMD64_IBSFETCHLINAD, val);
+			oprofile_write_reserve(&entry, regs, val,
 					       IBS_FETCH_CODE, IBS_FETCH_SIZE);
-			oprofile_add_data(&entry, (u32)msr);
-			oprofile_add_data(&entry, (u32)(msr >> 32));
-			oprofile_add_data(&entry, low);
-			oprofile_add_data(&entry, high);
-			rdmsrl(MSR_AMD64_IBSFETCHPHYSAD, msr);
-			oprofile_add_data(&entry, (u32)msr);
-			oprofile_add_data(&entry, (u32)(msr >> 32));
+			oprofile_add_data(&entry, (u32)val);
+			oprofile_add_data(&entry, (u32)(val >> 32));
+			oprofile_add_data(&entry, (u32)ctl);
+			oprofile_add_data(&entry, (u32)(ctl >> 32));
+			rdmsrl(MSR_AMD64_IBSFETCHPHYSAD, val);
+			oprofile_add_data(&entry, (u32)val);
+			oprofile_add_data(&entry, (u32)(val >> 32));
 			oprofile_write_commit(&entry);
 
 			/* reenable the IRQ */
-			high &= ~IBS_FETCH_HIGH_VALID_BIT;
-			high |= IBS_FETCH_HIGH_ENABLE;
-			low &= IBS_FETCH_LOW_MAX_CNT_MASK;
-			wrmsr(MSR_AMD64_IBSFETCHCTL, low, high);
+			ctl &= ~(IBS_FETCH_VAL | IBS_FETCH_CNT_MASK);
+			ctl |= IBS_FETCH_ENABLE;
+			wrmsrl(MSR_AMD64_IBSFETCHCTL, ctl);
 		}
 	}
 
 	if (ibs_config.op_enabled) {
-		rdmsr(MSR_AMD64_IBSOPCTL, low, high);
-		if (low & IBS_OP_LOW_VALID_BIT) {
-			rdmsrl(MSR_AMD64_IBSOPRIP, msr);
-			oprofile_write_reserve(&entry, regs, msr,
+		rdmsrl(MSR_AMD64_IBSOPCTL, ctl);
+		if (ctl & IBS_OP_VAL) {
+			rdmsrl(MSR_AMD64_IBSOPRIP, val);
+			oprofile_write_reserve(&entry, regs, val,
 					       IBS_OP_CODE, IBS_OP_SIZE);
-			oprofile_add_data(&entry, (u32)msr);
-			oprofile_add_data(&entry, (u32)(msr >> 32));
-			rdmsrl(MSR_AMD64_IBSOPDATA, msr);
-			oprofile_add_data(&entry, (u32)msr);
-			oprofile_add_data(&entry, (u32)(msr >> 32));
-			rdmsrl(MSR_AMD64_IBSOPDATA2, msr);
-			oprofile_add_data(&entry, (u32)msr);
-			oprofile_add_data(&entry, (u32)(msr >> 32));
-			rdmsrl(MSR_AMD64_IBSOPDATA3, msr);
-			oprofile_add_data(&entry, (u32)msr);
-			oprofile_add_data(&entry, (u32)(msr >> 32));
-			rdmsrl(MSR_AMD64_IBSDCLINAD, msr);
-			oprofile_add_data(&entry, (u32)msr);
-			oprofile_add_data(&entry, (u32)(msr >> 32));
-			rdmsrl(MSR_AMD64_IBSDCPHYSAD, msr);
-			oprofile_add_data(&entry, (u32)msr);
-			oprofile_add_data(&entry, (u32)(msr >> 32));
+			oprofile_add_data(&entry, (u32)val);
+			oprofile_add_data(&entry, (u32)(val >> 32));
+			rdmsrl(MSR_AMD64_IBSOPDATA, val);
+			oprofile_add_data(&entry, (u32)val);
+			oprofile_add_data(&entry, (u32)(val >> 32));
+			rdmsrl(MSR_AMD64_IBSOPDATA2, val);
+			oprofile_add_data(&entry, (u32)val);
+			oprofile_add_data(&entry, (u32)(val >> 32));
+			rdmsrl(MSR_AMD64_IBSOPDATA3, val);
+			oprofile_add_data(&entry, (u32)val);
+			oprofile_add_data(&entry, (u32)(val >> 32));
+			rdmsrl(MSR_AMD64_IBSDCLINAD, val);
+			oprofile_add_data(&entry, (u32)val);
+			oprofile_add_data(&entry, (u32)(val >> 32));
+			rdmsrl(MSR_AMD64_IBSDCPHYSAD, val);
+			oprofile_add_data(&entry, (u32)val);
+			oprofile_add_data(&entry, (u32)(val >> 32));
 			oprofile_write_commit(&entry);
 
 			/* reenable the IRQ */
-			high = 0;
-			low &= ~IBS_OP_LOW_VALID_BIT;
-			low |= IBS_OP_LOW_ENABLE;
-			wrmsr(MSR_AMD64_IBSOPCTL, low, high);
+			ctl &= ~IBS_OP_VAL & 0xFFFFFFFF;
+			ctl |= IBS_OP_ENABLE;
+			wrmsrl(MSR_AMD64_IBSOPCTL, ctl);
 		}
 	}
 
@@ -194,39 +193,31 @@ op_amd_handle_ibs(struct pt_regs * const regs,
 
 static inline void op_amd_start_ibs(void)
 {
-	unsigned int low, high;
+	u64 val;
 	if (has_ibs && ibs_config.fetch_enabled) {
-		low = (ibs_config.max_cnt_fetch >> 4) & 0xFFFF;
-		high = ((ibs_config.rand_en & 0x1) << 25) /* bit 57 */
-			+ IBS_FETCH_HIGH_ENABLE;
-		wrmsr(MSR_AMD64_IBSFETCHCTL, low, high);
+		val = (ibs_config.max_cnt_fetch >> 4) & 0xFFFF;
+		val |= ibs_config.rand_en ? IBS_FETCH_RAND_EN : 0;
+		val |= IBS_FETCH_ENABLE;
+		wrmsrl(MSR_AMD64_IBSFETCHCTL, val);
 	}
 
 	if (has_ibs && ibs_config.op_enabled) {
-		low = ((ibs_config.max_cnt_op >> 4) & 0xFFFF)
-			+ ((ibs_config.dispatched_ops & 0x1) << 19) /* bit 19 */
-			+ IBS_OP_LOW_ENABLE;
-		high = 0;
-		wrmsr(MSR_AMD64_IBSOPCTL, low, high);
+		val = (ibs_config.max_cnt_op >> 4) & 0xFFFF;
+		val |= ibs_config.dispatched_ops ? IBS_OP_CNT_CTL : 0;
+		val |= IBS_OP_ENABLE;
+		wrmsrl(MSR_AMD64_IBSOPCTL, val);
 	}
 }
 
 static void op_amd_stop_ibs(void)
 {
-	unsigned int low, high;
-	if (has_ibs && ibs_config.fetch_enabled) {
+	if (has_ibs && ibs_config.fetch_enabled)
 		/* clear max count and enable */
-		low = 0;
-		high = 0;
-		wrmsr(MSR_AMD64_IBSFETCHCTL, low, high);
-	}
+		wrmsrl(MSR_AMD64_IBSFETCHCTL, 0);
 
-	if (has_ibs && ibs_config.op_enabled) {
+	if (has_ibs && ibs_config.op_enabled)
 		/* clear max count and enable */
-		low = 0;
-		high = 0;
-		wrmsr(MSR_AMD64_IBSOPCTL, low, high);
-	}
+		wrmsrl(MSR_AMD64_IBSOPCTL, 0);
 }
 
 #else
-- 
cgit v1.2.3


From 51563a0e5650d0d76539625388d72d62b34c726e Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Wed, 3 Jun 2009 20:54:56 +0200
Subject: x86/oprofile: introduce oprofile_add_data64()

The IBS implemention writes 64 bit register values to the cpu buffer
by writing two 32 values using oprofile_add_data(). This patch
introduces oprofile_add_data64() to write a single 64 bit value to the
buffer.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_amd.c | 27 +++++++++------------------
 drivers/oprofile/cpu_buffer.c    | 15 +++++++++++++++
 include/linux/oprofile.h         |  1 +
 3 files changed, 25 insertions(+), 18 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index 6493ef7ae9ad..cc930467575d 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -140,13 +140,10 @@ op_amd_handle_ibs(struct pt_regs * const regs,
 			rdmsrl(MSR_AMD64_IBSFETCHLINAD, val);
 			oprofile_write_reserve(&entry, regs, val,
 					       IBS_FETCH_CODE, IBS_FETCH_SIZE);
-			oprofile_add_data(&entry, (u32)val);
-			oprofile_add_data(&entry, (u32)(val >> 32));
-			oprofile_add_data(&entry, (u32)ctl);
-			oprofile_add_data(&entry, (u32)(ctl >> 32));
+			oprofile_add_data64(&entry, val);
+			oprofile_add_data64(&entry, ctl);
 			rdmsrl(MSR_AMD64_IBSFETCHPHYSAD, val);
-			oprofile_add_data(&entry, (u32)val);
-			oprofile_add_data(&entry, (u32)(val >> 32));
+			oprofile_add_data64(&entry, val);
 			oprofile_write_commit(&entry);
 
 			/* reenable the IRQ */
@@ -162,23 +159,17 @@ op_amd_handle_ibs(struct pt_regs * const regs,
 			rdmsrl(MSR_AMD64_IBSOPRIP, val);
 			oprofile_write_reserve(&entry, regs, val,
 					       IBS_OP_CODE, IBS_OP_SIZE);
-			oprofile_add_data(&entry, (u32)val);
-			oprofile_add_data(&entry, (u32)(val >> 32));
+			oprofile_add_data64(&entry, val);
 			rdmsrl(MSR_AMD64_IBSOPDATA, val);
-			oprofile_add_data(&entry, (u32)val);
-			oprofile_add_data(&entry, (u32)(val >> 32));
+			oprofile_add_data64(&entry, val);
 			rdmsrl(MSR_AMD64_IBSOPDATA2, val);
-			oprofile_add_data(&entry, (u32)val);
-			oprofile_add_data(&entry, (u32)(val >> 32));
+			oprofile_add_data64(&entry, val);
 			rdmsrl(MSR_AMD64_IBSOPDATA3, val);
-			oprofile_add_data(&entry, (u32)val);
-			oprofile_add_data(&entry, (u32)(val >> 32));
+			oprofile_add_data64(&entry, val);
 			rdmsrl(MSR_AMD64_IBSDCLINAD, val);
-			oprofile_add_data(&entry, (u32)val);
-			oprofile_add_data(&entry, (u32)(val >> 32));
+			oprofile_add_data64(&entry, val);
 			rdmsrl(MSR_AMD64_IBSDCPHYSAD, val);
-			oprofile_add_data(&entry, (u32)val);
-			oprofile_add_data(&entry, (u32)(val >> 32));
+			oprofile_add_data64(&entry, val);
 			oprofile_write_commit(&entry);
 
 			/* reenable the IRQ */
diff --git a/drivers/oprofile/cpu_buffer.c b/drivers/oprofile/cpu_buffer.c
index 50640cc5eef2..a7aae24f2889 100644
--- a/drivers/oprofile/cpu_buffer.c
+++ b/drivers/oprofile/cpu_buffer.c
@@ -406,6 +406,21 @@ int oprofile_add_data(struct op_entry *entry, unsigned long val)
 	return op_cpu_buffer_add_data(entry, val);
 }
 
+int oprofile_add_data64(struct op_entry *entry, u64 val)
+{
+	if (!entry->event)
+		return 0;
+	if (op_cpu_buffer_get_size(entry) < 2)
+		/*
+		 * the function returns 0 to indicate a too small
+		 * buffer, even if there is some space left
+		 */
+		return 0;
+	if (!op_cpu_buffer_add_data(entry, (u32)val))
+		return 0;
+	return op_cpu_buffer_add_data(entry, (u32)(val >> 32));
+}
+
 int oprofile_write_commit(struct op_entry *entry)
 {
 	if (!entry->event)
diff --git a/include/linux/oprofile.h b/include/linux/oprofile.h
index dbbe2dbc4418..d68d2ed94f15 100644
--- a/include/linux/oprofile.h
+++ b/include/linux/oprofile.h
@@ -179,6 +179,7 @@ void oprofile_write_reserve(struct op_entry *entry,
 			    struct pt_regs * const regs,
 			    unsigned long pc, int code, int size);
 int oprofile_add_data(struct op_entry *entry, unsigned long val);
+int oprofile_add_data64(struct op_entry *entry, u64 val);
 int oprofile_write_commit(struct op_entry *entry);
 
 #endif /* OPROFILE_H */
-- 
cgit v1.2.3


From 802070f5474af1a49435a9528aede47bb18abd47 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Fri, 12 Jun 2009 18:32:07 +0200
Subject: x86/oprofile: fix initialization of arch_perfmon for core_i7

Commit:

 e419294 x86/oprofile: moving arch_perfmon counter setup to op_x86_model_spec.init

introduced a bug in the initialization of core_i7 leading to the
incorrect model setup to &op_ppro_spec. This patch fixes this.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index 7826dfcc8428..28ee490c1b80 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -406,6 +406,7 @@ module_param_call(cpu_type, force_cpu_type, NULL, NULL, 0);
 static int __init ppro_init(char **cpu_type)
 {
 	__u8 cpu_model = boot_cpu_data.x86_model;
+	struct op_x86_model_spec const *spec = &op_ppro_spec;	/* default */
 
 	if (force_arch_perfmon && cpu_has_arch_perfmon)
 		return 0;
@@ -432,7 +433,7 @@ static int __init ppro_init(char **cpu_type)
 		*cpu_type = "i386/core_2";
 		break;
 	case 26:
-		model = &op_arch_perfmon_spec;
+		spec = &op_arch_perfmon_spec;
 		*cpu_type = "i386/core_i7";
 		break;
 	case 28:
@@ -443,7 +444,7 @@ static int __init ppro_init(char **cpu_type)
 		return 0;
 	}
 
-	model = &op_ppro_spec;
+	model = spec;
 	return 1;
 }
 
-- 
cgit v1.2.3


From c64b04fe6e0cb7c78e01751a44ef56cf20344e87 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Sun, 14 Jun 2009 00:59:50 +0530
Subject: x86, cpu: cpu/proc.c display cache alignment and address sizes for 32
 bit

32 bits can also access x86_cache_alignment, x86_phys_bits and
x86_virt_bits, make them available to user space just as on 64 bits.

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
LKML-Reference: <1244921390.11733.30.camel@ht.satnam>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/kernel/cpu/proc.c | 2 --
 1 file changed, 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/proc.c b/arch/x86/kernel/cpu/proc.c
index d5e30397246b..f82706a3901d 100644
--- a/arch/x86/kernel/cpu/proc.c
+++ b/arch/x86/kernel/cpu/proc.c
@@ -116,11 +116,9 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 		seq_printf(m, "TLB size\t: %d 4K pages\n", c->x86_tlbsize);
 #endif
 	seq_printf(m, "clflush size\t: %u\n", c->x86_clflush_size);
-#ifdef CONFIG_X86_64
 	seq_printf(m, "cache_alignment\t: %d\n", c->x86_cache_alignment);
 	seq_printf(m, "address sizes\t: %u bits physical, %u bits virtual\n",
 		   c->x86_phys_bits, c->x86_virt_bits);
-#endif
 
 	seq_printf(m, "power management:");
 	for (i = 0; i < 32; i++) {
-- 
cgit v1.2.3


From ac5672f82c39ff2f8dce81bf3e68b1dfc41f366f Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Date: Tue, 14 Apr 2009 14:29:44 -0700
Subject: x86/paravirt: split paravirt definitions into paravirt_types.h

Split the monolithic asm/paravirt.h into separate paravirt.h (inlines and other
"active" definitions), and paravirt_types.h (types, constants and other "passive"
definitions).  This makes it easier to use the type/constant definitions without
pulling in everything else and causing circular dependency problems.

[ Impact: cleanup ]

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
---
 arch/x86/include/asm/paravirt.h       | 711 +--------------------------------
 arch/x86/include/asm/paravirt_types.h | 720 ++++++++++++++++++++++++++++++++++
 2 files changed, 721 insertions(+), 710 deletions(-)
 create mode 100644 arch/x86/include/asm/paravirt_types.h

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h
index 4fb37c8a0832..6a07af432c81 100644
--- a/arch/x86/include/asm/paravirt.h
+++ b/arch/x86/include/asm/paravirt.h
@@ -7,689 +7,11 @@
 #include <asm/pgtable_types.h>
 #include <asm/asm.h>
 
-/* Bitmask of what can be clobbered: usually at least eax. */
-#define CLBR_NONE 0
-#define CLBR_EAX  (1 << 0)
-#define CLBR_ECX  (1 << 1)
-#define CLBR_EDX  (1 << 2)
-#define CLBR_EDI  (1 << 3)
-
-#ifdef CONFIG_X86_32
-/* CLBR_ANY should match all regs platform has. For i386, that's just it */
-#define CLBR_ANY  ((1 << 4) - 1)
-
-#define CLBR_ARG_REGS	(CLBR_EAX | CLBR_EDX | CLBR_ECX)
-#define CLBR_RET_REG	(CLBR_EAX | CLBR_EDX)
-#define CLBR_SCRATCH	(0)
-#else
-#define CLBR_RAX  CLBR_EAX
-#define CLBR_RCX  CLBR_ECX
-#define CLBR_RDX  CLBR_EDX
-#define CLBR_RDI  CLBR_EDI
-#define CLBR_RSI  (1 << 4)
-#define CLBR_R8   (1 << 5)
-#define CLBR_R9   (1 << 6)
-#define CLBR_R10  (1 << 7)
-#define CLBR_R11  (1 << 8)
-
-#define CLBR_ANY  ((1 << 9) - 1)
-
-#define CLBR_ARG_REGS	(CLBR_RDI | CLBR_RSI | CLBR_RDX | \
-			 CLBR_RCX | CLBR_R8 | CLBR_R9)
-#define CLBR_RET_REG	(CLBR_RAX)
-#define CLBR_SCRATCH	(CLBR_R10 | CLBR_R11)
-
-#include <asm/desc_defs.h>
-#endif /* X86_64 */
-
-#define CLBR_CALLEE_SAVE ((CLBR_ARG_REGS | CLBR_SCRATCH) & ~CLBR_RET_REG)
+#include <asm/paravirt_types.h>
 
 #ifndef __ASSEMBLY__
 #include <linux/types.h>
 #include <linux/cpumask.h>
-#include <asm/kmap_types.h>
-#include <asm/desc_defs.h>
-
-struct page;
-struct thread_struct;
-struct desc_ptr;
-struct tss_struct;
-struct mm_struct;
-struct desc_struct;
-struct task_struct;
-
-/*
- * Wrapper type for pointers to code which uses the non-standard
- * calling convention.  See PV_CALL_SAVE_REGS_THUNK below.
- */
-struct paravirt_callee_save {
-	void *func;
-};
-
-/* general info */
-struct pv_info {
-	unsigned int kernel_rpl;
-	int shared_kernel_pmd;
-	int paravirt_enabled;
-	const char *name;
-};
-
-struct pv_init_ops {
-	/*
-	 * Patch may replace one of the defined code sequences with
-	 * arbitrary code, subject to the same register constraints.
-	 * This generally means the code is not free to clobber any
-	 * registers other than EAX.  The patch function should return
-	 * the number of bytes of code generated, as we nop pad the
-	 * rest in generic code.
-	 */
-	unsigned (*patch)(u8 type, u16 clobber, void *insnbuf,
-			  unsigned long addr, unsigned len);
-
-	/* Basic arch-specific setup */
-	void (*arch_setup)(void);
-	char *(*memory_setup)(void);
-	void (*post_allocator_init)(void);
-
-	/* Print a banner to identify the environment */
-	void (*banner)(void);
-};
-
-
-struct pv_lazy_ops {
-	/* Set deferred update mode, used for batching operations. */
-	void (*enter)(void);
-	void (*leave)(void);
-};
-
-struct pv_time_ops {
-	void (*time_init)(void);
-
-	/* Set and set time of day */
-	unsigned long (*get_wallclock)(void);
-	int (*set_wallclock)(unsigned long);
-
-	unsigned long long (*sched_clock)(void);
-	unsigned long (*get_tsc_khz)(void);
-};
-
-struct pv_cpu_ops {
-	/* hooks for various privileged instructions */
-	unsigned long (*get_debugreg)(int regno);
-	void (*set_debugreg)(int regno, unsigned long value);
-
-	void (*clts)(void);
-
-	unsigned long (*read_cr0)(void);
-	void (*write_cr0)(unsigned long);
-
-	unsigned long (*read_cr4_safe)(void);
-	unsigned long (*read_cr4)(void);
-	void (*write_cr4)(unsigned long);
-
-#ifdef CONFIG_X86_64
-	unsigned long (*read_cr8)(void);
-	void (*write_cr8)(unsigned long);
-#endif
-
-	/* Segment descriptor handling */
-	void (*load_tr_desc)(void);
-	void (*load_gdt)(const struct desc_ptr *);
-	void (*load_idt)(const struct desc_ptr *);
-	void (*store_gdt)(struct desc_ptr *);
-	void (*store_idt)(struct desc_ptr *);
-	void (*set_ldt)(const void *desc, unsigned entries);
-	unsigned long (*store_tr)(void);
-	void (*load_tls)(struct thread_struct *t, unsigned int cpu);
-#ifdef CONFIG_X86_64
-	void (*load_gs_index)(unsigned int idx);
-#endif
-	void (*write_ldt_entry)(struct desc_struct *ldt, int entrynum,
-				const void *desc);
-	void (*write_gdt_entry)(struct desc_struct *,
-				int entrynum, const void *desc, int size);
-	void (*write_idt_entry)(gate_desc *,
-				int entrynum, const gate_desc *gate);
-	void (*alloc_ldt)(struct desc_struct *ldt, unsigned entries);
-	void (*free_ldt)(struct desc_struct *ldt, unsigned entries);
-
-	void (*load_sp0)(struct tss_struct *tss, struct thread_struct *t);
-
-	void (*set_iopl_mask)(unsigned mask);
-
-	void (*wbinvd)(void);
-	void (*io_delay)(void);
-
-	/* cpuid emulation, mostly so that caps bits can be disabled */
-	void (*cpuid)(unsigned int *eax, unsigned int *ebx,
-		      unsigned int *ecx, unsigned int *edx);
-
-	/* MSR, PMC and TSR operations.
-	   err = 0/-EFAULT.  wrmsr returns 0/-EFAULT. */
-	u64 (*read_msr_amd)(unsigned int msr, int *err);
-	u64 (*read_msr)(unsigned int msr, int *err);
-	int (*write_msr)(unsigned int msr, unsigned low, unsigned high);
-
-	u64 (*read_tsc)(void);
-	u64 (*read_pmc)(int counter);
-	unsigned long long (*read_tscp)(unsigned int *aux);
-
-	/*
-	 * Atomically enable interrupts and return to userspace.  This
-	 * is only ever used to return to 32-bit processes; in a
-	 * 64-bit kernel, it's used for 32-on-64 compat processes, but
-	 * never native 64-bit processes.  (Jump, not call.)
-	 */
-	void (*irq_enable_sysexit)(void);
-
-	/*
-	 * Switch to usermode gs and return to 64-bit usermode using
-	 * sysret.  Only used in 64-bit kernels to return to 64-bit
-	 * processes.  Usermode register state, including %rsp, must
-	 * already be restored.
-	 */
-	void (*usergs_sysret64)(void);
-
-	/*
-	 * Switch to usermode gs and return to 32-bit usermode using
-	 * sysret.  Used to return to 32-on-64 compat processes.
-	 * Other usermode register state, including %esp, must already
-	 * be restored.
-	 */
-	void (*usergs_sysret32)(void);
-
-	/* Normal iret.  Jump to this with the standard iret stack
-	   frame set up. */
-	void (*iret)(void);
-
-	void (*swapgs)(void);
-
-	void (*start_context_switch)(struct task_struct *prev);
-	void (*end_context_switch)(struct task_struct *next);
-};
-
-struct pv_irq_ops {
-	void (*init_IRQ)(void);
-
-	/*
-	 * Get/set interrupt state.  save_fl and restore_fl are only
-	 * expected to use X86_EFLAGS_IF; all other bits
-	 * returned from save_fl are undefined, and may be ignored by
-	 * restore_fl.
-	 *
-	 * NOTE: These functions callers expect the callee to preserve
-	 * more registers than the standard C calling convention.
-	 */
-	struct paravirt_callee_save save_fl;
-	struct paravirt_callee_save restore_fl;
-	struct paravirt_callee_save irq_disable;
-	struct paravirt_callee_save irq_enable;
-
-	void (*safe_halt)(void);
-	void (*halt)(void);
-
-#ifdef CONFIG_X86_64
-	void (*adjust_exception_frame)(void);
-#endif
-};
-
-struct pv_apic_ops {
-#ifdef CONFIG_X86_LOCAL_APIC
-	void (*setup_boot_clock)(void);
-	void (*setup_secondary_clock)(void);
-
-	void (*startup_ipi_hook)(int phys_apicid,
-				 unsigned long start_eip,
-				 unsigned long start_esp);
-#endif
-};
-
-struct pv_mmu_ops {
-	/*
-	 * Called before/after init_mm pagetable setup. setup_start
-	 * may reset %cr3, and may pre-install parts of the pagetable;
-	 * pagetable setup is expected to preserve any existing
-	 * mapping.
-	 */
-	void (*pagetable_setup_start)(pgd_t *pgd_base);
-	void (*pagetable_setup_done)(pgd_t *pgd_base);
-
-	unsigned long (*read_cr2)(void);
-	void (*write_cr2)(unsigned long);
-
-	unsigned long (*read_cr3)(void);
-	void (*write_cr3)(unsigned long);
-
-	/*
-	 * Hooks for intercepting the creation/use/destruction of an
-	 * mm_struct.
-	 */
-	void (*activate_mm)(struct mm_struct *prev,
-			    struct mm_struct *next);
-	void (*dup_mmap)(struct mm_struct *oldmm,
-			 struct mm_struct *mm);
-	void (*exit_mmap)(struct mm_struct *mm);
-
-
-	/* TLB operations */
-	void (*flush_tlb_user)(void);
-	void (*flush_tlb_kernel)(void);
-	void (*flush_tlb_single)(unsigned long addr);
-	void (*flush_tlb_others)(const struct cpumask *cpus,
-				 struct mm_struct *mm,
-				 unsigned long va);
-
-	/* Hooks for allocating and freeing a pagetable top-level */
-	int  (*pgd_alloc)(struct mm_struct *mm);
-	void (*pgd_free)(struct mm_struct *mm, pgd_t *pgd);
-
-	/*
-	 * Hooks for allocating/releasing pagetable pages when they're
-	 * attached to a pagetable
-	 */
-	void (*alloc_pte)(struct mm_struct *mm, unsigned long pfn);
-	void (*alloc_pmd)(struct mm_struct *mm, unsigned long pfn);
-	void (*alloc_pmd_clone)(unsigned long pfn, unsigned long clonepfn, unsigned long start, unsigned long count);
-	void (*alloc_pud)(struct mm_struct *mm, unsigned long pfn);
-	void (*release_pte)(unsigned long pfn);
-	void (*release_pmd)(unsigned long pfn);
-	void (*release_pud)(unsigned long pfn);
-
-	/* Pagetable manipulation functions */
-	void (*set_pte)(pte_t *ptep, pte_t pteval);
-	void (*set_pte_at)(struct mm_struct *mm, unsigned long addr,
-			   pte_t *ptep, pte_t pteval);
-	void (*set_pmd)(pmd_t *pmdp, pmd_t pmdval);
-	void (*pte_update)(struct mm_struct *mm, unsigned long addr,
-			   pte_t *ptep);
-	void (*pte_update_defer)(struct mm_struct *mm,
-				 unsigned long addr, pte_t *ptep);
-
-	pte_t (*ptep_modify_prot_start)(struct mm_struct *mm, unsigned long addr,
-					pte_t *ptep);
-	void (*ptep_modify_prot_commit)(struct mm_struct *mm, unsigned long addr,
-					pte_t *ptep, pte_t pte);
-
-	struct paravirt_callee_save pte_val;
-	struct paravirt_callee_save make_pte;
-
-	struct paravirt_callee_save pgd_val;
-	struct paravirt_callee_save make_pgd;
-
-#if PAGETABLE_LEVELS >= 3
-#ifdef CONFIG_X86_PAE
-	void (*set_pte_atomic)(pte_t *ptep, pte_t pteval);
-	void (*pte_clear)(struct mm_struct *mm, unsigned long addr,
-			  pte_t *ptep);
-	void (*pmd_clear)(pmd_t *pmdp);
-
-#endif	/* CONFIG_X86_PAE */
-
-	void (*set_pud)(pud_t *pudp, pud_t pudval);
-
-	struct paravirt_callee_save pmd_val;
-	struct paravirt_callee_save make_pmd;
-
-#if PAGETABLE_LEVELS == 4
-	struct paravirt_callee_save pud_val;
-	struct paravirt_callee_save make_pud;
-
-	void (*set_pgd)(pgd_t *pudp, pgd_t pgdval);
-#endif	/* PAGETABLE_LEVELS == 4 */
-#endif	/* PAGETABLE_LEVELS >= 3 */
-
-#ifdef CONFIG_HIGHPTE
-	void *(*kmap_atomic_pte)(struct page *page, enum km_type type);
-#endif
-
-	struct pv_lazy_ops lazy_mode;
-
-	/* dom0 ops */
-
-	/* Sometimes the physical address is a pfn, and sometimes its
-	   an mfn.  We can tell which is which from the index. */
-	void (*set_fixmap)(unsigned /* enum fixed_addresses */ idx,
-			   phys_addr_t phys, pgprot_t flags);
-};
-
-struct raw_spinlock;
-struct pv_lock_ops {
-	int (*spin_is_locked)(struct raw_spinlock *lock);
-	int (*spin_is_contended)(struct raw_spinlock *lock);
-	void (*spin_lock)(struct raw_spinlock *lock);
-	void (*spin_lock_flags)(struct raw_spinlock *lock, unsigned long flags);
-	int (*spin_trylock)(struct raw_spinlock *lock);
-	void (*spin_unlock)(struct raw_spinlock *lock);
-};
-
-/* This contains all the paravirt structures: we get a convenient
- * number for each function using the offset which we use to indicate
- * what to patch. */
-struct paravirt_patch_template {
-	struct pv_init_ops pv_init_ops;
-	struct pv_time_ops pv_time_ops;
-	struct pv_cpu_ops pv_cpu_ops;
-	struct pv_irq_ops pv_irq_ops;
-	struct pv_apic_ops pv_apic_ops;
-	struct pv_mmu_ops pv_mmu_ops;
-	struct pv_lock_ops pv_lock_ops;
-};
-
-extern struct pv_info pv_info;
-extern struct pv_init_ops pv_init_ops;
-extern struct pv_time_ops pv_time_ops;
-extern struct pv_cpu_ops pv_cpu_ops;
-extern struct pv_irq_ops pv_irq_ops;
-extern struct pv_apic_ops pv_apic_ops;
-extern struct pv_mmu_ops pv_mmu_ops;
-extern struct pv_lock_ops pv_lock_ops;
-
-#define PARAVIRT_PATCH(x)					\
-	(offsetof(struct paravirt_patch_template, x) / sizeof(void *))
-
-#define paravirt_type(op)				\
-	[paravirt_typenum] "i" (PARAVIRT_PATCH(op)),	\
-	[paravirt_opptr] "i" (&(op))
-#define paravirt_clobber(clobber)		\
-	[paravirt_clobber] "i" (clobber)
-
-/*
- * Generate some code, and mark it as patchable by the
- * apply_paravirt() alternate instruction patcher.
- */
-#define _paravirt_alt(insn_string, type, clobber)	\
-	"771:\n\t" insn_string "\n" "772:\n"		\
-	".pushsection .parainstructions,\"a\"\n"	\
-	_ASM_ALIGN "\n"					\
-	_ASM_PTR " 771b\n"				\
-	"  .byte " type "\n"				\
-	"  .byte 772b-771b\n"				\
-	"  .short " clobber "\n"			\
-	".popsection\n"
-
-/* Generate patchable code, with the default asm parameters. */
-#define paravirt_alt(insn_string)					\
-	_paravirt_alt(insn_string, "%c[paravirt_typenum]", "%c[paravirt_clobber]")
-
-/* Simple instruction patching code. */
-#define DEF_NATIVE(ops, name, code) 					\
-	extern const char start_##ops##_##name[], end_##ops##_##name[];	\
-	asm("start_" #ops "_" #name ": " code "; end_" #ops "_" #name ":")
-
-unsigned paravirt_patch_nop(void);
-unsigned paravirt_patch_ident_32(void *insnbuf, unsigned len);
-unsigned paravirt_patch_ident_64(void *insnbuf, unsigned len);
-unsigned paravirt_patch_ignore(unsigned len);
-unsigned paravirt_patch_call(void *insnbuf,
-			     const void *target, u16 tgt_clobbers,
-			     unsigned long addr, u16 site_clobbers,
-			     unsigned len);
-unsigned paravirt_patch_jmp(void *insnbuf, const void *target,
-			    unsigned long addr, unsigned len);
-unsigned paravirt_patch_default(u8 type, u16 clobbers, void *insnbuf,
-				unsigned long addr, unsigned len);
-
-unsigned paravirt_patch_insns(void *insnbuf, unsigned len,
-			      const char *start, const char *end);
-
-unsigned native_patch(u8 type, u16 clobbers, void *ibuf,
-		      unsigned long addr, unsigned len);
-
-int paravirt_disable_iospace(void);
-
-/*
- * This generates an indirect call based on the operation type number.
- * The type number, computed in PARAVIRT_PATCH, is derived from the
- * offset into the paravirt_patch_template structure, and can therefore be
- * freely converted back into a structure offset.
- */
-#define PARAVIRT_CALL	"call *%c[paravirt_opptr];"
-
-/*
- * These macros are intended to wrap calls through one of the paravirt
- * ops structs, so that they can be later identified and patched at
- * runtime.
- *
- * Normally, a call to a pv_op function is a simple indirect call:
- * (pv_op_struct.operations)(args...).
- *
- * Unfortunately, this is a relatively slow operation for modern CPUs,
- * because it cannot necessarily determine what the destination
- * address is.  In this case, the address is a runtime constant, so at
- * the very least we can patch the call to e a simple direct call, or
- * ideally, patch an inline implementation into the callsite.  (Direct
- * calls are essentially free, because the call and return addresses
- * are completely predictable.)
- *
- * For i386, these macros rely on the standard gcc "regparm(3)" calling
- * convention, in which the first three arguments are placed in %eax,
- * %edx, %ecx (in that order), and the remaining arguments are placed
- * on the stack.  All caller-save registers (eax,edx,ecx) are expected
- * to be modified (either clobbered or used for return values).
- * X86_64, on the other hand, already specifies a register-based calling
- * conventions, returning at %rax, with parameteres going on %rdi, %rsi,
- * %rdx, and %rcx. Note that for this reason, x86_64 does not need any
- * special handling for dealing with 4 arguments, unlike i386.
- * However, x86_64 also have to clobber all caller saved registers, which
- * unfortunately, are quite a bit (r8 - r11)
- *
- * The call instruction itself is marked by placing its start address
- * and size into the .parainstructions section, so that
- * apply_paravirt() in arch/i386/kernel/alternative.c can do the
- * appropriate patching under the control of the backend pv_init_ops
- * implementation.
- *
- * Unfortunately there's no way to get gcc to generate the args setup
- * for the call, and then allow the call itself to be generated by an
- * inline asm.  Because of this, we must do the complete arg setup and
- * return value handling from within these macros.  This is fairly
- * cumbersome.
- *
- * There are 5 sets of PVOP_* macros for dealing with 0-4 arguments.
- * It could be extended to more arguments, but there would be little
- * to be gained from that.  For each number of arguments, there are
- * the two VCALL and CALL variants for void and non-void functions.
- *
- * When there is a return value, the invoker of the macro must specify
- * the return type.  The macro then uses sizeof() on that type to
- * determine whether its a 32 or 64 bit value, and places the return
- * in the right register(s) (just %eax for 32-bit, and %edx:%eax for
- * 64-bit). For x86_64 machines, it just returns at %rax regardless of
- * the return value size.
- *
- * 64-bit arguments are passed as a pair of adjacent 32-bit arguments
- * i386 also passes 64-bit arguments as a pair of adjacent 32-bit arguments
- * in low,high order
- *
- * Small structures are passed and returned in registers.  The macro
- * calling convention can't directly deal with this, so the wrapper
- * functions must do this.
- *
- * These PVOP_* macros are only defined within this header.  This
- * means that all uses must be wrapped in inline functions.  This also
- * makes sure the incoming and outgoing types are always correct.
- */
-#ifdef CONFIG_X86_32
-#define PVOP_VCALL_ARGS				\
-	unsigned long __eax = __eax, __edx = __edx, __ecx = __ecx
-#define PVOP_CALL_ARGS			PVOP_VCALL_ARGS
-
-#define PVOP_CALL_ARG1(x)		"a" ((unsigned long)(x))
-#define PVOP_CALL_ARG2(x)		"d" ((unsigned long)(x))
-#define PVOP_CALL_ARG3(x)		"c" ((unsigned long)(x))
-
-#define PVOP_VCALL_CLOBBERS		"=a" (__eax), "=d" (__edx),	\
-					"=c" (__ecx)
-#define PVOP_CALL_CLOBBERS		PVOP_VCALL_CLOBBERS
-
-#define PVOP_VCALLEE_CLOBBERS		"=a" (__eax), "=d" (__edx)
-#define PVOP_CALLEE_CLOBBERS		PVOP_VCALLEE_CLOBBERS
-
-#define EXTRA_CLOBBERS
-#define VEXTRA_CLOBBERS
-#else  /* CONFIG_X86_64 */
-#define PVOP_VCALL_ARGS					\
-	unsigned long __edi = __edi, __esi = __esi,	\
-		__edx = __edx, __ecx = __ecx
-#define PVOP_CALL_ARGS		PVOP_VCALL_ARGS, __eax
-
-#define PVOP_CALL_ARG1(x)		"D" ((unsigned long)(x))
-#define PVOP_CALL_ARG2(x)		"S" ((unsigned long)(x))
-#define PVOP_CALL_ARG3(x)		"d" ((unsigned long)(x))
-#define PVOP_CALL_ARG4(x)		"c" ((unsigned long)(x))
-
-#define PVOP_VCALL_CLOBBERS	"=D" (__edi),				\
-				"=S" (__esi), "=d" (__edx),		\
-				"=c" (__ecx)
-#define PVOP_CALL_CLOBBERS	PVOP_VCALL_CLOBBERS, "=a" (__eax)
-
-#define PVOP_VCALLEE_CLOBBERS	"=a" (__eax)
-#define PVOP_CALLEE_CLOBBERS	PVOP_VCALLEE_CLOBBERS
-
-#define EXTRA_CLOBBERS	 , "r8", "r9", "r10", "r11"
-#define VEXTRA_CLOBBERS	 , "rax", "r8", "r9", "r10", "r11"
-#endif	/* CONFIG_X86_32 */
-
-#ifdef CONFIG_PARAVIRT_DEBUG
-#define PVOP_TEST_NULL(op)	BUG_ON(op == NULL)
-#else
-#define PVOP_TEST_NULL(op)	((void)op)
-#endif
-
-#define ____PVOP_CALL(rettype, op, clbr, call_clbr, extra_clbr,		\
-		      pre, post, ...)					\
-	({								\
-		rettype __ret;						\
-		PVOP_CALL_ARGS;						\
-		PVOP_TEST_NULL(op);					\
-		/* This is 32-bit specific, but is okay in 64-bit */	\
-		/* since this condition will never hold */		\
-		if (sizeof(rettype) > sizeof(unsigned long)) {		\
-			asm volatile(pre				\
-				     paravirt_alt(PARAVIRT_CALL)	\
-				     post				\
-				     : call_clbr			\
-				     : paravirt_type(op),		\
-				       paravirt_clobber(clbr),		\
-				       ##__VA_ARGS__			\
-				     : "memory", "cc" extra_clbr);	\
-			__ret = (rettype)((((u64)__edx) << 32) | __eax); \
-		} else {						\
-			asm volatile(pre				\
-				     paravirt_alt(PARAVIRT_CALL)	\
-				     post				\
-				     : call_clbr			\
-				     : paravirt_type(op),		\
-				       paravirt_clobber(clbr),		\
-				       ##__VA_ARGS__			\
-				     : "memory", "cc" extra_clbr);	\
-			__ret = (rettype)__eax;				\
-		}							\
-		__ret;							\
-	})
-
-#define __PVOP_CALL(rettype, op, pre, post, ...)			\
-	____PVOP_CALL(rettype, op, CLBR_ANY, PVOP_CALL_CLOBBERS,	\
-		      EXTRA_CLOBBERS, pre, post, ##__VA_ARGS__)
-
-#define __PVOP_CALLEESAVE(rettype, op, pre, post, ...)			\
-	____PVOP_CALL(rettype, op.func, CLBR_RET_REG,			\
-		      PVOP_CALLEE_CLOBBERS, ,				\
-		      pre, post, ##__VA_ARGS__)
-
-
-#define ____PVOP_VCALL(op, clbr, call_clbr, extra_clbr, pre, post, ...)	\
-	({								\
-		PVOP_VCALL_ARGS;					\
-		PVOP_TEST_NULL(op);					\
-		asm volatile(pre					\
-			     paravirt_alt(PARAVIRT_CALL)		\
-			     post					\
-			     : call_clbr				\
-			     : paravirt_type(op),			\
-			       paravirt_clobber(clbr),			\
-			       ##__VA_ARGS__				\
-			     : "memory", "cc" extra_clbr);		\
-	})
-
-#define __PVOP_VCALL(op, pre, post, ...)				\
-	____PVOP_VCALL(op, CLBR_ANY, PVOP_VCALL_CLOBBERS,		\
-		       VEXTRA_CLOBBERS,					\
-		       pre, post, ##__VA_ARGS__)
-
-#define __PVOP_VCALLEESAVE(rettype, op, pre, post, ...)			\
-	____PVOP_CALL(rettype, op.func, CLBR_RET_REG,			\
-		      PVOP_VCALLEE_CLOBBERS, ,				\
-		      pre, post, ##__VA_ARGS__)
-
-
-
-#define PVOP_CALL0(rettype, op)						\
-	__PVOP_CALL(rettype, op, "", "")
-#define PVOP_VCALL0(op)							\
-	__PVOP_VCALL(op, "", "")
-
-#define PVOP_CALLEE0(rettype, op)					\
-	__PVOP_CALLEESAVE(rettype, op, "", "")
-#define PVOP_VCALLEE0(op)						\
-	__PVOP_VCALLEESAVE(op, "", "")
-
-
-#define PVOP_CALL1(rettype, op, arg1)					\
-	__PVOP_CALL(rettype, op, "", "", PVOP_CALL_ARG1(arg1))
-#define PVOP_VCALL1(op, arg1)						\
-	__PVOP_VCALL(op, "", "", PVOP_CALL_ARG1(arg1))
-
-#define PVOP_CALLEE1(rettype, op, arg1)					\
-	__PVOP_CALLEESAVE(rettype, op, "", "", PVOP_CALL_ARG1(arg1))
-#define PVOP_VCALLEE1(op, arg1)						\
-	__PVOP_VCALLEESAVE(op, "", "", PVOP_CALL_ARG1(arg1))
-
-
-#define PVOP_CALL2(rettype, op, arg1, arg2)				\
-	__PVOP_CALL(rettype, op, "", "", PVOP_CALL_ARG1(arg1),		\
-		    PVOP_CALL_ARG2(arg2))
-#define PVOP_VCALL2(op, arg1, arg2)					\
-	__PVOP_VCALL(op, "", "", PVOP_CALL_ARG1(arg1),			\
-		     PVOP_CALL_ARG2(arg2))
-
-#define PVOP_CALLEE2(rettype, op, arg1, arg2)				\
-	__PVOP_CALLEESAVE(rettype, op, "", "", PVOP_CALL_ARG1(arg1),	\
-			  PVOP_CALL_ARG2(arg2))
-#define PVOP_VCALLEE2(op, arg1, arg2)					\
-	__PVOP_VCALLEESAVE(op, "", "", PVOP_CALL_ARG1(arg1),		\
-			   PVOP_CALL_ARG2(arg2))
-
-
-#define PVOP_CALL3(rettype, op, arg1, arg2, arg3)			\
-	__PVOP_CALL(rettype, op, "", "", PVOP_CALL_ARG1(arg1),		\
-		    PVOP_CALL_ARG2(arg2), PVOP_CALL_ARG3(arg3))
-#define PVOP_VCALL3(op, arg1, arg2, arg3)				\
-	__PVOP_VCALL(op, "", "", PVOP_CALL_ARG1(arg1),			\
-		     PVOP_CALL_ARG2(arg2), PVOP_CALL_ARG3(arg3))
-
-/* This is the only difference in x86_64. We can make it much simpler */
-#ifdef CONFIG_X86_32
-#define PVOP_CALL4(rettype, op, arg1, arg2, arg3, arg4)			\
-	__PVOP_CALL(rettype, op,					\
-		    "push %[_arg4];", "lea 4(%%esp),%%esp;",		\
-		    PVOP_CALL_ARG1(arg1), PVOP_CALL_ARG2(arg2),		\
-		    PVOP_CALL_ARG3(arg3), [_arg4] "mr" ((u32)(arg4)))
-#define PVOP_VCALL4(op, arg1, arg2, arg3, arg4)				\
-	__PVOP_VCALL(op,						\
-		    "push %[_arg4];", "lea 4(%%esp),%%esp;",		\
-		    "0" ((u32)(arg1)), "1" ((u32)(arg2)),		\
-		    "2" ((u32)(arg3)), [_arg4] "mr" ((u32)(arg4)))
-#else
-#define PVOP_CALL4(rettype, op, arg1, arg2, arg3, arg4)			\
-	__PVOP_CALL(rettype, op, "", "",				\
-		    PVOP_CALL_ARG1(arg1), PVOP_CALL_ARG2(arg2),		\
-		    PVOP_CALL_ARG3(arg3), PVOP_CALL_ARG4(arg4))
-#define PVOP_VCALL4(op, arg1, arg2, arg3, arg4)				\
-	__PVOP_VCALL(op, "", "",					\
-		     PVOP_CALL_ARG1(arg1), PVOP_CALL_ARG2(arg2),	\
-		     PVOP_CALL_ARG3(arg3), PVOP_CALL_ARG4(arg4))
-#endif
 
 static inline int paravirt_enabled(void)
 {
@@ -1393,20 +715,6 @@ static inline void pmd_clear(pmd_t *pmdp)
 }
 #endif	/* CONFIG_X86_PAE */
 
-/* Lazy mode for batching updates / context switch */
-enum paravirt_lazy_mode {
-	PARAVIRT_LAZY_NONE,
-	PARAVIRT_LAZY_MMU,
-	PARAVIRT_LAZY_CPU,
-};
-
-enum paravirt_lazy_mode paravirt_get_lazy_mode(void);
-void paravirt_start_context_switch(struct task_struct *prev);
-void paravirt_end_context_switch(struct task_struct *next);
-
-void paravirt_enter_lazy_mmu(void);
-void paravirt_leave_lazy_mmu(void);
-
 #define  __HAVE_ARCH_START_CONTEXT_SWITCH
 static inline void arch_start_context_switch(struct task_struct *prev)
 {
@@ -1437,12 +745,6 @@ static inline void __set_fixmap(unsigned /* enum fixed_addresses */ idx,
 	pv_mmu_ops.set_fixmap(idx, phys, flags);
 }
 
-void _paravirt_nop(void);
-u32 _paravirt_ident_32(u32);
-u64 _paravirt_ident_64(u64);
-
-#define paravirt_nop	((void *)_paravirt_nop)
-
 #if defined(CONFIG_SMP) && defined(CONFIG_PARAVIRT_SPINLOCKS)
 
 static inline int __raw_spin_is_locked(struct raw_spinlock *lock)
@@ -1479,17 +781,6 @@ static __always_inline void __raw_spin_unlock(struct raw_spinlock *lock)
 
 #endif
 
-/* These all sit in the .parainstructions section to tell us what to patch. */
-struct paravirt_patch_site {
-	u8 *instr; 		/* original instructions */
-	u8 instrtype;		/* type of this instruction */
-	u8 len;			/* length of original instruction */
-	u16 clobbers;		/* what registers you may clobber */
-};
-
-extern struct paravirt_patch_site __parainstructions[],
-	__parainstructions_end[];
-
 #ifdef CONFIG_X86_32
 #define PV_SAVE_REGS "pushl %ecx; pushl %edx;"
 #define PV_RESTORE_REGS "popl %edx; popl %ecx;"
diff --git a/arch/x86/include/asm/paravirt_types.h b/arch/x86/include/asm/paravirt_types.h
new file mode 100644
index 000000000000..2b3371bae295
--- /dev/null
+++ b/arch/x86/include/asm/paravirt_types.h
@@ -0,0 +1,720 @@
+#ifndef _ASM_X86_PARAVIRT_TYPES_H
+#define _ASM_X86_PARAVIRT_TYPES_H
+
+/* Bitmask of what can be clobbered: usually at least eax. */
+#define CLBR_NONE 0
+#define CLBR_EAX  (1 << 0)
+#define CLBR_ECX  (1 << 1)
+#define CLBR_EDX  (1 << 2)
+#define CLBR_EDI  (1 << 3)
+
+#ifdef CONFIG_X86_32
+/* CLBR_ANY should match all regs platform has. For i386, that's just it */
+#define CLBR_ANY  ((1 << 4) - 1)
+
+#define CLBR_ARG_REGS	(CLBR_EAX | CLBR_EDX | CLBR_ECX)
+#define CLBR_RET_REG	(CLBR_EAX | CLBR_EDX)
+#define CLBR_SCRATCH	(0)
+#else
+#define CLBR_RAX  CLBR_EAX
+#define CLBR_RCX  CLBR_ECX
+#define CLBR_RDX  CLBR_EDX
+#define CLBR_RDI  CLBR_EDI
+#define CLBR_RSI  (1 << 4)
+#define CLBR_R8   (1 << 5)
+#define CLBR_R9   (1 << 6)
+#define CLBR_R10  (1 << 7)
+#define CLBR_R11  (1 << 8)
+
+#define CLBR_ANY  ((1 << 9) - 1)
+
+#define CLBR_ARG_REGS	(CLBR_RDI | CLBR_RSI | CLBR_RDX | \
+			 CLBR_RCX | CLBR_R8 | CLBR_R9)
+#define CLBR_RET_REG	(CLBR_RAX)
+#define CLBR_SCRATCH	(CLBR_R10 | CLBR_R11)
+
+#endif /* X86_64 */
+
+#define CLBR_CALLEE_SAVE ((CLBR_ARG_REGS | CLBR_SCRATCH) & ~CLBR_RET_REG)
+
+#ifndef __ASSEMBLY__
+
+#include <asm/desc_defs.h>
+#include <asm/kmap_types.h>
+
+struct page;
+struct thread_struct;
+struct desc_ptr;
+struct tss_struct;
+struct mm_struct;
+struct desc_struct;
+struct task_struct;
+struct cpumask;
+
+/*
+ * Wrapper type for pointers to code which uses the non-standard
+ * calling convention.  See PV_CALL_SAVE_REGS_THUNK below.
+ */
+struct paravirt_callee_save {
+	void *func;
+};
+
+/* general info */
+struct pv_info {
+	unsigned int kernel_rpl;
+	int shared_kernel_pmd;
+	int paravirt_enabled;
+	const char *name;
+};
+
+struct pv_init_ops {
+	/*
+	 * Patch may replace one of the defined code sequences with
+	 * arbitrary code, subject to the same register constraints.
+	 * This generally means the code is not free to clobber any
+	 * registers other than EAX.  The patch function should return
+	 * the number of bytes of code generated, as we nop pad the
+	 * rest in generic code.
+	 */
+	unsigned (*patch)(u8 type, u16 clobber, void *insnbuf,
+			  unsigned long addr, unsigned len);
+
+	/* Basic arch-specific setup */
+	void (*arch_setup)(void);
+	char *(*memory_setup)(void);
+	void (*post_allocator_init)(void);
+
+	/* Print a banner to identify the environment */
+	void (*banner)(void);
+};
+
+
+struct pv_lazy_ops {
+	/* Set deferred update mode, used for batching operations. */
+	void (*enter)(void);
+	void (*leave)(void);
+};
+
+struct pv_time_ops {
+	void (*time_init)(void);
+
+	/* Set and set time of day */
+	unsigned long (*get_wallclock)(void);
+	int (*set_wallclock)(unsigned long);
+
+	unsigned long long (*sched_clock)(void);
+	unsigned long (*get_tsc_khz)(void);
+};
+
+struct pv_cpu_ops {
+	/* hooks for various privileged instructions */
+	unsigned long (*get_debugreg)(int regno);
+	void (*set_debugreg)(int regno, unsigned long value);
+
+	void (*clts)(void);
+
+	unsigned long (*read_cr0)(void);
+	void (*write_cr0)(unsigned long);
+
+	unsigned long (*read_cr4_safe)(void);
+	unsigned long (*read_cr4)(void);
+	void (*write_cr4)(unsigned long);
+
+#ifdef CONFIG_X86_64
+	unsigned long (*read_cr8)(void);
+	void (*write_cr8)(unsigned long);
+#endif
+
+	/* Segment descriptor handling */
+	void (*load_tr_desc)(void);
+	void (*load_gdt)(const struct desc_ptr *);
+	void (*load_idt)(const struct desc_ptr *);
+	void (*store_gdt)(struct desc_ptr *);
+	void (*store_idt)(struct desc_ptr *);
+	void (*set_ldt)(const void *desc, unsigned entries);
+	unsigned long (*store_tr)(void);
+	void (*load_tls)(struct thread_struct *t, unsigned int cpu);
+#ifdef CONFIG_X86_64
+	void (*load_gs_index)(unsigned int idx);
+#endif
+	void (*write_ldt_entry)(struct desc_struct *ldt, int entrynum,
+				const void *desc);
+	void (*write_gdt_entry)(struct desc_struct *,
+				int entrynum, const void *desc, int size);
+	void (*write_idt_entry)(gate_desc *,
+				int entrynum, const gate_desc *gate);
+	void (*alloc_ldt)(struct desc_struct *ldt, unsigned entries);
+	void (*free_ldt)(struct desc_struct *ldt, unsigned entries);
+
+	void (*load_sp0)(struct tss_struct *tss, struct thread_struct *t);
+
+	void (*set_iopl_mask)(unsigned mask);
+
+	void (*wbinvd)(void);
+	void (*io_delay)(void);
+
+	/* cpuid emulation, mostly so that caps bits can be disabled */
+	void (*cpuid)(unsigned int *eax, unsigned int *ebx,
+		      unsigned int *ecx, unsigned int *edx);
+
+	/* MSR, PMC and TSR operations.
+	   err = 0/-EFAULT.  wrmsr returns 0/-EFAULT. */
+	u64 (*read_msr_amd)(unsigned int msr, int *err);
+	u64 (*read_msr)(unsigned int msr, int *err);
+	int (*write_msr)(unsigned int msr, unsigned low, unsigned high);
+
+	u64 (*read_tsc)(void);
+	u64 (*read_pmc)(int counter);
+	unsigned long long (*read_tscp)(unsigned int *aux);
+
+	/*
+	 * Atomically enable interrupts and return to userspace.  This
+	 * is only ever used to return to 32-bit processes; in a
+	 * 64-bit kernel, it's used for 32-on-64 compat processes, but
+	 * never native 64-bit processes.  (Jump, not call.)
+	 */
+	void (*irq_enable_sysexit)(void);
+
+	/*
+	 * Switch to usermode gs and return to 64-bit usermode using
+	 * sysret.  Only used in 64-bit kernels to return to 64-bit
+	 * processes.  Usermode register state, including %rsp, must
+	 * already be restored.
+	 */
+	void (*usergs_sysret64)(void);
+
+	/*
+	 * Switch to usermode gs and return to 32-bit usermode using
+	 * sysret.  Used to return to 32-on-64 compat processes.
+	 * Other usermode register state, including %esp, must already
+	 * be restored.
+	 */
+	void (*usergs_sysret32)(void);
+
+	/* Normal iret.  Jump to this with the standard iret stack
+	   frame set up. */
+	void (*iret)(void);
+
+	void (*swapgs)(void);
+
+	void (*start_context_switch)(struct task_struct *prev);
+	void (*end_context_switch)(struct task_struct *next);
+};
+
+struct pv_irq_ops {
+	void (*init_IRQ)(void);
+
+	/*
+	 * Get/set interrupt state.  save_fl and restore_fl are only
+	 * expected to use X86_EFLAGS_IF; all other bits
+	 * returned from save_fl are undefined, and may be ignored by
+	 * restore_fl.
+	 *
+	 * NOTE: These functions callers expect the callee to preserve
+	 * more registers than the standard C calling convention.
+	 */
+	struct paravirt_callee_save save_fl;
+	struct paravirt_callee_save restore_fl;
+	struct paravirt_callee_save irq_disable;
+	struct paravirt_callee_save irq_enable;
+
+	void (*safe_halt)(void);
+	void (*halt)(void);
+
+#ifdef CONFIG_X86_64
+	void (*adjust_exception_frame)(void);
+#endif
+};
+
+struct pv_apic_ops {
+#ifdef CONFIG_X86_LOCAL_APIC
+	void (*setup_boot_clock)(void);
+	void (*setup_secondary_clock)(void);
+
+	void (*startup_ipi_hook)(int phys_apicid,
+				 unsigned long start_eip,
+				 unsigned long start_esp);
+#endif
+};
+
+struct pv_mmu_ops {
+	/*
+	 * Called before/after init_mm pagetable setup. setup_start
+	 * may reset %cr3, and may pre-install parts of the pagetable;
+	 * pagetable setup is expected to preserve any existing
+	 * mapping.
+	 */
+	void (*pagetable_setup_start)(pgd_t *pgd_base);
+	void (*pagetable_setup_done)(pgd_t *pgd_base);
+
+	unsigned long (*read_cr2)(void);
+	void (*write_cr2)(unsigned long);
+
+	unsigned long (*read_cr3)(void);
+	void (*write_cr3)(unsigned long);
+
+	/*
+	 * Hooks for intercepting the creation/use/destruction of an
+	 * mm_struct.
+	 */
+	void (*activate_mm)(struct mm_struct *prev,
+			    struct mm_struct *next);
+	void (*dup_mmap)(struct mm_struct *oldmm,
+			 struct mm_struct *mm);
+	void (*exit_mmap)(struct mm_struct *mm);
+
+
+	/* TLB operations */
+	void (*flush_tlb_user)(void);
+	void (*flush_tlb_kernel)(void);
+	void (*flush_tlb_single)(unsigned long addr);
+	void (*flush_tlb_others)(const struct cpumask *cpus,
+				 struct mm_struct *mm,
+				 unsigned long va);
+
+	/* Hooks for allocating and freeing a pagetable top-level */
+	int  (*pgd_alloc)(struct mm_struct *mm);
+	void (*pgd_free)(struct mm_struct *mm, pgd_t *pgd);
+
+	/*
+	 * Hooks for allocating/releasing pagetable pages when they're
+	 * attached to a pagetable
+	 */
+	void (*alloc_pte)(struct mm_struct *mm, unsigned long pfn);
+	void (*alloc_pmd)(struct mm_struct *mm, unsigned long pfn);
+	void (*alloc_pmd_clone)(unsigned long pfn, unsigned long clonepfn, unsigned long start, unsigned long count);
+	void (*alloc_pud)(struct mm_struct *mm, unsigned long pfn);
+	void (*release_pte)(unsigned long pfn);
+	void (*release_pmd)(unsigned long pfn);
+	void (*release_pud)(unsigned long pfn);
+
+	/* Pagetable manipulation functions */
+	void (*set_pte)(pte_t *ptep, pte_t pteval);
+	void (*set_pte_at)(struct mm_struct *mm, unsigned long addr,
+			   pte_t *ptep, pte_t pteval);
+	void (*set_pmd)(pmd_t *pmdp, pmd_t pmdval);
+	void (*pte_update)(struct mm_struct *mm, unsigned long addr,
+			   pte_t *ptep);
+	void (*pte_update_defer)(struct mm_struct *mm,
+				 unsigned long addr, pte_t *ptep);
+
+	pte_t (*ptep_modify_prot_start)(struct mm_struct *mm, unsigned long addr,
+					pte_t *ptep);
+	void (*ptep_modify_prot_commit)(struct mm_struct *mm, unsigned long addr,
+					pte_t *ptep, pte_t pte);
+
+	struct paravirt_callee_save pte_val;
+	struct paravirt_callee_save make_pte;
+
+	struct paravirt_callee_save pgd_val;
+	struct paravirt_callee_save make_pgd;
+
+#if PAGETABLE_LEVELS >= 3
+#ifdef CONFIG_X86_PAE
+	void (*set_pte_atomic)(pte_t *ptep, pte_t pteval);
+	void (*pte_clear)(struct mm_struct *mm, unsigned long addr,
+			  pte_t *ptep);
+	void (*pmd_clear)(pmd_t *pmdp);
+
+#endif	/* CONFIG_X86_PAE */
+
+	void (*set_pud)(pud_t *pudp, pud_t pudval);
+
+	struct paravirt_callee_save pmd_val;
+	struct paravirt_callee_save make_pmd;
+
+#if PAGETABLE_LEVELS == 4
+	struct paravirt_callee_save pud_val;
+	struct paravirt_callee_save make_pud;
+
+	void (*set_pgd)(pgd_t *pudp, pgd_t pgdval);
+#endif	/* PAGETABLE_LEVELS == 4 */
+#endif	/* PAGETABLE_LEVELS >= 3 */
+
+#ifdef CONFIG_HIGHPTE
+	void *(*kmap_atomic_pte)(struct page *page, enum km_type type);
+#endif
+
+	struct pv_lazy_ops lazy_mode;
+
+	/* dom0 ops */
+
+	/* Sometimes the physical address is a pfn, and sometimes its
+	   an mfn.  We can tell which is which from the index. */
+	void (*set_fixmap)(unsigned /* enum fixed_addresses */ idx,
+			   phys_addr_t phys, pgprot_t flags);
+};
+
+struct raw_spinlock;
+struct pv_lock_ops {
+	int (*spin_is_locked)(struct raw_spinlock *lock);
+	int (*spin_is_contended)(struct raw_spinlock *lock);
+	void (*spin_lock)(struct raw_spinlock *lock);
+	void (*spin_lock_flags)(struct raw_spinlock *lock, unsigned long flags);
+	int (*spin_trylock)(struct raw_spinlock *lock);
+	void (*spin_unlock)(struct raw_spinlock *lock);
+};
+
+/* This contains all the paravirt structures: we get a convenient
+ * number for each function using the offset which we use to indicate
+ * what to patch. */
+struct paravirt_patch_template {
+	struct pv_init_ops pv_init_ops;
+	struct pv_time_ops pv_time_ops;
+	struct pv_cpu_ops pv_cpu_ops;
+	struct pv_irq_ops pv_irq_ops;
+	struct pv_apic_ops pv_apic_ops;
+	struct pv_mmu_ops pv_mmu_ops;
+	struct pv_lock_ops pv_lock_ops;
+};
+
+extern struct pv_info pv_info;
+extern struct pv_init_ops pv_init_ops;
+extern struct pv_time_ops pv_time_ops;
+extern struct pv_cpu_ops pv_cpu_ops;
+extern struct pv_irq_ops pv_irq_ops;
+extern struct pv_apic_ops pv_apic_ops;
+extern struct pv_mmu_ops pv_mmu_ops;
+extern struct pv_lock_ops pv_lock_ops;
+
+#define PARAVIRT_PATCH(x)					\
+	(offsetof(struct paravirt_patch_template, x) / sizeof(void *))
+
+#define paravirt_type(op)				\
+	[paravirt_typenum] "i" (PARAVIRT_PATCH(op)),	\
+	[paravirt_opptr] "i" (&(op))
+#define paravirt_clobber(clobber)		\
+	[paravirt_clobber] "i" (clobber)
+
+/*
+ * Generate some code, and mark it as patchable by the
+ * apply_paravirt() alternate instruction patcher.
+ */
+#define _paravirt_alt(insn_string, type, clobber)	\
+	"771:\n\t" insn_string "\n" "772:\n"		\
+	".pushsection .parainstructions,\"a\"\n"	\
+	_ASM_ALIGN "\n"					\
+	_ASM_PTR " 771b\n"				\
+	"  .byte " type "\n"				\
+	"  .byte 772b-771b\n"				\
+	"  .short " clobber "\n"			\
+	".popsection\n"
+
+/* Generate patchable code, with the default asm parameters. */
+#define paravirt_alt(insn_string)					\
+	_paravirt_alt(insn_string, "%c[paravirt_typenum]", "%c[paravirt_clobber]")
+
+/* Simple instruction patching code. */
+#define DEF_NATIVE(ops, name, code) 					\
+	extern const char start_##ops##_##name[], end_##ops##_##name[];	\
+	asm("start_" #ops "_" #name ": " code "; end_" #ops "_" #name ":")
+
+unsigned paravirt_patch_nop(void);
+unsigned paravirt_patch_ident_32(void *insnbuf, unsigned len);
+unsigned paravirt_patch_ident_64(void *insnbuf, unsigned len);
+unsigned paravirt_patch_ignore(unsigned len);
+unsigned paravirt_patch_call(void *insnbuf,
+			     const void *target, u16 tgt_clobbers,
+			     unsigned long addr, u16 site_clobbers,
+			     unsigned len);
+unsigned paravirt_patch_jmp(void *insnbuf, const void *target,
+			    unsigned long addr, unsigned len);
+unsigned paravirt_patch_default(u8 type, u16 clobbers, void *insnbuf,
+				unsigned long addr, unsigned len);
+
+unsigned paravirt_patch_insns(void *insnbuf, unsigned len,
+			      const char *start, const char *end);
+
+unsigned native_patch(u8 type, u16 clobbers, void *ibuf,
+		      unsigned long addr, unsigned len);
+
+int paravirt_disable_iospace(void);
+
+/*
+ * This generates an indirect call based on the operation type number.
+ * The type number, computed in PARAVIRT_PATCH, is derived from the
+ * offset into the paravirt_patch_template structure, and can therefore be
+ * freely converted back into a structure offset.
+ */
+#define PARAVIRT_CALL	"call *%c[paravirt_opptr];"
+
+/*
+ * These macros are intended to wrap calls through one of the paravirt
+ * ops structs, so that they can be later identified and patched at
+ * runtime.
+ *
+ * Normally, a call to a pv_op function is a simple indirect call:
+ * (pv_op_struct.operations)(args...).
+ *
+ * Unfortunately, this is a relatively slow operation for modern CPUs,
+ * because it cannot necessarily determine what the destination
+ * address is.  In this case, the address is a runtime constant, so at
+ * the very least we can patch the call to e a simple direct call, or
+ * ideally, patch an inline implementation into the callsite.  (Direct
+ * calls are essentially free, because the call and return addresses
+ * are completely predictable.)
+ *
+ * For i386, these macros rely on the standard gcc "regparm(3)" calling
+ * convention, in which the first three arguments are placed in %eax,
+ * %edx, %ecx (in that order), and the remaining arguments are placed
+ * on the stack.  All caller-save registers (eax,edx,ecx) are expected
+ * to be modified (either clobbered or used for return values).
+ * X86_64, on the other hand, already specifies a register-based calling
+ * conventions, returning at %rax, with parameteres going on %rdi, %rsi,
+ * %rdx, and %rcx. Note that for this reason, x86_64 does not need any
+ * special handling for dealing with 4 arguments, unlike i386.
+ * However, x86_64 also have to clobber all caller saved registers, which
+ * unfortunately, are quite a bit (r8 - r11)
+ *
+ * The call instruction itself is marked by placing its start address
+ * and size into the .parainstructions section, so that
+ * apply_paravirt() in arch/i386/kernel/alternative.c can do the
+ * appropriate patching under the control of the backend pv_init_ops
+ * implementation.
+ *
+ * Unfortunately there's no way to get gcc to generate the args setup
+ * for the call, and then allow the call itself to be generated by an
+ * inline asm.  Because of this, we must do the complete arg setup and
+ * return value handling from within these macros.  This is fairly
+ * cumbersome.
+ *
+ * There are 5 sets of PVOP_* macros for dealing with 0-4 arguments.
+ * It could be extended to more arguments, but there would be little
+ * to be gained from that.  For each number of arguments, there are
+ * the two VCALL and CALL variants for void and non-void functions.
+ *
+ * When there is a return value, the invoker of the macro must specify
+ * the return type.  The macro then uses sizeof() on that type to
+ * determine whether its a 32 or 64 bit value, and places the return
+ * in the right register(s) (just %eax for 32-bit, and %edx:%eax for
+ * 64-bit). For x86_64 machines, it just returns at %rax regardless of
+ * the return value size.
+ *
+ * 64-bit arguments are passed as a pair of adjacent 32-bit arguments
+ * i386 also passes 64-bit arguments as a pair of adjacent 32-bit arguments
+ * in low,high order
+ *
+ * Small structures are passed and returned in registers.  The macro
+ * calling convention can't directly deal with this, so the wrapper
+ * functions must do this.
+ *
+ * These PVOP_* macros are only defined within this header.  This
+ * means that all uses must be wrapped in inline functions.  This also
+ * makes sure the incoming and outgoing types are always correct.
+ */
+#ifdef CONFIG_X86_32
+#define PVOP_VCALL_ARGS				\
+	unsigned long __eax = __eax, __edx = __edx, __ecx = __ecx
+#define PVOP_CALL_ARGS			PVOP_VCALL_ARGS
+
+#define PVOP_CALL_ARG1(x)		"a" ((unsigned long)(x))
+#define PVOP_CALL_ARG2(x)		"d" ((unsigned long)(x))
+#define PVOP_CALL_ARG3(x)		"c" ((unsigned long)(x))
+
+#define PVOP_VCALL_CLOBBERS		"=a" (__eax), "=d" (__edx),	\
+					"=c" (__ecx)
+#define PVOP_CALL_CLOBBERS		PVOP_VCALL_CLOBBERS
+
+#define PVOP_VCALLEE_CLOBBERS		"=a" (__eax), "=d" (__edx)
+#define PVOP_CALLEE_CLOBBERS		PVOP_VCALLEE_CLOBBERS
+
+#define EXTRA_CLOBBERS
+#define VEXTRA_CLOBBERS
+#else  /* CONFIG_X86_64 */
+#define PVOP_VCALL_ARGS					\
+	unsigned long __edi = __edi, __esi = __esi,	\
+		__edx = __edx, __ecx = __ecx
+#define PVOP_CALL_ARGS		PVOP_VCALL_ARGS, __eax
+
+#define PVOP_CALL_ARG1(x)		"D" ((unsigned long)(x))
+#define PVOP_CALL_ARG2(x)		"S" ((unsigned long)(x))
+#define PVOP_CALL_ARG3(x)		"d" ((unsigned long)(x))
+#define PVOP_CALL_ARG4(x)		"c" ((unsigned long)(x))
+
+#define PVOP_VCALL_CLOBBERS	"=D" (__edi),				\
+				"=S" (__esi), "=d" (__edx),		\
+				"=c" (__ecx)
+#define PVOP_CALL_CLOBBERS	PVOP_VCALL_CLOBBERS, "=a" (__eax)
+
+#define PVOP_VCALLEE_CLOBBERS	"=a" (__eax)
+#define PVOP_CALLEE_CLOBBERS	PVOP_VCALLEE_CLOBBERS
+
+#define EXTRA_CLOBBERS	 , "r8", "r9", "r10", "r11"
+#define VEXTRA_CLOBBERS	 , "rax", "r8", "r9", "r10", "r11"
+#endif	/* CONFIG_X86_32 */
+
+#ifdef CONFIG_PARAVIRT_DEBUG
+#define PVOP_TEST_NULL(op)	BUG_ON(op == NULL)
+#else
+#define PVOP_TEST_NULL(op)	((void)op)
+#endif
+
+#define ____PVOP_CALL(rettype, op, clbr, call_clbr, extra_clbr,		\
+		      pre, post, ...)					\
+	({								\
+		rettype __ret;						\
+		PVOP_CALL_ARGS;						\
+		PVOP_TEST_NULL(op);					\
+		/* This is 32-bit specific, but is okay in 64-bit */	\
+		/* since this condition will never hold */		\
+		if (sizeof(rettype) > sizeof(unsigned long)) {		\
+			asm volatile(pre				\
+				     paravirt_alt(PARAVIRT_CALL)	\
+				     post				\
+				     : call_clbr			\
+				     : paravirt_type(op),		\
+				       paravirt_clobber(clbr),		\
+				       ##__VA_ARGS__			\
+				     : "memory", "cc" extra_clbr);	\
+			__ret = (rettype)((((u64)__edx) << 32) | __eax); \
+		} else {						\
+			asm volatile(pre				\
+				     paravirt_alt(PARAVIRT_CALL)	\
+				     post				\
+				     : call_clbr			\
+				     : paravirt_type(op),		\
+				       paravirt_clobber(clbr),		\
+				       ##__VA_ARGS__			\
+				     : "memory", "cc" extra_clbr);	\
+			__ret = (rettype)__eax;				\
+		}							\
+		__ret;							\
+	})
+
+#define __PVOP_CALL(rettype, op, pre, post, ...)			\
+	____PVOP_CALL(rettype, op, CLBR_ANY, PVOP_CALL_CLOBBERS,	\
+		      EXTRA_CLOBBERS, pre, post, ##__VA_ARGS__)
+
+#define __PVOP_CALLEESAVE(rettype, op, pre, post, ...)			\
+	____PVOP_CALL(rettype, op.func, CLBR_RET_REG,			\
+		      PVOP_CALLEE_CLOBBERS, ,				\
+		      pre, post, ##__VA_ARGS__)
+
+
+#define ____PVOP_VCALL(op, clbr, call_clbr, extra_clbr, pre, post, ...)	\
+	({								\
+		PVOP_VCALL_ARGS;					\
+		PVOP_TEST_NULL(op);					\
+		asm volatile(pre					\
+			     paravirt_alt(PARAVIRT_CALL)		\
+			     post					\
+			     : call_clbr				\
+			     : paravirt_type(op),			\
+			       paravirt_clobber(clbr),			\
+			       ##__VA_ARGS__				\
+			     : "memory", "cc" extra_clbr);		\
+	})
+
+#define __PVOP_VCALL(op, pre, post, ...)				\
+	____PVOP_VCALL(op, CLBR_ANY, PVOP_VCALL_CLOBBERS,		\
+		       VEXTRA_CLOBBERS,					\
+		       pre, post, ##__VA_ARGS__)
+
+#define __PVOP_VCALLEESAVE(rettype, op, pre, post, ...)			\
+	____PVOP_CALL(rettype, op.func, CLBR_RET_REG,			\
+		      PVOP_VCALLEE_CLOBBERS, ,				\
+		      pre, post, ##__VA_ARGS__)
+
+
+
+#define PVOP_CALL0(rettype, op)						\
+	__PVOP_CALL(rettype, op, "", "")
+#define PVOP_VCALL0(op)							\
+	__PVOP_VCALL(op, "", "")
+
+#define PVOP_CALLEE0(rettype, op)					\
+	__PVOP_CALLEESAVE(rettype, op, "", "")
+#define PVOP_VCALLEE0(op)						\
+	__PVOP_VCALLEESAVE(op, "", "")
+
+
+#define PVOP_CALL1(rettype, op, arg1)					\
+	__PVOP_CALL(rettype, op, "", "", PVOP_CALL_ARG1(arg1))
+#define PVOP_VCALL1(op, arg1)						\
+	__PVOP_VCALL(op, "", "", PVOP_CALL_ARG1(arg1))
+
+#define PVOP_CALLEE1(rettype, op, arg1)					\
+	__PVOP_CALLEESAVE(rettype, op, "", "", PVOP_CALL_ARG1(arg1))
+#define PVOP_VCALLEE1(op, arg1)						\
+	__PVOP_VCALLEESAVE(op, "", "", PVOP_CALL_ARG1(arg1))
+
+
+#define PVOP_CALL2(rettype, op, arg1, arg2)				\
+	__PVOP_CALL(rettype, op, "", "", PVOP_CALL_ARG1(arg1),		\
+		    PVOP_CALL_ARG2(arg2))
+#define PVOP_VCALL2(op, arg1, arg2)					\
+	__PVOP_VCALL(op, "", "", PVOP_CALL_ARG1(arg1),			\
+		     PVOP_CALL_ARG2(arg2))
+
+#define PVOP_CALLEE2(rettype, op, arg1, arg2)				\
+	__PVOP_CALLEESAVE(rettype, op, "", "", PVOP_CALL_ARG1(arg1),	\
+			  PVOP_CALL_ARG2(arg2))
+#define PVOP_VCALLEE2(op, arg1, arg2)					\
+	__PVOP_VCALLEESAVE(op, "", "", PVOP_CALL_ARG1(arg1),		\
+			   PVOP_CALL_ARG2(arg2))
+
+
+#define PVOP_CALL3(rettype, op, arg1, arg2, arg3)			\
+	__PVOP_CALL(rettype, op, "", "", PVOP_CALL_ARG1(arg1),		\
+		    PVOP_CALL_ARG2(arg2), PVOP_CALL_ARG3(arg3))
+#define PVOP_VCALL3(op, arg1, arg2, arg3)				\
+	__PVOP_VCALL(op, "", "", PVOP_CALL_ARG1(arg1),			\
+		     PVOP_CALL_ARG2(arg2), PVOP_CALL_ARG3(arg3))
+
+/* This is the only difference in x86_64. We can make it much simpler */
+#ifdef CONFIG_X86_32
+#define PVOP_CALL4(rettype, op, arg1, arg2, arg3, arg4)			\
+	__PVOP_CALL(rettype, op,					\
+		    "push %[_arg4];", "lea 4(%%esp),%%esp;",		\
+		    PVOP_CALL_ARG1(arg1), PVOP_CALL_ARG2(arg2),		\
+		    PVOP_CALL_ARG3(arg3), [_arg4] "mr" ((u32)(arg4)))
+#define PVOP_VCALL4(op, arg1, arg2, arg3, arg4)				\
+	__PVOP_VCALL(op,						\
+		    "push %[_arg4];", "lea 4(%%esp),%%esp;",		\
+		    "0" ((u32)(arg1)), "1" ((u32)(arg2)),		\
+		    "2" ((u32)(arg3)), [_arg4] "mr" ((u32)(arg4)))
+#else
+#define PVOP_CALL4(rettype, op, arg1, arg2, arg3, arg4)			\
+	__PVOP_CALL(rettype, op, "", "",				\
+		    PVOP_CALL_ARG1(arg1), PVOP_CALL_ARG2(arg2),		\
+		    PVOP_CALL_ARG3(arg3), PVOP_CALL_ARG4(arg4))
+#define PVOP_VCALL4(op, arg1, arg2, arg3, arg4)				\
+	__PVOP_VCALL(op, "", "",					\
+		     PVOP_CALL_ARG1(arg1), PVOP_CALL_ARG2(arg2),	\
+		     PVOP_CALL_ARG3(arg3), PVOP_CALL_ARG4(arg4))
+#endif
+
+/* Lazy mode for batching updates / context switch */
+enum paravirt_lazy_mode {
+	PARAVIRT_LAZY_NONE,
+	PARAVIRT_LAZY_MMU,
+	PARAVIRT_LAZY_CPU,
+};
+
+enum paravirt_lazy_mode paravirt_get_lazy_mode(void);
+void paravirt_start_context_switch(struct task_struct *prev);
+void paravirt_end_context_switch(struct task_struct *next);
+
+void paravirt_enter_lazy_mmu(void);
+void paravirt_leave_lazy_mmu(void);
+
+void _paravirt_nop(void);
+u32 _paravirt_ident_32(u32);
+u64 _paravirt_ident_64(u64);
+
+#define paravirt_nop	((void *)_paravirt_nop)
+
+/* These all sit in the .parainstructions section to tell us what to patch. */
+struct paravirt_patch_site {
+	u8 *instr; 		/* original instructions */
+	u8 instrtype;		/* type of this instruction */
+	u8 len;			/* length of original instruction */
+	u16 clobbers;		/* what registers you may clobber */
+};
+
+extern struct paravirt_patch_site __parainstructions[],
+	__parainstructions_end[];
+
+#endif	/* __ASSEMBLY__ */
+
+#endif	/* _ASM_X86_PARAVIRT_TYPES_H */
-- 
cgit v1.2.3


From e6e9cac8c3417b43498b243c1f8f11780e157168 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Date: Fri, 24 Apr 2009 00:40:59 -0700
Subject: x86: split out core __math_state_restore

Split the core fpu state restoration out into __math_state_restore, which
assumes that cr0.TS is clear and that the fpu context has been initialized.

This will be used during context switch.  There are two reasons this is
desireable:

- There's a small clarification.  When __switch_to() calls math_state_restore,
  it relies on the fact that tsk_used_math() returns true, and so will
  never do a blocking init_fpu().  __math_state_restore() does not have
  (or need) that logic, so the question never arises.

- It allows the clts() to be moved earler in __switch_to() so it can be performed
  while cpu context updates are batched (will be done in a later patch).

[ Impact: refactor code to make reuse cleaner; no functional change ]

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Alok Kataria <akataria@vmware.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
---
 arch/x86/include/asm/i387.h |  1 +
 arch/x86/kernel/traps.c     | 33 +++++++++++++++++++++++----------
 2 files changed, 24 insertions(+), 10 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h
index 175adf58dd4f..2e7529295f55 100644
--- a/arch/x86/include/asm/i387.h
+++ b/arch/x86/include/asm/i387.h
@@ -26,6 +26,7 @@ extern void fpu_init(void);
 extern void mxcsr_feature_mask_init(void);
 extern int init_fpu(struct task_struct *child);
 extern asmlinkage void math_state_restore(void);
+extern void __math_state_restore(void);
 extern void init_thread_xstate(void);
 extern int dump_fpu(struct pt_regs *, struct user_i387_struct *);
 
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index 5f935f0d5861..71b91669ad19 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -813,6 +813,28 @@ asmlinkage void __attribute__((weak)) smp_threshold_interrupt(void)
 {
 }
 
+/*
+ * __math_state_restore assumes that cr0.TS is already clear and the
+ * fpu state is all ready for use.  Used during context switch.
+ */
+void __math_state_restore(void)
+{
+	struct thread_info *thread = current_thread_info();
+	struct task_struct *tsk = thread->task;
+
+	/*
+	 * Paranoid restore. send a SIGSEGV if we fail to restore the state.
+	 */
+	if (unlikely(restore_fpu_checking(tsk))) {
+		stts();
+		force_sig(SIGSEGV, tsk);
+		return;
+	}
+
+	thread->status |= TS_USEDFPU;	/* So we fnsave on switch_to() */
+	tsk->fpu_counter++;
+}
+
 /*
  * 'math_state_restore()' saves the current math information in the
  * old math state array, and gets the new ones from the current task
@@ -844,17 +866,8 @@ asmlinkage void math_state_restore(void)
 	}
 
 	clts();				/* Allow maths ops (or we recurse) */
-	/*
-	 * Paranoid restore. send a SIGSEGV if we fail to restore the state.
-	 */
-	if (unlikely(restore_fpu_checking(tsk))) {
-		stts();
-		force_sig(SIGSEGV, tsk);
-		return;
-	}
 
-	thread->status |= TS_USEDFPU;	/* So we fnsave on switch_to() */
-	tsk->fpu_counter++;
+	__math_state_restore();
 }
 EXPORT_SYMBOL_GPL(math_state_restore);
 
-- 
cgit v1.2.3


From 2fcddce10f6771cfa0c56fd1e826d50d67d100b7 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Date: Fri, 24 Apr 2009 00:45:26 -0700
Subject: x86-32: make sure clts is batched during context switch

If we're preloading the fpu state during context switch, make sure the clts
happens while we're batching the cpu context update, then do the actual
__math_state_restore once the updates are flushed.

This allows more efficient context switches when running paravirtualized,
as all the hypercalls can be folded together into one.

[ Impact: optimise paravirtual FPU context switch ]

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Alok Kataria <akataria@vmware.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
---
 arch/x86/kernel/process_32.c | 27 ++++++++++++++++-----------
 1 file changed, 16 insertions(+), 11 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 59f4524984af..a80eddd41658 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -350,14 +350,21 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 				 *next = &next_p->thread;
 	int cpu = smp_processor_id();
 	struct tss_struct *tss = &per_cpu(init_tss, cpu);
+	bool preload_fpu;
 
 	/* never put a printk in __switch_to... printk() calls wake_up*() indirectly */
 
-	__unlazy_fpu(prev_p);
+	/*
+	 * If the task has used fpu the last 5 timeslices, just do a full
+	 * restore of the math state immediately to avoid the trap; the
+	 * chances of needing FPU soon are obviously high now
+	 */
+	preload_fpu = tsk_used_math(next_p) && next_p->fpu_counter > 5;
 
+	__unlazy_fpu(prev_p);
 
 	/* we're going to use this soon, after a few expensive things */
-	if (next_p->fpu_counter > 5)
+	if (preload_fpu)
 		prefetch(next->xstate);
 
 	/*
@@ -398,6 +405,11 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 		     task_thread_info(next_p)->flags & _TIF_WORK_CTXSW_NEXT))
 		__switch_to_xtra(prev_p, next_p, tss);
 
+	/* If we're going to preload the fpu context, make sure clts
+	   is run while we're batching the cpu state updates. */
+	if (preload_fpu)
+		clts();
+
 	/*
 	 * Leave lazy mode, flushing any hypercalls made here.
 	 * This must be done before restoring TLS segments so
@@ -407,15 +419,8 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	 */
 	arch_end_context_switch(next_p);
 
-	/* If the task has used fpu the last 5 timeslices, just do a full
-	 * restore of the math state immediately to avoid the trap; the
-	 * chances of needing FPU soon are obviously high now
-	 *
-	 * tsk_used_math() checks prevent calling math_state_restore(),
-	 * which can sleep in the case of !tsk_used_math()
-	 */
-	if (tsk_used_math(next_p) && next_p->fpu_counter > 5)
-		math_state_restore();
+	if (preload_fpu)
+		__math_state_restore();
 
 	/*
 	 * Restore %gs if needed (which is common)
-- 
cgit v1.2.3


From 16d9dbf0c2bd167fdd942b83592d59696c7b73bd Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Date: Fri, 24 Apr 2009 00:50:27 -0700
Subject: x86-64: move unlazy_fpu() into lazy cpu state part of context switch

Make sure that unlazy_fpu()'s stts gets batched along with the other
cpu state changes during context switch.  (32-bit already does this.)

This makes sure it gets batched when running paravirtualized.

[ Impact: optimise paravirtual FPU context switch ]

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Alok Kataria <akataria@vmware.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
---
 arch/x86/kernel/process_64.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index ebefb5407b9d..c9b8904736db 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -419,6 +419,9 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 
 	load_TLS(next, cpu);
 
+	/* Must be after DS reload */
+	unlazy_fpu(prev_p);
+
 	/*
 	 * Leave lazy mode, flushing any hypercalls made here.
 	 * This must be done before restoring TLS segments so
@@ -459,9 +462,6 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 		wrmsrl(MSR_KERNEL_GS_BASE, next->gs);
 	prev->gsindex = gsindex;
 
-	/* Must be after DS reload */
-	unlazy_fpu(prev_p);
-
 	/*
 	 * Switch the PDA and FPU contexts.
 	 */
-- 
cgit v1.2.3


From 17950c5b243f99cbabef173415ee988c52104d7e Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Date: Fri, 24 Apr 2009 01:01:01 -0700
Subject: x86-64: move clts into batch cpu state updates when preloading fpu

When a task is likely to be using the fpu, we preload its state during
the context switch, rather than waiting for it to run an fpu instruction.
Make sure the clts() happens while we're doing batched fpu state updates
to optimise paravirtualized context switches.

[ Impact: optimise paravirtual FPU context switch ]

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Alok Kataria <akataria@vmware.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
---
 arch/x86/kernel/process_64.c | 27 ++++++++++++++++++---------
 1 file changed, 18 insertions(+), 9 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index c9b8904736db..a28279dbb07c 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -386,9 +386,17 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	int cpu = smp_processor_id();
 	struct tss_struct *tss = &per_cpu(init_tss, cpu);
 	unsigned fsindex, gsindex;
+	bool preload_fpu;
+
+	/*
+	 * If the task has used fpu the last 5 timeslices, just do a full
+	 * restore of the math state immediately to avoid the trap; the
+	 * chances of needing FPU soon are obviously high now
+	 */
+	preload_fpu = tsk_used_math(next_p) && next_p->fpu_counter > 5;
 
 	/* we're going to use this soon, after a few expensive things */
-	if (next_p->fpu_counter > 5)
+	if (preload_fpu)
 		prefetch(next->xstate);
 
 	/*
@@ -422,6 +430,10 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	/* Must be after DS reload */
 	unlazy_fpu(prev_p);
 
+	/* Make sure cpu is ready for new context */
+	if (preload_fpu)
+		clts();
+
 	/*
 	 * Leave lazy mode, flushing any hypercalls made here.
 	 * This must be done before restoring TLS segments so
@@ -480,15 +492,12 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 		     task_thread_info(prev_p)->flags & _TIF_WORK_CTXSW_PREV))
 		__switch_to_xtra(prev_p, next_p, tss);
 
-	/* If the task has used fpu the last 5 timeslices, just do a full
-	 * restore of the math state immediately to avoid the trap; the
-	 * chances of needing FPU soon are obviously high now
-	 *
-	 * tsk_used_math() checks prevent calling math_state_restore(),
-	 * which can sleep in the case of !tsk_used_math()
+	/*
+	 * Preload the FPU context, now that we've determined that the
+	 * task is likely to be using it. 
 	 */
-	if (tsk_used_math(next_p) && next_p->fpu_counter > 5)
-		math_state_restore();
+	if (preload_fpu)
+		__math_state_restore();
 	return prev_p;
 }
 
-- 
cgit v1.2.3


From 21e70878215f620fe99ea7d7c74bc641aeec932f Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Thu, 18 Jun 2009 17:09:27 +0530
Subject: x86: oprofile/op_model_amd.c set return values for
 op_amd_handle_ibs()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

op_amd_handle_ibs() should return 0 when IBS is not present or not defined.

Fix compilation warning:
 CC [M]  arch/x86/oprofile/op_model_amd.o
 arch/x86/oprofile/op_model_amd.c: In function ‘op_amd_handle_ibs’:
 arch/x86/oprofile/op_model_amd.c:217: warning: no return statement in function returning non-void

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_amd.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index cc930467575d..e95268eb9220 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -132,7 +132,7 @@ op_amd_handle_ibs(struct pt_regs * const regs,
 	struct op_entry entry;
 
 	if (!has_ibs)
-		return 1;
+		return 0;
 
 	if (ibs_config.fetch_enabled) {
 		rdmsrl(MSR_AMD64_IBSFETCHCTL, ctl);
@@ -214,7 +214,10 @@ static void op_amd_stop_ibs(void)
 #else
 
 static inline int op_amd_handle_ibs(struct pt_regs * const regs,
-				    struct op_msrs const * const msrs) { }
+				    struct op_msrs const * const msrs)
+{
+	return 0;
+}
 static inline void op_amd_start_ibs(void) { }
 static inline void op_amd_stop_ibs(void) { }
 
-- 
cgit v1.2.3


From 4adc667593f83a18a8e54ce94f250fd166a272ac Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Thu, 18 Jun 2009 21:48:16 +0200
Subject: x86: add copies of some headers to convert to asm-generic

Just an intermediate step to make reviewing easier.
These files are identical copies of the existing headers.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
LKML-Reference: <cover.1245354003.git.arnd@arndb.de>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/include/asm/generic-mman.h        | 20 ++++++++
 arch/x86/include/asm/generic-module.h      | 80 ++++++++++++++++++++++++++++++
 arch/x86/include/asm/generic-scatterlist.h | 33 ++++++++++++
 arch/x86/include/asm/generic-types.h       | 30 +++++++++++
 arch/x86/include/asm/generic-ucontext.h    | 18 +++++++
 5 files changed, 181 insertions(+)
 create mode 100644 arch/x86/include/asm/generic-mman.h
 create mode 100644 arch/x86/include/asm/generic-module.h
 create mode 100644 arch/x86/include/asm/generic-scatterlist.h
 create mode 100644 arch/x86/include/asm/generic-types.h
 create mode 100644 arch/x86/include/asm/generic-ucontext.h

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/generic-mman.h b/arch/x86/include/asm/generic-mman.h
new file mode 100644
index 000000000000..751af2550ed9
--- /dev/null
+++ b/arch/x86/include/asm/generic-mman.h
@@ -0,0 +1,20 @@
+#ifndef _ASM_X86_MMAN_H
+#define _ASM_X86_MMAN_H
+
+#include <asm-generic/mman-common.h>
+
+#define MAP_32BIT	0x40		/* only give out 32bit addresses */
+
+#define MAP_GROWSDOWN	0x0100		/* stack-like segment */
+#define MAP_DENYWRITE	0x0800		/* ETXTBSY */
+#define MAP_EXECUTABLE	0x1000		/* mark it as an executable */
+#define MAP_LOCKED	0x2000		/* pages are locked */
+#define MAP_NORESERVE	0x4000		/* don't check for reservations */
+#define MAP_POPULATE	0x8000		/* populate (prefault) pagetables */
+#define MAP_NONBLOCK	0x10000		/* do not block on IO */
+#define MAP_STACK	0x20000		/* give out an address that is best suited for process/thread stacks */
+
+#define MCL_CURRENT	1		/* lock all current mappings */
+#define MCL_FUTURE	2		/* lock all future mappings */
+
+#endif /* _ASM_X86_MMAN_H */
diff --git a/arch/x86/include/asm/generic-module.h b/arch/x86/include/asm/generic-module.h
new file mode 100644
index 000000000000..47d62743c4d5
--- /dev/null
+++ b/arch/x86/include/asm/generic-module.h
@@ -0,0 +1,80 @@
+#ifndef _ASM_X86_MODULE_H
+#define _ASM_X86_MODULE_H
+
+/* x86_32/64 are simple */
+struct mod_arch_specific {};
+
+#ifdef CONFIG_X86_32
+# define Elf_Shdr Elf32_Shdr
+# define Elf_Sym Elf32_Sym
+# define Elf_Ehdr Elf32_Ehdr
+#else
+# define Elf_Shdr Elf64_Shdr
+# define Elf_Sym Elf64_Sym
+# define Elf_Ehdr Elf64_Ehdr
+#endif
+
+#ifdef CONFIG_X86_64
+/* X86_64 does not define MODULE_PROC_FAMILY */
+#elif defined CONFIG_M386
+#define MODULE_PROC_FAMILY "386 "
+#elif defined CONFIG_M486
+#define MODULE_PROC_FAMILY "486 "
+#elif defined CONFIG_M586
+#define MODULE_PROC_FAMILY "586 "
+#elif defined CONFIG_M586TSC
+#define MODULE_PROC_FAMILY "586TSC "
+#elif defined CONFIG_M586MMX
+#define MODULE_PROC_FAMILY "586MMX "
+#elif defined CONFIG_MCORE2
+#define MODULE_PROC_FAMILY "CORE2 "
+#elif defined CONFIG_M686
+#define MODULE_PROC_FAMILY "686 "
+#elif defined CONFIG_MPENTIUMII
+#define MODULE_PROC_FAMILY "PENTIUMII "
+#elif defined CONFIG_MPENTIUMIII
+#define MODULE_PROC_FAMILY "PENTIUMIII "
+#elif defined CONFIG_MPENTIUMM
+#define MODULE_PROC_FAMILY "PENTIUMM "
+#elif defined CONFIG_MPENTIUM4
+#define MODULE_PROC_FAMILY "PENTIUM4 "
+#elif defined CONFIG_MK6
+#define MODULE_PROC_FAMILY "K6 "
+#elif defined CONFIG_MK7
+#define MODULE_PROC_FAMILY "K7 "
+#elif defined CONFIG_MK8
+#define MODULE_PROC_FAMILY "K8 "
+#elif defined CONFIG_X86_ELAN
+#define MODULE_PROC_FAMILY "ELAN "
+#elif defined CONFIG_MCRUSOE
+#define MODULE_PROC_FAMILY "CRUSOE "
+#elif defined CONFIG_MEFFICEON
+#define MODULE_PROC_FAMILY "EFFICEON "
+#elif defined CONFIG_MWINCHIPC6
+#define MODULE_PROC_FAMILY "WINCHIPC6 "
+#elif defined CONFIG_MWINCHIP3D
+#define MODULE_PROC_FAMILY "WINCHIP3D "
+#elif defined CONFIG_MCYRIXIII
+#define MODULE_PROC_FAMILY "CYRIXIII "
+#elif defined CONFIG_MVIAC3_2
+#define MODULE_PROC_FAMILY "VIAC3-2 "
+#elif defined CONFIG_MVIAC7
+#define MODULE_PROC_FAMILY "VIAC7 "
+#elif defined CONFIG_MGEODEGX1
+#define MODULE_PROC_FAMILY "GEODEGX1 "
+#elif defined CONFIG_MGEODE_LX
+#define MODULE_PROC_FAMILY "GEODE "
+#else
+#error unknown processor family
+#endif
+
+#ifdef CONFIG_X86_32
+# ifdef CONFIG_4KSTACKS
+#  define MODULE_STACKSIZE "4KSTACKS "
+# else
+#  define MODULE_STACKSIZE ""
+# endif
+# define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY MODULE_STACKSIZE
+#endif
+
+#endif /* _ASM_X86_MODULE_H */
diff --git a/arch/x86/include/asm/generic-scatterlist.h b/arch/x86/include/asm/generic-scatterlist.h
new file mode 100644
index 000000000000..263d397d2eef
--- /dev/null
+++ b/arch/x86/include/asm/generic-scatterlist.h
@@ -0,0 +1,33 @@
+#ifndef _ASM_X86_SCATTERLIST_H
+#define _ASM_X86_SCATTERLIST_H
+
+#include <asm/types.h>
+
+struct scatterlist {
+#ifdef CONFIG_DEBUG_SG
+	unsigned long	sg_magic;
+#endif
+	unsigned long	page_link;
+	unsigned int	offset;
+	unsigned int	length;
+	dma_addr_t	dma_address;
+	unsigned int	dma_length;
+};
+
+#define ARCH_HAS_SG_CHAIN
+#define ISA_DMA_THRESHOLD (0x00ffffff)
+
+/*
+ * These macros should be used after a pci_map_sg call has been done
+ * to get bus addresses of each of the SG entries and their lengths.
+ * You should only work with the number of sg entries pci_map_sg
+ * returns.
+ */
+#define sg_dma_address(sg)	((sg)->dma_address)
+#ifdef CONFIG_X86_32
+# define sg_dma_len(sg)		((sg)->length)
+#else
+# define sg_dma_len(sg)		((sg)->dma_length)
+#endif
+
+#endif /* _ASM_X86_SCATTERLIST_H */
diff --git a/arch/x86/include/asm/generic-types.h b/arch/x86/include/asm/generic-types.h
new file mode 100644
index 000000000000..09b97745772f
--- /dev/null
+++ b/arch/x86/include/asm/generic-types.h
@@ -0,0 +1,30 @@
+#ifndef _ASM_X86_TYPES_H
+#define _ASM_X86_TYPES_H
+
+#include <asm-generic/int-ll64.h>
+
+#ifndef __ASSEMBLY__
+
+typedef unsigned short umode_t;
+
+#endif /* __ASSEMBLY__ */
+
+/*
+ * These aren't exported outside the kernel to avoid name space clashes
+ */
+#ifdef __KERNEL__
+
+#ifndef __ASSEMBLY__
+
+typedef u64 dma64_addr_t;
+#if defined(CONFIG_X86_64) || defined(CONFIG_HIGHMEM64G)
+/* DMA addresses come in 32-bit and 64-bit flavours. */
+typedef u64 dma_addr_t;
+#else
+typedef u32 dma_addr_t;
+#endif
+
+#endif /* __ASSEMBLY__ */
+#endif /* __KERNEL__ */
+
+#endif /* _ASM_X86_TYPES_H */
diff --git a/arch/x86/include/asm/generic-ucontext.h b/arch/x86/include/asm/generic-ucontext.h
new file mode 100644
index 000000000000..87324cf439d9
--- /dev/null
+++ b/arch/x86/include/asm/generic-ucontext.h
@@ -0,0 +1,18 @@
+#ifndef _ASM_X86_UCONTEXT_H
+#define _ASM_X86_UCONTEXT_H
+
+#define UC_FP_XSTATE	0x1	/* indicates the presence of extended state
+				 * information in the memory layout pointed
+				 * by the fpstate pointer in the ucontext's
+				 * sigcontext struct (uc_mcontext).
+				 */
+
+struct ucontext {
+	unsigned long	  uc_flags;
+	struct ucontext  *uc_link;
+	stack_t		  uc_stack;
+	struct sigcontext uc_mcontext;
+	sigset_t	  uc_sigmask;	/* mask last for extensibility */
+};
+
+#endif /* _ASM_X86_UCONTEXT_H */
-- 
cgit v1.2.3


From 7bfd124d6dae7d394e73753300594a81a022fe7d Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Thu, 18 Jun 2009 21:48:17 +0200
Subject: x86: convert trivial headers to asm-generic version

For these nine header files, the asm-generic version should
be semantically identical to what is in x86. Change the
contents to be binary identical, for better review.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
LKML-Reference: <cover.1245354003.git.arnd@arndb.de>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/include/asm/ioctls.h   | 40 ++++++++++++++++++++++++++++------------
 arch/x86/include/asm/ipcbuf.h   | 15 ++++++++++-----
 arch/x86/include/asm/msgbuf.h   | 30 +++++++++++++++++++-----------
 arch/x86/include/asm/param.h    |  8 +++++---
 arch/x86/include/asm/shmbuf.h   | 28 ++++++++++++++++++----------
 arch/x86/include/asm/socket.h   | 10 +++++-----
 arch/x86/include/asm/sockios.h  |  6 +++---
 arch/x86/include/asm/termbits.h |  8 ++++----
 8 files changed, 92 insertions(+), 53 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/ioctls.h b/arch/x86/include/asm/ioctls.h
index 0d5b23b7b06e..a799e20a769e 100644
--- a/arch/x86/include/asm/ioctls.h
+++ b/arch/x86/include/asm/ioctls.h
@@ -1,12 +1,23 @@
-#ifndef _ASM_X86_IOCTLS_H
-#define _ASM_X86_IOCTLS_H
+#ifndef __ASM_GENERIC_IOCTLS_H
+#define __ASM_GENERIC_IOCTLS_H
 
-#include <asm/ioctl.h>
+#include <linux/ioctl.h>
+
+/*
+ * These are the most common definitions for tty ioctl numbers.
+ * Most of them do not use the recommended _IOC(), but there is
+ * probably some source code out there hardcoding the number,
+ * so we might as well use them for all new platforms.
+ *
+ * The architectures that use different values here typically
+ * try to be compatible with some Unix variants for the same
+ * architecture.
+ */
 
 /* 0x54 is just a magic number to make these relatively unique ('T') */
 
 #define TCGETS		0x5401
-#define TCSETS		0x5402 /* Clashes with SNDCTL_TMR_START sound ioctl */
+#define TCSETS		0x5402
 #define TCSETSW		0x5403
 #define TCSETSF		0x5404
 #define TCGETA		0x5405
@@ -43,7 +54,6 @@
 #define TIOCSETD	0x5423
 #define TIOCGETD	0x5424
 #define TCSBRKP		0x5425	/* Needed for POSIX tcsendbreak() */
-/* #define TIOCTTYGSTRUCT 0x5426 - Former debugging-only ioctl */
 #define TIOCSBRK	0x5427  /* BSD compatibility */
 #define TIOCCBRK	0x5428  /* BSD compatibility */
 #define TIOCGSID	0x5429  /* Return the session ID of FD */
@@ -53,8 +63,7 @@
 #define TCSETSF2	_IOW('T', 0x2D, struct termios2)
 #define TIOCGRS485	0x542E
 #define TIOCSRS485	0x542F
-#define TIOCGPTN	_IOR('T', 0x30, unsigned int)
-				/* Get Pty Number (of pty-mux device) */
+#define TIOCGPTN	_IOR('T', 0x30, unsigned int) /* Get Pty Number (of pty-mux device) */
 #define TIOCSPTLCK	_IOW('T', 0x31, int)  /* Lock/unlock Pty */
 #define TCGETX		0x5432 /* SYS5 TCGETX compatibility */
 #define TCSETX		0x5433
@@ -76,9 +85,16 @@
 
 #define TIOCMIWAIT	0x545C	/* wait for a change on serial input line(s) */
 #define TIOCGICOUNT	0x545D	/* read serial port inline interrupt counts */
-#define TIOCGHAYESESP   0x545E  /* Get Hayes ESP configuration */
-#define TIOCSHAYESESP   0x545F  /* Set Hayes ESP configuration */
-#define FIOQSIZE	0x5460
+
+/*
+ * some architectures define FIOQSIZE as 0x545E, which is used for
+ * TIOCGHAYESESP on others
+ */
+#ifndef FIOQSIZE
+# define TIOCGHAYESESP	0x545E  /* Get Hayes ESP configuration */
+# define TIOCSHAYESESP	0x545F  /* Set Hayes ESP configuration */
+# define FIOQSIZE	0x5460
+#endif
 
 /* Used for packet mode */
 #define TIOCPKT_DATA		 0
@@ -89,6 +105,6 @@
 #define TIOCPKT_NOSTOP		16
 #define TIOCPKT_DOSTOP		32
 
-#define TIOCSER_TEMT    0x01	/* Transmitter physically empty */
+#define TIOCSER_TEMT	0x01	/* Transmitter physically empty */
 
-#endif /* _ASM_X86_IOCTLS_H */
+#endif /* __ASM_GENERIC_IOCTLS_H */
diff --git a/arch/x86/include/asm/ipcbuf.h b/arch/x86/include/asm/ipcbuf.h
index ee678fd51594..888ca1c8f951 100644
--- a/arch/x86/include/asm/ipcbuf.h
+++ b/arch/x86/include/asm/ipcbuf.h
@@ -1,13 +1,18 @@
-#ifndef _ASM_X86_IPCBUF_H
-#define _ASM_X86_IPCBUF_H
+#ifndef __ASM_GENERIC_IPCBUF_H
+#define __ASM_GENERIC_IPCBUF_H
 
 /*
- * The ipc64_perm structure for x86 architecture.
+ * The generic ipc64_perm structure:
  * Note extra padding because this structure is passed back and forth
  * between kernel and user space.
  *
+ * ipc64_perm was originally meant to be architecture specific, but
+ * everyone just ended up making identical copies without specific
+ * optimizations, so we may just as well all use the same one.
+ *
  * Pad space is left for:
- * - 32-bit mode_t and seq
+ * - 32-bit mode_t on architectures that only had 16 bit
+ * - 32-bit seq
  * - 2 miscellaneous 32-bit values
  */
 
@@ -25,4 +30,4 @@ struct ipc64_perm {
 	unsigned long		__unused2;
 };
 
-#endif /* _ASM_X86_IPCBUF_H */
+#endif /* __ASM_GENERIC_IPCBUF_H */
diff --git a/arch/x86/include/asm/msgbuf.h b/arch/x86/include/asm/msgbuf.h
index 7e4e9481f51c..aec850d9159e 100644
--- a/arch/x86/include/asm/msgbuf.h
+++ b/arch/x86/include/asm/msgbuf.h
@@ -1,30 +1,38 @@
-#ifndef _ASM_X86_MSGBUF_H
-#define _ASM_X86_MSGBUF_H
+#ifndef __ASM_GENERIC_MSGBUF_H
+#define __ASM_GENERIC_MSGBUF_H
 
+#include <asm/bitsperlong.h>
 /*
- * The msqid64_ds structure for i386 architecture.
+ * generic msqid64_ds structure.
+ *
  * Note extra padding because this structure is passed back and forth
  * between kernel and user space.
  *
- * Pad space on i386 is left for:
+ * msqid64_ds was originally meant to be architecture specific, but
+ * everyone just ended up making identical copies without specific
+ * optimizations, so we may just as well all use the same one.
+ *
+ * 64 bit architectures typically define a 64 bit __kernel_time_t,
+ * so they do not need the first three padding words.
+ * On big-endian systems, the padding is in the wrong place.
+ *
+ * Pad space is left for:
  * - 64-bit time_t to solve y2038 problem
  * - 2 miscellaneous 32-bit values
- *
- * Pad space on x8664 is left for:
- * - 2 miscellaneous 64-bit values
  */
+
 struct msqid64_ds {
 	struct ipc64_perm msg_perm;
 	__kernel_time_t msg_stime;	/* last msgsnd time */
-#ifdef __i386__
+#if __BITS_PER_LONG != 64
 	unsigned long	__unused1;
 #endif
 	__kernel_time_t msg_rtime;	/* last msgrcv time */
-#ifdef __i386__
+#if __BITS_PER_LONG != 64
 	unsigned long	__unused2;
 #endif
 	__kernel_time_t msg_ctime;	/* last change time */
-#ifdef __i386__
+#if __BITS_PER_LONG != 64
 	unsigned long	__unused3;
 #endif
 	unsigned long  msg_cbytes;	/* current number of bytes on queue */
@@ -36,4 +44,4 @@ struct msqid64_ds {
 	unsigned long  __unused5;
 };
 
-#endif /* _ASM_X86_MSGBUF_H */
+#endif /* __ASM_GENERIC_MSGBUF_H */
diff --git a/arch/x86/include/asm/param.h b/arch/x86/include/asm/param.h
index 6f0d0422f4ca..cdf8251bfb6c 100644
--- a/arch/x86/include/asm/param.h
+++ b/arch/x86/include/asm/param.h
@@ -1,5 +1,5 @@
-#ifndef _ASM_X86_PARAM_H
-#define _ASM_X86_PARAM_H
+#ifndef __ASM_GENERIC_PARAM_H
+#define __ASM_GENERIC_PARAM_H
 
 #ifdef __KERNEL__
 # define HZ		CONFIG_HZ	/* Internal kernel timer frequency */
@@ -11,7 +11,9 @@
 #define HZ 100
 #endif
 
+#ifndef EXEC_PAGESIZE
 #define EXEC_PAGESIZE	4096
+#endif
 
 #ifndef NOGROUP
 #define NOGROUP		(-1)
@@ -19,4 +21,4 @@
 
 #define MAXHOSTNAMELEN	64	/* max length of hostname */
 
-#endif /* _ASM_X86_PARAM_H */
+#endif /* __ASM_GENERIC_PARAM_H */
diff --git a/arch/x86/include/asm/shmbuf.h b/arch/x86/include/asm/shmbuf.h
index b51413b74971..5768fa60ac82 100644
--- a/arch/x86/include/asm/shmbuf.h
+++ b/arch/x86/include/asm/shmbuf.h
@@ -1,32 +1,40 @@
-#ifndef _ASM_X86_SHMBUF_H
-#define _ASM_X86_SHMBUF_H
+#ifndef __ASM_GENERIC_SHMBUF_H
+#define __ASM_GENERIC_SHMBUF_H
+
+#include <asm/bitsperlong.h>
 
 /*
  * The shmid64_ds structure for x86 architecture.
  * Note extra padding because this structure is passed back and forth
  * between kernel and user space.
  *
- * Pad space on 32 bit is left for:
+ * shmid64_ds was originally meant to be architecture specific, but
+ * everyone just ended up making identical copies without specific
+ * optimizations, so we may just as well all use the same one.
+ *
+ * 64 bit architectures typically define a 64 bit __kernel_time_t,
+ * so they do not need the first two padding words.
+ * On big-endian systems, the padding is in the wrong place.
+ *
+ *
+ * Pad space is left for:
  * - 64-bit time_t to solve y2038 problem
  * - 2 miscellaneous 32-bit values
- *
- * Pad space on 64 bit is left for:
- * - 2 miscellaneous 64-bit values
  */
 
 struct shmid64_ds {
 	struct ipc64_perm	shm_perm;	/* operation perms */
 	size_t			shm_segsz;	/* size of segment (bytes) */
 	__kernel_time_t		shm_atime;	/* last attach time */
-#ifdef __i386__
+#if __BITS_PER_LONG != 64
 	unsigned long		__unused1;
 #endif
 	__kernel_time_t		shm_dtime;	/* last detach time */
-#ifdef __i386__
+#if __BITS_PER_LONG != 64
 	unsigned long		__unused2;
 #endif
 	__kernel_time_t		shm_ctime;	/* last change time */
-#ifdef __i386__
+#if __BITS_PER_LONG != 64
 	unsigned long		__unused3;
 #endif
 	__kernel_pid_t		shm_cpid;	/* pid of creator */
@@ -48,4 +56,4 @@ struct shminfo64 {
 	unsigned long	__unused4;
 };
 
-#endif /* _ASM_X86_SHMBUF_H */
+#endif /* __ASM_GENERIC_SHMBUF_H */
diff --git a/arch/x86/include/asm/socket.h b/arch/x86/include/asm/socket.h
index ca8bf2cd0ba9..d4ae42a06a20 100644
--- a/arch/x86/include/asm/socket.h
+++ b/arch/x86/include/asm/socket.h
@@ -1,5 +1,5 @@
-#ifndef _ASM_X86_SOCKET_H
-#define _ASM_X86_SOCKET_H
+#ifndef __ASM_GENERIC_SOCKET_H
+#define __ASM_GENERIC_SOCKET_H
 
 #include <asm/sockios.h>
 
@@ -38,8 +38,8 @@
 #define SO_BINDTODEVICE	25
 
 /* Socket filtering */
-#define SO_ATTACH_FILTER        26
-#define SO_DETACH_FILTER        27
+#define SO_ATTACH_FILTER	26
+#define SO_DETACH_FILTER	27
 
 #define SO_PEERNAME		28
 #define SO_TIMESTAMP		29
@@ -57,4 +57,4 @@
 #define SO_TIMESTAMPING		37
 #define SCM_TIMESTAMPING	SO_TIMESTAMPING
 
-#endif /* _ASM_X86_SOCKET_H */
+#endif /* __ASM_GENERIC_SOCKET_H */
diff --git a/arch/x86/include/asm/sockios.h b/arch/x86/include/asm/sockios.h
index 49cc72b5d3c9..9a61a369b901 100644
--- a/arch/x86/include/asm/sockios.h
+++ b/arch/x86/include/asm/sockios.h
@@ -1,5 +1,5 @@
-#ifndef _ASM_X86_SOCKIOS_H
-#define _ASM_X86_SOCKIOS_H
+#ifndef __ASM_GENERIC_SOCKIOS_H
+#define __ASM_GENERIC_SOCKIOS_H
 
 /* Socket-level I/O control calls. */
 #define FIOSETOWN	0x8901
@@ -10,4 +10,4 @@
 #define SIOCGSTAMP	0x8906		/* Get stamp (timeval) */
 #define SIOCGSTAMPNS	0x8907		/* Get stamp (timespec) */
 
-#endif /* _ASM_X86_SOCKIOS_H */
+#endif /* __ASM_GENERIC_SOCKIOS_H */
diff --git a/arch/x86/include/asm/termbits.h b/arch/x86/include/asm/termbits.h
index af1b70ea440f..1c9773d48cb0 100644
--- a/arch/x86/include/asm/termbits.h
+++ b/arch/x86/include/asm/termbits.h
@@ -1,5 +1,5 @@
-#ifndef _ASM_X86_TERMBITS_H
-#define _ASM_X86_TERMBITS_H
+#ifndef __ASM_GENERIC_TERMBITS_H
+#define __ASM_GENERIC_TERMBITS_H
 
 #include <linux/posix_types.h>
 
@@ -140,7 +140,7 @@ struct ktermios {
 #define HUPCL	0002000
 #define CLOCAL	0004000
 #define CBAUDEX 0010000
-#define	   BOTHER 0010000		/* non standard rate */
+#define    BOTHER 0010000
 #define    B57600 0010001
 #define   B115200 0010002
 #define   B230400 0010003
@@ -195,4 +195,4 @@ struct ktermios {
 #define	TCSADRAIN	1
 #define	TCSAFLUSH	2
 
-#endif /* _ASM_X86_TERMBITS_H */
+#endif /* __ASM_GENERIC_TERMBITS_H */
-- 
cgit v1.2.3


From 06f5013aa8eb5895ced2c71d13f5114103605555 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Thu, 18 Jun 2009 21:48:18 +0200
Subject: x86: convert almost generic headers to asm-generic version

In x86, mman.h, module.h, scatterlist.h, types.h and ucontext.h
can use the asm-generic version by just defining the x86
specific parts locally and falling back on the generic code
for the common bits.

This patch illustrates the differences between the x86 and
asm-generic versions by changing a file that is initially
identical to the x86 version to one that is identical
to the asm-generic version.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
LKML-Reference: <cover.1245354003.git.arnd@arndb.de>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/include/asm/generic-mman.h        |  8 +--
 arch/x86/include/asm/generic-module.h      | 92 ++++++------------------------
 arch/x86/include/asm/generic-scatterlist.h | 34 +++++++----
 arch/x86/include/asm/generic-types.h       | 32 +++++++----
 arch/x86/include/asm/generic-ucontext.h    | 12 +---
 arch/x86/include/asm/mman.h                | 14 +----
 arch/x86/include/asm/module.h              | 13 +----
 arch/x86/include/asm/scatterlist.h         | 27 +--------
 arch/x86/include/asm/types.h               | 12 +---
 arch/x86/include/asm/ucontext.h            |  8 +--
 10 files changed, 73 insertions(+), 179 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/generic-mman.h b/arch/x86/include/asm/generic-mman.h
index 751af2550ed9..7cab4de2bca6 100644
--- a/arch/x86/include/asm/generic-mman.h
+++ b/arch/x86/include/asm/generic-mman.h
@@ -1,10 +1,8 @@
-#ifndef _ASM_X86_MMAN_H
-#define _ASM_X86_MMAN_H
+#ifndef __ASM_GENERIC_MMAN_H
+#define __ASM_GENERIC_MMAN_H
 
 #include <asm-generic/mman-common.h>
 
-#define MAP_32BIT	0x40		/* only give out 32bit addresses */
-
 #define MAP_GROWSDOWN	0x0100		/* stack-like segment */
 #define MAP_DENYWRITE	0x0800		/* ETXTBSY */
 #define MAP_EXECUTABLE	0x1000		/* mark it as an executable */
@@ -17,4 +15,4 @@
 #define MCL_CURRENT	1		/* lock all current mappings */
 #define MCL_FUTURE	2		/* lock all future mappings */
 
-#endif /* _ASM_X86_MMAN_H */
+#endif /* __ASM_GENERIC_MMAN_H */
diff --git a/arch/x86/include/asm/generic-module.h b/arch/x86/include/asm/generic-module.h
index 47d62743c4d5..ed5b44de4c91 100644
--- a/arch/x86/include/asm/generic-module.h
+++ b/arch/x86/include/asm/generic-module.h
@@ -1,80 +1,22 @@
-#ifndef _ASM_X86_MODULE_H
-#define _ASM_X86_MODULE_H
+#ifndef __ASM_GENERIC_MODULE_H
+#define __ASM_GENERIC_MODULE_H
 
-/* x86_32/64 are simple */
-struct mod_arch_specific {};
+/*
+ * Many architectures just need a simple module
+ * loader without arch specific data.
+ */
+struct mod_arch_specific
+{
+};
 
-#ifdef CONFIG_X86_32
-# define Elf_Shdr Elf32_Shdr
-# define Elf_Sym Elf32_Sym
-# define Elf_Ehdr Elf32_Ehdr
+#ifdef CONFIG_64BIT
+#define Elf_Shdr Elf64_Shdr
+#define Elf_Sym Elf64_Sym
+#define Elf_Ehdr Elf64_Ehdr
 #else
-# define Elf_Shdr Elf64_Shdr
-# define Elf_Sym Elf64_Sym
-# define Elf_Ehdr Elf64_Ehdr
+#define Elf_Shdr Elf32_Shdr
+#define Elf_Sym Elf32_Sym
+#define Elf_Ehdr Elf32_Ehdr
 #endif
 
-#ifdef CONFIG_X86_64
-/* X86_64 does not define MODULE_PROC_FAMILY */
-#elif defined CONFIG_M386
-#define MODULE_PROC_FAMILY "386 "
-#elif defined CONFIG_M486
-#define MODULE_PROC_FAMILY "486 "
-#elif defined CONFIG_M586
-#define MODULE_PROC_FAMILY "586 "
-#elif defined CONFIG_M586TSC
-#define MODULE_PROC_FAMILY "586TSC "
-#elif defined CONFIG_M586MMX
-#define MODULE_PROC_FAMILY "586MMX "
-#elif defined CONFIG_MCORE2
-#define MODULE_PROC_FAMILY "CORE2 "
-#elif defined CONFIG_M686
-#define MODULE_PROC_FAMILY "686 "
-#elif defined CONFIG_MPENTIUMII
-#define MODULE_PROC_FAMILY "PENTIUMII "
-#elif defined CONFIG_MPENTIUMIII
-#define MODULE_PROC_FAMILY "PENTIUMIII "
-#elif defined CONFIG_MPENTIUMM
-#define MODULE_PROC_FAMILY "PENTIUMM "
-#elif defined CONFIG_MPENTIUM4
-#define MODULE_PROC_FAMILY "PENTIUM4 "
-#elif defined CONFIG_MK6
-#define MODULE_PROC_FAMILY "K6 "
-#elif defined CONFIG_MK7
-#define MODULE_PROC_FAMILY "K7 "
-#elif defined CONFIG_MK8
-#define MODULE_PROC_FAMILY "K8 "
-#elif defined CONFIG_X86_ELAN
-#define MODULE_PROC_FAMILY "ELAN "
-#elif defined CONFIG_MCRUSOE
-#define MODULE_PROC_FAMILY "CRUSOE "
-#elif defined CONFIG_MEFFICEON
-#define MODULE_PROC_FAMILY "EFFICEON "
-#elif defined CONFIG_MWINCHIPC6
-#define MODULE_PROC_FAMILY "WINCHIPC6 "
-#elif defined CONFIG_MWINCHIP3D
-#define MODULE_PROC_FAMILY "WINCHIP3D "
-#elif defined CONFIG_MCYRIXIII
-#define MODULE_PROC_FAMILY "CYRIXIII "
-#elif defined CONFIG_MVIAC3_2
-#define MODULE_PROC_FAMILY "VIAC3-2 "
-#elif defined CONFIG_MVIAC7
-#define MODULE_PROC_FAMILY "VIAC7 "
-#elif defined CONFIG_MGEODEGX1
-#define MODULE_PROC_FAMILY "GEODEGX1 "
-#elif defined CONFIG_MGEODE_LX
-#define MODULE_PROC_FAMILY "GEODE "
-#else
-#error unknown processor family
-#endif
-
-#ifdef CONFIG_X86_32
-# ifdef CONFIG_4KSTACKS
-#  define MODULE_STACKSIZE "4KSTACKS "
-# else
-#  define MODULE_STACKSIZE ""
-# endif
-# define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY MODULE_STACKSIZE
-#endif
-
-#endif /* _ASM_X86_MODULE_H */
+#endif /* __ASM_GENERIC_MODULE_H */
diff --git a/arch/x86/include/asm/generic-scatterlist.h b/arch/x86/include/asm/generic-scatterlist.h
index 263d397d2eef..8b9454496a7c 100644
--- a/arch/x86/include/asm/generic-scatterlist.h
+++ b/arch/x86/include/asm/generic-scatterlist.h
@@ -1,7 +1,7 @@
-#ifndef _ASM_X86_SCATTERLIST_H
-#define _ASM_X86_SCATTERLIST_H
+#ifndef __ASM_GENERIC_SCATTERLIST_H
+#define __ASM_GENERIC_SCATTERLIST_H
 
-#include <asm/types.h>
+#include <linux/types.h>
 
 struct scatterlist {
 #ifdef CONFIG_DEBUG_SG
@@ -14,20 +14,30 @@ struct scatterlist {
 	unsigned int	dma_length;
 };
 
-#define ARCH_HAS_SG_CHAIN
-#define ISA_DMA_THRESHOLD (0x00ffffff)
-
 /*
- * These macros should be used after a pci_map_sg call has been done
+ * These macros should be used after a dma_map_sg call has been done
  * to get bus addresses of each of the SG entries and their lengths.
  * You should only work with the number of sg entries pci_map_sg
- * returns.
+ * returns, or alternatively stop on the first sg_dma_len(sg) which
+ * is 0.
  */
 #define sg_dma_address(sg)	((sg)->dma_address)
-#ifdef CONFIG_X86_32
-# define sg_dma_len(sg)		((sg)->length)
+#ifndef sg_dma_len
+/*
+ * Normally, you have an iommu on 64 bit machines, but not on 32 bit
+ * machines. Architectures that are differnt should override this.
+ */
+#if __BITS_PER_LONG == 64
+#define sg_dma_len(sg)		((sg)->dma_length)
 #else
-# define sg_dma_len(sg)		((sg)->dma_length)
+#define sg_dma_len(sg)		((sg)->length)
+#endif /* 64 bit */
+#endif /* sg_dma_len */
+
+#ifndef ISA_DMA_THRESHOLD
+#define ISA_DMA_THRESHOLD	(~0UL)
 #endif
 
-#endif /* _ASM_X86_SCATTERLIST_H */
+#define ARCH_HAS_SG_CHAIN
+
+#endif /* __ASM_GENERIC_SCATTERLIST_H */
diff --git a/arch/x86/include/asm/generic-types.h b/arch/x86/include/asm/generic-types.h
index 09b97745772f..fba7d33ca3f2 100644
--- a/arch/x86/include/asm/generic-types.h
+++ b/arch/x86/include/asm/generic-types.h
@@ -1,6 +1,9 @@
-#ifndef _ASM_X86_TYPES_H
-#define _ASM_X86_TYPES_H
-
+#ifndef _ASM_GENERIC_TYPES_H
+#define _ASM_GENERIC_TYPES_H
+/*
+ * int-ll64 is used practically everywhere now,
+ * so use it as a reasonable default.
+ */
 #include <asm-generic/int-ll64.h>
 
 #ifndef __ASSEMBLY__
@@ -13,18 +16,27 @@ typedef unsigned short umode_t;
  * These aren't exported outside the kernel to avoid name space clashes
  */
 #ifdef __KERNEL__
-
 #ifndef __ASSEMBLY__
-
-typedef u64 dma64_addr_t;
-#if defined(CONFIG_X86_64) || defined(CONFIG_HIGHMEM64G)
-/* DMA addresses come in 32-bit and 64-bit flavours. */
+/*
+ * DMA addresses may be very different from physical addresses
+ * and pointers. i386 and powerpc may have 64 bit DMA on 32 bit
+ * systems, while sparc64 uses 32 bit DMA addresses for 64 bit
+ * physical addresses.
+ * This default defines dma_addr_t to have the same size as
+ * phys_addr_t, which is the most common way.
+ * Do not define the dma64_addr_t type, which never really
+ * worked.
+ */
+#ifndef dma_addr_t
+#ifdef CONFIG_PHYS_ADDR_T_64BIT
 typedef u64 dma_addr_t;
 #else
 typedef u32 dma_addr_t;
-#endif
+#endif /* CONFIG_PHYS_ADDR_T_64BIT */
+#endif /* dma_addr_t */
 
 #endif /* __ASSEMBLY__ */
+
 #endif /* __KERNEL__ */
 
-#endif /* _ASM_X86_TYPES_H */
+#endif /* _ASM_GENERIC_TYPES_H */
diff --git a/arch/x86/include/asm/generic-ucontext.h b/arch/x86/include/asm/generic-ucontext.h
index 87324cf439d9..ad77343e8a9a 100644
--- a/arch/x86/include/asm/generic-ucontext.h
+++ b/arch/x86/include/asm/generic-ucontext.h
@@ -1,11 +1,5 @@
-#ifndef _ASM_X86_UCONTEXT_H
-#define _ASM_X86_UCONTEXT_H
-
-#define UC_FP_XSTATE	0x1	/* indicates the presence of extended state
-				 * information in the memory layout pointed
-				 * by the fpstate pointer in the ucontext's
-				 * sigcontext struct (uc_mcontext).
-				 */
+#ifndef __ASM_GENERIC_UCONTEXT_H
+#define __ASM_GENERIC_UCONTEXT_H
 
 struct ucontext {
 	unsigned long	  uc_flags;
@@ -15,4 +9,4 @@ struct ucontext {
 	sigset_t	  uc_sigmask;	/* mask last for extensibility */
 };
 
-#endif /* _ASM_X86_UCONTEXT_H */
+#endif /* __ASM_GENERIC_UCONTEXT_H */
diff --git a/arch/x86/include/asm/mman.h b/arch/x86/include/asm/mman.h
index 751af2550ed9..063d8c9e4d60 100644
--- a/arch/x86/include/asm/mman.h
+++ b/arch/x86/include/asm/mman.h
@@ -1,20 +1,8 @@
 #ifndef _ASM_X86_MMAN_H
 #define _ASM_X86_MMAN_H
 
-#include <asm-generic/mman-common.h>
-
 #define MAP_32BIT	0x40		/* only give out 32bit addresses */
 
-#define MAP_GROWSDOWN	0x0100		/* stack-like segment */
-#define MAP_DENYWRITE	0x0800		/* ETXTBSY */
-#define MAP_EXECUTABLE	0x1000		/* mark it as an executable */
-#define MAP_LOCKED	0x2000		/* pages are locked */
-#define MAP_NORESERVE	0x4000		/* don't check for reservations */
-#define MAP_POPULATE	0x8000		/* populate (prefault) pagetables */
-#define MAP_NONBLOCK	0x10000		/* do not block on IO */
-#define MAP_STACK	0x20000		/* give out an address that is best suited for process/thread stacks */
-
-#define MCL_CURRENT	1		/* lock all current mappings */
-#define MCL_FUTURE	2		/* lock all future mappings */
+#include <asm/generic-mman.h>
 
 #endif /* _ASM_X86_MMAN_H */
diff --git a/arch/x86/include/asm/module.h b/arch/x86/include/asm/module.h
index 47d62743c4d5..4a7a192910d0 100644
--- a/arch/x86/include/asm/module.h
+++ b/arch/x86/include/asm/module.h
@@ -1,18 +1,7 @@
 #ifndef _ASM_X86_MODULE_H
 #define _ASM_X86_MODULE_H
 
-/* x86_32/64 are simple */
-struct mod_arch_specific {};
-
-#ifdef CONFIG_X86_32
-# define Elf_Shdr Elf32_Shdr
-# define Elf_Sym Elf32_Sym
-# define Elf_Ehdr Elf32_Ehdr
-#else
-# define Elf_Shdr Elf64_Shdr
-# define Elf_Sym Elf64_Sym
-# define Elf_Ehdr Elf64_Ehdr
-#endif
+#include <asm/generic-module.h>
 
 #ifdef CONFIG_X86_64
 /* X86_64 does not define MODULE_PROC_FAMILY */
diff --git a/arch/x86/include/asm/scatterlist.h b/arch/x86/include/asm/scatterlist.h
index 263d397d2eef..2097d686471d 100644
--- a/arch/x86/include/asm/scatterlist.h
+++ b/arch/x86/include/asm/scatterlist.h
@@ -1,33 +1,8 @@
 #ifndef _ASM_X86_SCATTERLIST_H
 #define _ASM_X86_SCATTERLIST_H
 
-#include <asm/types.h>
-
-struct scatterlist {
-#ifdef CONFIG_DEBUG_SG
-	unsigned long	sg_magic;
-#endif
-	unsigned long	page_link;
-	unsigned int	offset;
-	unsigned int	length;
-	dma_addr_t	dma_address;
-	unsigned int	dma_length;
-};
-
-#define ARCH_HAS_SG_CHAIN
 #define ISA_DMA_THRESHOLD (0x00ffffff)
 
-/*
- * These macros should be used after a pci_map_sg call has been done
- * to get bus addresses of each of the SG entries and their lengths.
- * You should only work with the number of sg entries pci_map_sg
- * returns.
- */
-#define sg_dma_address(sg)	((sg)->dma_address)
-#ifdef CONFIG_X86_32
-# define sg_dma_len(sg)		((sg)->length)
-#else
-# define sg_dma_len(sg)		((sg)->dma_length)
-#endif
+#include <asm/generic-scatterlist.h>
 
 #endif /* _ASM_X86_SCATTERLIST_H */
diff --git a/arch/x86/include/asm/types.h b/arch/x86/include/asm/types.h
index 09b97745772f..f2fe528e9682 100644
--- a/arch/x86/include/asm/types.h
+++ b/arch/x86/include/asm/types.h
@@ -1,19 +1,11 @@
 #ifndef _ASM_X86_TYPES_H
 #define _ASM_X86_TYPES_H
 
-#include <asm-generic/int-ll64.h>
+#define dma_addr_t	dma_addr_t
 
-#ifndef __ASSEMBLY__
-
-typedef unsigned short umode_t;
+#include <asm/generic-types.h>
 
-#endif /* __ASSEMBLY__ */
-
-/*
- * These aren't exported outside the kernel to avoid name space clashes
- */
 #ifdef __KERNEL__
-
 #ifndef __ASSEMBLY__
 
 typedef u64 dma64_addr_t;
diff --git a/arch/x86/include/asm/ucontext.h b/arch/x86/include/asm/ucontext.h
index 87324cf439d9..7cfc436f86da 100644
--- a/arch/x86/include/asm/ucontext.h
+++ b/arch/x86/include/asm/ucontext.h
@@ -7,12 +7,6 @@
 				 * sigcontext struct (uc_mcontext).
 				 */
 
-struct ucontext {
-	unsigned long	  uc_flags;
-	struct ucontext  *uc_link;
-	stack_t		  uc_stack;
-	struct sigcontext uc_mcontext;
-	sigset_t	  uc_sigmask;	/* mask last for extensibility */
-};
+#include <asm/generic-ucontext.h>
 
 #endif /* _ASM_X86_UCONTEXT_H */
-- 
cgit v1.2.3


From 69d5ffdaad7b77b97229b55c36afb20e5bebd29e Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Thu, 18 Jun 2009 21:48:19 +0200
Subject: x86: convert termios.h to the asm-generic version

This patch turned out more controversial than expected
and may get dropped in the future. I'm including it
for reference anyway.

The user_termio_to_kernel_termios and kernel_termios_to_user_termio
functions on x86 are lacking error checking from get_user and
are not portable to big-endian systems, so the asm-generic
header has to differ in this regard.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
LKML-Reference: <cover.1245354003.git.arnd@arndb.de>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/include/asm/termios.h | 86 +++++++++++++++++++++++++++++++-----------
 1 file changed, 63 insertions(+), 23 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/termios.h b/arch/x86/include/asm/termios.h
index c4ee8056baca..d0922adc56d4 100644
--- a/arch/x86/include/asm/termios.h
+++ b/arch/x86/include/asm/termios.h
@@ -1,5 +1,12 @@
-#ifndef _ASM_X86_TERMIOS_H
-#define _ASM_X86_TERMIOS_H
+#ifndef _ASM_GENERIC_TERMIOS_H
+#define _ASM_GENERIC_TERMIOS_H
+/*
+ * Most architectures have straight copies of the x86 code, with
+ * varying levels of bug fixes on top. Usually it's a good idea
+ * to use this generic version instead, but be careful to avoid
+ * ABI changes.
+ * New architectures should not provide their own version.
+ */
 
 #include <asm/termbits.h>
 #include <asm/ioctls.h>
@@ -54,37 +61,57 @@ struct termio {
 /*
  * Translate a "termio" structure into a "termios". Ugh.
  */
-#define SET_LOW_TERMIOS_BITS(termios, termio, x) { \
-	unsigned short __tmp; \
-	get_user(__tmp,&(termio)->x); \
-	*(unsigned short *) &(termios)->x = __tmp; \
-}
-
 static inline int user_termio_to_kernel_termios(struct ktermios *termios,
-						struct termio __user *termio)
+						const struct termio __user *termio)
 {
-	SET_LOW_TERMIOS_BITS(termios, termio, c_iflag);
-	SET_LOW_TERMIOS_BITS(termios, termio, c_oflag);
-	SET_LOW_TERMIOS_BITS(termios, termio, c_cflag);
-	SET_LOW_TERMIOS_BITS(termios, termio, c_lflag);
-	get_user(termios->c_line, &termio->c_line);
-	return copy_from_user(termios->c_cc, termio->c_cc, NCC);
+	unsigned short tmp;
+
+	if (get_user(tmp, &termio->c_iflag) < 0)
+		goto fault;
+	termios->c_iflag = (0xffff0000 & termios->c_iflag) | tmp;
+
+	if (get_user(tmp, &termio->c_oflag) < 0)
+		goto fault;
+	termios->c_oflag = (0xffff0000 & termios->c_oflag) | tmp;
+
+	if (get_user(tmp, &termio->c_cflag) < 0)
+		goto fault;
+	termios->c_cflag = (0xffff0000 & termios->c_cflag) | tmp;
+
+	if (get_user(tmp, &termio->c_lflag) < 0)
+		goto fault;
+	termios->c_lflag = (0xffff0000 & termios->c_lflag) | tmp;
+
+	if (get_user(termios->c_line, &termio->c_line) < 0)
+		goto fault;
+
+	if (copy_from_user(termios->c_cc, termio->c_cc, NCC) != 0)
+		goto fault;
+
+	return 0;
+
+ fault:
+	return -EFAULT;
 }
 
 /*
  * Translate a "termios" structure into a "termio". Ugh.
  */
 static inline int kernel_termios_to_user_termio(struct termio __user *termio,
-					    struct ktermios *termios)
+						struct ktermios *termios)
 {
-	put_user((termios)->c_iflag, &(termio)->c_iflag);
-	put_user((termios)->c_oflag, &(termio)->c_oflag);
-	put_user((termios)->c_cflag, &(termio)->c_cflag);
-	put_user((termios)->c_lflag, &(termio)->c_lflag);
-	put_user((termios)->c_line,  &(termio)->c_line);
-	return copy_to_user((termio)->c_cc, (termios)->c_cc, NCC);
+	if (put_user(termios->c_iflag, &termio->c_iflag) < 0 ||
+	    put_user(termios->c_oflag, &termio->c_oflag) < 0 ||
+	    put_user(termios->c_cflag, &termio->c_cflag) < 0 ||
+	    put_user(termios->c_lflag, &termio->c_lflag) < 0 ||
+	    put_user(termios->c_line,  &termio->c_line) < 0 ||
+	    copy_to_user(termio->c_cc, termios->c_cc, NCC) != 0)
+		return -EFAULT;
+
+	return 0;
 }
 
+#ifdef TCGETS2
 static inline int user_termios_to_kernel_termios(struct ktermios *k,
 						 struct termios2 __user *u)
 {
@@ -108,7 +135,20 @@ static inline int kernel_termios_to_user_termios_1(struct termios __user *u,
 {
 	return copy_to_user(u, k, sizeof(struct termios));
 }
+#else /* TCGETS2 */
+static inline int user_termios_to_kernel_termios(struct ktermios *k,
+						 struct termios __user *u)
+{
+	return copy_from_user(k, u, sizeof(struct termios));
+}
+
+static inline int kernel_termios_to_user_termios(struct termios __user *u,
+						 struct ktermios *k)
+{
+	return copy_to_user(u, k, sizeof(struct termios));
+}
+#endif /* TCGETS2 */
 
 #endif	/* __KERNEL__ */
 
-#endif /* _ASM_X86_TERMIOS_H */
+#endif /* _ASM_GENERIC_TERMIOS_H */
-- 
cgit v1.2.3


From 73a2d096fdf23aa841f7595d114a11ec85a85e4d Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Thu, 18 Jun 2009 21:48:20 +0200
Subject: x86: remove all now-duplicate header files

All files that have been made identical to the asm-generic
version in the previous patches can now be removed,
guaranteeing that this does not introduce semantic changes.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
LKML-Reference: <cover.1245354003.git.arnd@arndb.de>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/include/asm/generic-mman.h        |  18 ---
 arch/x86/include/asm/generic-module.h      |  22 ----
 arch/x86/include/asm/generic-scatterlist.h |  43 -------
 arch/x86/include/asm/generic-types.h       |  42 ------
 arch/x86/include/asm/generic-ucontext.h    |  12 --
 arch/x86/include/asm/ioctls.h              | 111 +---------------
 arch/x86/include/asm/ipcbuf.h              |  34 +----
 arch/x86/include/asm/mman.h                |   2 +-
 arch/x86/include/asm/module.h              |   2 +-
 arch/x86/include/asm/msgbuf.h              |  48 +------
 arch/x86/include/asm/param.h               |  25 +---
 arch/x86/include/asm/scatterlist.h         |   2 +-
 arch/x86/include/asm/shmbuf.h              |  60 +--------
 arch/x86/include/asm/socket.h              |  61 +--------
 arch/x86/include/asm/sockios.h             |  14 +-
 arch/x86/include/asm/termbits.h            | 199 +----------------------------
 arch/x86/include/asm/termios.h             | 155 +---------------------
 arch/x86/include/asm/types.h               |   2 +-
 arch/x86/include/asm/ucontext.h            |   2 +-
 19 files changed, 14 insertions(+), 840 deletions(-)
 delete mode 100644 arch/x86/include/asm/generic-mman.h
 delete mode 100644 arch/x86/include/asm/generic-module.h
 delete mode 100644 arch/x86/include/asm/generic-scatterlist.h
 delete mode 100644 arch/x86/include/asm/generic-types.h
 delete mode 100644 arch/x86/include/asm/generic-ucontext.h

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/generic-mman.h b/arch/x86/include/asm/generic-mman.h
deleted file mode 100644
index 7cab4de2bca6..000000000000
--- a/arch/x86/include/asm/generic-mman.h
+++ /dev/null
@@ -1,18 +0,0 @@
-#ifndef __ASM_GENERIC_MMAN_H
-#define __ASM_GENERIC_MMAN_H
-
-#include <asm-generic/mman-common.h>
-
-#define MAP_GROWSDOWN	0x0100		/* stack-like segment */
-#define MAP_DENYWRITE	0x0800		/* ETXTBSY */
-#define MAP_EXECUTABLE	0x1000		/* mark it as an executable */
-#define MAP_LOCKED	0x2000		/* pages are locked */
-#define MAP_NORESERVE	0x4000		/* don't check for reservations */
-#define MAP_POPULATE	0x8000		/* populate (prefault) pagetables */
-#define MAP_NONBLOCK	0x10000		/* do not block on IO */
-#define MAP_STACK	0x20000		/* give out an address that is best suited for process/thread stacks */
-
-#define MCL_CURRENT	1		/* lock all current mappings */
-#define MCL_FUTURE	2		/* lock all future mappings */
-
-#endif /* __ASM_GENERIC_MMAN_H */
diff --git a/arch/x86/include/asm/generic-module.h b/arch/x86/include/asm/generic-module.h
deleted file mode 100644
index ed5b44de4c91..000000000000
--- a/arch/x86/include/asm/generic-module.h
+++ /dev/null
@@ -1,22 +0,0 @@
-#ifndef __ASM_GENERIC_MODULE_H
-#define __ASM_GENERIC_MODULE_H
-
-/*
- * Many architectures just need a simple module
- * loader without arch specific data.
- */
-struct mod_arch_specific
-{
-};
-
-#ifdef CONFIG_64BIT
-#define Elf_Shdr Elf64_Shdr
-#define Elf_Sym Elf64_Sym
-#define Elf_Ehdr Elf64_Ehdr
-#else
-#define Elf_Shdr Elf32_Shdr
-#define Elf_Sym Elf32_Sym
-#define Elf_Ehdr Elf32_Ehdr
-#endif
-
-#endif /* __ASM_GENERIC_MODULE_H */
diff --git a/arch/x86/include/asm/generic-scatterlist.h b/arch/x86/include/asm/generic-scatterlist.h
deleted file mode 100644
index 8b9454496a7c..000000000000
--- a/arch/x86/include/asm/generic-scatterlist.h
+++ /dev/null
@@ -1,43 +0,0 @@
-#ifndef __ASM_GENERIC_SCATTERLIST_H
-#define __ASM_GENERIC_SCATTERLIST_H
-
-#include <linux/types.h>
-
-struct scatterlist {
-#ifdef CONFIG_DEBUG_SG
-	unsigned long	sg_magic;
-#endif
-	unsigned long	page_link;
-	unsigned int	offset;
-	unsigned int	length;
-	dma_addr_t	dma_address;
-	unsigned int	dma_length;
-};
-
-/*
- * These macros should be used after a dma_map_sg call has been done
- * to get bus addresses of each of the SG entries and their lengths.
- * You should only work with the number of sg entries pci_map_sg
- * returns, or alternatively stop on the first sg_dma_len(sg) which
- * is 0.
- */
-#define sg_dma_address(sg)	((sg)->dma_address)
-#ifndef sg_dma_len
-/*
- * Normally, you have an iommu on 64 bit machines, but not on 32 bit
- * machines. Architectures that are differnt should override this.
- */
-#if __BITS_PER_LONG == 64
-#define sg_dma_len(sg)		((sg)->dma_length)
-#else
-#define sg_dma_len(sg)		((sg)->length)
-#endif /* 64 bit */
-#endif /* sg_dma_len */
-
-#ifndef ISA_DMA_THRESHOLD
-#define ISA_DMA_THRESHOLD	(~0UL)
-#endif
-
-#define ARCH_HAS_SG_CHAIN
-
-#endif /* __ASM_GENERIC_SCATTERLIST_H */
diff --git a/arch/x86/include/asm/generic-types.h b/arch/x86/include/asm/generic-types.h
deleted file mode 100644
index fba7d33ca3f2..000000000000
--- a/arch/x86/include/asm/generic-types.h
+++ /dev/null
@@ -1,42 +0,0 @@
-#ifndef _ASM_GENERIC_TYPES_H
-#define _ASM_GENERIC_TYPES_H
-/*
- * int-ll64 is used practically everywhere now,
- * so use it as a reasonable default.
- */
-#include <asm-generic/int-ll64.h>
-
-#ifndef __ASSEMBLY__
-
-typedef unsigned short umode_t;
-
-#endif /* __ASSEMBLY__ */
-
-/*
- * These aren't exported outside the kernel to avoid name space clashes
- */
-#ifdef __KERNEL__
-#ifndef __ASSEMBLY__
-/*
- * DMA addresses may be very different from physical addresses
- * and pointers. i386 and powerpc may have 64 bit DMA on 32 bit
- * systems, while sparc64 uses 32 bit DMA addresses for 64 bit
- * physical addresses.
- * This default defines dma_addr_t to have the same size as
- * phys_addr_t, which is the most common way.
- * Do not define the dma64_addr_t type, which never really
- * worked.
- */
-#ifndef dma_addr_t
-#ifdef CONFIG_PHYS_ADDR_T_64BIT
-typedef u64 dma_addr_t;
-#else
-typedef u32 dma_addr_t;
-#endif /* CONFIG_PHYS_ADDR_T_64BIT */
-#endif /* dma_addr_t */
-
-#endif /* __ASSEMBLY__ */
-
-#endif /* __KERNEL__ */
-
-#endif /* _ASM_GENERIC_TYPES_H */
diff --git a/arch/x86/include/asm/generic-ucontext.h b/arch/x86/include/asm/generic-ucontext.h
deleted file mode 100644
index ad77343e8a9a..000000000000
--- a/arch/x86/include/asm/generic-ucontext.h
+++ /dev/null
@@ -1,12 +0,0 @@
-#ifndef __ASM_GENERIC_UCONTEXT_H
-#define __ASM_GENERIC_UCONTEXT_H
-
-struct ucontext {
-	unsigned long	  uc_flags;
-	struct ucontext  *uc_link;
-	stack_t		  uc_stack;
-	struct sigcontext uc_mcontext;
-	sigset_t	  uc_sigmask;	/* mask last for extensibility */
-};
-
-#endif /* __ASM_GENERIC_UCONTEXT_H */
diff --git a/arch/x86/include/asm/ioctls.h b/arch/x86/include/asm/ioctls.h
index a799e20a769e..ec34c760665e 100644
--- a/arch/x86/include/asm/ioctls.h
+++ b/arch/x86/include/asm/ioctls.h
@@ -1,110 +1 @@
-#ifndef __ASM_GENERIC_IOCTLS_H
-#define __ASM_GENERIC_IOCTLS_H
-
-#include <linux/ioctl.h>
-
-/*
- * These are the most common definitions for tty ioctl numbers.
- * Most of them do not use the recommended _IOC(), but there is
- * probably some source code out there hardcoding the number,
- * so we might as well use them for all new platforms.
- *
- * The architectures that use different values here typically
- * try to be compatible with some Unix variants for the same
- * architecture.
- */
-
-/* 0x54 is just a magic number to make these relatively unique ('T') */
-
-#define TCGETS		0x5401
-#define TCSETS		0x5402
-#define TCSETSW		0x5403
-#define TCSETSF		0x5404
-#define TCGETA		0x5405
-#define TCSETA		0x5406
-#define TCSETAW		0x5407
-#define TCSETAF		0x5408
-#define TCSBRK		0x5409
-#define TCXONC		0x540A
-#define TCFLSH		0x540B
-#define TIOCEXCL	0x540C
-#define TIOCNXCL	0x540D
-#define TIOCSCTTY	0x540E
-#define TIOCGPGRP	0x540F
-#define TIOCSPGRP	0x5410
-#define TIOCOUTQ	0x5411
-#define TIOCSTI		0x5412
-#define TIOCGWINSZ	0x5413
-#define TIOCSWINSZ	0x5414
-#define TIOCMGET	0x5415
-#define TIOCMBIS	0x5416
-#define TIOCMBIC	0x5417
-#define TIOCMSET	0x5418
-#define TIOCGSOFTCAR	0x5419
-#define TIOCSSOFTCAR	0x541A
-#define FIONREAD	0x541B
-#define TIOCINQ		FIONREAD
-#define TIOCLINUX	0x541C
-#define TIOCCONS	0x541D
-#define TIOCGSERIAL	0x541E
-#define TIOCSSERIAL	0x541F
-#define TIOCPKT		0x5420
-#define FIONBIO		0x5421
-#define TIOCNOTTY	0x5422
-#define TIOCSETD	0x5423
-#define TIOCGETD	0x5424
-#define TCSBRKP		0x5425	/* Needed for POSIX tcsendbreak() */
-#define TIOCSBRK	0x5427  /* BSD compatibility */
-#define TIOCCBRK	0x5428  /* BSD compatibility */
-#define TIOCGSID	0x5429  /* Return the session ID of FD */
-#define TCGETS2		_IOR('T', 0x2A, struct termios2)
-#define TCSETS2		_IOW('T', 0x2B, struct termios2)
-#define TCSETSW2	_IOW('T', 0x2C, struct termios2)
-#define TCSETSF2	_IOW('T', 0x2D, struct termios2)
-#define TIOCGRS485	0x542E
-#define TIOCSRS485	0x542F
-#define TIOCGPTN	_IOR('T', 0x30, unsigned int) /* Get Pty Number (of pty-mux device) */
-#define TIOCSPTLCK	_IOW('T', 0x31, int)  /* Lock/unlock Pty */
-#define TCGETX		0x5432 /* SYS5 TCGETX compatibility */
-#define TCSETX		0x5433
-#define TCSETXF		0x5434
-#define TCSETXW		0x5435
-
-#define FIONCLEX	0x5450
-#define FIOCLEX		0x5451
-#define FIOASYNC	0x5452
-#define TIOCSERCONFIG	0x5453
-#define TIOCSERGWILD	0x5454
-#define TIOCSERSWILD	0x5455
-#define TIOCGLCKTRMIOS	0x5456
-#define TIOCSLCKTRMIOS	0x5457
-#define TIOCSERGSTRUCT	0x5458 /* For debugging only */
-#define TIOCSERGETLSR   0x5459 /* Get line status register */
-#define TIOCSERGETMULTI 0x545A /* Get multiport config  */
-#define TIOCSERSETMULTI 0x545B /* Set multiport config */
-
-#define TIOCMIWAIT	0x545C	/* wait for a change on serial input line(s) */
-#define TIOCGICOUNT	0x545D	/* read serial port inline interrupt counts */
-
-/*
- * some architectures define FIOQSIZE as 0x545E, which is used for
- * TIOCGHAYESESP on others
- */
-#ifndef FIOQSIZE
-# define TIOCGHAYESESP	0x545E  /* Get Hayes ESP configuration */
-# define TIOCSHAYESESP	0x545F  /* Set Hayes ESP configuration */
-# define FIOQSIZE	0x5460
-#endif
-
-/* Used for packet mode */
-#define TIOCPKT_DATA		 0
-#define TIOCPKT_FLUSHREAD	 1
-#define TIOCPKT_FLUSHWRITE	 2
-#define TIOCPKT_STOP		 4
-#define TIOCPKT_START		 8
-#define TIOCPKT_NOSTOP		16
-#define TIOCPKT_DOSTOP		32
-
-#define TIOCSER_TEMT	0x01	/* Transmitter physically empty */
-
-#endif /* __ASM_GENERIC_IOCTLS_H */
+#include <asm-generic/ioctls.h>
diff --git a/arch/x86/include/asm/ipcbuf.h b/arch/x86/include/asm/ipcbuf.h
index 888ca1c8f951..84c7e51cb6d0 100644
--- a/arch/x86/include/asm/ipcbuf.h
+++ b/arch/x86/include/asm/ipcbuf.h
@@ -1,33 +1 @@
-#ifndef __ASM_GENERIC_IPCBUF_H
-#define __ASM_GENERIC_IPCBUF_H
-
-/*
- * The generic ipc64_perm structure:
- * Note extra padding because this structure is passed back and forth
- * between kernel and user space.
- *
- * ipc64_perm was originally meant to be architecture specific, but
- * everyone just ended up making identical copies without specific
- * optimizations, so we may just as well all use the same one.
- *
- * Pad space is left for:
- * - 32-bit mode_t on architectures that only had 16 bit
- * - 32-bit seq
- * - 2 miscellaneous 32-bit values
- */
-
-struct ipc64_perm {
-	__kernel_key_t		key;
-	__kernel_uid32_t	uid;
-	__kernel_gid32_t	gid;
-	__kernel_uid32_t	cuid;
-	__kernel_gid32_t	cgid;
-	__kernel_mode_t		mode;
-	unsigned short		__pad1;
-	unsigned short		seq;
-	unsigned short		__pad2;
-	unsigned long		__unused1;
-	unsigned long		__unused2;
-};
-
-#endif /* __ASM_GENERIC_IPCBUF_H */
+#include <asm-generic/ipcbuf.h>
diff --git a/arch/x86/include/asm/mman.h b/arch/x86/include/asm/mman.h
index 063d8c9e4d60..593e51d4643f 100644
--- a/arch/x86/include/asm/mman.h
+++ b/arch/x86/include/asm/mman.h
@@ -3,6 +3,6 @@
 
 #define MAP_32BIT	0x40		/* only give out 32bit addresses */
 
-#include <asm/generic-mman.h>
+#include <asm-generic/mman.h>
 
 #endif /* _ASM_X86_MMAN_H */
diff --git a/arch/x86/include/asm/module.h b/arch/x86/include/asm/module.h
index 4a7a192910d0..555bc12bdcd6 100644
--- a/arch/x86/include/asm/module.h
+++ b/arch/x86/include/asm/module.h
@@ -1,7 +1,7 @@
 #ifndef _ASM_X86_MODULE_H
 #define _ASM_X86_MODULE_H
 
-#include <asm/generic-module.h>
+#include <asm-generic/module.h>
 
 #ifdef CONFIG_X86_64
 /* X86_64 does not define MODULE_PROC_FAMILY */
diff --git a/arch/x86/include/asm/msgbuf.h b/arch/x86/include/asm/msgbuf.h
index aec850d9159e..809134c644a6 100644
--- a/arch/x86/include/asm/msgbuf.h
+++ b/arch/x86/include/asm/msgbuf.h
@@ -1,47 +1 @@
-#ifndef __ASM_GENERIC_MSGBUF_H
-#define __ASM_GENERIC_MSGBUF_H
-
-#include <asm/bitsperlong.h>
-/*
- * generic msqid64_ds structure.
- *
- * Note extra padding because this structure is passed back and forth
- * between kernel and user space.
- *
- * msqid64_ds was originally meant to be architecture specific, but
- * everyone just ended up making identical copies without specific
- * optimizations, so we may just as well all use the same one.
- *
- * 64 bit architectures typically define a 64 bit __kernel_time_t,
- * so they do not need the first three padding words.
- * On big-endian systems, the padding is in the wrong place.
- *
- * Pad space is left for:
- * - 64-bit time_t to solve y2038 problem
- * - 2 miscellaneous 32-bit values
- */
-
-struct msqid64_ds {
-	struct ipc64_perm msg_perm;
-	__kernel_time_t msg_stime;	/* last msgsnd time */
-#if __BITS_PER_LONG != 64
-	unsigned long	__unused1;
-#endif
-	__kernel_time_t msg_rtime;	/* last msgrcv time */
-#if __BITS_PER_LONG != 64
-	unsigned long	__unused2;
-#endif
-	__kernel_time_t msg_ctime;	/* last change time */
-#if __BITS_PER_LONG != 64
-	unsigned long	__unused3;
-#endif
-	unsigned long  msg_cbytes;	/* current number of bytes on queue */
-	unsigned long  msg_qnum;	/* number of messages in queue */
-	unsigned long  msg_qbytes;	/* max number of bytes on queue */
-	__kernel_pid_t msg_lspid;	/* pid of last msgsnd */
-	__kernel_pid_t msg_lrpid;	/* last receive pid */
-	unsigned long  __unused4;
-	unsigned long  __unused5;
-};
-
-#endif /* __ASM_GENERIC_MSGBUF_H */
+#include <asm-generic/msgbuf.h>
diff --git a/arch/x86/include/asm/param.h b/arch/x86/include/asm/param.h
index cdf8251bfb6c..965d45427975 100644
--- a/arch/x86/include/asm/param.h
+++ b/arch/x86/include/asm/param.h
@@ -1,24 +1 @@
-#ifndef __ASM_GENERIC_PARAM_H
-#define __ASM_GENERIC_PARAM_H
-
-#ifdef __KERNEL__
-# define HZ		CONFIG_HZ	/* Internal kernel timer frequency */
-# define USER_HZ	100		/* some user interfaces are */
-# define CLOCKS_PER_SEC	(USER_HZ)       /* in "ticks" like times() */
-#endif
-
-#ifndef HZ
-#define HZ 100
-#endif
-
-#ifndef EXEC_PAGESIZE
-#define EXEC_PAGESIZE	4096
-#endif
-
-#ifndef NOGROUP
-#define NOGROUP		(-1)
-#endif
-
-#define MAXHOSTNAMELEN	64	/* max length of hostname */
-
-#endif /* __ASM_GENERIC_PARAM_H */
+#include <asm-generic/param.h>
diff --git a/arch/x86/include/asm/scatterlist.h b/arch/x86/include/asm/scatterlist.h
index 2097d686471d..75af592677ec 100644
--- a/arch/x86/include/asm/scatterlist.h
+++ b/arch/x86/include/asm/scatterlist.h
@@ -3,6 +3,6 @@
 
 #define ISA_DMA_THRESHOLD (0x00ffffff)
 
-#include <asm/generic-scatterlist.h>
+#include <asm-generic/scatterlist.h>
 
 #endif /* _ASM_X86_SCATTERLIST_H */
diff --git a/arch/x86/include/asm/shmbuf.h b/arch/x86/include/asm/shmbuf.h
index 5768fa60ac82..83c05fc2de38 100644
--- a/arch/x86/include/asm/shmbuf.h
+++ b/arch/x86/include/asm/shmbuf.h
@@ -1,59 +1 @@
-#ifndef __ASM_GENERIC_SHMBUF_H
-#define __ASM_GENERIC_SHMBUF_H
-
-#include <asm/bitsperlong.h>
-
-/*
- * The shmid64_ds structure for x86 architecture.
- * Note extra padding because this structure is passed back and forth
- * between kernel and user space.
- *
- * shmid64_ds was originally meant to be architecture specific, but
- * everyone just ended up making identical copies without specific
- * optimizations, so we may just as well all use the same one.
- *
- * 64 bit architectures typically define a 64 bit __kernel_time_t,
- * so they do not need the first two padding words.
- * On big-endian systems, the padding is in the wrong place.
- *
- *
- * Pad space is left for:
- * - 64-bit time_t to solve y2038 problem
- * - 2 miscellaneous 32-bit values
- */
-
-struct shmid64_ds {
-	struct ipc64_perm	shm_perm;	/* operation perms */
-	size_t			shm_segsz;	/* size of segment (bytes) */
-	__kernel_time_t		shm_atime;	/* last attach time */
-#if __BITS_PER_LONG != 64
-	unsigned long		__unused1;
-#endif
-	__kernel_time_t		shm_dtime;	/* last detach time */
-#if __BITS_PER_LONG != 64
-	unsigned long		__unused2;
-#endif
-	__kernel_time_t		shm_ctime;	/* last change time */
-#if __BITS_PER_LONG != 64
-	unsigned long		__unused3;
-#endif
-	__kernel_pid_t		shm_cpid;	/* pid of creator */
-	__kernel_pid_t		shm_lpid;	/* pid of last operator */
-	unsigned long		shm_nattch;	/* no. of current attaches */
-	unsigned long		__unused4;
-	unsigned long		__unused5;
-};
-
-struct shminfo64 {
-	unsigned long	shmmax;
-	unsigned long	shmmin;
-	unsigned long	shmmni;
-	unsigned long	shmseg;
-	unsigned long	shmall;
-	unsigned long	__unused1;
-	unsigned long	__unused2;
-	unsigned long	__unused3;
-	unsigned long	__unused4;
-};
-
-#endif /* __ASM_GENERIC_SHMBUF_H */
+#include <asm-generic/shmbuf.h>
diff --git a/arch/x86/include/asm/socket.h b/arch/x86/include/asm/socket.h
index d4ae42a06a20..6b71384b9d8b 100644
--- a/arch/x86/include/asm/socket.h
+++ b/arch/x86/include/asm/socket.h
@@ -1,60 +1 @@
-#ifndef __ASM_GENERIC_SOCKET_H
-#define __ASM_GENERIC_SOCKET_H
-
-#include <asm/sockios.h>
-
-/* For setsockopt(2) */
-#define SOL_SOCKET	1
-
-#define SO_DEBUG	1
-#define SO_REUSEADDR	2
-#define SO_TYPE		3
-#define SO_ERROR	4
-#define SO_DONTROUTE	5
-#define SO_BROADCAST	6
-#define SO_SNDBUF	7
-#define SO_RCVBUF	8
-#define SO_SNDBUFFORCE	32
-#define SO_RCVBUFFORCE	33
-#define SO_KEEPALIVE	9
-#define SO_OOBINLINE	10
-#define SO_NO_CHECK	11
-#define SO_PRIORITY	12
-#define SO_LINGER	13
-#define SO_BSDCOMPAT	14
-/* To add :#define SO_REUSEPORT 15 */
-#define SO_PASSCRED	16
-#define SO_PEERCRED	17
-#define SO_RCVLOWAT	18
-#define SO_SNDLOWAT	19
-#define SO_RCVTIMEO	20
-#define SO_SNDTIMEO	21
-
-/* Security levels - as per NRL IPv6 - don't actually do anything */
-#define SO_SECURITY_AUTHENTICATION		22
-#define SO_SECURITY_ENCRYPTION_TRANSPORT	23
-#define SO_SECURITY_ENCRYPTION_NETWORK		24
-
-#define SO_BINDTODEVICE	25
-
-/* Socket filtering */
-#define SO_ATTACH_FILTER	26
-#define SO_DETACH_FILTER	27
-
-#define SO_PEERNAME		28
-#define SO_TIMESTAMP		29
-#define SCM_TIMESTAMP		SO_TIMESTAMP
-
-#define SO_ACCEPTCONN		30
-
-#define SO_PEERSEC		31
-#define SO_PASSSEC		34
-#define SO_TIMESTAMPNS		35
-#define SCM_TIMESTAMPNS		SO_TIMESTAMPNS
-
-#define SO_MARK			36
-
-#define SO_TIMESTAMPING		37
-#define SCM_TIMESTAMPING	SO_TIMESTAMPING
-
-#endif /* __ASM_GENERIC_SOCKET_H */
+#include <asm-generic/socket.h>
diff --git a/arch/x86/include/asm/sockios.h b/arch/x86/include/asm/sockios.h
index 9a61a369b901..def6d4746ee7 100644
--- a/arch/x86/include/asm/sockios.h
+++ b/arch/x86/include/asm/sockios.h
@@ -1,13 +1 @@
-#ifndef __ASM_GENERIC_SOCKIOS_H
-#define __ASM_GENERIC_SOCKIOS_H
-
-/* Socket-level I/O control calls. */
-#define FIOSETOWN	0x8901
-#define SIOCSPGRP	0x8902
-#define FIOGETOWN	0x8903
-#define SIOCGPGRP	0x8904
-#define SIOCATMARK	0x8905
-#define SIOCGSTAMP	0x8906		/* Get stamp (timeval) */
-#define SIOCGSTAMPNS	0x8907		/* Get stamp (timespec) */
-
-#endif /* __ASM_GENERIC_SOCKIOS_H */
+#include <asm-generic/sockios.h>
diff --git a/arch/x86/include/asm/termbits.h b/arch/x86/include/asm/termbits.h
index 1c9773d48cb0..3935b106de79 100644
--- a/arch/x86/include/asm/termbits.h
+++ b/arch/x86/include/asm/termbits.h
@@ -1,198 +1 @@
-#ifndef __ASM_GENERIC_TERMBITS_H
-#define __ASM_GENERIC_TERMBITS_H
-
-#include <linux/posix_types.h>
-
-typedef unsigned char	cc_t;
-typedef unsigned int	speed_t;
-typedef unsigned int	tcflag_t;
-
-#define NCCS 19
-struct termios {
-	tcflag_t c_iflag;		/* input mode flags */
-	tcflag_t c_oflag;		/* output mode flags */
-	tcflag_t c_cflag;		/* control mode flags */
-	tcflag_t c_lflag;		/* local mode flags */
-	cc_t c_line;			/* line discipline */
-	cc_t c_cc[NCCS];		/* control characters */
-};
-
-struct termios2 {
-	tcflag_t c_iflag;		/* input mode flags */
-	tcflag_t c_oflag;		/* output mode flags */
-	tcflag_t c_cflag;		/* control mode flags */
-	tcflag_t c_lflag;		/* local mode flags */
-	cc_t c_line;			/* line discipline */
-	cc_t c_cc[NCCS];		/* control characters */
-	speed_t c_ispeed;		/* input speed */
-	speed_t c_ospeed;		/* output speed */
-};
-
-struct ktermios {
-	tcflag_t c_iflag;		/* input mode flags */
-	tcflag_t c_oflag;		/* output mode flags */
-	tcflag_t c_cflag;		/* control mode flags */
-	tcflag_t c_lflag;		/* local mode flags */
-	cc_t c_line;			/* line discipline */
-	cc_t c_cc[NCCS];		/* control characters */
-	speed_t c_ispeed;		/* input speed */
-	speed_t c_ospeed;		/* output speed */
-};
-
-/* c_cc characters */
-#define VINTR 0
-#define VQUIT 1
-#define VERASE 2
-#define VKILL 3
-#define VEOF 4
-#define VTIME 5
-#define VMIN 6
-#define VSWTC 7
-#define VSTART 8
-#define VSTOP 9
-#define VSUSP 10
-#define VEOL 11
-#define VREPRINT 12
-#define VDISCARD 13
-#define VWERASE 14
-#define VLNEXT 15
-#define VEOL2 16
-
-/* c_iflag bits */
-#define IGNBRK	0000001
-#define BRKINT	0000002
-#define IGNPAR	0000004
-#define PARMRK	0000010
-#define INPCK	0000020
-#define ISTRIP	0000040
-#define INLCR	0000100
-#define IGNCR	0000200
-#define ICRNL	0000400
-#define IUCLC	0001000
-#define IXON	0002000
-#define IXANY	0004000
-#define IXOFF	0010000
-#define IMAXBEL	0020000
-#define IUTF8	0040000
-
-/* c_oflag bits */
-#define OPOST	0000001
-#define OLCUC	0000002
-#define ONLCR	0000004
-#define OCRNL	0000010
-#define ONOCR	0000020
-#define ONLRET	0000040
-#define OFILL	0000100
-#define OFDEL	0000200
-#define NLDLY	0000400
-#define   NL0	0000000
-#define   NL1	0000400
-#define CRDLY	0003000
-#define   CR0	0000000
-#define   CR1	0001000
-#define   CR2	0002000
-#define   CR3	0003000
-#define TABDLY	0014000
-#define   TAB0	0000000
-#define   TAB1	0004000
-#define   TAB2	0010000
-#define   TAB3	0014000
-#define   XTABS	0014000
-#define BSDLY	0020000
-#define   BS0	0000000
-#define   BS1	0020000
-#define VTDLY	0040000
-#define   VT0	0000000
-#define   VT1	0040000
-#define FFDLY	0100000
-#define   FF0	0000000
-#define   FF1	0100000
-
-/* c_cflag bit meaning */
-#define CBAUD	0010017
-#define  B0	0000000		/* hang up */
-#define  B50	0000001
-#define  B75	0000002
-#define  B110	0000003
-#define  B134	0000004
-#define  B150	0000005
-#define  B200	0000006
-#define  B300	0000007
-#define  B600	0000010
-#define  B1200	0000011
-#define  B1800	0000012
-#define  B2400	0000013
-#define  B4800	0000014
-#define  B9600	0000015
-#define  B19200	0000016
-#define  B38400	0000017
-#define EXTA B19200
-#define EXTB B38400
-#define CSIZE	0000060
-#define   CS5	0000000
-#define   CS6	0000020
-#define   CS7	0000040
-#define   CS8	0000060
-#define CSTOPB	0000100
-#define CREAD	0000200
-#define PARENB	0000400
-#define PARODD	0001000
-#define HUPCL	0002000
-#define CLOCAL	0004000
-#define CBAUDEX 0010000
-#define    BOTHER 0010000
-#define    B57600 0010001
-#define   B115200 0010002
-#define   B230400 0010003
-#define   B460800 0010004
-#define   B500000 0010005
-#define   B576000 0010006
-#define   B921600 0010007
-#define  B1000000 0010010
-#define  B1152000 0010011
-#define  B1500000 0010012
-#define  B2000000 0010013
-#define  B2500000 0010014
-#define  B3000000 0010015
-#define  B3500000 0010016
-#define  B4000000 0010017
-#define CIBAUD	  002003600000	/* input baud rate */
-#define CMSPAR	  010000000000	/* mark or space (stick) parity */
-#define CRTSCTS	  020000000000	/* flow control */
-
-#define IBSHIFT	  16		/* Shift from CBAUD to CIBAUD */
-
-/* c_lflag bits */
-#define ISIG	0000001
-#define ICANON	0000002
-#define XCASE	0000004
-#define ECHO	0000010
-#define ECHOE	0000020
-#define ECHOK	0000040
-#define ECHONL	0000100
-#define NOFLSH	0000200
-#define TOSTOP	0000400
-#define ECHOCTL	0001000
-#define ECHOPRT	0002000
-#define ECHOKE	0004000
-#define FLUSHO	0010000
-#define PENDIN	0040000
-#define IEXTEN	0100000
-
-/* tcflow() and TCXONC use these */
-#define	TCOOFF		0
-#define	TCOON		1
-#define	TCIOFF		2
-#define	TCION		3
-
-/* tcflush() and TCFLSH use these */
-#define	TCIFLUSH	0
-#define	TCOFLUSH	1
-#define	TCIOFLUSH	2
-
-/* tcsetattr uses these */
-#define	TCSANOW		0
-#define	TCSADRAIN	1
-#define	TCSAFLUSH	2
-
-#endif /* __ASM_GENERIC_TERMBITS_H */
+#include <asm-generic/termbits.h>
diff --git a/arch/x86/include/asm/termios.h b/arch/x86/include/asm/termios.h
index d0922adc56d4..280d78a9d966 100644
--- a/arch/x86/include/asm/termios.h
+++ b/arch/x86/include/asm/termios.h
@@ -1,154 +1 @@
-#ifndef _ASM_GENERIC_TERMIOS_H
-#define _ASM_GENERIC_TERMIOS_H
-/*
- * Most architectures have straight copies of the x86 code, with
- * varying levels of bug fixes on top. Usually it's a good idea
- * to use this generic version instead, but be careful to avoid
- * ABI changes.
- * New architectures should not provide their own version.
- */
-
-#include <asm/termbits.h>
-#include <asm/ioctls.h>
-
-struct winsize {
-	unsigned short ws_row;
-	unsigned short ws_col;
-	unsigned short ws_xpixel;
-	unsigned short ws_ypixel;
-};
-
-#define NCC 8
-struct termio {
-	unsigned short c_iflag;		/* input mode flags */
-	unsigned short c_oflag;		/* output mode flags */
-	unsigned short c_cflag;		/* control mode flags */
-	unsigned short c_lflag;		/* local mode flags */
-	unsigned char c_line;		/* line discipline */
-	unsigned char c_cc[NCC];	/* control characters */
-};
-
-/* modem lines */
-#define TIOCM_LE	0x001
-#define TIOCM_DTR	0x002
-#define TIOCM_RTS	0x004
-#define TIOCM_ST	0x008
-#define TIOCM_SR	0x010
-#define TIOCM_CTS	0x020
-#define TIOCM_CAR	0x040
-#define TIOCM_RNG	0x080
-#define TIOCM_DSR	0x100
-#define TIOCM_CD	TIOCM_CAR
-#define TIOCM_RI	TIOCM_RNG
-#define TIOCM_OUT1	0x2000
-#define TIOCM_OUT2	0x4000
-#define TIOCM_LOOP	0x8000
-
-/* ioctl (fd, TIOCSERGETLSR, &result) where result may be as below */
-
-#ifdef __KERNEL__
-
-#include <asm/uaccess.h>
-
-/*	intr=^C		quit=^\		erase=del	kill=^U
-	eof=^D		vtime=\0	vmin=\1		sxtc=\0
-	start=^Q	stop=^S		susp=^Z		eol=\0
-	reprint=^R	discard=^U	werase=^W	lnext=^V
-	eol2=\0
-*/
-#define INIT_C_CC "\003\034\177\025\004\0\1\0\021\023\032\0\022\017\027\026\0"
-
-/*
- * Translate a "termio" structure into a "termios". Ugh.
- */
-static inline int user_termio_to_kernel_termios(struct ktermios *termios,
-						const struct termio __user *termio)
-{
-	unsigned short tmp;
-
-	if (get_user(tmp, &termio->c_iflag) < 0)
-		goto fault;
-	termios->c_iflag = (0xffff0000 & termios->c_iflag) | tmp;
-
-	if (get_user(tmp, &termio->c_oflag) < 0)
-		goto fault;
-	termios->c_oflag = (0xffff0000 & termios->c_oflag) | tmp;
-
-	if (get_user(tmp, &termio->c_cflag) < 0)
-		goto fault;
-	termios->c_cflag = (0xffff0000 & termios->c_cflag) | tmp;
-
-	if (get_user(tmp, &termio->c_lflag) < 0)
-		goto fault;
-	termios->c_lflag = (0xffff0000 & termios->c_lflag) | tmp;
-
-	if (get_user(termios->c_line, &termio->c_line) < 0)
-		goto fault;
-
-	if (copy_from_user(termios->c_cc, termio->c_cc, NCC) != 0)
-		goto fault;
-
-	return 0;
-
- fault:
-	return -EFAULT;
-}
-
-/*
- * Translate a "termios" structure into a "termio". Ugh.
- */
-static inline int kernel_termios_to_user_termio(struct termio __user *termio,
-						struct ktermios *termios)
-{
-	if (put_user(termios->c_iflag, &termio->c_iflag) < 0 ||
-	    put_user(termios->c_oflag, &termio->c_oflag) < 0 ||
-	    put_user(termios->c_cflag, &termio->c_cflag) < 0 ||
-	    put_user(termios->c_lflag, &termio->c_lflag) < 0 ||
-	    put_user(termios->c_line,  &termio->c_line) < 0 ||
-	    copy_to_user(termio->c_cc, termios->c_cc, NCC) != 0)
-		return -EFAULT;
-
-	return 0;
-}
-
-#ifdef TCGETS2
-static inline int user_termios_to_kernel_termios(struct ktermios *k,
-						 struct termios2 __user *u)
-{
-	return copy_from_user(k, u, sizeof(struct termios2));
-}
-
-static inline int kernel_termios_to_user_termios(struct termios2 __user *u,
-						 struct ktermios *k)
-{
-	return copy_to_user(u, k, sizeof(struct termios2));
-}
-
-static inline int user_termios_to_kernel_termios_1(struct ktermios *k,
-						   struct termios __user *u)
-{
-	return copy_from_user(k, u, sizeof(struct termios));
-}
-
-static inline int kernel_termios_to_user_termios_1(struct termios __user *u,
-						   struct ktermios *k)
-{
-	return copy_to_user(u, k, sizeof(struct termios));
-}
-#else /* TCGETS2 */
-static inline int user_termios_to_kernel_termios(struct ktermios *k,
-						 struct termios __user *u)
-{
-	return copy_from_user(k, u, sizeof(struct termios));
-}
-
-static inline int kernel_termios_to_user_termios(struct termios __user *u,
-						 struct ktermios *k)
-{
-	return copy_to_user(u, k, sizeof(struct termios));
-}
-#endif /* TCGETS2 */
-
-#endif	/* __KERNEL__ */
-
-#endif /* _ASM_GENERIC_TERMIOS_H */
+#include <asm-generic/termios.h>
diff --git a/arch/x86/include/asm/types.h b/arch/x86/include/asm/types.h
index f2fe528e9682..df1da20f4534 100644
--- a/arch/x86/include/asm/types.h
+++ b/arch/x86/include/asm/types.h
@@ -3,7 +3,7 @@
 
 #define dma_addr_t	dma_addr_t
 
-#include <asm/generic-types.h>
+#include <asm-generic/types.h>
 
 #ifdef __KERNEL__
 #ifndef __ASSEMBLY__
diff --git a/arch/x86/include/asm/ucontext.h b/arch/x86/include/asm/ucontext.h
index 7cfc436f86da..b7c29c8017f2 100644
--- a/arch/x86/include/asm/ucontext.h
+++ b/arch/x86/include/asm/ucontext.h
@@ -7,6 +7,6 @@
 				 * sigcontext struct (uc_mcontext).
 				 */
 
-#include <asm/generic-ucontext.h>
+#include <asm-generic/ucontext.h>
 
 #endif /* _ASM_X86_UCONTEXT_H */
-- 
cgit v1.2.3


From feaa0457ec8351cae855edc9a3052ac49322538e Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Sat, 20 Jun 2009 16:15:40 +0530
Subject: x86: ds.c fix invalid assignment

Fixes the type mixups that cause the following sparse warnings:

  CHECK   arch/x86/kernel/ds.c
arch/x86/kernel/ds.c:549:19: warning: incorrect type in argument 2 (invalid types)
arch/x86/kernel/ds.c:549:19:    expected bad type enum bts_field field
arch/x86/kernel/ds.c:549:19:    got int
arch/x86/kernel/ds.c:514:35: error: incompatible types for operation (*)
arch/x86/kernel/ds.c:514:35:    left side has type unsigned char static [unsigned] [toplevel] sizeof_ptr_field
arch/x86/kernel/ds.c:514:35:    right side has type bad type enum bts_field field
arch/x86/kernel/ds.c:514:7: error: invalid assignment
arch/x86/kernel/ds.c:514:35: error: incompatible types for operation (*)
arch/x86/kernel/ds.c:514:35:    left side has type unsigned char static [unsigned] [toplevel] sizeof_ptr_field
arch/x86/kernel/ds.c:514:35:    right side has type bad type enum bts_field field
arch/x86/kernel/ds.c:514:7: error: invalid assignment
arch/x86/kernel/ds.c:514:35: error: incompatible types for operation (*)
arch/x86/kernel/ds.c:514:35:    left side has type unsigned char static [unsigned] [toplevel] sizeof_ptr_field
arch/x86/kernel/ds.c:514:35:    right side has type bad type enum bts_field field
arch/x86/kernel/ds.c:514:7: error: invalid assignment
arch/x86/kernel/ds.c:514:35: error: incompatible types for operation (*)
arch/x86/kernel/ds.c:514:35:    left side has type unsigned char static [unsigned] [toplevel] sizeof_ptr_field
arch/x86/kernel/ds.c:514:35:    right side has type bad type enum bts_field field
arch/x86/kernel/ds.c:514:7: error: invalid assignment
arch/x86/kernel/ds.c:514:35: error: incompatible types for operation (*)
arch/x86/kernel/ds.c:514:35:    left side has type unsigned char static [unsigned] [toplevel] sizeof_ptr_field
arch/x86/kernel/ds.c:514:35:    right side has type bad type enum bts_field field
arch/x86/kernel/ds.c:514:7: error: invalid assignment
arch/x86/kernel/ds.c:514:35: error: incompatible types for operation (*)
arch/x86/kernel/ds.c:514:35:    left side has type unsigned char static [unsigned] [toplevel] sizeof_ptr_field
arch/x86/kernel/ds.c:514:35:    right side has type bad type enum bts_field field
arch/x86/kernel/ds.c:514:7: error: invalid assignment
arch/x86/kernel/ds.c:520:35: error: incompatible types for operation (*)
arch/x86/kernel/ds.c:520:35:    left side has type unsigned char static [unsigned] [toplevel] sizeof_ptr_field
arch/x86/kernel/ds.c:520:35:    right side has type bad type enum bts_field field
arch/x86/kernel/ds.c:520:7: error: invalid assignment
arch/x86/kernel/ds.c:520:35: error: incompatible types for operation (*)

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Markus Metzger <markus.t.metzger@intel.com>
LKML-Reference: <1245494740.8613.12.camel@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/ds.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c
index 48bfe1386038..ef42a038f1a6 100644
--- a/arch/x86/kernel/ds.c
+++ b/arch/x86/kernel/ds.c
@@ -509,15 +509,15 @@ enum bts_field {
 	bts_escape		= ((unsigned long)-1 & ~bts_qual_mask)
 };
 
-static inline unsigned long bts_get(const char *base, enum bts_field field)
+static inline unsigned long bts_get(const char *base, unsigned long field)
 {
 	base += (ds_cfg.sizeof_ptr_field * field);
 	return *(unsigned long *)base;
 }
 
-static inline void bts_set(char *base, enum bts_field field, unsigned long val)
+static inline void bts_set(char *base, unsigned long field, unsigned long val)
 {
-	base += (ds_cfg.sizeof_ptr_field * field);;
+	base += (ds_cfg.sizeof_ptr_field * field);
 	(*(unsigned long *)base) = val;
 }
 
-- 
cgit v1.2.3


From c9944881acf02b6f25fa62a0441a98b7dc0d7ae6 Mon Sep 17 00:00:00 2001
From: Roland Dreier <rolandd@cisco.com>
Date: Wed, 24 Jun 2009 13:42:40 +0800
Subject: crypto: aes-ni - Don't print message with KERN_ERR on old system

When the aes-intel module is loaded on a system that does not have the
AES instructions, it prints

    Intel AES-NI instructions are not detected.

at level KERN_ERR.  Since aes-intel is aliased to "aes" it will be tried
whenever anything uses AES and spam the console.  This doesn't match
existing practice for how to handle "no hardware" when initializing a
module, so downgrade the message to KERN_INFO.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 arch/x86/crypto/aesni-intel_glue.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/crypto/aesni-intel_glue.c b/arch/x86/crypto/aesni-intel_glue.c
index c580c5ec1cad..d3ec8d588d4b 100644
--- a/arch/x86/crypto/aesni-intel_glue.c
+++ b/arch/x86/crypto/aesni-intel_glue.c
@@ -636,7 +636,7 @@ static int __init aesni_init(void)
 	int err;
 
 	if (!cpu_has_aes) {
-		printk(KERN_ERR "Intel AES-NI instructions are not detected.\n");
+		printk(KERN_INFO "Intel AES-NI instructions are not detected.\n");
 		return -ENODEV;
 	}
 	if ((err = crypto_register_alg(&aesni_alg)))
-- 
cgit v1.2.3


From 2495fbf7effa6868f5d74124ae9b22a57980755b Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Fri, 26 Jun 2009 10:53:57 -0700
Subject: x86, setup: remove obsolete pre-Kconfig CONFIG_VIDEO_ variables

There were a set of pre-Kconfig configuration variables defined in the
video code.  There is absolutely no evidence that they have been
tweaked by anybody in modern history, so just get rid of them and hope
nobody notices.  If someone does complain, these should be made real
Kconfig variables.

Reported-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/boot/video-vesa.c |  7 +------
 arch/x86/boot/video-vga.c  | 10 ----------
 arch/x86/boot/video.c      |  5 -----
 arch/x86/boot/video.h      | 20 ++------------------
 4 files changed, 3 insertions(+), 39 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/boot/video-vesa.c b/arch/x86/boot/video-vesa.c
index c700147d6ffb..d7ef26ba4540 100644
--- a/arch/x86/boot/video-vesa.c
+++ b/arch/x86/boot/video-vesa.c
@@ -31,7 +31,6 @@ static inline void vesa_store_mode_params_graphics(void) {}
 
 static int vesa_probe(void)
 {
-#if defined(CONFIG_VIDEO_VESA) || defined(CONFIG_FIRMWARE_EDID)
 	struct biosregs ireg, oreg;
 	u16 mode;
 	addr_t mode_ptr;
@@ -49,8 +48,7 @@ static int vesa_probe(void)
 	    vginfo.signature != VESA_MAGIC ||
 	    vginfo.version < 0x0102)
 		return 0;	/* Not present */
-#endif /* CONFIG_VIDEO_VESA || CONFIG_FIRMWARE_EDID */
-#ifdef CONFIG_VIDEO_VESA
+
 	set_fs(vginfo.video_mode_ptr.seg);
 	mode_ptr = vginfo.video_mode_ptr.off;
 
@@ -102,9 +100,6 @@ static int vesa_probe(void)
 	}
 
 	return nmodes;
-#else
-	return 0;
-#endif /* CONFIG_VIDEO_VESA */
 }
 
 static int vesa_set_mode(struct mode_info *mode)
diff --git a/arch/x86/boot/video-vga.c b/arch/x86/boot/video-vga.c
index 8f8d827e254d..819caa1f2008 100644
--- a/arch/x86/boot/video-vga.c
+++ b/arch/x86/boot/video-vga.c
@@ -47,14 +47,6 @@ static u8 vga_set_basic_mode(void)
 
 	initregs(&ireg);
 
-#ifdef CONFIG_VIDEO_400_HACK
-	if (adapter >= ADAPTER_VGA) {
-		ireg.ax = 0x1202;
-		ireg.bx = 0x0030;
-		intcall(0x10, &ireg, NULL);
-	}
-#endif
-
 	ax = 0x0f00;
 	intcall(0x10, &ireg, &oreg);
 	mode = oreg.al;
@@ -62,11 +54,9 @@ static u8 vga_set_basic_mode(void)
 	set_fs(0);
 	rows = rdfs8(0x484);	/* rows minus one */
 
-#ifndef CONFIG_VIDEO_400_HACK
 	if ((oreg.ax == 0x5003 || oreg.ax == 0x5007) &&
 	    (rows == 0 || rows == 24))
 		return mode;
-#endif
 
 	if (mode != 3 && mode != 7)
 		mode = 3;
diff --git a/arch/x86/boot/video.c b/arch/x86/boot/video.c
index bad728b76fc2..d42da3802499 100644
--- a/arch/x86/boot/video.c
+++ b/arch/x86/boot/video.c
@@ -221,7 +221,6 @@ static unsigned int mode_menu(void)
 	}
 }
 
-#ifdef CONFIG_VIDEO_RETAIN
 /* Save screen content to the heap */
 static struct saved_screen {
 	int x, y;
@@ -299,10 +298,6 @@ static void restore_screen(void)
 	ireg.dl = saved.curx;
 	intcall(0x10, &ireg, NULL);
 }
-#else
-#define save_screen()		((void)0)
-#define restore_screen()	((void)0)
-#endif
 
 void set_video(void)
 {
diff --git a/arch/x86/boot/video.h b/arch/x86/boot/video.h
index 5bb174a997fc..ff339c5db311 100644
--- a/arch/x86/boot/video.h
+++ b/arch/x86/boot/video.h
@@ -17,19 +17,8 @@
 
 #include <linux/types.h>
 
-/* Enable autodetection of SVGA adapters and modes. */
-#undef CONFIG_VIDEO_SVGA
-
-/* Enable autodetection of VESA modes */
-#define CONFIG_VIDEO_VESA
-
-/* Retain screen contents when switching modes */
-#define CONFIG_VIDEO_RETAIN
-
-/* Force 400 scan lines for standard modes (hack to fix bad BIOS behaviour */
-#undef CONFIG_VIDEO_400_HACK
-
-/* This code uses an extended set of video mode numbers. These include:
+/*
+ * This code uses an extended set of video mode numbers. These include:
  * Aliases for standard modes
  *      NORMAL_VGA (-1)
  *      EXTENDED_VGA (-2)
@@ -67,13 +56,8 @@
 /* The "recalculate timings" flag */
 #define VIDEO_RECALC 0x8000
 
-/* Define DO_STORE according to CONFIG_VIDEO_RETAIN */
-#ifdef CONFIG_VIDEO_RETAIN
 void store_screen(void);
 #define DO_STORE() store_screen()
-#else
-#define DO_STORE() ((void)0)
-#endif /* CONFIG_VIDEO_RETAIN */
 
 /*
  * Mode table structures
-- 
cgit v1.2.3


From 087975b06b00af9bf888fab6f94ae113c5cd80bd Mon Sep 17 00:00:00 2001
From: Akinobu Mita <akinobu.mita@gmail.com>
Date: Sat, 27 Jun 2009 15:35:15 +0900
Subject: x86: Clean up dump_pagetable()

Use pgtable access helpers for 32-bit version dump_pagetable()
and get rid of __typeof__() operators. This needs to make
pmd_pfn() available for 2-level pgtable.

Also, remove some casts for 64-bit version dump_pagetable().

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
LKML-Reference: <20090627063514.GA2834@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/include/asm/pgtable.h | 10 ++++-----
 arch/x86/mm/fault.c            | 51 +++++++++++++++++-------------------------
 2 files changed, 26 insertions(+), 35 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h
index 3cc06e3fceb8..af5481e94863 100644
--- a/arch/x86/include/asm/pgtable.h
+++ b/arch/x86/include/asm/pgtable.h
@@ -134,6 +134,11 @@ static inline unsigned long pte_pfn(pte_t pte)
 	return (pte_val(pte) & PTE_PFN_MASK) >> PAGE_SHIFT;
 }
 
+static inline unsigned long pmd_pfn(pmd_t pmd)
+{
+	return (pmd_val(pmd) & PTE_PFN_MASK) >> PAGE_SHIFT;
+}
+
 #define pte_page(pte)	pfn_to_page(pte_pfn(pte))
 
 static inline int pmd_large(pmd_t pte)
@@ -422,11 +427,6 @@ static inline pmd_t *pmd_offset(pud_t *pud, unsigned long address)
 	return (pmd_t *)pud_page_vaddr(*pud) + pmd_index(address);
 }
 
-static inline unsigned long pmd_pfn(pmd_t pmd)
-{
-	return (pmd_val(pmd) & PTE_PFN_MASK) >> PAGE_SHIFT;
-}
-
 static inline int pud_large(pud_t pud)
 {
 	return (pud_val(pud) & (_PAGE_PSE | _PAGE_PRESENT)) ==
diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c
index 78a5fff857be..9bf7e52c2869 100644
--- a/arch/x86/mm/fault.c
+++ b/arch/x86/mm/fault.c
@@ -285,26 +285,25 @@ check_v8086_mode(struct pt_regs *regs, unsigned long address,
 		tsk->thread.screen_bitmap |= 1 << bit;
 }
 
-static void dump_pagetable(unsigned long address)
+static bool low_pfn(unsigned long pfn)
 {
-	__typeof__(pte_val(__pte(0))) page;
+	return pfn < max_low_pfn;
+}
 
-	page = read_cr3();
-	page = ((__typeof__(page) *) __va(page))[address >> PGDIR_SHIFT];
+static void dump_pagetable(unsigned long address)
+{
+	pgd_t *base = __va(read_cr3());
+	pgd_t *pgd = &base[pgd_index(address)];
+	pmd_t *pmd;
+	pte_t *pte;
 
 #ifdef CONFIG_X86_PAE
-	printk("*pdpt = %016Lx ", page);
-	if ((page >> PAGE_SHIFT) < max_low_pfn
-	    && page & _PAGE_PRESENT) {
-		page &= PAGE_MASK;
-		page = ((__typeof__(page) *) __va(page))[(address >> PMD_SHIFT)
-							& (PTRS_PER_PMD - 1)];
-		printk(KERN_CONT "*pde = %016Lx ", page);
-		page &= ~_PAGE_NX;
-	}
-#else
-	printk("*pde = %08lx ", page);
+	printk("*pdpt = %016Lx ", pgd_val(*pgd));
+	if (!low_pfn(pgd_val(*pgd) >> PAGE_SHIFT) || !pgd_present(*pgd))
+		goto out;
 #endif
+	pmd = pmd_offset(pud_offset(pgd, address), address);
+	printk(KERN_CONT "*pde = %0*Lx ", sizeof(*pmd) * 2, (u64)pmd_val(*pmd));
 
 	/*
 	 * We must not directly access the pte in the highpte
@@ -312,16 +311,12 @@ static void dump_pagetable(unsigned long address)
 	 * And let's rather not kmap-atomic the pte, just in case
 	 * it's allocated already:
 	 */
-	if ((page >> PAGE_SHIFT) < max_low_pfn
-	    && (page & _PAGE_PRESENT)
-	    && !(page & _PAGE_PSE)) {
-
-		page &= PAGE_MASK;
-		page = ((__typeof__(page) *) __va(page))[(address >> PAGE_SHIFT)
-							& (PTRS_PER_PTE - 1)];
-		printk("*pte = %0*Lx ", sizeof(page)*2, (u64)page);
-	}
+	if (!low_pfn(pmd_pfn(*pmd)) || !pmd_present(*pmd) || pmd_large(*pmd))
+		goto out;
 
+	pte = pte_offset_kernel(pmd, address);
+	printk("*pte = %0*Lx ", sizeof(*pte) * 2, (u64)pte_val(*pte));
+out:
 	printk("\n");
 }
 
@@ -449,16 +444,12 @@ static int bad_address(void *p)
 
 static void dump_pagetable(unsigned long address)
 {
-	pgd_t *pgd;
+	pgd_t *base = __va(read_cr3() & PHYSICAL_PAGE_MASK);
+	pgd_t *pgd = base + pgd_index(address);
 	pud_t *pud;
 	pmd_t *pmd;
 	pte_t *pte;
 
-	pgd = (pgd_t *)read_cr3();
-
-	pgd = __va((unsigned long)pgd & PHYSICAL_PAGE_MASK);
-
-	pgd += pgd_index(address);
 	if (bad_address(pgd))
 		goto bad;
 
-- 
cgit v1.2.3


From ce0c0f9eec2f377055e8b23c6fa192202381e022 Mon Sep 17 00:00:00 2001
From: "Figo.zhang" <figo1802@gmail.com>
Date: Sun, 28 Jun 2009 18:07:39 +0800
Subject: x86, pgtable.h: Clean up types

Use "unsigned long" consistently, not "unsigned".

Signed-off-by: Figo.zhang <figo1802@gmail.com>
LKML-Reference: <1246183659.2530.4.camel@myhost>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/include/asm/pgtable.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h
index af5481e94863..9de8729c1c8f 100644
--- a/arch/x86/include/asm/pgtable.h
+++ b/arch/x86/include/asm/pgtable.h
@@ -356,7 +356,7 @@ static inline unsigned long pmd_page_vaddr(pmd_t pmd)
  * this macro returns the index of the entry in the pmd page which would
  * control the given virtual address
  */
-static inline unsigned pmd_index(unsigned long address)
+static inline unsigned long pmd_index(unsigned long address)
 {
 	return (address >> PMD_SHIFT) & (PTRS_PER_PMD - 1);
 }
@@ -376,7 +376,7 @@ static inline unsigned pmd_index(unsigned long address)
  * this function returns the index of the entry in the pte page which would
  * control the given virtual address
  */
-static inline unsigned pte_index(unsigned long address)
+static inline unsigned long pte_index(unsigned long address)
 {
 	return (address >> PAGE_SHIFT) & (PTRS_PER_PTE - 1);
 }
@@ -462,7 +462,7 @@ static inline unsigned long pgd_page_vaddr(pgd_t pgd)
 #define pgd_page(pgd)		pfn_to_page(pgd_val(pgd) >> PAGE_SHIFT)
 
 /* to find an entry in a page-table-directory. */
-static inline unsigned pud_index(unsigned long address)
+static inline unsigned long pud_index(unsigned long address)
 {
 	return (address >> PUD_SHIFT) & (PTRS_PER_PUD - 1);
 }
-- 
cgit v1.2.3


From 565b0c1f100408ccbcb04ba458a14da454cb271d Mon Sep 17 00:00:00 2001
From: "Figo.zhang" <figo1802@gmail.com>
Date: Mon, 29 Jun 2009 12:02:55 +0800
Subject: x86, highmem_32.c: Clean up comment

Signed-off-by: Figo.zhang <figo1802@gmail.com>
Cc: Andrew Morton <akpm@osdl.org>
LKML-Reference: <1246248175.5759.12.camel@myhost>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/highmem_32.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/mm/highmem_32.c b/arch/x86/mm/highmem_32.c
index 58f621e81919..0c6f43cee25d 100644
--- a/arch/x86/mm/highmem_32.c
+++ b/arch/x86/mm/highmem_32.c
@@ -24,7 +24,7 @@ void kunmap(struct page *page)
  * no global lock is needed and because the kmap code must perform a global TLB
  * invalidation when the kmap pool wraps.
  *
- * However when holding an atomic kmap is is not legal to sleep, so atomic
+ * However when holding an atomic kmap it is not legal to sleep, so atomic
  * kmaps are appropriate for short, tight code paths only.
  */
 void *kmap_atomic_prot(struct page *page, enum km_type type, pgprot_t prot)
-- 
cgit v1.2.3


From 42204455f160dab0c47f19e1be23f5c927af2d17 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Sat, 4 Jul 2009 07:50:00 +0530
Subject: x86: Clean up mtrr/amd.c:

Fix trivial style problems :

  ERROR: trailing whitespace
  WARNING: line over 80 characters
  ERROR: do not use C99 // comments

arch/x86/kernel/cpu/mtrr/amd.o:

   text	   data	    bss	    dec	    hex	filename
    501	     32	      0	    533	    215	amd.o.before
    501	     32	      0	    533	    215	amd.o.after

md5:
   62f795eb840ee2d17b03df89e789e76c  amd.o.before.asm
   62f795eb840ee2d17b03df89e789e76c  amd.o.after.asm

Suggested-by: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <20090703164225.GA21447@elte.hu>
[ Also restructured comments to be standard, removed stray return,
  converted function description to DocBook style, etc. ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/mtrr/amd.c | 97 ++++++++++++++++++++++--------------------
 1 file changed, 51 insertions(+), 46 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/mtrr/amd.c b/arch/x86/kernel/cpu/mtrr/amd.c
index ee2331b0e58f..33af14110dfd 100644
--- a/arch/x86/kernel/cpu/mtrr/amd.c
+++ b/arch/x86/kernel/cpu/mtrr/amd.c
@@ -7,15 +7,15 @@
 
 static void
 amd_get_mtrr(unsigned int reg, unsigned long *base,
-	     unsigned long *size, mtrr_type * type)
+	     unsigned long *size, mtrr_type *type)
 {
 	unsigned long low, high;
 
 	rdmsr(MSR_K6_UWCCR, low, high);
-	/*  Upper dword is region 1, lower is region 0  */
+	/* Upper dword is region 1, lower is region 0 */
 	if (reg == 1)
 		low = high;
-	/*  The base masks off on the right alignment  */
+	/* The base masks off on the right alignment */
 	*base = (low & 0xFFFE0000) >> PAGE_SHIFT;
 	*type = 0;
 	if (low & 1)
@@ -27,74 +27,81 @@ amd_get_mtrr(unsigned int reg, unsigned long *base,
 		return;
 	}
 	/*
-	 *  This needs a little explaining. The size is stored as an
-	 *  inverted mask of bits of 128K granularity 15 bits long offset
-	 *  2 bits
+	 * This needs a little explaining. The size is stored as an
+	 * inverted mask of bits of 128K granularity 15 bits long offset
+	 * 2 bits.
 	 *
-	 *  So to get a size we do invert the mask and add 1 to the lowest
-	 *  mask bit (4 as its 2 bits in). This gives us a size we then shift
-	 *  to turn into 128K blocks
+	 * So to get a size we do invert the mask and add 1 to the lowest
+	 * mask bit (4 as its 2 bits in). This gives us a size we then shift
+	 * to turn into 128K blocks.
 	 *
-	 *  eg              111 1111 1111 1100      is 512K
+	 * eg              111 1111 1111 1100      is 512K
 	 *
-	 *  invert          000 0000 0000 0011
-	 *  +1              000 0000 0000 0100
-	 *  *128K   ...
+	 * invert          000 0000 0000 0011
+	 * +1              000 0000 0000 0100
+	 * *128K   ...
 	 */
 	low = (~low) & 0x1FFFC;
 	*size = (low + 4) << (15 - PAGE_SHIFT);
-	return;
 }
 
-static void amd_set_mtrr(unsigned int reg, unsigned long base,
-			 unsigned long size, mtrr_type type)
-/*  [SUMMARY] Set variable MTRR register on the local CPU.
-    <reg> The register to set.
-    <base> The base address of the region.
-    <size> The size of the region. If this is 0 the region is disabled.
-    <type> The type of the region.
-    [RETURNS] Nothing.
-*/
+/**
+ * amd_set_mtrr - Set variable MTRR register on the local CPU.
+ *
+ * @reg The register to set.
+ * @base The base address of the region.
+ * @size The size of the region. If this is 0 the region is disabled.
+ * @type The type of the region.
+ *
+ * Returns nothing.
+ */
+static void
+amd_set_mtrr(unsigned int reg, unsigned long base, unsigned long size, mtrr_type type)
 {
 	u32 regs[2];
 
 	/*
-	 *  Low is MTRR0 , High MTRR 1
+	 * Low is MTRR0, High MTRR 1
 	 */
 	rdmsr(MSR_K6_UWCCR, regs[0], regs[1]);
 	/*
-	 *  Blank to disable
+	 * Blank to disable
 	 */
-	if (size == 0)
+	if (size == 0) {
 		regs[reg] = 0;
-	else
-		/* Set the register to the base, the type (off by one) and an
-		   inverted bitmask of the size The size is the only odd
-		   bit. We are fed say 512K We invert this and we get 111 1111
-		   1111 1011 but if you subtract one and invert you get the   
-		   desired 111 1111 1111 1100 mask
-
-		   But ~(x - 1) == ~x + 1 == -x. Two's complement rocks!  */
+	} else {
+		/*
+		 * Set the register to the base, the type (off by one) and an
+		 * inverted bitmask of the size The size is the only odd
+		 * bit. We are fed say 512K We invert this and we get 111 1111
+		 * 1111 1011 but if you subtract one and invert you get the
+		 * desired 111 1111 1111 1100 mask
+		 *
+		 *  But ~(x - 1) == ~x + 1 == -x. Two's complement rocks!
+		 */
 		regs[reg] = (-size >> (15 - PAGE_SHIFT) & 0x0001FFFC)
 		    | (base << PAGE_SHIFT) | (type + 1);
+	}
 
 	/*
-	 *  The writeback rule is quite specific. See the manual. Its
-	 *  disable local interrupts, write back the cache, set the mtrr
+	 * The writeback rule is quite specific. See the manual. Its
+	 * disable local interrupts, write back the cache, set the mtrr
 	 */
 	wbinvd();
 	wrmsr(MSR_K6_UWCCR, regs[0], regs[1]);
 }
 
-static int amd_validate_add_page(unsigned long base, unsigned long size, unsigned int type)
+static int
+amd_validate_add_page(unsigned long base, unsigned long size, unsigned int type)
 {
-	/* Apply the K6 block alignment and size rules
-	   In order
-	   o Uncached or gathering only
-	   o 128K or bigger block
-	   o Power of 2 block
-	   o base suitably aligned to the power
-	*/
+	/*
+	 * Apply the K6 block alignment and size rules
+	 * In order
+	 * o Uncached or gathering only
+	 * o 128K or bigger block
+	 * o Power of 2 block
+	 * o base suitably aligned to the power
+	 */
 	if (type > MTRR_TYPE_WRCOMB || size < (1 << (17 - PAGE_SHIFT))
 	    || (size & ~(size - 1)) - size || (base & (size - 1)))
 		return -EINVAL;
@@ -115,5 +122,3 @@ int __init amd_init_mtrr(void)
 	set_mtrr_ops(&amd_mtrr_ops);
 	return 0;
 }
-
-//arch_initcall(amd_mtrr_init);
-- 
cgit v1.2.3


From 6c4caa1ab737502190e416b76e6c10d2bf24276a Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Sat, 4 Jul 2009 07:50:44 +0530
Subject: x86: Clean up mtrr/centaur.c

Remove dead code and fix trivial style problems:

  ERROR: trailing whitespace X 2
  WARNING: line over 80 characters X 3
  ROR: trailing whitespace
  ERROR: do not use C99 // comments X 2

arch/x86/kernel/cpu/mtrr/centaur.o:

   text	   data	    bss	    dec	    hex	filename
    605	     32	     68	    705	    2c1	centaur.o.before
    605	     32	     68	    705	    2c1	centaur.o.after

md5:
   a4865ea98ce3c163bb1d376a3949b3e3  centaur.o.before.asm
   a4865ea98ce3c163bb1d376a3949b3e3  centaur.o.after.asm

Suggested-by: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <20090703164225.GA21447@elte.hu>
[ Standardized comments, DocBook, curly braces, newlines. ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/mtrr/centaur.c | 168 ++++++++-----------------------------
 1 file changed, 35 insertions(+), 133 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/mtrr/centaur.c b/arch/x86/kernel/cpu/mtrr/centaur.c
index cb9aa3a7a7ab..de89f14eff3a 100644
--- a/arch/x86/kernel/cpu/mtrr/centaur.c
+++ b/arch/x86/kernel/cpu/mtrr/centaur.c
@@ -1,7 +1,9 @@
 #include <linux/init.h>
 #include <linux/mm.h>
+
 #include <asm/mtrr.h>
 #include <asm/msr.h>
+
 #include "mtrr.h"
 
 static struct {
@@ -12,25 +14,25 @@ static struct {
 static u8 centaur_mcr_reserved;
 static u8 centaur_mcr_type;	/* 0 for winchip, 1 for winchip2 */
 
-/*
- *	Report boot time MCR setups 
+/**
+ * centaur_get_free_region - Get a free MTRR.
+ *
+ * @base: The starting (base) address of the region.
+ * @size: The size (in bytes) of the region.
+ *
+ * Returns: the index of the region on success, else -1 on error.
  */
-
 static int
 centaur_get_free_region(unsigned long base, unsigned long size, int replace_reg)
-/*  [SUMMARY] Get a free MTRR.
-    <base> The starting (base) address of the region.
-    <size> The size (in bytes) of the region.
-    [RETURNS] The index of the region on success, else -1 on error.
-*/
 {
-	int i, max;
-	mtrr_type ltype;
 	unsigned long lbase, lsize;
+	mtrr_type ltype;
+	int i, max;
 
 	max = num_var_ranges;
 	if (replace_reg >= 0 && replace_reg < max)
 		return replace_reg;
+
 	for (i = 0; i < max; ++i) {
 		if (centaur_mcr_reserved & (1 << i))
 			continue;
@@ -38,11 +40,14 @@ centaur_get_free_region(unsigned long base, unsigned long size, int replace_reg)
 		if (lsize == 0)
 			return i;
 	}
+
 	return -ENOSPC;
 }
 
-void
-mtrr_centaur_report_mcr(int mcr, u32 lo, u32 hi)
+/*
+ * Report boot time MCR setups
+ */
+void mtrr_centaur_report_mcr(int mcr, u32 lo, u32 hi)
 {
 	centaur_mcr[mcr].low = lo;
 	centaur_mcr[mcr].high = hi;
@@ -54,33 +59,35 @@ centaur_get_mcr(unsigned int reg, unsigned long *base,
 {
 	*base = centaur_mcr[reg].high >> PAGE_SHIFT;
 	*size = -(centaur_mcr[reg].low & 0xfffff000) >> PAGE_SHIFT;
-	*type = MTRR_TYPE_WRCOMB;	/*  If it is there, it is write-combining  */
+	*type = MTRR_TYPE_WRCOMB;		/* write-combining  */
+
 	if (centaur_mcr_type == 1 && ((centaur_mcr[reg].low & 31) & 2))
 		*type = MTRR_TYPE_UNCACHABLE;
 	if (centaur_mcr_type == 1 && (centaur_mcr[reg].low & 31) == 25)
 		*type = MTRR_TYPE_WRBACK;
 	if (centaur_mcr_type == 0 && (centaur_mcr[reg].low & 31) == 31)
 		*type = MTRR_TYPE_WRBACK;
-
 }
 
-static void centaur_set_mcr(unsigned int reg, unsigned long base,
-			    unsigned long size, mtrr_type type)
+static void
+centaur_set_mcr(unsigned int reg, unsigned long base,
+		unsigned long size, mtrr_type type)
 {
 	unsigned long low, high;
 
 	if (size == 0) {
-		/*  Disable  */
+		/* Disable */
 		high = low = 0;
 	} else {
 		high = base << PAGE_SHIFT;
-		if (centaur_mcr_type == 0)
-			low = -size << PAGE_SHIFT | 0x1f;	/* only support write-combining... */
-		else {
+		if (centaur_mcr_type == 0) {
+			/* Only support write-combining... */
+			low = -size << PAGE_SHIFT | 0x1f;
+		} else {
 			if (type == MTRR_TYPE_UNCACHABLE)
-				low = -size << PAGE_SHIFT | 0x02;	/* NC */
+				low = -size << PAGE_SHIFT | 0x02; /* NC */
 			else
-				low = -size << PAGE_SHIFT | 0x09;	/* WWO,WC */
+				low = -size << PAGE_SHIFT | 0x09; /* WWO, WC */
 		}
 	}
 	centaur_mcr[reg].high = high;
@@ -88,118 +95,16 @@ static void centaur_set_mcr(unsigned int reg, unsigned long base,
 	wrmsr(MSR_IDT_MCR0 + reg, low, high);
 }
 
-#if 0
-/*
- *	Initialise the later (saner) Winchip MCR variant. In this version
- *	the BIOS can pass us the registers it has used (but not their values)
- *	and the control register is read/write
- */
-
-static void __init
-centaur_mcr1_init(void)
-{
-	unsigned i;
-	u32 lo, hi;
-
-	/* Unfortunately, MCR's are read-only, so there is no way to
-	 * find out what the bios might have done.
-	 */
-
-	rdmsr(MSR_IDT_MCR_CTRL, lo, hi);
-	if (((lo >> 17) & 7) == 1) {	/* Type 1 Winchip2 MCR */
-		lo &= ~0x1C0;	/* clear key */
-		lo |= 0x040;	/* set key to 1 */
-		wrmsr(MSR_IDT_MCR_CTRL, lo, hi);	/* unlock MCR */
-	}
-
-	centaur_mcr_type = 1;
-
-	/*
-	 *  Clear any unconfigured MCR's.
-	 */
-
-	for (i = 0; i < 8; ++i) {
-		if (centaur_mcr[i].high == 0 && centaur_mcr[i].low == 0) {
-			if (!(lo & (1 << (9 + i))))
-				wrmsr(MSR_IDT_MCR0 + i, 0, 0);
-			else
-				/*
-				 *      If the BIOS set up an MCR we cannot see it
-				 *      but we don't wish to obliterate it
-				 */
-				centaur_mcr_reserved |= (1 << i);
-		}
-	}
-	/*  
-	 *  Throw the main write-combining switch... 
-	 *  However if OOSTORE is enabled then people have already done far
-	 *  cleverer things and we should behave. 
-	 */
-
-	lo |= 15;		/* Write combine enables */
-	wrmsr(MSR_IDT_MCR_CTRL, lo, hi);
-}
-
-/*
- *	Initialise the original winchip with read only MCR registers
- *	no used bitmask for the BIOS to pass on and write only control
- */
-
-static void __init
-centaur_mcr0_init(void)
-{
-	unsigned i;
-
-	/* Unfortunately, MCR's are read-only, so there is no way to
-	 * find out what the bios might have done.
-	 */
-
-	/* Clear any unconfigured MCR's.
-	 * This way we are sure that the centaur_mcr array contains the actual
-	 * values. The disadvantage is that any BIOS tweaks are thus undone.
-	 *
-	 */
-	for (i = 0; i < 8; ++i) {
-		if (centaur_mcr[i].high == 0 && centaur_mcr[i].low == 0)
-			wrmsr(MSR_IDT_MCR0 + i, 0, 0);
-	}
-
-	wrmsr(MSR_IDT_MCR_CTRL, 0x01F0001F, 0);	/* Write only */
-}
-
-/*
- *	Initialise Winchip series MCR registers
- */
-
-static void __init
-centaur_mcr_init(void)
-{
-	struct set_mtrr_context ctxt;
-
-	set_mtrr_prepare_save(&ctxt);
-	set_mtrr_cache_disable(&ctxt);
-
-	if (boot_cpu_data.x86_model == 4)
-		centaur_mcr0_init();
-	else if (boot_cpu_data.x86_model == 8 || boot_cpu_data.x86_model == 9)
-		centaur_mcr1_init();
-
-	set_mtrr_done(&ctxt);
-}
-#endif
-
-static int centaur_validate_add_page(unsigned long base, 
-				     unsigned long size, unsigned int type)
+static int
+centaur_validate_add_page(unsigned long base, unsigned long size, unsigned int type)
 {
 	/*
-	 *  FIXME: Winchip2 supports uncached
+	 * FIXME: Winchip2 supports uncached
 	 */
-	if (type != MTRR_TYPE_WRCOMB && 
+	if (type != MTRR_TYPE_WRCOMB &&
 	    (centaur_mcr_type == 0 || type != MTRR_TYPE_UNCACHABLE)) {
-		printk(KERN_WARNING
-		       "mtrr: only write-combining%s supported\n",
-		       centaur_mcr_type ? " and uncacheable are"
-		       : " is");
+		pr_warning("mtrr: only write-combining%s supported\n",
+			   centaur_mcr_type ? " and uncacheable are" : " is");
 		return -EINVAL;
 	}
 	return 0;
@@ -207,7 +112,6 @@ static int centaur_validate_add_page(unsigned long base,
 
 static struct mtrr_ops centaur_mtrr_ops = {
 	.vendor            = X86_VENDOR_CENTAUR,
-//	.init              = centaur_mcr_init,
 	.set               = centaur_set_mcr,
 	.get               = centaur_get_mcr,
 	.get_free_region   = centaur_get_free_region,
@@ -220,5 +124,3 @@ int __init centaur_init_mtrr(void)
 	set_mtrr_ops(&centaur_mtrr_ops);
 	return 0;
 }
-
-//arch_initcall(centaur_init_mtrr);
-- 
cgit v1.2.3


From 63f9600fadb10ea739108ae93e3e842d9843c58b Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Sat, 4 Jul 2009 07:51:32 +0530
Subject: x86: Clean up mtrr/cleanup.c

Fix trivial style problems:

  WARNING: Use #include <linux/uaccess.h> instead of <asm/uaccess.h>
  WARNING: Use #include <linux/kvm_para.h> instead of <asm/kvm_para.h>

Also, nr_mtrr_spare_reg should be unsigned long.

arch/x86/kernel/cpu/mtrr/cleanup.o:

   text	   data	    bss	    dec	    hex	filename
   6241	   8992	   2056	  17289	   4389	cleanup.o.before
   6241	   8992	   2056	  17289	   4389	cleanup.o.after

The md5 has changed:
   1a7a27513aef1825236daf29110fe657  cleanup.o.before.asm
   bcea358efa2532b6020e338e158447af  cleanup.o.after.asm

Because a WARN_ON()'s __LINE__ value changed by 3 lines.

Suggested-by: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <20090703164225.GA21447@elte.hu>
[ Did lots of other cleanups to make the code look more consistent. ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/mtrr/cleanup.c | 350 +++++++++++++++++++------------------
 1 file changed, 176 insertions(+), 174 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/mtrr/cleanup.c b/arch/x86/kernel/cpu/mtrr/cleanup.c
index 1d584a18a50d..b8aba811b60e 100644
--- a/arch/x86/kernel/cpu/mtrr/cleanup.c
+++ b/arch/x86/kernel/cpu/mtrr/cleanup.c
@@ -1,51 +1,52 @@
-/*  MTRR (Memory Type Range Register) cleanup
-
-    Copyright (C) 2009 Yinghai Lu
-
-    This library is free software; you can redistribute it and/or
-    modify it under the terms of the GNU Library General Public
-    License as published by the Free Software Foundation; either
-    version 2 of the License, or (at your option) any later version.
-
-    This library 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
-    Library General Public License for more details.
-
-    You should have received a copy of the GNU Library General Public
-    License along with this library; if not, write to the Free
-    Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
-*/
-
+/*
+ * MTRR (Memory Type Range Register) cleanup
+ *
+ *  Copyright (C) 2009 Yinghai Lu
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Library General Public
+ * License as published by the Free Software Foundation; either
+ * version 2 of the License, or (at your option) any later version.
+ *
+ * This library 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
+ * Library General Public License for more details.
+ *
+ * You should have received a copy of the GNU Library General Public
+ * License along with this library; if not, write to the Free
+ * Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
 #include <linux/module.h>
 #include <linux/init.h>
 #include <linux/pci.h>
 #include <linux/smp.h>
 #include <linux/cpu.h>
-#include <linux/mutex.h>
 #include <linux/sort.h>
+#include <linux/mutex.h>
+#include <linux/uaccess.h>
+#include <linux/kvm_para.h>
 
+#include <asm/processor.h>
 #include <asm/e820.h>
 #include <asm/mtrr.h>
-#include <asm/uaccess.h>
-#include <asm/processor.h>
 #include <asm/msr.h>
-#include <asm/kvm_para.h>
+
 #include "mtrr.h"
 
-/* should be related to MTRR_VAR_RANGES nums */
+/* Should be related to MTRR_VAR_RANGES nums */
 #define RANGE_NUM 256
 
 struct res_range {
-	unsigned long start;
-	unsigned long end;
+	unsigned long	start;
+	unsigned long	end;
 };
 
 static int __init
-add_range(struct res_range *range, int nr_range, unsigned long start,
-			      unsigned long end)
+add_range(struct res_range *range, int nr_range,
+	  unsigned long start, unsigned long end)
 {
-	/* out of slots */
+	/* Out of slots: */
 	if (nr_range >= RANGE_NUM)
 		return nr_range;
 
@@ -58,12 +59,12 @@ add_range(struct res_range *range, int nr_range, unsigned long start,
 }
 
 static int __init
-add_range_with_merge(struct res_range *range, int nr_range, unsigned long start,
-			      unsigned long end)
+add_range_with_merge(struct res_range *range, int nr_range,
+		     unsigned long start, unsigned long end)
 {
 	int i;
 
-	/* try to merge it with old one */
+	/* Try to merge it with old one: */
 	for (i = 0; i < nr_range; i++) {
 		unsigned long final_start, final_end;
 		unsigned long common_start, common_end;
@@ -84,7 +85,7 @@ add_range_with_merge(struct res_range *range, int nr_range, unsigned long start,
 		return nr_range;
 	}
 
-	/* need to add that */
+	/* Need to add it: */
 	return add_range(range, nr_range, start, end);
 }
 
@@ -117,7 +118,7 @@ subtract_range(struct res_range *range, unsigned long start, unsigned long end)
 		}
 
 		if (start > range[j].start && end < range[j].end) {
-			/* find the new spare */
+			/* Find the new spare: */
 			for (i = 0; i < RANGE_NUM; i++) {
 				if (range[i].end == 0)
 					break;
@@ -147,13 +148,19 @@ static int __init cmp_range(const void *x1, const void *x2)
 }
 
 struct var_mtrr_range_state {
-	unsigned long base_pfn;
-	unsigned long size_pfn;
-	mtrr_type type;
+	unsigned long	base_pfn;
+	unsigned long	size_pfn;
+	mtrr_type	type;
 };
 
 static struct var_mtrr_range_state __initdata range_state[RANGE_NUM];
+
 static int __initdata debug_print;
+#define Dprintk(x...) do { if (debug_print) printk(KERN_DEBUG x); } while (0)
+
+
+#define BIOS_BUG_MSG KERN_WARNING \
+	"WARNING: BIOS bug: VAR MTRR %d contains strange UC entry under 1M, check with your system vendor!\n"
 
 static int __init
 x86_get_mtrr_mem_range(struct res_range *range, int nr_range,
@@ -180,7 +187,7 @@ x86_get_mtrr_mem_range(struct res_range *range, int nr_range,
 				 range[i].start, range[i].end + 1);
 	}
 
-	/* take out UC ranges */
+	/* Take out UC ranges: */
 	for (i = 0; i < num_var_ranges; i++) {
 		type = range_state[i].type;
 		if (type != MTRR_TYPE_UNCACHABLE &&
@@ -244,10 +251,9 @@ static int __initdata nr_range;
 
 static unsigned long __init sum_ranges(struct res_range *range, int nr_range)
 {
-	unsigned long sum;
+	unsigned long sum = 0;
 	int i;
 
-	sum = 0;
 	for (i = 0; i < nr_range; i++)
 		sum += range[i].end + 1 - range[i].start;
 
@@ -288,7 +294,7 @@ struct var_mtrr_state {
 
 static void __init
 set_var_mtrr(unsigned int reg, unsigned long basek, unsigned long sizek,
-		unsigned char type, unsigned int address_bits)
+	     unsigned char type, unsigned int address_bits)
 {
 	u32 base_lo, base_hi, mask_lo, mask_hi;
 	u64 base, mask;
@@ -301,7 +307,7 @@ set_var_mtrr(unsigned int reg, unsigned long basek, unsigned long sizek,
 	mask = (1ULL << address_bits) - 1;
 	mask &= ~((((u64)sizek) << 10) - 1);
 
-	base  = ((u64)basek) << 10;
+	base = ((u64)basek) << 10;
 
 	base |= type;
 	mask |= 0x800;
@@ -317,15 +323,14 @@ set_var_mtrr(unsigned int reg, unsigned long basek, unsigned long sizek,
 
 static void __init
 save_var_mtrr(unsigned int reg, unsigned long basek, unsigned long sizek,
-		unsigned char type)
+	      unsigned char type)
 {
 	range_state[reg].base_pfn = basek >> (PAGE_SHIFT - 10);
 	range_state[reg].size_pfn = sizek >> (PAGE_SHIFT - 10);
 	range_state[reg].type = type;
 }
 
-static void __init
-set_var_mtrr_all(unsigned int address_bits)
+static void __init set_var_mtrr_all(unsigned int address_bits)
 {
 	unsigned long basek, sizek;
 	unsigned char type;
@@ -342,11 +347,11 @@ set_var_mtrr_all(unsigned int address_bits)
 
 static unsigned long to_size_factor(unsigned long sizek, char *factorp)
 {
-	char factor;
 	unsigned long base = sizek;
+	char factor;
 
 	if (base & ((1<<10) - 1)) {
-		/* not MB alignment */
+		/* Not MB-aligned: */
 		factor = 'K';
 	} else if (base & ((1<<20) - 1)) {
 		factor = 'M';
@@ -372,11 +377,12 @@ range_to_mtrr(unsigned int reg, unsigned long range_startk,
 		unsigned long max_align, align;
 		unsigned long sizek;
 
-		/* Compute the maximum size I can make a range */
+		/* Compute the maximum size with which we can make a range: */
 		if (range_startk)
 			max_align = ffs(range_startk) - 1;
 		else
 			max_align = 32;
+
 		align = fls(range_sizek) - 1;
 		if (align > max_align)
 			align = max_align;
@@ -386,11 +392,10 @@ range_to_mtrr(unsigned int reg, unsigned long range_startk,
 			char start_factor = 'K', size_factor = 'K';
 			unsigned long start_base, size_base;
 
-			start_base = to_size_factor(range_startk,
-							 &start_factor),
-			size_base = to_size_factor(sizek, &size_factor),
+			start_base = to_size_factor(range_startk, &start_factor);
+			size_base = to_size_factor(sizek, &size_factor);
 
-			printk(KERN_DEBUG "Setting variable MTRR %d, "
+			Dprintk("Setting variable MTRR %d, "
 				"base: %ld%cB, range: %ld%cB, type %s\n",
 				reg, start_base, start_factor,
 				size_base, size_factor,
@@ -425,10 +430,11 @@ range_to_mtrr_with_hole(struct var_mtrr_state *state, unsigned long basek,
 	chunk_sizek = state->chunk_sizek;
 	gran_sizek = state->gran_sizek;
 
-	/* align with gran size, prevent small block used up MTRRs */
+	/* Align with gran size, prevent small block used up MTRRs: */
 	range_basek = ALIGN(state->range_startk, gran_sizek);
 	if ((range_basek > basek) && basek)
 		return second_sizek;
+
 	state->range_sizek -= (range_basek - state->range_startk);
 	range_sizek = ALIGN(state->range_sizek, gran_sizek);
 
@@ -439,22 +445,21 @@ range_to_mtrr_with_hole(struct var_mtrr_state *state, unsigned long basek,
 	}
 	state->range_sizek = range_sizek;
 
-	/* try to append some small hole */
+	/* Try to append some small hole: */
 	range0_basek = state->range_startk;
 	range0_sizek = ALIGN(state->range_sizek, chunk_sizek);
 
-	/* no increase */
+	/* No increase: */
 	if (range0_sizek == state->range_sizek) {
-		if (debug_print)
-			printk(KERN_DEBUG "rangeX: %016lx - %016lx\n",
-				range0_basek<<10,
-				(range0_basek + state->range_sizek)<<10);
+		Dprintk("rangeX: %016lx - %016lx\n",
+			range0_basek<<10,
+			(range0_basek + state->range_sizek)<<10);
 		state->reg = range_to_mtrr(state->reg, range0_basek,
 				state->range_sizek, MTRR_TYPE_WRBACK);
 		return 0;
 	}
 
-	/* only cut back, when it is not the last */
+	/* Only cut back when it is not the last: */
 	if (sizek) {
 		while (range0_basek + range0_sizek > (basek + sizek)) {
 			if (range0_sizek >= chunk_sizek)
@@ -470,16 +475,16 @@ range_to_mtrr_with_hole(struct var_mtrr_state *state, unsigned long basek,
 second_try:
 	range_basek = range0_basek + range0_sizek;
 
-	/* one hole in the middle */
+	/* One hole in the middle: */
 	if (range_basek > basek && range_basek <= (basek + sizek))
 		second_sizek = range_basek - basek;
 
 	if (range0_sizek > state->range_sizek) {
 
-		/* one hole in middle or at end */
+		/* One hole in middle or at the end: */
 		hole_sizek = range0_sizek - state->range_sizek - second_sizek;
 
-		/* hole size should be less than half of range0 size */
+		/* Hole size should be less than half of range0 size: */
 		if (hole_sizek >= (range0_sizek >> 1) &&
 		    range0_sizek >= chunk_sizek) {
 			range0_sizek -= chunk_sizek;
@@ -491,32 +496,30 @@ second_try:
 	}
 
 	if (range0_sizek) {
-		if (debug_print)
-			printk(KERN_DEBUG "range0: %016lx - %016lx\n",
-				range0_basek<<10,
-				(range0_basek + range0_sizek)<<10);
+		Dprintk("range0: %016lx - %016lx\n",
+			range0_basek<<10,
+			(range0_basek + range0_sizek)<<10);
 		state->reg = range_to_mtrr(state->reg, range0_basek,
 				range0_sizek, MTRR_TYPE_WRBACK);
 	}
 
 	if (range0_sizek < state->range_sizek) {
-		/* need to handle left over */
+		/* Need to handle left over range: */
 		range_sizek = state->range_sizek - range0_sizek;
 
-		if (debug_print)
-			printk(KERN_DEBUG "range: %016lx - %016lx\n",
-				 range_basek<<10,
-				 (range_basek + range_sizek)<<10);
+		Dprintk("range: %016lx - %016lx\n",
+			 range_basek<<10,
+			 (range_basek + range_sizek)<<10);
+
 		state->reg = range_to_mtrr(state->reg, range_basek,
 				 range_sizek, MTRR_TYPE_WRBACK);
 	}
 
 	if (hole_sizek) {
 		hole_basek = range_basek - hole_sizek - second_sizek;
-		if (debug_print)
-			printk(KERN_DEBUG "hole: %016lx - %016lx\n",
-				 hole_basek<<10,
-				 (hole_basek + hole_sizek)<<10);
+		Dprintk("hole: %016lx - %016lx\n",
+			 hole_basek<<10,
+			 (hole_basek + hole_sizek)<<10);
 		state->reg = range_to_mtrr(state->reg, hole_basek,
 				 hole_sizek, MTRR_TYPE_UNCACHABLE);
 	}
@@ -537,23 +540,23 @@ set_var_mtrr_range(struct var_mtrr_state *state, unsigned long base_pfn,
 	basek = base_pfn << (PAGE_SHIFT - 10);
 	sizek = size_pfn << (PAGE_SHIFT - 10);
 
-	/* See if I can merge with the last range */
+	/* See if I can merge with the last range: */
 	if ((basek <= 1024) ||
 	    (state->range_startk + state->range_sizek == basek)) {
 		unsigned long endk = basek + sizek;
 		state->range_sizek = endk - state->range_startk;
 		return;
 	}
-	/* Write the range mtrrs */
+	/* Write the range mtrrs: */
 	if (state->range_sizek != 0)
 		second_sizek = range_to_mtrr_with_hole(state, basek, sizek);
 
-	/* Allocate an msr */
+	/* Allocate an msr: */
 	state->range_startk = basek + second_sizek;
 	state->range_sizek  = sizek - second_sizek;
 }
 
-/* mininum size of mtrr block that can take hole */
+/* Mininum size of mtrr block that can take hole: */
 static u64 mtrr_chunk_size __initdata = (256ULL<<20);
 
 static int __init parse_mtrr_chunk_size_opt(char *p)
@@ -565,7 +568,7 @@ static int __init parse_mtrr_chunk_size_opt(char *p)
 }
 early_param("mtrr_chunk_size", parse_mtrr_chunk_size_opt);
 
-/* granity of mtrr of block */
+/* Granularity of mtrr of block: */
 static u64 mtrr_gran_size __initdata;
 
 static int __init parse_mtrr_gran_size_opt(char *p)
@@ -577,7 +580,7 @@ static int __init parse_mtrr_gran_size_opt(char *p)
 }
 early_param("mtrr_gran_size", parse_mtrr_gran_size_opt);
 
-static int nr_mtrr_spare_reg __initdata =
+static unsigned long nr_mtrr_spare_reg __initdata =
 				 CONFIG_MTRR_SANITIZER_SPARE_REG_NR_DEFAULT;
 
 static int __init parse_mtrr_spare_reg(char *arg)
@@ -586,7 +589,6 @@ static int __init parse_mtrr_spare_reg(char *arg)
 		nr_mtrr_spare_reg = simple_strtoul(arg, NULL, 0);
 	return 0;
 }
-
 early_param("mtrr_spare_reg_nr", parse_mtrr_spare_reg);
 
 static int __init
@@ -594,8 +596,8 @@ x86_setup_var_mtrrs(struct res_range *range, int nr_range,
 		    u64 chunk_size, u64 gran_size)
 {
 	struct var_mtrr_state var_state;
-	int i;
 	int num_reg;
+	int i;
 
 	var_state.range_startk	= 0;
 	var_state.range_sizek	= 0;
@@ -605,17 +607,18 @@ x86_setup_var_mtrrs(struct res_range *range, int nr_range,
 
 	memset(range_state, 0, sizeof(range_state));
 
-	/* Write the range etc */
-	for (i = 0; i < nr_range; i++)
+	/* Write the range: */
+	for (i = 0; i < nr_range; i++) {
 		set_var_mtrr_range(&var_state, range[i].start,
 				   range[i].end - range[i].start + 1);
+	}
 
-	/* Write the last range */
+	/* Write the last range: */
 	if (var_state.range_sizek != 0)
 		range_to_mtrr_with_hole(&var_state, 0, 0);
 
 	num_reg = var_state.reg;
-	/* Clear out the extra MTRR's */
+	/* Clear out the extra MTRR's: */
 	while (var_state.reg < num_var_ranges) {
 		save_var_mtrr(var_state.reg, 0, 0, 0);
 		var_state.reg++;
@@ -625,11 +628,11 @@ x86_setup_var_mtrrs(struct res_range *range, int nr_range,
 }
 
 struct mtrr_cleanup_result {
-	unsigned long gran_sizek;
-	unsigned long chunk_sizek;
-	unsigned long lose_cover_sizek;
-	unsigned int num_reg;
-	int bad;
+	unsigned long	gran_sizek;
+	unsigned long	chunk_sizek;
+	unsigned long	lose_cover_sizek;
+	unsigned int	num_reg;
+	int		bad;
 };
 
 /*
@@ -645,10 +648,10 @@ static unsigned long __initdata min_loss_pfn[RANGE_NUM];
 
 static void __init print_out_mtrr_range_state(void)
 {
-	int i;
 	char start_factor = 'K', size_factor = 'K';
 	unsigned long start_base, size_base;
 	mtrr_type type;
+	int i;
 
 	for (i = 0; i < num_var_ranges; i++) {
 
@@ -676,10 +679,10 @@ static int __init mtrr_need_cleanup(void)
 	int i;
 	mtrr_type type;
 	unsigned long size;
-	/* extra one for all 0 */
+	/* Extra one for all 0: */
 	int num[MTRR_NUM_TYPES + 1];
 
-	/* check entries number */
+	/* Check entries number: */
 	memset(num, 0, sizeof(num));
 	for (i = 0; i < num_var_ranges; i++) {
 		type = range_state[i].type;
@@ -693,88 +696,86 @@ static int __init mtrr_need_cleanup(void)
 		num[type]++;
 	}
 
-	/* check if we got UC entries */
+	/* Check if we got UC entries: */
 	if (!num[MTRR_TYPE_UNCACHABLE])
 		return 0;
 
-	/* check if we only had WB and UC */
+	/* Check if we only had WB and UC */
 	if (num[MTRR_TYPE_WRBACK] + num[MTRR_TYPE_UNCACHABLE] !=
-		num_var_ranges - num[MTRR_NUM_TYPES])
+	    num_var_ranges - num[MTRR_NUM_TYPES])
 		return 0;
 
 	return 1;
 }
 
 static unsigned long __initdata range_sums;
-static void __init mtrr_calc_range_state(u64 chunk_size, u64 gran_size,
-					 unsigned long extra_remove_base,
-					 unsigned long extra_remove_size,
-					 int i)
+
+static void __init
+mtrr_calc_range_state(u64 chunk_size, u64 gran_size,
+		      unsigned long x_remove_base,
+		      unsigned long x_remove_size, int i)
 {
-	int num_reg;
 	static struct res_range range_new[RANGE_NUM];
-	static int nr_range_new;
 	unsigned long range_sums_new;
+	static int nr_range_new;
+	int num_reg;
 
-	/* convert ranges to var ranges state */
-	num_reg = x86_setup_var_mtrrs(range, nr_range,
-						chunk_size, gran_size);
+	/* Convert ranges to var ranges state: */
+	num_reg = x86_setup_var_mtrrs(range, nr_range, chunk_size, gran_size);
 
-	/* we got new setting in range_state, check it */
+	/* We got new setting in range_state, check it: */
 	memset(range_new, 0, sizeof(range_new));
 	nr_range_new = x86_get_mtrr_mem_range(range_new, 0,
-				extra_remove_base, extra_remove_size);
+				x_remove_base, x_remove_size);
 	range_sums_new = sum_ranges(range_new, nr_range_new);
 
 	result[i].chunk_sizek = chunk_size >> 10;
 	result[i].gran_sizek = gran_size >> 10;
 	result[i].num_reg = num_reg;
+
 	if (range_sums < range_sums_new) {
-		result[i].lose_cover_sizek =
-			(range_sums_new - range_sums) << PSHIFT;
+		result[i].lose_cover_sizek = (range_sums_new - range_sums) << PSHIFT;
 		result[i].bad = 1;
-	} else
-		result[i].lose_cover_sizek =
-			(range_sums - range_sums_new) << PSHIFT;
+	} else {
+		result[i].lose_cover_sizek = (range_sums - range_sums_new) << PSHIFT;
+	}
 
-	/* double check it */
+	/* Double check it: */
 	if (!result[i].bad && !result[i].lose_cover_sizek) {
-		if (nr_range_new != nr_range ||
-			memcmp(range, range_new, sizeof(range)))
-				result[i].bad = 1;
+		if (nr_range_new != nr_range || memcmp(range, range_new, sizeof(range)))
+			result[i].bad = 1;
 	}
 
-	if (!result[i].bad && (range_sums - range_sums_new <
-				min_loss_pfn[num_reg])) {
-		min_loss_pfn[num_reg] =
-			range_sums - range_sums_new;
-	}
+	if (!result[i].bad && (range_sums - range_sums_new < min_loss_pfn[num_reg]))
+		min_loss_pfn[num_reg] = range_sums - range_sums_new;
 }
 
 static void __init mtrr_print_out_one_result(int i)
 {
-	char gran_factor, chunk_factor, lose_factor;
 	unsigned long gran_base, chunk_base, lose_base;
+	char gran_factor, chunk_factor, lose_factor;
 
 	gran_base = to_size_factor(result[i].gran_sizek, &gran_factor),
 	chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor),
 	lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor),
-	printk(KERN_INFO "%sgran_size: %ld%c \tchunk_size: %ld%c \t",
-			result[i].bad ? "*BAD*" : " ",
-			gran_base, gran_factor, chunk_base, chunk_factor);
-	printk(KERN_CONT "num_reg: %d  \tlose cover RAM: %s%ld%c\n",
-			result[i].num_reg, result[i].bad ? "-" : "",
-			lose_base, lose_factor);
+
+	pr_info("%sgran_size: %ld%c \tchunk_size: %ld%c \t",
+		result[i].bad ? "*BAD*" : " ",
+		gran_base, gran_factor, chunk_base, chunk_factor);
+	pr_cont("num_reg: %d  \tlose cover RAM: %s%ld%c\n",
+		result[i].num_reg, result[i].bad ? "-" : "",
+		lose_base, lose_factor);
 }
 
 static int __init mtrr_search_optimal_index(void)
 {
-	int i;
 	int num_reg_good;
 	int index_good;
+	int i;
 
 	if (nr_mtrr_spare_reg >= num_var_ranges)
 		nr_mtrr_spare_reg = num_var_ranges - 1;
+
 	num_reg_good = -1;
 	for (i = num_var_ranges - nr_mtrr_spare_reg; i > 0; i--) {
 		if (!min_loss_pfn[i])
@@ -796,24 +797,24 @@ static int __init mtrr_search_optimal_index(void)
 	return index_good;
 }
 
-
 int __init mtrr_cleanup(unsigned address_bits)
 {
-	unsigned long extra_remove_base, extra_remove_size;
+	unsigned long x_remove_base, x_remove_size;
 	unsigned long base, size, def, dummy;
-	mtrr_type type;
 	u64 chunk_size, gran_size;
+	mtrr_type type;
 	int index_good;
 	int i;
 
 	if (!is_cpu(INTEL) || enable_mtrr_cleanup < 1)
 		return 0;
+
 	rdmsr(MSR_MTRRdefType, def, dummy);
 	def &= 0xff;
 	if (def != MTRR_TYPE_UNCACHABLE)
 		return 0;
 
-	/* get it and store it aside */
+	/* Get it and store it aside: */
 	memset(range_state, 0, sizeof(range_state));
 	for (i = 0; i < num_var_ranges; i++) {
 		mtrr_if->get(i, &base, &size, &type);
@@ -822,29 +823,28 @@ int __init mtrr_cleanup(unsigned address_bits)
 		range_state[i].type = type;
 	}
 
-	/* check if we need handle it and can handle it */
+	/* Check if we need handle it and can handle it: */
 	if (!mtrr_need_cleanup())
 		return 0;
 
-	/* print original var MTRRs at first, for debugging: */
+	/* Print original var MTRRs at first, for debugging: */
 	printk(KERN_DEBUG "original variable MTRRs\n");
 	print_out_mtrr_range_state();
 
 	memset(range, 0, sizeof(range));
-	extra_remove_size = 0;
-	extra_remove_base = 1 << (32 - PAGE_SHIFT);
+	x_remove_size = 0;
+	x_remove_base = 1 << (32 - PAGE_SHIFT);
 	if (mtrr_tom2)
-		extra_remove_size =
-			(mtrr_tom2 >> PAGE_SHIFT) - extra_remove_base;
-	nr_range = x86_get_mtrr_mem_range(range, 0, extra_remove_base,
-					  extra_remove_size);
+		x_remove_size = (mtrr_tom2 >> PAGE_SHIFT) - x_remove_base;
+
+	nr_range = x86_get_mtrr_mem_range(range, 0, x_remove_base, x_remove_size);
 	/*
-	 * [0, 1M) should always be coverred by var mtrr with WB
-	 * and fixed mtrrs should take effective before var mtrr for it
+	 * [0, 1M) should always be covered by var mtrr with WB
+	 * and fixed mtrrs should take effect before var mtrr for it:
 	 */
 	nr_range = add_range_with_merge(range, nr_range, 0,
 					(1ULL<<(20 - PAGE_SHIFT)) - 1);
-	/* sort the ranges */
+	/* Sort the ranges: */
 	sort(range, nr_range, sizeof(struct res_range), cmp_range, NULL);
 
 	range_sums = sum_ranges(range, nr_range);
@@ -854,7 +854,7 @@ int __init mtrr_cleanup(unsigned address_bits)
 	if (mtrr_chunk_size && mtrr_gran_size) {
 		i = 0;
 		mtrr_calc_range_state(mtrr_chunk_size, mtrr_gran_size,
-				      extra_remove_base, extra_remove_size, i);
+				      x_remove_base, x_remove_size, i);
 
 		mtrr_print_out_one_result(i);
 
@@ -880,7 +880,7 @@ int __init mtrr_cleanup(unsigned address_bits)
 				continue;
 
 			mtrr_calc_range_state(chunk_size, gran_size,
-				      extra_remove_base, extra_remove_size, i);
+				      x_remove_base, x_remove_size, i);
 			if (debug_print) {
 				mtrr_print_out_one_result(i);
 				printk(KERN_INFO "\n");
@@ -890,7 +890,7 @@ int __init mtrr_cleanup(unsigned address_bits)
 		}
 	}
 
-	/* try to find the optimal index */
+	/* Try to find the optimal index: */
 	index_good = mtrr_search_optimal_index();
 
 	if (index_good != -1) {
@@ -898,7 +898,7 @@ int __init mtrr_cleanup(unsigned address_bits)
 		i = index_good;
 		mtrr_print_out_one_result(i);
 
-		/* convert ranges to var ranges state */
+		/* Convert ranges to var ranges state: */
 		chunk_size = result[i].chunk_sizek;
 		chunk_size <<= 10;
 		gran_size = result[i].gran_sizek;
@@ -941,8 +941,8 @@ early_param("disable_mtrr_trim", disable_mtrr_trim_setup);
  * Note this won't check if the MTRRs < 4GB where the magic bit doesn't
  * apply to are wrong, but so far we don't know of any such case in the wild.
  */
-#define Tom2Enabled (1U << 21)
-#define Tom2ForceMemTypeWB (1U << 22)
+#define Tom2Enabled		(1U << 21)
+#define Tom2ForceMemTypeWB	(1U << 22)
 
 int __init amd_special_default_mtrr(void)
 {
@@ -952,7 +952,7 @@ int __init amd_special_default_mtrr(void)
 		return 0;
 	if (boot_cpu_data.x86 < 0xf || boot_cpu_data.x86 > 0x11)
 		return 0;
-	/* In case some hypervisor doesn't pass SYSCFG through */
+	/* In case some hypervisor doesn't pass SYSCFG through: */
 	if (rdmsr_safe(MSR_K8_SYSCFG, &l, &h) < 0)
 		return 0;
 	/*
@@ -965,19 +965,21 @@ int __init amd_special_default_mtrr(void)
 	return 0;
 }
 
-static u64 __init real_trim_memory(unsigned long start_pfn,
-				   unsigned long limit_pfn)
+static u64 __init
+real_trim_memory(unsigned long start_pfn, unsigned long limit_pfn)
 {
 	u64 trim_start, trim_size;
+
 	trim_start = start_pfn;
 	trim_start <<= PAGE_SHIFT;
+
 	trim_size = limit_pfn;
 	trim_size <<= PAGE_SHIFT;
 	trim_size -= trim_start;
 
-	return e820_update_range(trim_start, trim_size, E820_RAM,
-				E820_RESERVED);
+	return e820_update_range(trim_start, trim_size, E820_RAM, E820_RESERVED);
 }
+
 /**
  * mtrr_trim_uncached_memory - trim RAM not covered by MTRRs
  * @end_pfn: ending page frame number
@@ -985,7 +987,7 @@ static u64 __init real_trim_memory(unsigned long start_pfn,
  * Some buggy BIOSes don't setup the MTRRs properly for systems with certain
  * memory configurations.  This routine checks that the highest MTRR matches
  * the end of memory, to make sure the MTRRs having a write back type cover
- * all of the memory the kernel is intending to use. If not, it'll trim any
+ * all of the memory the kernel is intending to use.  If not, it'll trim any
  * memory off the end by adjusting end_pfn, removing it from the kernel's
  * allocation pools, warning the user with an obnoxious message.
  */
@@ -994,21 +996,22 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn)
 	unsigned long i, base, size, highest_pfn = 0, def, dummy;
 	mtrr_type type;
 	u64 total_trim_size;
-
 	/* extra one for all 0 */
 	int num[MTRR_NUM_TYPES + 1];
+
 	/*
 	 * Make sure we only trim uncachable memory on machines that
 	 * support the Intel MTRR architecture:
 	 */
 	if (!is_cpu(INTEL) || disable_mtrr_trim)
 		return 0;
+
 	rdmsr(MSR_MTRRdefType, def, dummy);
 	def &= 0xff;
 	if (def != MTRR_TYPE_UNCACHABLE)
 		return 0;
 
-	/* get it and store it aside */
+	/* Get it and store it aside: */
 	memset(range_state, 0, sizeof(range_state));
 	for (i = 0; i < num_var_ranges; i++) {
 		mtrr_if->get(i, &base, &size, &type);
@@ -1017,7 +1020,7 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn)
 		range_state[i].type = type;
 	}
 
-	/* Find highest cached pfn */
+	/* Find highest cached pfn: */
 	for (i = 0; i < num_var_ranges; i++) {
 		type = range_state[i].type;
 		if (type != MTRR_TYPE_WRBACK)
@@ -1028,13 +1031,13 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn)
 			highest_pfn = base + size;
 	}
 
-	/* kvm/qemu doesn't have mtrr set right, don't trim them all */
+	/* kvm/qemu doesn't have mtrr set right, don't trim them all: */
 	if (!highest_pfn) {
 		printk(KERN_INFO "CPU MTRRs all blank - virtualized system.\n");
 		return 0;
 	}
 
-	/* check entries number */
+	/* Check entries number: */
 	memset(num, 0, sizeof(num));
 	for (i = 0; i < num_var_ranges; i++) {
 		type = range_state[i].type;
@@ -1046,11 +1049,11 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn)
 		num[type]++;
 	}
 
-	/* no entry for WB? */
+	/* No entry for WB? */
 	if (!num[MTRR_TYPE_WRBACK])
 		return 0;
 
-	/* check if we only had WB and UC */
+	/* Check if we only had WB and UC: */
 	if (num[MTRR_TYPE_WRBACK] + num[MTRR_TYPE_UNCACHABLE] !=
 		num_var_ranges - num[MTRR_NUM_TYPES])
 		return 0;
@@ -1066,31 +1069,31 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn)
 	}
 	nr_range = x86_get_mtrr_mem_range(range, nr_range, 0, 0);
 
+	/* Check the head: */
 	total_trim_size = 0;
-	/* check the head */
 	if (range[0].start)
 		total_trim_size += real_trim_memory(0, range[0].start);
-	/* check the holes */
+
+	/* Check the holes: */
 	for (i = 0; i < nr_range - 1; i++) {
 		if (range[i].end + 1 < range[i+1].start)
 			total_trim_size += real_trim_memory(range[i].end + 1,
 							    range[i+1].start);
 	}
-	/* check the top */
+
+	/* Check the top: */
 	i = nr_range - 1;
 	if (range[i].end + 1 < end_pfn)
 		total_trim_size += real_trim_memory(range[i].end + 1,
 							 end_pfn);
 
 	if (total_trim_size) {
-		printk(KERN_WARNING "WARNING: BIOS bug: CPU MTRRs don't cover"
-			" all of memory, losing %lluMB of RAM.\n",
-			total_trim_size >> 20);
+		pr_warning("WARNING: BIOS bug: CPU MTRRs don't cover all of memory, losing %lluMB of RAM.\n", total_trim_size >> 20);
 
 		if (!changed_by_mtrr_cleanup)
 			WARN_ON(1);
 
-		printk(KERN_INFO "update e820 for mtrr\n");
+		pr_info("update e820 for mtrr\n");
 		update_e820();
 
 		return 1;
@@ -1098,4 +1101,3 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn)
 
 	return 0;
 }
-
-- 
cgit v1.2.3


From 2311037708c170977506fbcbe0a2ba0c6d221940 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Sat, 4 Jul 2009 07:52:08 +0530
Subject: x86: Clean up mtrr/cyrix.c

Fix trivial style problems:

  WARNING: Use #include <linux/io.h> instead of <asm/io.h>
  WARNING: line over 80 characters
  ERROR: do not initialise statics to 0 or NULL
  ERROR: space prohibited after that open parenthesis '(' X 2
  ERROR: space prohibited before that close parenthesis ')' X 2
  ERROR: trailing whitespace X 2
  ERROR: trailing statements should be on next line
  ERROR: do not use C99 // comments X 2

arch/x86/kernel/cpu/mtrr/cyrix.o:

   text	   data	    bss	    dec	    hex	filename
   1637	     32	      8	   1677	    68d	cyrix.o.before
   1637	     32	      8	   1677	    68d	cyrix.o.after

md5:
   6f52abd06905be3f4cabb5239f9b0ff0  cyrix.o.before.asm
   6f52abd06905be3f4cabb5239f9b0ff0  cyrix.o.after.asm

Suggested-by: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <20090703164225.GA21447@elte.hu>
[ Made the code more consistent ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/mtrr/cyrix.c | 94 ++++++++++++++++++++++------------------
 1 file changed, 51 insertions(+), 43 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/mtrr/cyrix.c b/arch/x86/kernel/cpu/mtrr/cyrix.c
index ff14c320040c..228d982ce09c 100644
--- a/arch/x86/kernel/cpu/mtrr/cyrix.c
+++ b/arch/x86/kernel/cpu/mtrr/cyrix.c
@@ -1,38 +1,40 @@
 #include <linux/init.h>
+#include <linux/io.h>
 #include <linux/mm.h>
-#include <asm/mtrr.h>
-#include <asm/msr.h>
-#include <asm/io.h>
+
 #include <asm/processor-cyrix.h>
 #include <asm/processor-flags.h>
+#include <asm/mtrr.h>
+#include <asm/msr.h>
+
 #include "mtrr.h"
 
 static void
 cyrix_get_arr(unsigned int reg, unsigned long *base,
 	      unsigned long *size, mtrr_type * type)
 {
-	unsigned long flags;
 	unsigned char arr, ccr3, rcr, shift;
+	unsigned long flags;
 
 	arr = CX86_ARR_BASE + (reg << 1) + reg;	/* avoid multiplication by 3 */
 
-	/* Save flags and disable interrupts */
 	local_irq_save(flags);
 
 	ccr3 = getCx86(CX86_CCR3);
 	setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10);	/* enable MAPEN */
-	((unsigned char *) base)[3] = getCx86(arr);
-	((unsigned char *) base)[2] = getCx86(arr + 1);
-	((unsigned char *) base)[1] = getCx86(arr + 2);
+	((unsigned char *)base)[3] = getCx86(arr);
+	((unsigned char *)base)[2] = getCx86(arr + 1);
+	((unsigned char *)base)[1] = getCx86(arr + 2);
 	rcr = getCx86(CX86_RCR_BASE + reg);
-	setCx86(CX86_CCR3, ccr3);	/* disable MAPEN */
+	setCx86(CX86_CCR3, ccr3);			/* disable MAPEN */
 
-	/* Enable interrupts if it was enabled previously */
 	local_irq_restore(flags);
+
 	shift = ((unsigned char *) base)[1] & 0x0f;
 	*base >>= PAGE_SHIFT;
 
-	/* Power of two, at least 4K on ARR0-ARR6, 256K on ARR7
+	/*
+	 * Power of two, at least 4K on ARR0-ARR6, 256K on ARR7
 	 * Note: shift==0xf means 4G, this is unsupported.
 	 */
 	if (shift)
@@ -76,17 +78,20 @@ cyrix_get_arr(unsigned int reg, unsigned long *base,
 	}
 }
 
+/*
+ * cyrix_get_free_region - get a free ARR.
+ *
+ * @base: the starting (base) address of the region.
+ * @size: the size (in bytes) of the region.
+ *
+ * Returns: the index of the region on success, else -1 on error.
+*/
 static int
 cyrix_get_free_region(unsigned long base, unsigned long size, int replace_reg)
-/*  [SUMMARY] Get a free ARR.
-    <base> The starting (base) address of the region.
-    <size> The size (in bytes) of the region.
-    [RETURNS] The index of the region on success, else -1 on error.
-*/
 {
-	int i;
-	mtrr_type ltype;
 	unsigned long lbase, lsize;
+	mtrr_type ltype;
+	int i;
 
 	switch (replace_reg) {
 	case 7:
@@ -107,14 +112,17 @@ cyrix_get_free_region(unsigned long base, unsigned long size, int replace_reg)
 		cyrix_get_arr(7, &lbase, &lsize, &ltype);
 		if (lsize == 0)
 			return 7;
-		/*  Else try ARR0-ARR6 first  */
+		/* Else try ARR0-ARR6 first  */
 	} else {
 		for (i = 0; i < 7; i++) {
 			cyrix_get_arr(i, &lbase, &lsize, &ltype);
 			if (lsize == 0)
 				return i;
 		}
-		/* ARR0-ARR6 isn't free, try ARR7 but its size must be at least 256K */
+		/*
+		 * ARR0-ARR6 isn't free
+		 * try ARR7 but its size must be at least 256K
+		 */
 		cyrix_get_arr(i, &lbase, &lsize, &ltype);
 		if ((lsize == 0) && (size >= 0x40))
 			return i;
@@ -122,21 +130,22 @@ cyrix_get_free_region(unsigned long base, unsigned long size, int replace_reg)
 	return -ENOSPC;
 }
 
-static u32 cr4 = 0;
-static u32 ccr3;
+static u32 cr4, ccr3;
 
 static void prepare_set(void)
 {
 	u32 cr0;
 
 	/*  Save value of CR4 and clear Page Global Enable (bit 7)  */
-	if ( cpu_has_pge ) {
+	if (cpu_has_pge) {
 		cr4 = read_cr4();
 		write_cr4(cr4 & ~X86_CR4_PGE);
 	}
 
-	/*  Disable and flush caches. Note that wbinvd flushes the TLBs as
-	    a side-effect  */
+	/*
+	 * Disable and flush caches.
+	 * Note that wbinvd flushes the TLBs as a side-effect
+	 */
 	cr0 = read_cr0() | X86_CR0_CD;
 	wbinvd();
 	write_cr0(cr0);
@@ -147,22 +156,21 @@ static void prepare_set(void)
 
 	/* Cyrix ARRs - everything else was excluded at the top */
 	setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10);
-
 }
 
 static void post_set(void)
 {
-	/*  Flush caches and TLBs  */
+	/* Flush caches and TLBs */
 	wbinvd();
 
 	/* Cyrix ARRs - everything else was excluded at the top */
 	setCx86(CX86_CCR3, ccr3);
-		
-	/*  Enable caches  */
+
+	/* Enable caches */
 	write_cr0(read_cr0() & 0xbfffffff);
 
-	/*  Restore value of CR4  */
-	if ( cpu_has_pge )
+	/* Restore value of CR4 */
+	if (cpu_has_pge)
 		write_cr4(cr4);
 }
 
@@ -178,7 +186,8 @@ static void cyrix_set_arr(unsigned int reg, unsigned long base,
 		size >>= 6;
 
 	size &= 0x7fff;		/* make sure arr_size <= 14 */
-	for (arr_size = 0; size; arr_size++, size >>= 1) ;
+	for (arr_size = 0; size; arr_size++, size >>= 1)
+		;
 
 	if (reg < 7) {
 		switch (type) {
@@ -215,18 +224,18 @@ static void cyrix_set_arr(unsigned int reg, unsigned long base,
 	prepare_set();
 
 	base <<= PAGE_SHIFT;
-	setCx86(arr, ((unsigned char *) &base)[3]);
-	setCx86(arr + 1, ((unsigned char *) &base)[2]);
-	setCx86(arr + 2, (((unsigned char *) &base)[1]) | arr_size);
+	setCx86(arr + 0,  ((unsigned char *)&base)[3]);
+	setCx86(arr + 1,  ((unsigned char *)&base)[2]);
+	setCx86(arr + 2, (((unsigned char *)&base)[1]) | arr_size);
 	setCx86(CX86_RCR_BASE + reg, arr_type);
 
 	post_set();
 }
 
 typedef struct {
-	unsigned long base;
-	unsigned long size;
-	mtrr_type type;
+	unsigned long	base;
+	unsigned long	size;
+	mtrr_type	type;
 } arr_state_t;
 
 static arr_state_t arr_state[8] = {
@@ -247,16 +256,17 @@ static void cyrix_set_all(void)
 		setCx86(CX86_CCR0 + i, ccr_state[i]);
 	for (; i < 7; i++)
 		setCx86(CX86_CCR4 + i, ccr_state[i]);
-	for (i = 0; i < 8; i++)
-		cyrix_set_arr(i, arr_state[i].base, 
+
+	for (i = 0; i < 8; i++) {
+		cyrix_set_arr(i, arr_state[i].base,
 			      arr_state[i].size, arr_state[i].type);
+	}
 
 	post_set();
 }
 
 static struct mtrr_ops cyrix_mtrr_ops = {
 	.vendor            = X86_VENDOR_CYRIX,
-//	.init              = cyrix_arr_init,
 	.set_all	   = cyrix_set_all,
 	.set               = cyrix_set_arr,
 	.get               = cyrix_get_arr,
@@ -270,5 +280,3 @@ int __init cyrix_init_mtrr(void)
 	set_mtrr_ops(&cyrix_mtrr_ops);
 	return 0;
 }
-
-//arch_initcall(cyrix_init_mtrr);
-- 
cgit v1.2.3


From a1a499a39911fcfecbebaba1f38588088909f918 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Sat, 4 Jul 2009 07:53:00 +0530
Subject: x86: Clean up mtrr/generic.c

Fix following trivial style problems:

  ERROR: trailing whitespace X 4
  WARNING: Use #include <linux/io.h> instead of <asm/io.h>
  WARNING: braces {} are not necessary for single statement blocks X 3
  ERROR: "foo * bar" should be "foo *bar"
  WARNING: line over 80 characters X 6
  ERROR: "foo * bar" should be "foo *bar"
  ERROR: spaces required around that '=' (ctx:VxO)
  ERROR: space required before that '-' (ctx:OxV)
  WARNING: suspect code indent for conditional statements (8, 12)
  ERROR: spaces required around that '=' (ctx:VxV)
  ERROR: do not initialise statics to 0 or NULL
  ERROR: space prohibited after that open parenthesis '(' X 2
  ERROR: space prohibited before that close parenthesis ')' X 2
  ERROR: trailing statements should be on next line
  ERROR: return is not a function, parentheses are not required

Also use pr_debug and pr_warning where possible.

arch/x86/kernel/cpu/mtrr/generic.o:

   text	   data	    bss	    dec	    hex	filename
   5652	     77	   4224	   9953	   26e1	generic.o.before
   5652	     77	   4220	   9949	   26dd	generic.o.after

The md5 changed:
   b34d6c045f06daa4ed092b90cc760e8f  generic.o.before.asm
   a490c6251cfd8442fbffecc0e09a573d  generic.o.after.asm

Because mtrr_state moved from data to bss, changing its
offsets - and also because __LINE__ numbers changed.

Suggested-by: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <20090703164225.GA21447@elte.hu>
[ Further cleanups to make the code more consistent ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/mtrr/generic.c | 304 +++++++++++++++++++++----------------
 1 file changed, 169 insertions(+), 135 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c
index 0543f69f0b27..55da0c5f68dd 100644
--- a/arch/x86/kernel/cpu/mtrr/generic.c
+++ b/arch/x86/kernel/cpu/mtrr/generic.c
@@ -1,28 +1,34 @@
-/* This only handles 32bit MTRR on 32bit hosts. This is strictly wrong
-   because MTRRs can span upto 40 bits (36bits on most modern x86) */ 
+/*
+ * This only handles 32bit MTRR on 32bit hosts. This is strictly wrong
+ * because MTRRs can span upto 40 bits (36bits on most modern x86)
+ */
+#define DEBUG
+
+#include <linux/module.h>
 #include <linux/init.h>
 #include <linux/slab.h>
+#include <linux/io.h>
 #include <linux/mm.h>
-#include <linux/module.h>
-#include <asm/io.h>
-#include <asm/mtrr.h>
-#include <asm/msr.h>
-#include <asm/system.h>
-#include <asm/cpufeature.h>
+
 #include <asm/processor-flags.h>
+#include <asm/cpufeature.h>
 #include <asm/tlbflush.h>
+#include <asm/system.h>
+#include <asm/mtrr.h>
+#include <asm/msr.h>
 #include <asm/pat.h>
+
 #include "mtrr.h"
 
 struct fixed_range_block {
-	int base_msr; /* start address of an MTRR block */
-	int ranges;   /* number of MTRRs in this block  */
+	int base_msr;		/* start address of an MTRR block */
+	int ranges;		/* number of MTRRs in this block  */
 };
 
 static struct fixed_range_block fixed_range_blocks[] = {
-	{ MSR_MTRRfix64K_00000, 1 }, /* one  64k MTRR  */
-	{ MSR_MTRRfix16K_80000, 2 }, /* two  16k MTRRs */
-	{ MSR_MTRRfix4K_C0000,  8 }, /* eight 4k MTRRs */
+	{ MSR_MTRRfix64K_00000, 1 }, /* one   64k MTRR  */
+	{ MSR_MTRRfix16K_80000, 2 }, /* two   16k MTRRs */
+	{ MSR_MTRRfix4K_C0000,  8 }, /* eight  4k MTRRs */
 	{}
 };
 
@@ -30,10 +36,10 @@ static unsigned long smp_changes_mask;
 static int mtrr_state_set;
 u64 mtrr_tom2;
 
-struct mtrr_state_type mtrr_state = {};
+struct mtrr_state_type mtrr_state;
 EXPORT_SYMBOL_GPL(mtrr_state);
 
-/**
+/*
  * BIOS is expected to clear MtrrFixDramModEn bit, see for example
  * "BIOS and Kernel Developer's Guide for the AMD Athlon 64 and AMD
  * Opteron Processors" (26094 Rev. 3.30 February 2006), section
@@ -104,9 +110,8 @@ u8 mtrr_type_lookup(u64 start, u64 end)
 	 * Look of multiple ranges matching this address and pick type
 	 * as per MTRR precedence
 	 */
-	if (!(mtrr_state.enabled & 2)) {
+	if (!(mtrr_state.enabled & 2))
 		return mtrr_state.def_type;
-	}
 
 	prev_match = 0xFF;
 	for (i = 0; i < num_var_ranges; ++i) {
@@ -125,9 +130,8 @@ u8 mtrr_type_lookup(u64 start, u64 end)
 		if (start_state != end_state)
 			return 0xFE;
 
-		if ((start & mask) != (base & mask)) {
+		if ((start & mask) != (base & mask))
 			continue;
-		}
 
 		curr_match = mtrr_state.var_ranges[i].base_lo & 0xff;
 		if (prev_match == 0xFF) {
@@ -148,9 +152,8 @@ u8 mtrr_type_lookup(u64 start, u64 end)
 			curr_match = MTRR_TYPE_WRTHROUGH;
 		}
 
-		if (prev_match != curr_match) {
+		if (prev_match != curr_match)
 			return MTRR_TYPE_UNCACHABLE;
-		}
 	}
 
 	if (mtrr_tom2) {
@@ -164,7 +167,7 @@ u8 mtrr_type_lookup(u64 start, u64 end)
 	return mtrr_state.def_type;
 }
 
-/*  Get the MSR pair relating to a var range  */
+/* Get the MSR pair relating to a var range */
 static void
 get_mtrr_var_range(unsigned int index, struct mtrr_var_range *vr)
 {
@@ -172,7 +175,7 @@ get_mtrr_var_range(unsigned int index, struct mtrr_var_range *vr)
 	rdmsr(MTRRphysMask_MSR(index), vr->mask_lo, vr->mask_hi);
 }
 
-/*  fill the MSR pair relating to a var range  */
+/* Fill the MSR pair relating to a var range */
 void fill_mtrr_var_range(unsigned int index,
 		u32 base_lo, u32 base_hi, u32 mask_lo, u32 mask_hi)
 {
@@ -186,10 +189,9 @@ void fill_mtrr_var_range(unsigned int index,
 	vr[index].mask_hi = mask_hi;
 }
 
-static void
-get_fixed_ranges(mtrr_type * frs)
+static void get_fixed_ranges(mtrr_type *frs)
 {
-	unsigned int *p = (unsigned int *) frs;
+	unsigned int *p = (unsigned int *)frs;
 	int i;
 
 	k8_check_syscfg_dram_mod_en();
@@ -217,22 +219,22 @@ static void __init print_fixed_last(void)
 	if (!last_fixed_end)
 		return;
 
-	printk(KERN_DEBUG "  %05X-%05X %s\n", last_fixed_start,
-		last_fixed_end - 1, mtrr_attrib_to_str(last_fixed_type));
+	pr_debug("  %05X-%05X %s\n", last_fixed_start,
+		 last_fixed_end - 1, mtrr_attrib_to_str(last_fixed_type));
 
 	last_fixed_end = 0;
 }
 
 static void __init update_fixed_last(unsigned base, unsigned end,
-				       mtrr_type type)
+				     mtrr_type type)
 {
 	last_fixed_start = base;
 	last_fixed_end = end;
 	last_fixed_type = type;
 }
 
-static void __init print_fixed(unsigned base, unsigned step,
-			       const mtrr_type *types)
+static void __init
+print_fixed(unsigned base, unsigned step, const mtrr_type *types)
 {
 	unsigned i;
 
@@ -259,54 +261,55 @@ static void __init print_mtrr_state(void)
 	unsigned int i;
 	int high_width;
 
-	printk(KERN_DEBUG "MTRR default type: %s\n",
-			 mtrr_attrib_to_str(mtrr_state.def_type));
+	pr_debug("MTRR default type: %s\n",
+		 mtrr_attrib_to_str(mtrr_state.def_type));
 	if (mtrr_state.have_fixed) {
-		printk(KERN_DEBUG "MTRR fixed ranges %sabled:\n",
-		       mtrr_state.enabled & 1 ? "en" : "dis");
+		pr_debug("MTRR fixed ranges %sabled:\n",
+			 mtrr_state.enabled & 1 ? "en" : "dis");
 		print_fixed(0x00000, 0x10000, mtrr_state.fixed_ranges + 0);
 		for (i = 0; i < 2; ++i)
-			print_fixed(0x80000 + i * 0x20000, 0x04000, mtrr_state.fixed_ranges + (i + 1) * 8);
+			print_fixed(0x80000 + i * 0x20000, 0x04000,
+				    mtrr_state.fixed_ranges + (i + 1) * 8);
 		for (i = 0; i < 8; ++i)
-			print_fixed(0xC0000 + i * 0x08000, 0x01000, mtrr_state.fixed_ranges + (i + 3) * 8);
+			print_fixed(0xC0000 + i * 0x08000, 0x01000,
+				    mtrr_state.fixed_ranges + (i + 3) * 8);
 
 		/* tail */
 		print_fixed_last();
 	}
-	printk(KERN_DEBUG "MTRR variable ranges %sabled:\n",
-	       mtrr_state.enabled & 2 ? "en" : "dis");
+	pr_debug("MTRR variable ranges %sabled:\n",
+		 mtrr_state.enabled & 2 ? "en" : "dis");
 	if (size_or_mask & 0xffffffffUL)
 		high_width = ffs(size_or_mask & 0xffffffffUL) - 1;
 	else
 		high_width = ffs(size_or_mask>>32) + 32 - 1;
 	high_width = (high_width - (32 - PAGE_SHIFT) + 3) / 4;
+
 	for (i = 0; i < num_var_ranges; ++i) {
 		if (mtrr_state.var_ranges[i].mask_lo & (1 << 11))
-			printk(KERN_DEBUG "  %u base %0*X%05X000 mask %0*X%05X000 %s\n",
-			       i,
-			       high_width,
-			       mtrr_state.var_ranges[i].base_hi,
-			       mtrr_state.var_ranges[i].base_lo >> 12,
-			       high_width,
-			       mtrr_state.var_ranges[i].mask_hi,
-			       mtrr_state.var_ranges[i].mask_lo >> 12,
-			       mtrr_attrib_to_str(mtrr_state.var_ranges[i].base_lo & 0xff));
+			pr_debug("  %u base %0*X%05X000 mask %0*X%05X000 %s\n",
+				 i,
+				 high_width,
+				 mtrr_state.var_ranges[i].base_hi,
+				 mtrr_state.var_ranges[i].base_lo >> 12,
+				 high_width,
+				 mtrr_state.var_ranges[i].mask_hi,
+				 mtrr_state.var_ranges[i].mask_lo >> 12,
+				 mtrr_attrib_to_str(mtrr_state.var_ranges[i].base_lo & 0xff));
 		else
-			printk(KERN_DEBUG "  %u disabled\n", i);
-	}
-	if (mtrr_tom2) {
-		printk(KERN_DEBUG "TOM2: %016llx aka %lldM\n",
-				  mtrr_tom2, mtrr_tom2>>20);
+			pr_debug("  %u disabled\n", i);
 	}
+	if (mtrr_tom2)
+		pr_debug("TOM2: %016llx aka %lldM\n", mtrr_tom2, mtrr_tom2>>20);
 }
 
-/*  Grab all of the MTRR state for this CPU into *state  */
+/* Grab all of the MTRR state for this CPU into *state */
 void __init get_mtrr_state(void)
 {
-	unsigned int i;
 	struct mtrr_var_range *vrs;
-	unsigned lo, dummy;
 	unsigned long flags;
+	unsigned lo, dummy;
+	unsigned int i;
 
 	vrs = mtrr_state.var_ranges;
 
@@ -324,6 +327,7 @@ void __init get_mtrr_state(void)
 
 	if (amd_special_default_mtrr()) {
 		unsigned low, high;
+
 		/* TOP_MEM2 */
 		rdmsr(MSR_K8_TOP_MEM2, low, high);
 		mtrr_tom2 = high;
@@ -344,10 +348,9 @@ void __init get_mtrr_state(void)
 
 	post_set();
 	local_irq_restore(flags);
-
 }
 
-/*  Some BIOS's are fucked and don't set all MTRRs the same!  */
+/* Some BIOS's are messed up and don't set all MTRRs the same! */
 void __init mtrr_state_warn(void)
 {
 	unsigned long mask = smp_changes_mask;
@@ -355,28 +358,33 @@ void __init mtrr_state_warn(void)
 	if (!mask)
 		return;
 	if (mask & MTRR_CHANGE_MASK_FIXED)
-		printk(KERN_WARNING "mtrr: your CPUs had inconsistent fixed MTRR settings\n");
+		pr_warning("mtrr: your CPUs had inconsistent fixed MTRR settings\n");
 	if (mask & MTRR_CHANGE_MASK_VARIABLE)
-		printk(KERN_WARNING "mtrr: your CPUs had inconsistent variable MTRR settings\n");
+		pr_warning("mtrr: your CPUs had inconsistent variable MTRR settings\n");
 	if (mask & MTRR_CHANGE_MASK_DEFTYPE)
-		printk(KERN_WARNING "mtrr: your CPUs had inconsistent MTRRdefType settings\n");
+		pr_warning("mtrr: your CPUs had inconsistent MTRRdefType settings\n");
+
 	printk(KERN_INFO "mtrr: probably your BIOS does not setup all CPUs.\n");
 	printk(KERN_INFO "mtrr: corrected configuration.\n");
 }
 
-/* Doesn't attempt to pass an error out to MTRR users
-   because it's quite complicated in some cases and probably not
-   worth it because the best error handling is to ignore it. */
+/*
+ * Doesn't attempt to pass an error out to MTRR users
+ * because it's quite complicated in some cases and probably not
+ * worth it because the best error handling is to ignore it.
+ */
 void mtrr_wrmsr(unsigned msr, unsigned a, unsigned b)
 {
-	if (wrmsr_safe(msr, a, b) < 0)
+	if (wrmsr_safe(msr, a, b) < 0) {
 		printk(KERN_ERR
 			"MTRR: CPU %u: Writing MSR %x to %x:%x failed\n",
 			smp_processor_id(), msr, a, b);
+	}
 }
 
 /**
- * set_fixed_range - checks & updates a fixed-range MTRR if it differs from the value it should have
+ * set_fixed_range - checks & updates a fixed-range MTRR if it
+ *		     differs from the value it should have
  * @msr: MSR address of the MTTR which should be checked and updated
  * @changed: pointer which indicates whether the MTRR needed to be changed
  * @msrwords: pointer to the MSR values which the MSR should have
@@ -401,20 +409,23 @@ static void set_fixed_range(int msr, bool *changed, unsigned int *msrwords)
  *
  * Returns: The index of the region on success, else negative on error.
  */
-int generic_get_free_region(unsigned long base, unsigned long size, int replace_reg)
+int
+generic_get_free_region(unsigned long base, unsigned long size, int replace_reg)
 {
-	int i, max;
-	mtrr_type ltype;
 	unsigned long lbase, lsize;
+	mtrr_type ltype;
+	int i, max;
 
 	max = num_var_ranges;
 	if (replace_reg >= 0 && replace_reg < max)
 		return replace_reg;
+
 	for (i = 0; i < max; ++i) {
 		mtrr_if->get(i, &lbase, &lsize, &ltype);
 		if (lsize == 0)
 			return i;
 	}
+
 	return -ENOSPC;
 }
 
@@ -434,7 +445,7 @@ static void generic_get_mtrr(unsigned int reg, unsigned long *base,
 	rdmsr(MTRRphysMask_MSR(reg), mask_lo, mask_hi);
 
 	if ((mask_lo & 0x800) == 0) {
-		/*  Invalid (i.e. free) range  */
+		/*  Invalid (i.e. free) range */
 		*base = 0;
 		*size = 0;
 		*type = 0;
@@ -471,27 +482,31 @@ out_put_cpu:
 }
 
 /**
- * set_fixed_ranges - checks & updates the fixed-range MTRRs if they differ from the saved set
+ * set_fixed_ranges - checks & updates the fixed-range MTRRs if they
+ *		      differ from the saved set
  * @frs: pointer to fixed-range MTRR values, saved by get_fixed_ranges()
  */
-static int set_fixed_ranges(mtrr_type * frs)
+static int set_fixed_ranges(mtrr_type *frs)
 {
-	unsigned long long *saved = (unsigned long long *) frs;
+	unsigned long long *saved = (unsigned long long *)frs;
 	bool changed = false;
-	int block=-1, range;
+	int block = -1, range;
 
 	k8_check_syscfg_dram_mod_en();
 
-	while (fixed_range_blocks[++block].ranges)
-	    for (range=0; range < fixed_range_blocks[block].ranges; range++)
-		set_fixed_range(fixed_range_blocks[block].base_msr + range,
-		    &changed, (unsigned int *) saved++);
+	while (fixed_range_blocks[++block].ranges) {
+		for (range = 0; range < fixed_range_blocks[block].ranges; range++)
+			set_fixed_range(fixed_range_blocks[block].base_msr + range,
+					&changed, (unsigned int *)saved++);
+	}
 
 	return changed;
 }
 
-/*  Set the MSR pair relating to a var range. Returns TRUE if
-    changes are made  */
+/*
+ * Set the MSR pair relating to a var range.
+ * Returns true if changes are made.
+ */
 static bool set_mtrr_var_ranges(unsigned int index, struct mtrr_var_range *vr)
 {
 	unsigned int lo, hi;
@@ -501,6 +516,7 @@ static bool set_mtrr_var_ranges(unsigned int index, struct mtrr_var_range *vr)
 	if ((vr->base_lo & 0xfffff0ffUL) != (lo & 0xfffff0ffUL)
 	    || (vr->base_hi & (size_and_mask >> (32 - PAGE_SHIFT))) !=
 		(hi & (size_and_mask >> (32 - PAGE_SHIFT)))) {
+
 		mtrr_wrmsr(MTRRphysBase_MSR(index), vr->base_lo, vr->base_hi);
 		changed = true;
 	}
@@ -526,21 +542,26 @@ static u32 deftype_lo, deftype_hi;
  */
 static unsigned long set_mtrr_state(void)
 {
-	unsigned int i;
 	unsigned long change_mask = 0;
+	unsigned int i;
 
-	for (i = 0; i < num_var_ranges; i++)
+	for (i = 0; i < num_var_ranges; i++) {
 		if (set_mtrr_var_ranges(i, &mtrr_state.var_ranges[i]))
 			change_mask |= MTRR_CHANGE_MASK_VARIABLE;
+	}
 
 	if (mtrr_state.have_fixed && set_fixed_ranges(mtrr_state.fixed_ranges))
 		change_mask |= MTRR_CHANGE_MASK_FIXED;
 
-	/*  Set_mtrr_restore restores the old value of MTRRdefType,
-	   so to set it we fiddle with the saved value  */
+	/*
+	 * Set_mtrr_restore restores the old value of MTRRdefType,
+	 * so to set it we fiddle with the saved value:
+	 */
 	if ((deftype_lo & 0xff) != mtrr_state.def_type
 	    || ((deftype_lo & 0xc00) >> 10) != mtrr_state.enabled) {
-		deftype_lo = (deftype_lo & ~0xcff) | mtrr_state.def_type | (mtrr_state.enabled << 10);
+
+		deftype_lo = (deftype_lo & ~0xcff) | mtrr_state.def_type |
+			     (mtrr_state.enabled << 10);
 		change_mask |= MTRR_CHANGE_MASK_DEFTYPE;
 	}
 
@@ -548,33 +569,36 @@ static unsigned long set_mtrr_state(void)
 }
 
 
-static unsigned long cr4 = 0;
+static unsigned long cr4;
 static DEFINE_SPINLOCK(set_atomicity_lock);
 
 /*
- * Since we are disabling the cache don't allow any interrupts - they
- * would run extremely slow and would only increase the pain.  The caller must
- * ensure that local interrupts are disabled and are reenabled after post_set()
- * has been called.
+ * Since we are disabling the cache don't allow any interrupts,
+ * they would run extremely slow and would only increase the pain.
+ *
+ * The caller must ensure that local interrupts are disabled and
+ * are reenabled after post_set() has been called.
  */
-
 static void prepare_set(void) __acquires(set_atomicity_lock)
 {
 	unsigned long cr0;
 
-	/*  Note that this is not ideal, since the cache is only flushed/disabled
-	   for this CPU while the MTRRs are changed, but changing this requires
-	   more invasive changes to the way the kernel boots  */
+	/*
+	 * Note that this is not ideal
+	 * since the cache is only flushed/disabled for this CPU while the
+	 * MTRRs are changed, but changing this requires more invasive
+	 * changes to the way the kernel boots
+	 */
 
 	spin_lock(&set_atomicity_lock);
 
-	/*  Enter the no-fill (CD=1, NW=0) cache mode and flush caches. */
+	/* Enter the no-fill (CD=1, NW=0) cache mode and flush caches. */
 	cr0 = read_cr0() | X86_CR0_CD;
 	write_cr0(cr0);
 	wbinvd();
 
-	/*  Save value of CR4 and clear Page Global Enable (bit 7)  */
-	if ( cpu_has_pge ) {
+	/* Save value of CR4 and clear Page Global Enable (bit 7) */
+	if (cpu_has_pge) {
 		cr4 = read_cr4();
 		write_cr4(cr4 & ~X86_CR4_PGE);
 	}
@@ -582,26 +606,26 @@ static void prepare_set(void) __acquires(set_atomicity_lock)
 	/* Flush all TLBs via a mov %cr3, %reg; mov %reg, %cr3 */
 	__flush_tlb();
 
-	/*  Save MTRR state */
+	/* Save MTRR state */
 	rdmsr(MSR_MTRRdefType, deftype_lo, deftype_hi);
 
-	/*  Disable MTRRs, and set the default type to uncached  */
+	/* Disable MTRRs, and set the default type to uncached */
 	mtrr_wrmsr(MSR_MTRRdefType, deftype_lo & ~0xcff, deftype_hi);
 }
 
 static void post_set(void) __releases(set_atomicity_lock)
 {
-	/*  Flush TLBs (no need to flush caches - they are disabled)  */
+	/* Flush TLBs (no need to flush caches - they are disabled) */
 	__flush_tlb();
 
 	/* Intel (P6) standard MTRRs */
 	mtrr_wrmsr(MSR_MTRRdefType, deftype_lo, deftype_hi);
-		
-	/*  Enable caches  */
+
+	/* Enable caches */
 	write_cr0(read_cr0() & 0xbfffffff);
 
-	/*  Restore value of CR4  */
-	if ( cpu_has_pge )
+	/* Restore value of CR4 */
+	if (cpu_has_pge)
 		write_cr4(cr4);
 	spin_unlock(&set_atomicity_lock);
 }
@@ -623,24 +647,27 @@ static void generic_set_all(void)
 	post_set();
 	local_irq_restore(flags);
 
-	/*  Use the atomic bitops to update the global mask  */
+	/* Use the atomic bitops to update the global mask */
 	for (count = 0; count < sizeof mask * 8; ++count) {
 		if (mask & 0x01)
 			set_bit(count, &smp_changes_mask);
 		mask >>= 1;
 	}
-	
+
 }
 
+/**
+ * generic_set_mtrr - set variable MTRR register on the local CPU.
+ *
+ * @reg: The register to set.
+ * @base: The base address of the region.
+ * @size: The size of the region. If this is 0 the region is disabled.
+ * @type: The type of the region.
+ *
+ * Returns nothing.
+ */
 static void generic_set_mtrr(unsigned int reg, unsigned long base,
 			     unsigned long size, mtrr_type type)
-/*  [SUMMARY] Set variable MTRR register on the local CPU.
-    <reg> The register to set.
-    <base> The base address of the region.
-    <size> The size of the region. If this is 0 the region is disabled.
-    <type> The type of the region.
-    [RETURNS] Nothing.
-*/
 {
 	unsigned long flags;
 	struct mtrr_var_range *vr;
@@ -651,8 +678,10 @@ static void generic_set_mtrr(unsigned int reg, unsigned long base,
 	prepare_set();
 
 	if (size == 0) {
-		/* The invalid bit is kept in the mask, so we simply clear the
-		   relevant mask register to disable a range. */
+		/*
+		 * The invalid bit is kept in the mask, so we simply
+		 * clear the relevant mask register to disable a range.
+		 */
 		mtrr_wrmsr(MTRRphysMask_MSR(reg), 0, 0);
 		memset(vr, 0, sizeof(struct mtrr_var_range));
 	} else {
@@ -669,46 +698,50 @@ static void generic_set_mtrr(unsigned int reg, unsigned long base,
 	local_irq_restore(flags);
 }
 
-int generic_validate_add_page(unsigned long base, unsigned long size, unsigned int type)
+int generic_validate_add_page(unsigned long base, unsigned long size,
+			      unsigned int type)
 {
 	unsigned long lbase, last;
 
-	/*  For Intel PPro stepping <= 7, must be 4 MiB aligned 
-	    and not touch 0x70000000->0x7003FFFF */
+	/*
+	 * For Intel PPro stepping <= 7
+	 * must be 4 MiB aligned and not touch 0x70000000 -> 0x7003FFFF
+	 */
 	if (is_cpu(INTEL) && boot_cpu_data.x86 == 6 &&
 	    boot_cpu_data.x86_model == 1 &&
 	    boot_cpu_data.x86_mask <= 7) {
 		if (base & ((1 << (22 - PAGE_SHIFT)) - 1)) {
-			printk(KERN_WARNING "mtrr: base(0x%lx000) is not 4 MiB aligned\n", base);
+			pr_warning("mtrr: base(0x%lx000) is not 4 MiB aligned\n", base);
 			return -EINVAL;
 		}
 		if (!(base + size < 0x70000 || base > 0x7003F) &&
 		    (type == MTRR_TYPE_WRCOMB
 		     || type == MTRR_TYPE_WRBACK)) {
-			printk(KERN_WARNING "mtrr: writable mtrr between 0x70000000 and 0x7003FFFF may hang the CPU.\n");
+			pr_warning("mtrr: writable mtrr between 0x70000000 and 0x7003FFFF may hang the CPU.\n");
 			return -EINVAL;
 		}
 	}
 
-	/*  Check upper bits of base and last are equal and lower bits are 0
-	    for base and 1 for last  */
+	/*
+	 * Check upper bits of base and last are equal and lower bits are 0
+	 * for base and 1 for last
+	 */
 	last = base + size - 1;
 	for (lbase = base; !(lbase & 1) && (last & 1);
-	     lbase = lbase >> 1, last = last >> 1) ;
+	     lbase = lbase >> 1, last = last >> 1)
+		;
 	if (lbase != last) {
-		printk(KERN_WARNING "mtrr: base(0x%lx000) is not aligned on a size(0x%lx000) boundary\n",
-		       base, size);
+		pr_warning("mtrr: base(0x%lx000) is not aligned on a size(0x%lx000) boundary\n", base, size);
 		return -EINVAL;
 	}
 	return 0;
 }
 
-
 static int generic_have_wrcomb(void)
 {
 	unsigned long config, dummy;
 	rdmsr(MSR_MTRRcap, config, dummy);
-	return (config & (1 << 10));
+	return config & (1 << 10);
 }
 
 int positive_have_wrcomb(void)
@@ -716,14 +749,15 @@ int positive_have_wrcomb(void)
 	return 1;
 }
 
-/* generic structure...
+/*
+ * Generic structure...
  */
 struct mtrr_ops generic_mtrr_ops = {
-	.use_intel_if      = 1,
-	.set_all	   = generic_set_all,
-	.get               = generic_get_mtrr,
-	.get_free_region   = generic_get_free_region,
-	.set               = generic_set_mtrr,
-	.validate_add_page = generic_validate_add_page,
-	.have_wrcomb       = generic_have_wrcomb,
+	.use_intel_if		= 1,
+	.set_all		= generic_set_all,
+	.get			= generic_get_mtrr,
+	.get_free_region	= generic_get_free_region,
+	.set			= generic_set_mtrr,
+	.validate_add_page	= generic_validate_add_page,
+	.have_wrcomb		= generic_have_wrcomb,
 };
-- 
cgit v1.2.3


From 26dc67eda19beafb7e5ef2770cec5b3ee5995a8e Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Sat, 4 Jul 2009 07:53:40 +0530
Subject: x86: Clean up mtrr/if.c

Fix:

  WARNING: Use #include <linux/uaccess.h> instead of <asm/uaccess.h>
  ERROR: trailing whitespace X 7
  ERROR: trailing statements should be on next line X 3
  WARNING: line over 80 characters X 5
  ERROR: space required before the open parenthesis '('

arch/x86/kernel/cpu/mtrr/if.o:

   text	   data	    bss	    dec	    hex	filename
   2239	      4	      0	   2243	    8c3	if.o.before
   2239	      4	      0	   2243	    8c3	if.o.after

md5:
   78d1f2aa4843ec6509c18e2dee54bc7f  if.o.before.asm
   78d1f2aa4843ec6509c18e2dee54bc7f  if.o.after.asm

Suggested-by: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <20090703164225.GA21447@elte.hu>
[ More cleanups to make the code more consistent. ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/mtrr/if.c | 135 ++++++++++++++++++++++++------------------
 1 file changed, 76 insertions(+), 59 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/mtrr/if.c b/arch/x86/kernel/cpu/mtrr/if.c
index fb73a52913a4..08b6ea4c62b4 100644
--- a/arch/x86/kernel/cpu/mtrr/if.c
+++ b/arch/x86/kernel/cpu/mtrr/if.c
@@ -1,27 +1,28 @@
-#include <linux/init.h>
-#include <linux/proc_fs.h>
 #include <linux/capability.h>
-#include <linux/ctype.h>
-#include <linux/module.h>
 #include <linux/seq_file.h>
-#include <asm/uaccess.h>
+#include <linux/uaccess.h>
+#include <linux/proc_fs.h>
+#include <linux/module.h>
+#include <linux/ctype.h>
+#include <linux/init.h>
 
 #define LINE_SIZE 80
 
 #include <asm/mtrr.h>
+
 #include "mtrr.h"
 
 #define FILE_FCOUNT(f) (((struct seq_file *)((f)->private_data))->private)
 
 static const char *const mtrr_strings[MTRR_NUM_TYPES] =
 {
-    "uncachable",               /* 0 */
-    "write-combining",          /* 1 */
-    "?",                        /* 2 */
-    "?",                        /* 3 */
-    "write-through",            /* 4 */
-    "write-protect",            /* 5 */
-    "write-back",               /* 6 */
+	"uncachable",		/* 0 */
+	"write-combining",	/* 1 */
+	"?",			/* 2 */
+	"?",			/* 3 */
+	"write-through",	/* 4 */
+	"write-protect",	/* 5 */
+	"write-back",		/* 6 */
 };
 
 const char *mtrr_attrib_to_str(int x)
@@ -35,8 +36,8 @@ static int
 mtrr_file_add(unsigned long base, unsigned long size,
 	      unsigned int type, bool increment, struct file *file, int page)
 {
+	unsigned int *fcount = FILE_FCOUNT(file);
 	int reg, max;
-	unsigned int *fcount = FILE_FCOUNT(file); 
 
 	max = num_var_ranges;
 	if (fcount == NULL) {
@@ -61,8 +62,8 @@ static int
 mtrr_file_del(unsigned long base, unsigned long size,
 	      struct file *file, int page)
 {
-	int reg;
 	unsigned int *fcount = FILE_FCOUNT(file);
+	int reg;
 
 	if (!page) {
 		if ((base & (PAGE_SIZE - 1)) || (size & (PAGE_SIZE - 1)))
@@ -81,13 +82,14 @@ mtrr_file_del(unsigned long base, unsigned long size,
 	return reg;
 }
 
-/* RED-PEN: seq_file can seek now. this is ignored. */
+/*
+ * seq_file can seek but we ignore it.
+ *
+ * Format of control line:
+ *    "base=%Lx size=%Lx type=%s" or "disable=%d"
+ */
 static ssize_t
 mtrr_write(struct file *file, const char __user *buf, size_t len, loff_t * ppos)
-/*  Format of control line:
-    "base=%Lx size=%Lx type=%s"     OR:
-    "disable=%d"
-*/
 {
 	int i, err;
 	unsigned long reg;
@@ -100,15 +102,18 @@ mtrr_write(struct file *file, const char __user *buf, size_t len, loff_t * ppos)
 		return -EPERM;
 	if (!len)
 		return -EINVAL;
+
 	memset(line, 0, LINE_SIZE);
 	if (len > LINE_SIZE)
 		len = LINE_SIZE;
 	if (copy_from_user(line, buf, len - 1))
 		return -EFAULT;
+
 	linelen = strlen(line);
 	ptr = line + linelen - 1;
 	if (linelen && *ptr == '\n')
 		*ptr = '\0';
+
 	if (!strncmp(line, "disable=", 8)) {
 		reg = simple_strtoul(line + 8, &ptr, 0);
 		err = mtrr_del_page(reg, 0, 0);
@@ -116,28 +121,35 @@ mtrr_write(struct file *file, const char __user *buf, size_t len, loff_t * ppos)
 			return err;
 		return len;
 	}
+
 	if (strncmp(line, "base=", 5))
 		return -EINVAL;
+
 	base = simple_strtoull(line + 5, &ptr, 0);
-	for (; isspace(*ptr); ++ptr) ;
+	for (; isspace(*ptr); ++ptr)
+		;
+
 	if (strncmp(ptr, "size=", 5))
 		return -EINVAL;
+
 	size = simple_strtoull(ptr + 5, &ptr, 0);
 	if ((base & 0xfff) || (size & 0xfff))
 		return -EINVAL;
-	for (; isspace(*ptr); ++ptr) ;
+	for (; isspace(*ptr); ++ptr)
+		;
+
 	if (strncmp(ptr, "type=", 5))
 		return -EINVAL;
 	ptr += 5;
-	for (; isspace(*ptr); ++ptr) ;
+	for (; isspace(*ptr); ++ptr)
+		;
+
 	for (i = 0; i < MTRR_NUM_TYPES; ++i) {
 		if (strcmp(ptr, mtrr_strings[i]))
 			continue;
 		base >>= PAGE_SHIFT;
 		size >>= PAGE_SHIFT;
-		err =
-		    mtrr_add_page((unsigned long) base, (unsigned long) size, i,
-				  true);
+		err = mtrr_add_page((unsigned long)base, (unsigned long)size, i, true);
 		if (err < 0)
 			return err;
 		return len;
@@ -181,7 +193,9 @@ mtrr_ioctl(struct file *file, unsigned int cmd, unsigned long __arg)
 	case MTRRIOC32_SET_PAGE_ENTRY:
 	case MTRRIOC32_DEL_PAGE_ENTRY:
 	case MTRRIOC32_KILL_PAGE_ENTRY: {
-		struct mtrr_sentry32 __user *s32 = (struct mtrr_sentry32 __user *)__arg;
+		struct mtrr_sentry32 __user *s32;
+
+		s32 = (struct mtrr_sentry32 __user *)__arg;
 		err = get_user(sentry.base, &s32->base);
 		err |= get_user(sentry.size, &s32->size);
 		err |= get_user(sentry.type, &s32->type);
@@ -191,7 +205,9 @@ mtrr_ioctl(struct file *file, unsigned int cmd, unsigned long __arg)
 	}
 	case MTRRIOC32_GET_ENTRY:
 	case MTRRIOC32_GET_PAGE_ENTRY: {
-		struct mtrr_gentry32 __user *g32 = (struct mtrr_gentry32 __user *)__arg;
+		struct mtrr_gentry32 __user *g32;
+
+		g32 = (struct mtrr_gentry32 __user *)__arg;
 		err = get_user(gentry.regnum, &g32->regnum);
 		err |= get_user(gentry.base, &g32->base);
 		err |= get_user(gentry.size, &g32->size);
@@ -314,7 +330,7 @@ mtrr_ioctl(struct file *file, unsigned int cmd, unsigned long __arg)
 	if (err)
 		return err;
 
-	switch(cmd) {
+	switch (cmd) {
 	case MTRRIOC_GET_ENTRY:
 	case MTRRIOC_GET_PAGE_ENTRY:
 		if (copy_to_user(arg, &gentry, sizeof gentry))
@@ -323,7 +339,9 @@ mtrr_ioctl(struct file *file, unsigned int cmd, unsigned long __arg)
 #ifdef CONFIG_COMPAT
 	case MTRRIOC32_GET_ENTRY:
 	case MTRRIOC32_GET_PAGE_ENTRY: {
-		struct mtrr_gentry32 __user *g32 = (struct mtrr_gentry32 __user *)__arg;
+		struct mtrr_gentry32 __user *g32;
+
+		g32 = (struct mtrr_gentry32 __user *)__arg;
 		err = put_user(gentry.base, &g32->base);
 		err |= put_user(gentry.size, &g32->size);
 		err |= put_user(gentry.regnum, &g32->regnum);
@@ -335,11 +353,10 @@ mtrr_ioctl(struct file *file, unsigned int cmd, unsigned long __arg)
 	return err;
 }
 
-static int
-mtrr_close(struct inode *ino, struct file *file)
+static int mtrr_close(struct inode *ino, struct file *file)
 {
-	int i, max;
 	unsigned int *fcount = FILE_FCOUNT(file);
+	int i, max;
 
 	if (fcount != NULL) {
 		max = num_var_ranges;
@@ -359,22 +376,22 @@ static int mtrr_seq_show(struct seq_file *seq, void *offset);
 
 static int mtrr_open(struct inode *inode, struct file *file)
 {
-	if (!mtrr_if) 
+	if (!mtrr_if)
 		return -EIO;
-	if (!mtrr_if->get) 
-		return -ENXIO; 
+	if (!mtrr_if->get)
+		return -ENXIO;
 	return single_open(file, mtrr_seq_show, NULL);
 }
 
 static const struct file_operations mtrr_fops = {
-	.owner   = THIS_MODULE,
-	.open	 = mtrr_open, 
-	.read    = seq_read,
-	.llseek  = seq_lseek,
-	.write   = mtrr_write,
-	.unlocked_ioctl = mtrr_ioctl,
-	.compat_ioctl = mtrr_ioctl,
-	.release = mtrr_close,
+	.owner			= THIS_MODULE,
+	.open			= mtrr_open,
+	.read			= seq_read,
+	.llseek			= seq_lseek,
+	.write			= mtrr_write,
+	.unlocked_ioctl		= mtrr_ioctl,
+	.compat_ioctl		= mtrr_ioctl,
+	.release		= mtrr_close,
 };
 
 static int mtrr_seq_show(struct seq_file *seq, void *offset)
@@ -388,23 +405,24 @@ static int mtrr_seq_show(struct seq_file *seq, void *offset)
 	max = num_var_ranges;
 	for (i = 0; i < max; i++) {
 		mtrr_if->get(i, &base, &size, &type);
-		if (size == 0)
+		if (size == 0) {
 			mtrr_usage_table[i] = 0;
-		else {
-			if (size < (0x100000 >> PAGE_SHIFT)) {
-				/* less than 1MB */
-				factor = 'K';
-				size <<= PAGE_SHIFT - 10;
-			} else {
-				factor = 'M';
-				size >>= 20 - PAGE_SHIFT;
-			}
-			/* RED-PEN: base can be > 32bit */ 
-			len += seq_printf(seq, 
-				   "reg%02i: base=0x%06lx000 (%5luMB), size=%5lu%cB, count=%d: %s\n",
-			     i, base, base >> (20 - PAGE_SHIFT), size, factor,
-			     mtrr_usage_table[i], mtrr_attrib_to_str(type));
+			continue;
 		}
+		if (size < (0x100000 >> PAGE_SHIFT)) {
+			/* less than 1MB */
+			factor = 'K';
+			size <<= PAGE_SHIFT - 10;
+		} else {
+			factor = 'M';
+			size >>= 20 - PAGE_SHIFT;
+		}
+		/* Base can be > 32bit */
+		len += seq_printf(seq, "reg%02i: base=0x%06lx000 "
+			"(%5luMB), size=%5lu%cB, count=%d: %s\n",
+			i, base, base >> (20 - PAGE_SHIFT), size,
+			factor, mtrr_usage_table[i],
+			mtrr_attrib_to_str(type));
 	}
 	return 0;
 }
@@ -422,6 +440,5 @@ static int __init mtrr_if_init(void)
 	proc_create("mtrr", S_IWUSR | S_IRUGO, NULL, &mtrr_fops);
 	return 0;
 }
-
 arch_initcall(mtrr_if_init);
 #endif			/*  CONFIG_PROC_FS  */
-- 
cgit v1.2.3


From 3ec8dbcb09bb6df83993ca03e88cb85e3aaa8edb Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Sat, 4 Jul 2009 07:54:16 +0530
Subject: x86: Clean up mtrr/mtrr.h

Fix:

  ERROR: do not use C99 // comments
  ERROR: "foo * bar" should be "foo *bar" X 2

Suggested-by: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <20090703164225.GA21447@elte.hu>
[ More tidyups ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/mtrr/mtrr.h | 19 +++++++++----------
 1 file changed, 9 insertions(+), 10 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/mtrr/mtrr.h b/arch/x86/kernel/cpu/mtrr/mtrr.h
index 7538b767f206..a501dee9a87a 100644
--- a/arch/x86/kernel/cpu/mtrr/mtrr.h
+++ b/arch/x86/kernel/cpu/mtrr/mtrr.h
@@ -1,5 +1,5 @@
 /*
- * local mtrr defines.
+ * local MTRR defines.
  */
 
 #include <linux/types.h>
@@ -14,13 +14,12 @@ extern unsigned int mtrr_usage_table[MTRR_MAX_VAR_RANGES];
 struct mtrr_ops {
 	u32	vendor;
 	u32	use_intel_if;
-//	void	(*init)(void);
 	void	(*set)(unsigned int reg, unsigned long base,
 		       unsigned long size, mtrr_type type);
 	void	(*set_all)(void);
 
 	void	(*get)(unsigned int reg, unsigned long *base,
-		       unsigned long *size, mtrr_type * type);
+		       unsigned long *size, mtrr_type *type);
 	int	(*get_free_region)(unsigned long base, unsigned long size,
 				   int replace_reg);
 	int	(*validate_add_page)(unsigned long base, unsigned long size,
@@ -39,11 +38,11 @@ extern int positive_have_wrcomb(void);
 
 /* library functions for processor-specific routines */
 struct set_mtrr_context {
-	unsigned long flags;
-	unsigned long cr4val;
-	u32 deftype_lo;
-	u32 deftype_hi;
-	u32 ccr3;
+	unsigned long	flags;
+	unsigned long	cr4val;
+	u32		deftype_lo;
+	u32		deftype_hi;
+	u32		ccr3;
 };
 
 void set_mtrr_done(struct set_mtrr_context *ctxt);
@@ -54,10 +53,10 @@ void fill_mtrr_var_range(unsigned int index,
 		u32 base_lo, u32 base_hi, u32 mask_lo, u32 mask_hi);
 void get_mtrr_state(void);
 
-extern void set_mtrr_ops(struct mtrr_ops * ops);
+extern void set_mtrr_ops(struct mtrr_ops *ops);
 
 extern u64 size_or_mask, size_and_mask;
-extern struct mtrr_ops * mtrr_if;
+extern struct mtrr_ops *mtrr_if;
 
 #define is_cpu(vnd)	(mtrr_if && mtrr_if->vendor == X86_VENDOR_##vnd)
 #define use_intel()	(mtrr_if && mtrr_if->use_intel_if == 1)
-- 
cgit v1.2.3


From 09b22c85d59dd935fdfa71655a443785e3f99c18 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Sat, 4 Jul 2009 07:54:53 +0530
Subject: x86: Clean up mtrr/state.c

Fix:

  WARNING: Use #include <linux/io.h> instead of <asm/io.h>
  WARNING: line over 80 characters X 4

arch/x86/kernel/cpu/mtrr/state.o:

   text	   data	    bss	    dec	    hex	filename
    864	      0	      0	    864	    360	state.o.before
    864	      0	      0	    864	    360	state.o.after

md5:
   c5c4364b9aeac74d70111e1e49667a2c  state.o.before.asm
   c5c4364b9aeac74d70111e1e49667a2c  state.o.after.asm

Suggested-by: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <20090703164225.GA21447@elte.hu>
[ More cleanups ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/mtrr/state.c | 68 +++++++++++++++++++++++-----------------
 1 file changed, 40 insertions(+), 28 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/mtrr/state.c b/arch/x86/kernel/cpu/mtrr/state.c
index 1f5fb1588d1f..dfc80b4e6b0d 100644
--- a/arch/x86/kernel/cpu/mtrr/state.c
+++ b/arch/x86/kernel/cpu/mtrr/state.c
@@ -1,24 +1,25 @@
-#include <linux/mm.h>
 #include <linux/init.h>
-#include <asm/io.h>
-#include <asm/mtrr.h>
-#include <asm/msr.h>
+#include <linux/io.h>
+#include <linux/mm.h>
+
 #include <asm/processor-cyrix.h>
 #include <asm/processor-flags.h>
-#include "mtrr.h"
+#include <asm/mtrr.h>
+#include <asm/msr.h>
 
+#include "mtrr.h"
 
-/*  Put the processor into a state where MTRRs can be safely set  */
+/* Put the processor into a state where MTRRs can be safely set */
 void set_mtrr_prepare_save(struct set_mtrr_context *ctxt)
 {
 	unsigned int cr0;
 
-	/*  Disable interrupts locally  */
+	/* Disable interrupts locally */
 	local_irq_save(ctxt->flags);
 
 	if (use_intel() || is_cpu(CYRIX)) {
 
-		/*  Save value of CR4 and clear Page Global Enable (bit 7)  */
+		/* Save value of CR4 and clear Page Global Enable (bit 7) */
 		if (cpu_has_pge) {
 			ctxt->cr4val = read_cr4();
 			write_cr4(ctxt->cr4val & ~X86_CR4_PGE);
@@ -33,50 +34,61 @@ void set_mtrr_prepare_save(struct set_mtrr_context *ctxt)
 		write_cr0(cr0);
 		wbinvd();
 
-		if (use_intel())
-			/*  Save MTRR state */
+		if (use_intel()) {
+			/* Save MTRR state */
 			rdmsr(MSR_MTRRdefType, ctxt->deftype_lo, ctxt->deftype_hi);
-		else
-			/* Cyrix ARRs - everything else were excluded at the top */
+		} else {
+			/*
+			 * Cyrix ARRs -
+			 * everything else were excluded at the top
+			 */
 			ctxt->ccr3 = getCx86(CX86_CCR3);
+		}
 	}
 }
 
 void set_mtrr_cache_disable(struct set_mtrr_context *ctxt)
 {
-	if (use_intel())
-		/*  Disable MTRRs, and set the default type to uncached  */
+	if (use_intel()) {
+		/* Disable MTRRs, and set the default type to uncached */
 		mtrr_wrmsr(MSR_MTRRdefType, ctxt->deftype_lo & 0xf300UL,
 		      ctxt->deftype_hi);
-	else if (is_cpu(CYRIX))
-		/* Cyrix ARRs - everything else were excluded at the top */
-		setCx86(CX86_CCR3, (ctxt->ccr3 & 0x0f) | 0x10);
+	} else {
+		if (is_cpu(CYRIX)) {
+			/* Cyrix ARRs - everything else were excluded at the top */
+			setCx86(CX86_CCR3, (ctxt->ccr3 & 0x0f) | 0x10);
+		}
+	}
 }
 
-/*  Restore the processor after a set_mtrr_prepare  */
+/* Restore the processor after a set_mtrr_prepare */
 void set_mtrr_done(struct set_mtrr_context *ctxt)
 {
 	if (use_intel() || is_cpu(CYRIX)) {
 
-		/*  Flush caches and TLBs  */
+		/* Flush caches and TLBs */
 		wbinvd();
 
-		/*  Restore MTRRdefType  */
-		if (use_intel())
+		/* Restore MTRRdefType */
+		if (use_intel()) {
 			/* Intel (P6) standard MTRRs */
-			mtrr_wrmsr(MSR_MTRRdefType, ctxt->deftype_lo, ctxt->deftype_hi);
-		else
-			/* Cyrix ARRs - everything else was excluded at the top */
+			mtrr_wrmsr(MSR_MTRRdefType, ctxt->deftype_lo,
+				   ctxt->deftype_hi);
+		} else {
+			/*
+			 * Cyrix ARRs -
+			 * everything else was excluded at the top
+			 */
 			setCx86(CX86_CCR3, ctxt->ccr3);
+		}
 
-		/*  Enable caches  */
+		/* Enable caches */
 		write_cr0(read_cr0() & 0xbfffffff);
 
-		/*  Restore value of CR4  */
+		/* Restore value of CR4 */
 		if (cpu_has_pge)
 			write_cr4(ctxt->cr4val);
 	}
-	/*  Re-enable interrupts locally (if enabled previously)  */
+	/* Re-enable interrupts locally (if enabled previously) */
 	local_irq_restore(ctxt->flags);
 }
-
-- 
cgit v1.2.3


From dbd51be026eaf84088fdee7fab9f38fa92eef26d Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Sat, 4 Jul 2009 07:56:28 +0530
Subject: x86: Clean up mtrr/main.c

Fix following trivial style problems:

  ERROR: trailing whitespace X 25
  WARNING: Use #include <linux/uaccess.h> instead of <asm/uaccess.h>
  WARNING: Use #include <linux/kvm_para.h> instead of <asm/kvm_para.h>
  ERROR: do not initialise externals to 0 or NULL X 2
  ERROR: "foo * bar" should be "foo *bar" X 5
  ERROR: do not use assignment in if condition X 2
  WARNING: line over 80 characters X 8
  ERROR: return is not a function, parentheses are not required
  WARNING: braces {} are not necessary for any arm of this statement
  ERROR: space required before the open parenthesis '(' X 2
  ERROR: open brace '{' following function declarations go on the next line
  ERROR: space required after that ',' (ctx:VxV) X 8
  ERROR: space required before the open parenthesis '(' X 3
  ERROR: else should follow close brace '}'
  WARNING: space prohibited between function name and open parenthesis '('
  WARNING: EXPORT_SYMBOL(foo); should immediately follow its function/variable X 2

Also use pr_debug and pr_warning where possible.

total: 50 errors, 14 warnings

arch/x86/kernel/cpu/mtrr/main.o:

   text	   data	    bss	    dec	    hex	filename
   3668	    116	   4156	   7940	   1f04	main.o.before
   3668	    116	   4156	   7940	   1f04	main.o.after

md5:
   e01af2fd28deef77c8d01e71acfbd365  main.o.before.asm
   e01af2fd28deef77c8d01e71acfbd365  main.o.after.asm

Suggested-by: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <20090703164225.GA21447@elte.hu>
Cc: Avi Kivity <avi@redhat.com> # Avi, please have a look at the kvm_para.h bit
[ More cleanups ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/mtrr/main.c | 455 +++++++++++++++++++++-------------------
 1 file changed, 242 insertions(+), 213 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c
index 8fc248b5aeaf..7af0f88a4163 100644
--- a/arch/x86/kernel/cpu/mtrr/main.c
+++ b/arch/x86/kernel/cpu/mtrr/main.c
@@ -25,43 +25,48 @@
     Operating System Writer's Guide" (Intel document number 242692),
     section 11.11.7
 
-    This was cleaned and made readable by Patrick Mochel <mochel@osdl.org> 
-    on 6-7 March 2002. 
-    Source: Intel Architecture Software Developers Manual, Volume 3: 
+    This was cleaned and made readable by Patrick Mochel <mochel@osdl.org>
+    on 6-7 March 2002.
+    Source: Intel Architecture Software Developers Manual, Volume 3:
     System Programming Guide; Section 9.11. (1997 edition - PPro).
 */
 
+#define DEBUG
+
+#include <linux/types.h> /* FIXME: kvm_para.h needs this */
+
+#include <linux/kvm_para.h>
+#include <linux/uaccess.h>
 #include <linux/module.h>
+#include <linux/mutex.h>
 #include <linux/init.h>
+#include <linux/sort.h>
+#include <linux/cpu.h>
 #include <linux/pci.h>
 #include <linux/smp.h>
-#include <linux/cpu.h>
-#include <linux/mutex.h>
-#include <linux/sort.h>
 
+#include <asm/processor.h>
 #include <asm/e820.h>
 #include <asm/mtrr.h>
-#include <asm/uaccess.h>
-#include <asm/processor.h>
 #include <asm/msr.h>
-#include <asm/kvm_para.h>
+
 #include "mtrr.h"
 
-u32 num_var_ranges = 0;
+u32 num_var_ranges;
 
 unsigned int mtrr_usage_table[MTRR_MAX_VAR_RANGES];
 static DEFINE_MUTEX(mtrr_mutex);
 
 u64 size_or_mask, size_and_mask;
 
-static struct mtrr_ops * mtrr_ops[X86_VENDOR_NUM] = {};
+static struct mtrr_ops *mtrr_ops[X86_VENDOR_NUM];
 
-struct mtrr_ops * mtrr_if = NULL;
+struct mtrr_ops *mtrr_if;
 
 static void set_mtrr(unsigned int reg, unsigned long base,
 		     unsigned long size, mtrr_type type);
 
-void set_mtrr_ops(struct mtrr_ops * ops)
+void set_mtrr_ops(struct mtrr_ops *ops)
 {
 	if (ops->vendor && ops->vendor < X86_VENDOR_NUM)
 		mtrr_ops[ops->vendor] = ops;
@@ -72,30 +77,36 @@ static int have_wrcomb(void)
 {
 	struct pci_dev *dev;
 	u8 rev;
-	
-	if ((dev = pci_get_class(PCI_CLASS_BRIDGE_HOST << 8, NULL)) != NULL) {
-		/* ServerWorks LE chipsets < rev 6 have problems with write-combining
-		   Don't allow it and leave room for other chipsets to be tagged */
+
+	dev = pci_get_class(PCI_CLASS_BRIDGE_HOST << 8, NULL);
+	if (dev != NULL) {
+		/*
+		 * ServerWorks LE chipsets < rev 6 have problems with
+		 * write-combining. Don't allow it and leave room for other
+		 * chipsets to be tagged
+		 */
 		if (dev->vendor == PCI_VENDOR_ID_SERVERWORKS &&
 		    dev->device == PCI_DEVICE_ID_SERVERWORKS_LE) {
 			pci_read_config_byte(dev, PCI_CLASS_REVISION, &rev);
 			if (rev <= 5) {
-				printk(KERN_INFO "mtrr: Serverworks LE rev < 6 detected. Write-combining disabled.\n");
+				pr_info("mtrr: Serverworks LE rev < 6 detected. Write-combining disabled.\n");
 				pci_dev_put(dev);
 				return 0;
 			}
 		}
-		/* Intel 450NX errata # 23. Non ascending cacheline evictions to
-		   write combining memory may resulting in data corruption */
+		/*
+		 * Intel 450NX errata # 23. Non ascending cacheline evictions to
+		 * write combining memory may resulting in data corruption
+		 */
 		if (dev->vendor == PCI_VENDOR_ID_INTEL &&
 		    dev->device == PCI_DEVICE_ID_INTEL_82451NX) {
-			printk(KERN_INFO "mtrr: Intel 450NX MMC detected. Write-combining disabled.\n");
+			pr_info("mtrr: Intel 450NX MMC detected. Write-combining disabled.\n");
 			pci_dev_put(dev);
 			return 0;
 		}
 		pci_dev_put(dev);
-	}		
-	return (mtrr_if->have_wrcomb ? mtrr_if->have_wrcomb() : 0);
+	}
+	return mtrr_if->have_wrcomb ? mtrr_if->have_wrcomb() : 0;
 }
 
 /*  This function returns the number of variable MTRRs  */
@@ -103,12 +114,13 @@ static void __init set_num_var_ranges(void)
 {
 	unsigned long config = 0, dummy;
 
-	if (use_intel()) {
+	if (use_intel())
 		rdmsr(MSR_MTRRcap, config, dummy);
-	} else if (is_cpu(AMD))
+	else if (is_cpu(AMD))
 		config = 2;
 	else if (is_cpu(CYRIX) || is_cpu(CENTAUR))
 		config = 8;
+
 	num_var_ranges = config & 0xff;
 }
 
@@ -130,10 +142,12 @@ struct set_mtrr_data {
 	mtrr_type	smp_type;
 };
 
+/**
+ * ipi_handler - Synchronisation handler. Executed by "other" CPUs.
+ *
+ * Returns nothing.
+ */
 static void ipi_handler(void *info)
-/*  [SUMMARY] Synchronisation handler. Executed by "other" CPUs.
-    [RETURNS] Nothing.
-*/
 {
 #ifdef CONFIG_SMP
 	struct set_mtrr_data *data = info;
@@ -142,18 +156,19 @@ static void ipi_handler(void *info)
 	local_irq_save(flags);
 
 	atomic_dec(&data->count);
-	while(!atomic_read(&data->gate))
+	while (!atomic_read(&data->gate))
 		cpu_relax();
 
 	/*  The master has cleared me to execute  */
-	if (data->smp_reg != ~0U) 
-		mtrr_if->set(data->smp_reg, data->smp_base, 
+	if (data->smp_reg != ~0U) {
+		mtrr_if->set(data->smp_reg, data->smp_base,
 			     data->smp_size, data->smp_type);
-	else
+	} else {
 		mtrr_if->set_all();
+	}
 
 	atomic_dec(&data->count);
-	while(atomic_read(&data->gate))
+	while (atomic_read(&data->gate))
 		cpu_relax();
 
 	atomic_dec(&data->count);
@@ -161,7 +176,8 @@ static void ipi_handler(void *info)
 #endif
 }
 
-static inline int types_compatible(mtrr_type type1, mtrr_type type2) {
+static inline int types_compatible(mtrr_type type1, mtrr_type type2)
+{
 	return type1 == MTRR_TYPE_UNCACHABLE ||
 	       type2 == MTRR_TYPE_UNCACHABLE ||
 	       (type1 == MTRR_TYPE_WRTHROUGH && type2 == MTRR_TYPE_WRBACK) ||
@@ -176,10 +192,10 @@ static inline int types_compatible(mtrr_type type1, mtrr_type type2) {
  * @type:	mtrr type
  *
  * This is kinda tricky, but fortunately, Intel spelled it out for us cleanly:
- * 
+ *
  * 1. Send IPI to do the following:
  * 2. Disable Interrupts
- * 3. Wait for all procs to do so 
+ * 3. Wait for all procs to do so
  * 4. Enter no-fill cache mode
  * 5. Flush caches
  * 6. Clear PGE bit
@@ -189,26 +205,27 @@ static inline int types_compatible(mtrr_type type1, mtrr_type type2) {
  * 10. Enable all range registers
  * 11. Flush all TLBs and caches again
  * 12. Enter normal cache mode and reenable caching
- * 13. Set PGE 
+ * 13. Set PGE
  * 14. Wait for buddies to catch up
  * 15. Enable interrupts.
- * 
+ *
  * What does that mean for us? Well, first we set data.count to the number
  * of CPUs. As each CPU disables interrupts, it'll decrement it once. We wait
  * until it hits 0 and proceed. We set the data.gate flag and reset data.count.
- * Meanwhile, they are waiting for that flag to be set. Once it's set, each 
- * CPU goes through the transition of updating MTRRs. The CPU vendors may each do it 
- * differently, so we call mtrr_if->set() callback and let them take care of it.
- * When they're done, they again decrement data->count and wait for data.gate to 
- * be reset. 
- * When we finish, we wait for data.count to hit 0 and toggle the data.gate flag.
+ * Meanwhile, they are waiting for that flag to be set. Once it's set, each
+ * CPU goes through the transition of updating MTRRs.
+ * The CPU vendors may each do it differently,
+ * so we call mtrr_if->set() callback and let them take care of it.
+ * When they're done, they again decrement data->count and wait for data.gate
+ * to be reset.
+ * When we finish, we wait for data.count to hit 0 and toggle the data.gate flag
  * Everyone then enables interrupts and we all continue on.
  *
  * Note that the mechanism is the same for UP systems, too; all the SMP stuff
  * becomes nops.
  */
-static void set_mtrr(unsigned int reg, unsigned long base,
-		     unsigned long size, mtrr_type type)
+static void
+set_mtrr(unsigned int reg, unsigned long base, unsigned long size, mtrr_type type)
 {
 	struct set_mtrr_data data;
 	unsigned long flags;
@@ -218,121 +235,122 @@ static void set_mtrr(unsigned int reg, unsigned long base,
 	data.smp_size = size;
 	data.smp_type = type;
 	atomic_set(&data.count, num_booting_cpus() - 1);
-	/* make sure data.count is visible before unleashing other CPUs */
+
+	/* Make sure data.count is visible before unleashing other CPUs */
 	smp_wmb();
-	atomic_set(&data.gate,0);
+	atomic_set(&data.gate, 0);
 
-	/*  Start the ball rolling on other CPUs  */
+	/* Start the ball rolling on other CPUs */
 	if (smp_call_function(ipi_handler, &data, 0) != 0)
 		panic("mtrr: timed out waiting for other CPUs\n");
 
 	local_irq_save(flags);
 
-	while(atomic_read(&data.count))
+	while (atomic_read(&data.count))
 		cpu_relax();
 
-	/* ok, reset count and toggle gate */
+	/* Ok, reset count and toggle gate */
 	atomic_set(&data.count, num_booting_cpus() - 1);
 	smp_wmb();
-	atomic_set(&data.gate,1);
+	atomic_set(&data.gate, 1);
 
-	/* do our MTRR business */
+	/* Do our MTRR business */
 
-	/* HACK!
+	/*
+	 * HACK!
 	 * We use this same function to initialize the mtrrs on boot.
 	 * The state of the boot cpu's mtrrs has been saved, and we want
-	 * to replicate across all the APs. 
+	 * to replicate across all the APs.
 	 * If we're doing that @reg is set to something special...
 	 */
-	if (reg != ~0U) 
-		mtrr_if->set(reg,base,size,type);
+	if (reg != ~0U)
+		mtrr_if->set(reg, base, size, type);
 
-	/* wait for the others */
-	while(atomic_read(&data.count))
+	/* Wait for the others */
+	while (atomic_read(&data.count))
 		cpu_relax();
 
 	atomic_set(&data.count, num_booting_cpus() - 1);
 	smp_wmb();
-	atomic_set(&data.gate,0);
+	atomic_set(&data.gate, 0);
 
 	/*
 	 * Wait here for everyone to have seen the gate change
 	 * So we're the last ones to touch 'data'
 	 */
-	while(atomic_read(&data.count))
+	while (atomic_read(&data.count))
 		cpu_relax();
 
 	local_irq_restore(flags);
 }
 
 /**
- *	mtrr_add_page - Add a memory type region
- *	@base: Physical base address of region in pages (in units of 4 kB!)
- *	@size: Physical size of region in pages (4 kB)
- *	@type: Type of MTRR desired
- *	@increment: If this is true do usage counting on the region
+ * mtrr_add_page - Add a memory type region
+ * @base: Physical base address of region in pages (in units of 4 kB!)
+ * @size: Physical size of region in pages (4 kB)
+ * @type: Type of MTRR desired
+ * @increment: If this is true do usage counting on the region
  *
- *	Memory type region registers control the caching on newer Intel and
- *	non Intel processors. This function allows drivers to request an
- *	MTRR is added. The details and hardware specifics of each processor's
- *	implementation are hidden from the caller, but nevertheless the 
- *	caller should expect to need to provide a power of two size on an
- *	equivalent power of two boundary.
+ * Memory type region registers control the caching on newer Intel and
+ * non Intel processors. This function allows drivers to request an
+ * MTRR is added. The details and hardware specifics of each processor's
+ * implementation are hidden from the caller, but nevertheless the
+ * caller should expect to need to provide a power of two size on an
+ * equivalent power of two boundary.
  *
- *	If the region cannot be added either because all regions are in use
- *	or the CPU cannot support it a negative value is returned. On success
- *	the register number for this entry is returned, but should be treated
- *	as a cookie only.
+ * If the region cannot be added either because all regions are in use
+ * or the CPU cannot support it a negative value is returned. On success
+ * the register number for this entry is returned, but should be treated
+ * as a cookie only.
  *
- *	On a multiprocessor machine the changes are made to all processors.
- *	This is required on x86 by the Intel processors.
+ * On a multiprocessor machine the changes are made to all processors.
+ * This is required on x86 by the Intel processors.
  *
- *	The available types are
+ * The available types are
  *
- *	%MTRR_TYPE_UNCACHABLE	-	No caching
+ * %MTRR_TYPE_UNCACHABLE - No caching
  *
- *	%MTRR_TYPE_WRBACK	-	Write data back in bursts whenever
+ * %MTRR_TYPE_WRBACK - Write data back in bursts whenever
  *
- *	%MTRR_TYPE_WRCOMB	-	Write data back soon but allow bursts
+ * %MTRR_TYPE_WRCOMB - Write data back soon but allow bursts
  *
- *	%MTRR_TYPE_WRTHROUGH	-	Cache reads but not writes
+ * %MTRR_TYPE_WRTHROUGH - Cache reads but not writes
  *
- *	BUGS: Needs a quiet flag for the cases where drivers do not mind
- *	failures and do not wish system log messages to be sent.
+ * BUGS: Needs a quiet flag for the cases where drivers do not mind
+ * failures and do not wish system log messages to be sent.
  */
-
-int mtrr_add_page(unsigned long base, unsigned long size, 
+int mtrr_add_page(unsigned long base, unsigned long size,
 		  unsigned int type, bool increment)
 {
+	unsigned long lbase, lsize;
 	int i, replace, error;
 	mtrr_type ltype;
-	unsigned long lbase, lsize;
 
 	if (!mtrr_if)
 		return -ENXIO;
-		
-	if ((error = mtrr_if->validate_add_page(base,size,type)))
+
+	error = mtrr_if->validate_add_page(base, size, type);
+	if (error)
 		return error;
 
 	if (type >= MTRR_NUM_TYPES) {
-		printk(KERN_WARNING "mtrr: type: %u invalid\n", type);
+		pr_warning("mtrr: type: %u invalid\n", type);
 		return -EINVAL;
 	}
 
-	/*  If the type is WC, check that this processor supports it  */
+	/* If the type is WC, check that this processor supports it */
 	if ((type == MTRR_TYPE_WRCOMB) && !have_wrcomb()) {
-		printk(KERN_WARNING
-		       "mtrr: your processor doesn't support write-combining\n");
+		pr_warning("mtrr: your processor doesn't support write-combining\n");
 		return -ENOSYS;
 	}
 
 	if (!size) {
-		printk(KERN_WARNING "mtrr: zero sized request\n");
+		pr_warning("mtrr: zero sized request\n");
 		return -EINVAL;
 	}
 
 	if (base & size_or_mask || size & size_or_mask) {
-		printk(KERN_WARNING "mtrr: base or size exceeds the MTRR width\n");
+		pr_warning("mtrr: base or size exceeds the MTRR width\n");
 		return -EINVAL;
 	}
 
@@ -341,36 +359,40 @@ int mtrr_add_page(unsigned long base, unsigned long size,
 
 	/* No CPU hotplug when we change MTRR entries */
 	get_online_cpus();
-	/*  Search for existing MTRR  */
+
+	/* Search for existing MTRR  */
 	mutex_lock(&mtrr_mutex);
 	for (i = 0; i < num_var_ranges; ++i) {
 		mtrr_if->get(i, &lbase, &lsize, &ltype);
-		if (!lsize || base > lbase + lsize - 1 || base + size - 1 < lbase)
+		if (!lsize || base > lbase + lsize - 1 ||
+		    base + size - 1 < lbase)
 			continue;
-		/*  At this point we know there is some kind of overlap/enclosure  */
+		/*
+		 * At this point we know there is some kind of
+		 * overlap/enclosure
+		 */
 		if (base < lbase || base + size - 1 > lbase + lsize - 1) {
-			if (base <= lbase && base + size - 1 >= lbase + lsize - 1) {
+			if (base <= lbase &&
+			    base + size - 1 >= lbase + lsize - 1) {
 				/*  New region encloses an existing region  */
 				if (type == ltype) {
 					replace = replace == -1 ? i : -2;
 					continue;
-				}
-				else if (types_compatible(type, ltype))
+				} else if (types_compatible(type, ltype))
 					continue;
 			}
-			printk(KERN_WARNING
-			       "mtrr: 0x%lx000,0x%lx000 overlaps existing"
-			       " 0x%lx000,0x%lx000\n", base, size, lbase,
-			       lsize);
+			pr_warning("mtrr: 0x%lx000,0x%lx000 overlaps existing"
+				" 0x%lx000,0x%lx000\n", base, size, lbase,
+				lsize);
 			goto out;
 		}
-		/*  New region is enclosed by an existing region  */
+		/* New region is enclosed by an existing region */
 		if (ltype != type) {
 			if (types_compatible(type, ltype))
 				continue;
-			printk (KERN_WARNING "mtrr: type mismatch for %lx000,%lx000 old: %s new: %s\n",
-			     base, size, mtrr_attrib_to_str(ltype),
-			     mtrr_attrib_to_str(type));
+			pr_warning("mtrr: type mismatch for %lx000,%lx000 old: %s new: %s\n",
+				base, size, mtrr_attrib_to_str(ltype),
+				mtrr_attrib_to_str(type));
 			goto out;
 		}
 		if (increment)
@@ -378,7 +400,7 @@ int mtrr_add_page(unsigned long base, unsigned long size,
 		error = i;
 		goto out;
 	}
-	/*  Search for an empty MTRR  */
+	/* Search for an empty MTRR */
 	i = mtrr_if->get_free_region(base, size, replace);
 	if (i >= 0) {
 		set_mtrr(i, base, size, type);
@@ -393,8 +415,9 @@ int mtrr_add_page(unsigned long base, unsigned long size,
 				mtrr_usage_table[replace] = 0;
 			}
 		}
-	} else
-		printk(KERN_INFO "mtrr: no more MTRRs available\n");
+	} else {
+		pr_info("mtrr: no more MTRRs available\n");
+	}
 	error = i;
  out:
 	mutex_unlock(&mtrr_mutex);
@@ -405,10 +428,8 @@ int mtrr_add_page(unsigned long base, unsigned long size,
 static int mtrr_check(unsigned long base, unsigned long size)
 {
 	if ((base & (PAGE_SIZE - 1)) || (size & (PAGE_SIZE - 1))) {
-		printk(KERN_WARNING
-			"mtrr: size and base must be multiples of 4 kiB\n");
-		printk(KERN_DEBUG
-			"mtrr: size: 0x%lx  base: 0x%lx\n", size, base);
+		pr_warning("mtrr: size and base must be multiples of 4 kiB\n");
+		pr_debug("mtrr: size: 0x%lx  base: 0x%lx\n", size, base);
 		dump_stack();
 		return -1;
 	}
@@ -416,66 +437,64 @@ static int mtrr_check(unsigned long base, unsigned long size)
 }
 
 /**
- *	mtrr_add - Add a memory type region
- *	@base: Physical base address of region
- *	@size: Physical size of region
- *	@type: Type of MTRR desired
- *	@increment: If this is true do usage counting on the region
+ * mtrr_add - Add a memory type region
+ * @base: Physical base address of region
+ * @size: Physical size of region
+ * @type: Type of MTRR desired
+ * @increment: If this is true do usage counting on the region
  *
- *	Memory type region registers control the caching on newer Intel and
- *	non Intel processors. This function allows drivers to request an
- *	MTRR is added. The details and hardware specifics of each processor's
- *	implementation are hidden from the caller, but nevertheless the 
- *	caller should expect to need to provide a power of two size on an
- *	equivalent power of two boundary.
+ * Memory type region registers control the caching on newer Intel and
+ * non Intel processors. This function allows drivers to request an
+ * MTRR is added. The details and hardware specifics of each processor's
+ * implementation are hidden from the caller, but nevertheless the
+ * caller should expect to need to provide a power of two size on an
+ * equivalent power of two boundary.
  *
- *	If the region cannot be added either because all regions are in use
- *	or the CPU cannot support it a negative value is returned. On success
- *	the register number for this entry is returned, but should be treated
- *	as a cookie only.
+ * If the region cannot be added either because all regions are in use
+ * or the CPU cannot support it a negative value is returned. On success
+ * the register number for this entry is returned, but should be treated
+ * as a cookie only.
  *
- *	On a multiprocessor machine the changes are made to all processors.
- *	This is required on x86 by the Intel processors.
+ * On a multiprocessor machine the changes are made to all processors.
+ * This is required on x86 by the Intel processors.
  *
- *	The available types are
+ * The available types are
  *
- *	%MTRR_TYPE_UNCACHABLE	-	No caching
+ * %MTRR_TYPE_UNCACHABLE - No caching
  *
- *	%MTRR_TYPE_WRBACK	-	Write data back in bursts whenever
+ * %MTRR_TYPE_WRBACK - Write data back in bursts whenever
  *
- *	%MTRR_TYPE_WRCOMB	-	Write data back soon but allow bursts
+ * %MTRR_TYPE_WRCOMB - Write data back soon but allow bursts
  *
- *	%MTRR_TYPE_WRTHROUGH	-	Cache reads but not writes
+ * %MTRR_TYPE_WRTHROUGH - Cache reads but not writes
  *
- *	BUGS: Needs a quiet flag for the cases where drivers do not mind
- *	failures and do not wish system log messages to be sent.
+ * BUGS: Needs a quiet flag for the cases where drivers do not mind
+ * failures and do not wish system log messages to be sent.
  */
-
-int
-mtrr_add(unsigned long base, unsigned long size, unsigned int type,
-	 bool increment)
+int mtrr_add(unsigned long base, unsigned long size, unsigned int type,
+	     bool increment)
 {
 	if (mtrr_check(base, size))
 		return -EINVAL;
 	return mtrr_add_page(base >> PAGE_SHIFT, size >> PAGE_SHIFT, type,
 			     increment);
 }
+EXPORT_SYMBOL(mtrr_add);
 
 /**
- *	mtrr_del_page - delete a memory type region
- *	@reg: Register returned by mtrr_add
- *	@base: Physical base address
- *	@size: Size of region
+ * mtrr_del_page - delete a memory type region
+ * @reg: Register returned by mtrr_add
+ * @base: Physical base address
+ * @size: Size of region
  *
- *	If register is supplied then base and size are ignored. This is
- *	how drivers should call it.
+ * If register is supplied then base and size are ignored. This is
+ * how drivers should call it.
  *
- *	Releases an MTRR region. If the usage count drops to zero the 
- *	register is freed and the region returns to default state.
- *	On success the register is returned, on failure a negative error
- *	code.
+ * Releases an MTRR region. If the usage count drops to zero the
+ * register is freed and the region returns to default state.
+ * On success the register is returned, on failure a negative error
+ * code.
  */
-
 int mtrr_del_page(int reg, unsigned long base, unsigned long size)
 {
 	int i, max;
@@ -500,22 +519,22 @@ int mtrr_del_page(int reg, unsigned long base, unsigned long size)
 			}
 		}
 		if (reg < 0) {
-			printk(KERN_DEBUG "mtrr: no MTRR for %lx000,%lx000 found\n", base,
-			       size);
+			pr_debug("mtrr: no MTRR for %lx000,%lx000 found\n",
+				 base, size);
 			goto out;
 		}
 	}
 	if (reg >= max) {
-		printk(KERN_WARNING "mtrr: register: %d too big\n", reg);
+		pr_warning("mtrr: register: %d too big\n", reg);
 		goto out;
 	}
 	mtrr_if->get(reg, &lbase, &lsize, &ltype);
 	if (lsize < 1) {
-		printk(KERN_WARNING "mtrr: MTRR %d not used\n", reg);
+		pr_warning("mtrr: MTRR %d not used\n", reg);
 		goto out;
 	}
 	if (mtrr_usage_table[reg] < 1) {
-		printk(KERN_WARNING "mtrr: reg: %d has count=0\n", reg);
+		pr_warning("mtrr: reg: %d has count=0\n", reg);
 		goto out;
 	}
 	if (--mtrr_usage_table[reg] < 1)
@@ -526,33 +545,31 @@ int mtrr_del_page(int reg, unsigned long base, unsigned long size)
 	put_online_cpus();
 	return error;
 }
+
 /**
- *	mtrr_del - delete a memory type region
- *	@reg: Register returned by mtrr_add
- *	@base: Physical base address
- *	@size: Size of region
+ * mtrr_del - delete a memory type region
+ * @reg: Register returned by mtrr_add
+ * @base: Physical base address
+ * @size: Size of region
  *
- *	If register is supplied then base and size are ignored. This is
- *	how drivers should call it.
+ * If register is supplied then base and size are ignored. This is
+ * how drivers should call it.
  *
- *	Releases an MTRR region. If the usage count drops to zero the 
- *	register is freed and the region returns to default state.
- *	On success the register is returned, on failure a negative error
- *	code.
+ * Releases an MTRR region. If the usage count drops to zero the
+ * register is freed and the region returns to default state.
+ * On success the register is returned, on failure a negative error
+ * code.
  */
-
-int
-mtrr_del(int reg, unsigned long base, unsigned long size)
+int mtrr_del(int reg, unsigned long base, unsigned long size)
 {
 	if (mtrr_check(base, size))
 		return -EINVAL;
 	return mtrr_del_page(reg, base >> PAGE_SHIFT, size >> PAGE_SHIFT);
 }
-
-EXPORT_SYMBOL(mtrr_add);
 EXPORT_SYMBOL(mtrr_del);
 
-/* HACK ALERT!
+/*
+ * HACK ALERT!
  * These should be called implicitly, but we can't yet until all the initcall
  * stuff is done...
  */
@@ -576,29 +593,28 @@ struct mtrr_value {
 
 static struct mtrr_value mtrr_value[MTRR_MAX_VAR_RANGES];
 
-static int mtrr_save(struct sys_device * sysdev, pm_message_t state)
+static int mtrr_save(struct sys_device *sysdev, pm_message_t state)
 {
 	int i;
 
 	for (i = 0; i < num_var_ranges; i++) {
-		mtrr_if->get(i,
-			     &mtrr_value[i].lbase,
-			     &mtrr_value[i].lsize,
-			     &mtrr_value[i].ltype);
+		mtrr_if->get(i, &mtrr_value[i].lbase,
+				&mtrr_value[i].lsize,
+				&mtrr_value[i].ltype);
 	}
 	return 0;
 }
 
-static int mtrr_restore(struct sys_device * sysdev)
+static int mtrr_restore(struct sys_device *sysdev)
 {
 	int i;
 
 	for (i = 0; i < num_var_ranges; i++) {
-		if (mtrr_value[i].lsize)
-			set_mtrr(i,
-				 mtrr_value[i].lbase,
-				 mtrr_value[i].lsize,
-				 mtrr_value[i].ltype);
+		if (mtrr_value[i].lsize) {
+			set_mtrr(i, mtrr_value[i].lbase,
+				    mtrr_value[i].lsize,
+				    mtrr_value[i].ltype);
+		}
 	}
 	return 0;
 }
@@ -615,26 +631,29 @@ int __initdata changed_by_mtrr_cleanup;
 /**
  * mtrr_bp_init - initialize mtrrs on the boot CPU
  *
- * This needs to be called early; before any of the other CPUs are 
+ * This needs to be called early; before any of the other CPUs are
  * initialized (i.e. before smp_init()).
- * 
+ *
  */
 void __init mtrr_bp_init(void)
 {
 	u32 phys_addr;
+
 	init_ifs();
 
 	phys_addr = 32;
 
 	if (cpu_has_mtrr) {
 		mtrr_if = &generic_mtrr_ops;
-		size_or_mask = 0xff000000;	/* 36 bits */
+		size_or_mask = 0xff000000;			/* 36 bits */
 		size_and_mask = 0x00f00000;
 		phys_addr = 36;
 
-		/* This is an AMD specific MSR, but we assume(hope?) that
-		   Intel will implement it to when they extend the address
-		   bus of the Xeon. */
+		/*
+		 * This is an AMD specific MSR, but we assume(hope?) that
+		 * Intel will implement it to when they extend the address
+		 * bus of the Xeon.
+		 */
 		if (cpuid_eax(0x80000000) >= 0x80000008) {
 			phys_addr = cpuid_eax(0x80000008) & 0xff;
 			/* CPUID workaround for Intel 0F33/0F34 CPU */
@@ -649,9 +668,11 @@ void __init mtrr_bp_init(void)
 			size_and_mask = ~size_or_mask & 0xfffff00000ULL;
 		} else if (boot_cpu_data.x86_vendor == X86_VENDOR_CENTAUR &&
 			   boot_cpu_data.x86 == 6) {
-			/* VIA C* family have Intel style MTRRs, but
-			   don't support PAE */
-			size_or_mask = 0xfff00000;	/* 32 bits */
+			/*
+			 * VIA C* family have Intel style MTRRs,
+			 * but don't support PAE
+			 */
+			size_or_mask = 0xfff00000;		/* 32 bits */
 			size_and_mask = 0;
 			phys_addr = 32;
 		}
@@ -694,7 +715,6 @@ void __init mtrr_bp_init(void)
 				changed_by_mtrr_cleanup = 1;
 				mtrr_if->set_all();
 			}
-
 		}
 	}
 }
@@ -706,12 +726,17 @@ void mtrr_ap_init(void)
 	if (!mtrr_if || !use_intel())
 		return;
 	/*
-	 * Ideally we should hold mtrr_mutex here to avoid mtrr entries changed,
-	 * but this routine will be called in cpu boot time, holding the lock
-	 * breaks it. This routine is called in two cases: 1.very earily time
-	 * of software resume, when there absolutely isn't mtrr entry changes;
-	 * 2.cpu hotadd time. We let mtrr_add/del_page hold cpuhotplug lock to
-	 * prevent mtrr entry changes
+	 * Ideally we should hold mtrr_mutex here to avoid mtrr entries
+	 * changed, but this routine will be called in cpu boot time,
+	 * holding the lock breaks it.
+	 *
+	 * This routine is called in two cases:
+	 *
+	 *   1. very earily time of software resume, when there absolutely
+	 *      isn't mtrr entry changes;
+	 *
+	 *   2. cpu hotadd time. We let mtrr_add/del_page hold cpuhotplug
+	 *      lock to prevent mtrr entry changes
 	 */
 	local_irq_save(flags);
 
@@ -732,19 +757,23 @@ static int __init mtrr_init_finialize(void)
 {
 	if (!mtrr_if)
 		return 0;
+
 	if (use_intel()) {
 		if (!changed_by_mtrr_cleanup)
 			mtrr_state_warn();
-	} else {
-		/* The CPUs haven't MTRR and seem to not support SMP. They have
-		 * specific drivers, we use a tricky method to support
-		 * suspend/resume for them.
-		 * TBD: is there any system with such CPU which supports
-		 * suspend/resume?  if no, we should remove the code.
-		 */
-		sysdev_driver_register(&cpu_sysdev_class,
-			&mtrr_sysdev_driver);
+		return 0;
 	}
+
+	/*
+	 * The CPU has no MTRR and seems to not support SMP. They have
+	 * specific drivers, we use a tricky method to support
+	 * suspend/resume for them.
+	 *
+	 * TBD: is there any system with such CPU which supports
+	 * suspend/resume? If no, we should remove the code.
+	 */
+	sysdev_driver_register(&cpu_sysdev_class, &mtrr_sysdev_driver);
+
 	return 0;
 }
 subsys_initcall(mtrr_init_finialize);
-- 
cgit v1.2.3


From e3d0e69268dffb9676bf0800a60fb3573a723480 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Sun, 5 Jul 2009 09:44:11 +0200
Subject: x86: Further clean up of mtrr/generic.c

Yinghai noticed that i defined BIOS_BUG_MSG but added no
usage for it. The usage is to clean up this turd in generic.c:

			printk(KERN_WARNING "WARNING: BIOS bug: VAR MTRR %d "
				"contains strange UC entry under 1M, check "
				"with your system vendor!\n", i);

Breaking printk lines in the middle looks ugly, is hard to read
and breaks 'git grep'. Use the BIOS_BUG_MSG instead.

Also complete the moving of structure definitions and variables
to the top of the file.

Reported-by: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <20090703164225.GA21447@elte.hu>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/mtrr/cleanup.c | 56 ++++++++++++++++++--------------------
 1 file changed, 27 insertions(+), 29 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/mtrr/cleanup.c b/arch/x86/kernel/cpu/mtrr/cleanup.c
index b8aba811b60e..315738c74aad 100644
--- a/arch/x86/kernel/cpu/mtrr/cleanup.c
+++ b/arch/x86/kernel/cpu/mtrr/cleanup.c
@@ -34,14 +34,37 @@
 
 #include "mtrr.h"
 
-/* Should be related to MTRR_VAR_RANGES nums */
-#define RANGE_NUM 256
-
 struct res_range {
 	unsigned long	start;
 	unsigned long	end;
 };
 
+struct var_mtrr_range_state {
+	unsigned long	base_pfn;
+	unsigned long	size_pfn;
+	mtrr_type	type;
+};
+
+struct var_mtrr_state {
+	unsigned long	range_startk;
+	unsigned long	range_sizek;
+	unsigned long	chunk_sizek;
+	unsigned long	gran_sizek;
+	unsigned int	reg;
+};
+
+/* Should be related to MTRR_VAR_RANGES nums */
+#define RANGE_NUM				256
+
+static struct res_range __initdata		range[RANGE_NUM];
+static int __initdata				nr_range;
+
+static struct var_mtrr_range_state __initdata	range_state[RANGE_NUM];
+
+static int __initdata debug_print;
+#define Dprintk(x...) do { if (debug_print) printk(KERN_DEBUG x); } while (0)
+
+
 static int __init
 add_range(struct res_range *range, int nr_range,
 	  unsigned long start, unsigned long end)
@@ -147,18 +170,6 @@ static int __init cmp_range(const void *x1, const void *x2)
 	return start1 - start2;
 }
 
-struct var_mtrr_range_state {
-	unsigned long	base_pfn;
-	unsigned long	size_pfn;
-	mtrr_type	type;
-};
-
-static struct var_mtrr_range_state __initdata range_state[RANGE_NUM];
-
-static int __initdata debug_print;
-#define Dprintk(x...) do { if (debug_print) printk(KERN_DEBUG x); } while (0)
-
-
 #define BIOS_BUG_MSG KERN_WARNING \
 	"WARNING: BIOS bug: VAR MTRR %d contains strange UC entry under 1M, check with your system vendor!\n"
 
@@ -200,9 +211,7 @@ x86_get_mtrr_mem_range(struct res_range *range, int nr_range,
 		if (base < (1<<(20-PAGE_SHIFT)) && mtrr_state.have_fixed &&
 		    (mtrr_state.enabled & 1)) {
 			/* Var MTRR contains UC entry below 1M? Skip it: */
-			printk(KERN_WARNING "WARNING: BIOS bug: VAR MTRR %d "
-				"contains strange UC entry under 1M, check "
-				"with your system vendor!\n", i);
+			printk(BIOS_BUG_MSG, i);
 			if (base + size <= (1<<(20-PAGE_SHIFT)))
 				continue;
 			size -= (1<<(20-PAGE_SHIFT)) - base;
@@ -244,9 +253,6 @@ x86_get_mtrr_mem_range(struct res_range *range, int nr_range,
 	return nr_range;
 }
 
-static struct res_range __initdata range[RANGE_NUM];
-static int __initdata nr_range;
-
 #ifdef CONFIG_MTRR_SANITIZER
 
 static unsigned long __init sum_ranges(struct res_range *range, int nr_range)
@@ -284,14 +290,6 @@ static int __init mtrr_cleanup_debug_setup(char *str)
 }
 early_param("mtrr_cleanup_debug", mtrr_cleanup_debug_setup);
 
-struct var_mtrr_state {
-	unsigned long	range_startk;
-	unsigned long	range_sizek;
-	unsigned long	chunk_sizek;
-	unsigned long	gran_sizek;
-	unsigned int	reg;
-};
-
 static void __init
 set_var_mtrr(unsigned int reg, unsigned long basek, unsigned long sizek,
 	     unsigned char type, unsigned int address_bits)
-- 
cgit v1.2.3


From 9ff80942992cd5abd0779c815f310f65b7b83860 Mon Sep 17 00:00:00 2001
From: Cyrill Gorcunov <gorcunov@openvz.org>
Date: Wed, 8 Jul 2009 22:03:53 +0400
Subject: x86: Clean up idt_descr and idt_tableby using NR_VECTORS instead of
 hardcoded number

Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <20090708180353.GH5301@lenovo>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/common.c | 2 +-
 arch/x86/kernel/traps.c      | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index f1961c07af9a..d6f27c92854b 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -982,7 +982,7 @@ static __init int setup_disablecpuid(char *arg)
 __setup("clearcpuid=", setup_disablecpuid);
 
 #ifdef CONFIG_X86_64
-struct desc_ptr idt_descr = { 256 * 16 - 1, (unsigned long) idt_table };
+struct desc_ptr idt_descr = { NR_VECTORS * 16 - 1, (unsigned long) idt_table };
 
 DEFINE_PER_CPU_FIRST(union irq_stack_union,
 		     irq_stack_union) __aligned(PAGE_SIZE);
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index 5204332f475d..7e4b1f5dec8e 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -76,7 +76,7 @@ char ignore_fpu_irq;
  * F0 0F bug workaround.. We have a special link segment
  * for this.
  */
-gate_desc idt_table[256]
+gate_desc idt_table[NR_VECTORS]
 	__attribute__((__section__(".data.idt"))) = { { { { 0, 0 } } }, };
 #endif
 
-- 
cgit v1.2.3


From a1b4f1a5b7f57be2593a9f1fca465a529c95fc07 Mon Sep 17 00:00:00 2001
From: Cyrill Gorcunov <gorcunov@openvz.org>
Date: Sun, 5 Jul 2009 20:01:54 +0400
Subject: x86, ipi: Clean up safe_smp_processor_id() by using the
 cpu_has_apic() macro helper

We already use a lot of cpu_has_ helpers.
Lets do here the same for consistency.

Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <20090705160154.GB4791@lenovo>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic/ipi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apic/ipi.c b/arch/x86/kernel/apic/ipi.c
index dbf5445727a9..e6b4f517fcfe 100644
--- a/arch/x86/kernel/apic/ipi.c
+++ b/arch/x86/kernel/apic/ipi.c
@@ -150,7 +150,7 @@ int safe_smp_processor_id(void)
 {
 	int apicid, cpuid;
 
-	if (!boot_cpu_has(X86_FEATURE_APIC))
+	if (!cpu_has_apic)
 		return 0;
 
 	apicid = hard_smp_processor_id();
-- 
cgit v1.2.3


From e90476d3bab4322070c0afb3e3b55671de8664ea Mon Sep 17 00:00:00 2001
From: Huang Weiyi <weiyi.huang@gmail.com>
Date: Sat, 11 Jul 2009 09:32:46 +0800
Subject: x86: Remove duplicated #include

Remove duplicated #include in:

  arch/x86/kernel/dumpstack.c

Signed-off-by: Huang Weiyi <weiyi.huang@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/dumpstack.c | 1 -
 1 file changed, 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c
index c8405718a4c3..2d8a371d4339 100644
--- a/arch/x86/kernel/dumpstack.c
+++ b/arch/x86/kernel/dumpstack.c
@@ -15,7 +15,6 @@
 #include <linux/bug.h>
 #include <linux/nmi.h>
 #include <linux/sysfs.h>
-#include <linux/ftrace.h>
 
 #include <asm/stacktrace.h>
 
-- 
cgit v1.2.3


From 8bdbd962ecfcbdd96f9dbb02d780b4553afd2543 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@linux.intel.com>
Date: Sat, 4 Jul 2009 00:35:45 +0100
Subject: x86/cpu: Clean up various files a bit

No code changes except printk levels (although some of the K6
mtrr code might be clearer if there were a few as would
splitting out some of the intel cache code).

Signed-off-by: Alan Cox <alan@linux.intel.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/amd.c              |  37 ++++++-----
 arch/x86/kernel/cpu/bugs.c             |  10 +--
 arch/x86/kernel/cpu/bugs_64.c          |   2 +-
 arch/x86/kernel/cpu/common.c           |   8 +--
 arch/x86/kernel/cpu/cyrix.c            |  19 ++++--
 arch/x86/kernel/cpu/hypervisor.c       |   5 +-
 arch/x86/kernel/cpu/intel.c            |  11 ++--
 arch/x86/kernel/cpu/intel_cacheinfo.c  | 116 +++++++++++++++++----------------
 arch/x86/kernel/cpu/perfctr-watchdog.c |  45 ++++++-------
 arch/x86/kernel/cpu/proc.c             |   2 +-
 arch/x86/kernel/cpu/vmware.c           |  18 ++---
 11 files changed, 144 insertions(+), 129 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index 28e5f5956042..c6eb02e69875 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -2,7 +2,7 @@
 #include <linux/bitops.h>
 #include <linux/mm.h>
 
-#include <asm/io.h>
+#include <linux/io.h>
 #include <asm/processor.h>
 #include <asm/apic.h>
 #include <asm/cpu.h>
@@ -45,8 +45,8 @@ static void __cpuinit init_amd_k5(struct cpuinfo_x86 *c)
 #define CBAR_ENB	(0x80000000)
 #define CBAR_KEY	(0X000000CB)
 	if (c->x86_model == 9 || c->x86_model == 10) {
-		if (inl (CBAR) & CBAR_ENB)
-			outl (0 | CBAR_KEY, CBAR);
+		if (inl(CBAR) & CBAR_ENB)
+			outl(0 | CBAR_KEY, CBAR);
 	}
 }
 
@@ -87,9 +87,10 @@ static void __cpuinit init_amd_k6(struct cpuinfo_x86 *c)
 		d = d2-d;
 
 		if (d > 20*K6_BUG_LOOP)
-			printk("system stability may be impaired when more than 32 MB are used.\n");
+			printk(KERN_CONT
+				"system stability may be impaired when more than 32 MB are used.\n");
 		else
-			printk("probably OK (after B9730xxxx).\n");
+			printk(KERN_CONT "probably OK (after B9730xxxx).\n");
 		printk(KERN_INFO "Please see http://membres.lycos.fr/poulot/k6bug.html\n");
 	}
 
@@ -219,8 +220,9 @@ static void __cpuinit init_amd_k7(struct cpuinfo_x86 *c)
 	if ((c->x86_model == 8 && c->x86_mask >= 1) || (c->x86_model > 8)) {
 		rdmsr(MSR_K7_CLK_CTL, l, h);
 		if ((l & 0xfff00000) != 0x20000000) {
-			printk ("CPU: CLK_CTL MSR was %x. Reprogramming to %x\n", l,
-				((l & 0x000fffff)|0x20000000));
+			printk(KERN_INFO
+			    "CPU: CLK_CTL MSR was %x. Reprogramming to %x\n",
+					l, ((l & 0x000fffff)|0x20000000));
 			wrmsr(MSR_K7_CLK_CTL, (l & 0x000fffff)|0x20000000, h);
 		}
 	}
@@ -398,7 +400,7 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 		u32 level;
 
 		level = cpuid_eax(1);
-		if((level >= 0x0f48 && level < 0x0f50) || level >= 0x0f58)
+		if ((level >= 0x0f48 && level < 0x0f50) || level >= 0x0f58)
 			set_cpu_cap(c, X86_FEATURE_REP_GOOD);
 	}
 	if (c->x86 == 0x10 || c->x86 == 0x11)
@@ -487,27 +489,30 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 		 * benefit in doing so.
 		 */
 		if (!rdmsrl_safe(MSR_K8_TSEG_ADDR, &tseg)) {
-		    printk(KERN_DEBUG "tseg: %010llx\n", tseg);
-		    if ((tseg>>PMD_SHIFT) <
+			printk(KERN_DEBUG "tseg: %010llx\n", tseg);
+			if ((tseg>>PMD_SHIFT) <
 				(max_low_pfn_mapped>>(PMD_SHIFT-PAGE_SHIFT)) ||
-			((tseg>>PMD_SHIFT) <
+				((tseg>>PMD_SHIFT) <
 				(max_pfn_mapped>>(PMD_SHIFT-PAGE_SHIFT)) &&
-			 (tseg>>PMD_SHIFT) >= (1ULL<<(32 - PMD_SHIFT))))
-			set_memory_4k((unsigned long)__va(tseg), 1);
+				(tseg>>PMD_SHIFT) >= (1ULL<<(32 - PMD_SHIFT))))
+				set_memory_4k((unsigned long)__va(tseg), 1);
 		}
 	}
 #endif
 }
 
 #ifdef CONFIG_X86_32
-static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 *c, unsigned int size)
+static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 *c,
+							unsigned int size)
 {
 	/* AMD errata T13 (order #21922) */
 	if ((c->x86 == 6)) {
-		if (c->x86_model == 3 && c->x86_mask == 0)	/* Duron Rev A0 */
+		/* Duron Rev A0 */
+		if (c->x86_model == 3 && c->x86_mask == 0)
 			size = 64;
+		/* Tbird rev A1/A2 */
 		if (c->x86_model == 4 &&
-		    (c->x86_mask == 0 || c->x86_mask == 1))	/* Tbird rev A1/A2 */
+			(c->x86_mask == 0 || c->x86_mask == 1))
 			size = 256;
 	}
 	return size;
diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c
index c8e315f1aa83..01a265212395 100644
--- a/arch/x86/kernel/cpu/bugs.c
+++ b/arch/x86/kernel/cpu/bugs.c
@@ -81,7 +81,7 @@ static void __init check_fpu(void)
 
 	boot_cpu_data.fdiv_bug = fdiv_bug;
 	if (boot_cpu_data.fdiv_bug)
-		printk("Hmm, FPU with FDIV bug.\n");
+		printk(KERN_WARNING "Hmm, FPU with FDIV bug.\n");
 }
 
 static void __init check_hlt(void)
@@ -98,7 +98,7 @@ static void __init check_hlt(void)
 	halt();
 	halt();
 	halt();
-	printk("OK.\n");
+	printk(KERN_CONT "OK.\n");
 }
 
 /*
@@ -122,9 +122,9 @@ static void __init check_popad(void)
 	 * CPU hard. Too bad.
 	 */
 	if (res != 12345678)
-		printk("Buggy.\n");
+		printk(KERN_CONT "Buggy.\n");
 	else
-		printk("OK.\n");
+		printk(KERN_CONT "OK.\n");
 #endif
 }
 
@@ -156,7 +156,7 @@ void __init check_bugs(void)
 {
 	identify_boot_cpu();
 #ifndef CONFIG_SMP
-	printk("CPU: ");
+	printk(KERN_INFO "CPU: ");
 	print_cpu_info(&boot_cpu_data);
 #endif
 	check_config();
diff --git a/arch/x86/kernel/cpu/bugs_64.c b/arch/x86/kernel/cpu/bugs_64.c
index 9a3ed0649d4e..04f0fe5af83e 100644
--- a/arch/x86/kernel/cpu/bugs_64.c
+++ b/arch/x86/kernel/cpu/bugs_64.c
@@ -15,7 +15,7 @@ void __init check_bugs(void)
 {
 	identify_boot_cpu();
 #if !defined(CONFIG_SMP)
-	printk("CPU: ");
+	printk(KERN_INFO "CPU: ");
 	print_cpu_info(&boot_cpu_data);
 #endif
 	alternative_instructions();
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index d6f27c92854b..c96ea44928bf 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -18,8 +18,8 @@
 #include <asm/hypervisor.h>
 #include <asm/processor.h>
 #include <asm/sections.h>
-#include <asm/topology.h>
-#include <asm/cpumask.h>
+#include <linux/topology.h>
+#include <linux/cpumask.h>
 #include <asm/pgtable.h>
 #include <asm/atomic.h>
 #include <asm/proto.h>
@@ -28,13 +28,13 @@
 #include <asm/desc.h>
 #include <asm/i387.h>
 #include <asm/mtrr.h>
-#include <asm/numa.h>
+#include <linux/numa.h>
 #include <asm/asm.h>
 #include <asm/cpu.h>
 #include <asm/mce.h>
 #include <asm/msr.h>
 #include <asm/pat.h>
-#include <asm/smp.h>
+#include <linux/smp.h>
 
 #ifdef CONFIG_X86_LOCAL_APIC
 #include <asm/uv/uv.h>
diff --git a/arch/x86/kernel/cpu/cyrix.c b/arch/x86/kernel/cpu/cyrix.c
index 593171e967ef..19807b89f058 100644
--- a/arch/x86/kernel/cpu/cyrix.c
+++ b/arch/x86/kernel/cpu/cyrix.c
@@ -3,10 +3,10 @@
 #include <linux/delay.h>
 #include <linux/pci.h>
 #include <asm/dma.h>
-#include <asm/io.h>
+#include <linux/io.h>
 #include <asm/processor-cyrix.h>
 #include <asm/processor-flags.h>
-#include <asm/timer.h>
+#include <linux/timer.h>
 #include <asm/pci-direct.h>
 #include <asm/tsc.h>
 
@@ -282,7 +282,8 @@ static void __cpuinit init_cyrix(struct cpuinfo_x86 *c)
 		 *  The 5510/5520 companion chips have a funky PIT.
 		 */
 		if (vendor == PCI_VENDOR_ID_CYRIX &&
-	 (device == PCI_DEVICE_ID_CYRIX_5510 || device == PCI_DEVICE_ID_CYRIX_5520))
+			(device == PCI_DEVICE_ID_CYRIX_5510 ||
+					device == PCI_DEVICE_ID_CYRIX_5520))
 			mark_tsc_unstable("cyrix 5510/5520 detected");
 	}
 #endif
@@ -299,7 +300,8 @@ static void __cpuinit init_cyrix(struct cpuinfo_x86 *c)
 			 *  ?  : 0x7x
 			 * GX1 : 0x8x          GX1  datasheet 56
 			 */
-			if ((0x30 <= dir1 && dir1 <= 0x6f) || (0x80 <= dir1 && dir1 <= 0x8f))
+			if ((0x30 <= dir1 && dir1 <= 0x6f) ||
+					(0x80 <= dir1 && dir1 <= 0x8f))
 				geode_configure();
 			return;
 		} else { /* MediaGX */
@@ -427,9 +429,12 @@ static void __cpuinit cyrix_identify(struct cpuinfo_x86 *c)
 			printk(KERN_INFO "Enabling CPUID on Cyrix processor.\n");
 			local_irq_save(flags);
 			ccr3 = getCx86(CX86_CCR3);
-			setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10);       /* enable MAPEN  */
-			setCx86_old(CX86_CCR4, getCx86_old(CX86_CCR4) | 0x80);  /* enable cpuid  */
-			setCx86(CX86_CCR3, ccr3);                       /* disable MAPEN */
+			/* enable MAPEN  */
+			setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10);
+			/* enable cpuid  */
+			setCx86_old(CX86_CCR4, getCx86_old(CX86_CCR4) | 0x80);
+			/* disable MAPEN */
+			setCx86(CX86_CCR3, ccr3);
 			local_irq_restore(flags);
 		}
 	}
diff --git a/arch/x86/kernel/cpu/hypervisor.c b/arch/x86/kernel/cpu/hypervisor.c
index fb5b86af0b01..93ba8eeb100a 100644
--- a/arch/x86/kernel/cpu/hypervisor.c
+++ b/arch/x86/kernel/cpu/hypervisor.c
@@ -28,11 +28,10 @@
 static inline void __cpuinit
 detect_hypervisor_vendor(struct cpuinfo_x86 *c)
 {
-	if (vmware_platform()) {
+	if (vmware_platform())
 		c->x86_hyper_vendor = X86_HYPER_VENDOR_VMWARE;
-	} else {
+	else
 		c->x86_hyper_vendor = X86_HYPER_VENDOR_NONE;
-	}
 }
 
 unsigned long get_hypervisor_tsc_freq(void)
diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c
index 3260ab044996..80a722a071b5 100644
--- a/arch/x86/kernel/cpu/intel.c
+++ b/arch/x86/kernel/cpu/intel.c
@@ -7,17 +7,17 @@
 #include <linux/sched.h>
 #include <linux/thread_info.h>
 #include <linux/module.h>
+#include <linux/uaccess.h>
 
 #include <asm/processor.h>
 #include <asm/pgtable.h>
 #include <asm/msr.h>
-#include <asm/uaccess.h>
 #include <asm/ds.h>
 #include <asm/bugs.h>
 #include <asm/cpu.h>
 
 #ifdef CONFIG_X86_64
-#include <asm/topology.h>
+#include <linux/topology.h>
 #include <asm/numa_64.h>
 #endif
 
@@ -174,7 +174,8 @@ static void __cpuinit intel_workarounds(struct cpuinfo_x86 *c)
 #ifdef CONFIG_X86_F00F_BUG
 	/*
 	 * All current models of Pentium and Pentium with MMX technology CPUs
-	 * have the F0 0F bug, which lets nonprivileged users lock up the system.
+	 * have the F0 0F bug, which lets nonprivileged users lock up the
+	 * system.
 	 * Note that the workaround only should be initialized once...
 	 */
 	c->f00f_bug = 0;
@@ -207,7 +208,7 @@ static void __cpuinit intel_workarounds(struct cpuinfo_x86 *c)
 			printk (KERN_INFO "CPU: C0 stepping P4 Xeon detected.\n");
 			printk (KERN_INFO "CPU: Disabling hardware prefetching (Errata 037)\n");
 			lo |= MSR_IA32_MISC_ENABLE_PREFETCH_DISABLE;
-			wrmsr (MSR_IA32_MISC_ENABLE, lo, hi);
+			wrmsr(MSR_IA32_MISC_ENABLE, lo, hi);
 		}
 	}
 
@@ -283,7 +284,7 @@ static int __cpuinit intel_num_cpu_cores(struct cpuinfo_x86 *c)
 	/* Intel has a non-standard dependency on %ecx for this CPUID level. */
 	cpuid_count(4, 0, &eax, &ebx, &ecx, &edx);
 	if (eax & 0x1f)
-		return ((eax >> 26) + 1);
+		return (eax >> 26) + 1;
 	else
 		return 1;
 }
diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c
index 789efe217e1a..306bf0dca061 100644
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -3,7 +3,7 @@
  *
  *	Changes:
  *	Venkatesh Pallipadi	: Adding cache identification through cpuid(4)
- *		Ashok Raj <ashok.raj@intel.com>: Work with CPU hotplug infrastructure.
+ *	Ashok Raj <ashok.raj@intel.com>: Work with CPU hotplug infrastructure.
  *	Andi Kleen / Andreas Herrmann	: CPUID4 emulation on AMD.
  */
 
@@ -16,7 +16,7 @@
 #include <linux/pci.h>
 
 #include <asm/processor.h>
-#include <asm/smp.h>
+#include <linux/smp.h>
 #include <asm/k8.h>
 
 #define LVL_1_INST	1
@@ -25,14 +25,15 @@
 #define LVL_3		4
 #define LVL_TRACE	5
 
-struct _cache_table
-{
+struct _cache_table {
 	unsigned char descriptor;
 	char cache_type;
 	short size;
 };
 
-/* all the cache descriptor types we care about (no TLB or trace cache entries) */
+/* All the cache descriptor types we care about (no TLB or
+   trace cache entries) */
+
 static const struct _cache_table __cpuinitconst cache_table[] =
 {
 	{ 0x06, LVL_1_INST, 8 },	/* 4-way set assoc, 32 byte line size */
@@ -105,8 +106,7 @@ static const struct _cache_table __cpuinitconst cache_table[] =
 };
 
 
-enum _cache_type
-{
+enum _cache_type {
 	CACHE_TYPE_NULL	= 0,
 	CACHE_TYPE_DATA = 1,
 	CACHE_TYPE_INST = 2,
@@ -170,31 +170,31 @@ unsigned short			num_cache_leaves;
    Maybe later */
 union l1_cache {
 	struct {
-		unsigned line_size : 8;
-		unsigned lines_per_tag : 8;
-		unsigned assoc : 8;
-		unsigned size_in_kb : 8;
+		unsigned line_size:8;
+		unsigned lines_per_tag:8;
+		unsigned assoc:8;
+		unsigned size_in_kb:8;
 	};
 	unsigned val;
 };
 
 union l2_cache {
 	struct {
-		unsigned line_size : 8;
-		unsigned lines_per_tag : 4;
-		unsigned assoc : 4;
-		unsigned size_in_kb : 16;
+		unsigned line_size:8;
+		unsigned lines_per_tag:4;
+		unsigned assoc:4;
+		unsigned size_in_kb:16;
 	};
 	unsigned val;
 };
 
 union l3_cache {
 	struct {
-		unsigned line_size : 8;
-		unsigned lines_per_tag : 4;
-		unsigned assoc : 4;
-		unsigned res : 2;
-		unsigned size_encoded : 14;
+		unsigned line_size:8;
+		unsigned lines_per_tag:4;
+		unsigned assoc:4;
+		unsigned res:2;
+		unsigned size_encoded:14;
 	};
 	unsigned val;
 };
@@ -350,7 +350,8 @@ static int __cpuinit find_num_cache_leaves(void)
 
 unsigned int __cpuinit init_intel_cacheinfo(struct cpuinfo_x86 *c)
 {
-	unsigned int trace = 0, l1i = 0, l1d = 0, l2 = 0, l3 = 0; /* Cache sizes */
+	/* Cache sizes */
+	unsigned int trace = 0, l1i = 0, l1d = 0, l2 = 0, l3 = 0;
 	unsigned int new_l1d = 0, new_l1i = 0; /* Cache sizes from cpuid(4) */
 	unsigned int new_l2 = 0, new_l3 = 0, i; /* Cache sizes from cpuid(4) */
 	unsigned int l2_id = 0, l3_id = 0, num_threads_sharing, index_msb;
@@ -377,8 +378,8 @@ unsigned int __cpuinit init_intel_cacheinfo(struct cpuinfo_x86 *c)
 
 			retval = cpuid4_cache_lookup_regs(i, &this_leaf);
 			if (retval >= 0) {
-				switch(this_leaf.eax.split.level) {
-				    case 1:
+				switch (this_leaf.eax.split.level) {
+				case 1:
 					if (this_leaf.eax.split.type ==
 							CACHE_TYPE_DATA)
 						new_l1d = this_leaf.size/1024;
@@ -386,19 +387,20 @@ unsigned int __cpuinit init_intel_cacheinfo(struct cpuinfo_x86 *c)
 							CACHE_TYPE_INST)
 						new_l1i = this_leaf.size/1024;
 					break;
-				    case 2:
+				case 2:
 					new_l2 = this_leaf.size/1024;
 					num_threads_sharing = 1 + this_leaf.eax.split.num_threads_sharing;
 					index_msb = get_count_order(num_threads_sharing);
 					l2_id = c->apicid >> index_msb;
 					break;
-				    case 3:
+				case 3:
 					new_l3 = this_leaf.size/1024;
 					num_threads_sharing = 1 + this_leaf.eax.split.num_threads_sharing;
-					index_msb = get_count_order(num_threads_sharing);
+					index_msb = get_count_order(
+							num_threads_sharing);
 					l3_id = c->apicid >> index_msb;
 					break;
-				    default:
+				default:
 					break;
 				}
 			}
@@ -421,22 +423,21 @@ unsigned int __cpuinit init_intel_cacheinfo(struct cpuinfo_x86 *c)
 		/* Number of times to iterate */
 		n = cpuid_eax(2) & 0xFF;
 
-		for ( i = 0 ; i < n ; i++ ) {
+		for (i = 0 ; i < n ; i++) {
 			cpuid(2, &regs[0], &regs[1], &regs[2], &regs[3]);
 
 			/* If bit 31 is set, this is an unknown format */
-			for ( j = 0 ; j < 3 ; j++ ) {
-				if (regs[j] & (1 << 31)) regs[j] = 0;
-			}
+			for (j = 0 ; j < 3 ; j++)
+				if (regs[j] & (1 << 31))
+					regs[j] = 0;
 
 			/* Byte 0 is level count, not a descriptor */
-			for ( j = 1 ; j < 16 ; j++ ) {
+			for (j = 1 ; j < 16 ; j++) {
 				unsigned char des = dp[j];
 				unsigned char k = 0;
 
 				/* look up this descriptor in the table */
-				while (cache_table[k].descriptor != 0)
-				{
+				while (cache_table[k].descriptor != 0) {
 					if (cache_table[k].descriptor == des) {
 						if (only_trace && cache_table[k].cache_type != LVL_TRACE)
 							break;
@@ -488,14 +489,14 @@ unsigned int __cpuinit init_intel_cacheinfo(struct cpuinfo_x86 *c)
 	}
 
 	if (trace)
-		printk (KERN_INFO "CPU: Trace cache: %dK uops", trace);
-	else if ( l1i )
-		printk (KERN_INFO "CPU: L1 I cache: %dK", l1i);
+		printk(KERN_INFO "CPU: Trace cache: %dK uops", trace);
+	else if (l1i)
+		printk(KERN_INFO "CPU: L1 I cache: %dK", l1i);
 
 	if (l1d)
-		printk(", L1 D cache: %dK\n", l1d);
+		printk(KERN_CONT ", L1 D cache: %dK\n", l1d);
 	else
-		printk("\n");
+		printk(KERN_CONT "\n");
 
 	if (l2)
 		printk(KERN_INFO "CPU: L2 cache: %dK\n", l2);
@@ -558,8 +559,13 @@ static void __cpuinit cache_remove_shared_cpu_map(unsigned int cpu, int index)
 	}
 }
 #else
-static void __cpuinit cache_shared_cpu_map_setup(unsigned int cpu, int index) {}
-static void __cpuinit cache_remove_shared_cpu_map(unsigned int cpu, int index) {}
+static void __cpuinit cache_shared_cpu_map_setup(unsigned int cpu, int index)
+{
+}
+
+static void __cpuinit cache_remove_shared_cpu_map(unsigned int cpu, int index)
+{
+}
 #endif
 
 static void __cpuinit free_cache_attributes(unsigned int cpu)
@@ -645,7 +651,7 @@ static DEFINE_PER_CPU(struct _index_kobject *, index_kobject);
 static ssize_t show_##file_name						\
 			(struct _cpuid4_info *this_leaf, char *buf)	\
 {									\
-	return sprintf (buf, "%lu\n", (unsigned long)this_leaf->object + val); \
+	return sprintf(buf, "%lu\n", (unsigned long)this_leaf->object + val); \
 }
 
 show_one_plus(level, eax.split.level, 0);
@@ -656,7 +662,7 @@ show_one_plus(number_of_sets, ecx.split.number_of_sets, 1);
 
 static ssize_t show_size(struct _cpuid4_info *this_leaf, char *buf)
 {
-	return sprintf (buf, "%luK\n", this_leaf->size / 1024);
+	return sprintf(buf, "%luK\n", this_leaf->size / 1024);
 }
 
 static ssize_t show_shared_cpu_map_func(struct _cpuid4_info *this_leaf,
@@ -669,7 +675,7 @@ static ssize_t show_shared_cpu_map_func(struct _cpuid4_info *this_leaf,
 		const struct cpumask *mask;
 
 		mask = to_cpumask(this_leaf->shared_cpu_map);
-		n = type?
+		n = type ?
 			cpulist_scnprintf(buf, len-2, mask) :
 			cpumask_scnprintf(buf, len-2, mask);
 		buf[n++] = '\n';
@@ -800,7 +806,7 @@ static struct _cache_attr cache_disable_0 = __ATTR(cache_disable_0, 0644,
 static struct _cache_attr cache_disable_1 = __ATTR(cache_disable_1, 0644,
 		show_cache_disable_1, store_cache_disable_1);
 
-static struct attribute * default_attrs[] = {
+static struct attribute *default_attrs[] = {
 	&type.attr,
 	&level.attr,
 	&coherency_line_size.attr,
@@ -815,7 +821,7 @@ static struct attribute * default_attrs[] = {
 	NULL
 };
 
-static ssize_t show(struct kobject * kobj, struct attribute * attr, char * buf)
+static ssize_t show(struct kobject *kobj, struct attribute *attr, char *buf)
 {
 	struct _cache_attr *fattr = to_attr(attr);
 	struct _index_kobject *this_leaf = to_object(kobj);
@@ -828,8 +834,8 @@ static ssize_t show(struct kobject * kobj, struct attribute * attr, char * buf)
 	return ret;
 }
 
-static ssize_t store(struct kobject * kobj, struct attribute * attr,
-		     const char * buf, size_t count)
+static ssize_t store(struct kobject *kobj, struct attribute *attr,
+		     const char *buf, size_t count)
 {
 	struct _cache_attr *fattr = to_attr(attr);
 	struct _index_kobject *this_leaf = to_object(kobj);
@@ -883,7 +889,7 @@ static int __cpuinit cpuid4_cache_sysfs_init(unsigned int cpu)
 		goto err_out;
 
 	per_cpu(index_kobject, cpu) = kzalloc(
-	    sizeof(struct _index_kobject ) * num_cache_leaves, GFP_KERNEL);
+	    sizeof(struct _index_kobject) * num_cache_leaves, GFP_KERNEL);
 	if (unlikely(per_cpu(index_kobject, cpu) == NULL))
 		goto err_out;
 
@@ -917,7 +923,7 @@ static int __cpuinit cache_add_dev(struct sys_device * sys_dev)
 	}
 
 	for (i = 0; i < num_cache_leaves; i++) {
-		this_object = INDEX_KOBJECT_PTR(cpu,i);
+		this_object = INDEX_KOBJECT_PTR(cpu, i);
 		this_object->cpu = cpu;
 		this_object->index = i;
 		retval = kobject_init_and_add(&(this_object->kobj),
@@ -925,9 +931,8 @@ static int __cpuinit cache_add_dev(struct sys_device * sys_dev)
 					      per_cpu(cache_kobject, cpu),
 					      "index%1lu", i);
 		if (unlikely(retval)) {
-			for (j = 0; j < i; j++) {
-				kobject_put(&(INDEX_KOBJECT_PTR(cpu,j)->kobj));
-			}
+			for (j = 0; j < i; j++)
+				kobject_put(&(INDEX_KOBJECT_PTR(cpu, j)->kobj));
 			kobject_put(per_cpu(cache_kobject, cpu));
 			cpuid4_cache_sysfs_exit(cpu);
 			return retval;
@@ -952,7 +957,7 @@ static void __cpuinit cache_remove_dev(struct sys_device * sys_dev)
 	cpumask_clear_cpu(cpu, to_cpumask(cache_dev_map));
 
 	for (i = 0; i < num_cache_leaves; i++)
-		kobject_put(&(INDEX_KOBJECT_PTR(cpu,i)->kobj));
+		kobject_put(&(INDEX_KOBJECT_PTR(cpu, i)->kobj));
 	kobject_put(per_cpu(cache_kobject, cpu));
 	cpuid4_cache_sysfs_exit(cpu);
 }
@@ -977,8 +982,7 @@ static int __cpuinit cacheinfo_cpu_callback(struct notifier_block *nfb,
 	return NOTIFY_OK;
 }
 
-static struct notifier_block __cpuinitdata cacheinfo_cpu_notifier =
-{
+static struct notifier_block __cpuinitdata cacheinfo_cpu_notifier = {
 	.notifier_call = cacheinfo_cpu_callback,
 };
 
diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c
index 5c481f6205bf..8100a29c854f 100644
--- a/arch/x86/kernel/cpu/perfctr-watchdog.c
+++ b/arch/x86/kernel/cpu/perfctr-watchdog.c
@@ -68,16 +68,16 @@ static inline unsigned int nmi_perfctr_msr_to_bit(unsigned int msr)
 	/* returns the bit offset of the performance counter register */
 	switch (boot_cpu_data.x86_vendor) {
 	case X86_VENDOR_AMD:
-		return (msr - MSR_K7_PERFCTR0);
+		return msr - MSR_K7_PERFCTR0;
 	case X86_VENDOR_INTEL:
 		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
-			return (msr - MSR_ARCH_PERFMON_PERFCTR0);
+			return msr - MSR_ARCH_PERFMON_PERFCTR0;
 
 		switch (boot_cpu_data.x86) {
 		case 6:
-			return (msr - MSR_P6_PERFCTR0);
+			return msr - MSR_P6_PERFCTR0;
 		case 15:
-			return (msr - MSR_P4_BPU_PERFCTR0);
+			return msr - MSR_P4_BPU_PERFCTR0;
 		}
 	}
 	return 0;
@@ -92,16 +92,16 @@ static inline unsigned int nmi_evntsel_msr_to_bit(unsigned int msr)
 	/* returns the bit offset of the event selection register */
 	switch (boot_cpu_data.x86_vendor) {
 	case X86_VENDOR_AMD:
-		return (msr - MSR_K7_EVNTSEL0);
+		return msr - MSR_K7_EVNTSEL0;
 	case X86_VENDOR_INTEL:
 		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
-			return (msr - MSR_ARCH_PERFMON_EVENTSEL0);
+			return msr - MSR_ARCH_PERFMON_EVENTSEL0;
 
 		switch (boot_cpu_data.x86) {
 		case 6:
-			return (msr - MSR_P6_EVNTSEL0);
+			return msr - MSR_P6_EVNTSEL0;
 		case 15:
-			return (msr - MSR_P4_BSU_ESCR0);
+			return msr - MSR_P4_BSU_ESCR0;
 		}
 	}
 	return 0;
@@ -113,7 +113,7 @@ int avail_to_resrv_perfctr_nmi_bit(unsigned int counter)
 {
 	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
 
-	return (!test_bit(counter, perfctr_nmi_owner));
+	return !test_bit(counter, perfctr_nmi_owner);
 }
 
 /* checks the an msr for availability */
@@ -124,7 +124,7 @@ int avail_to_resrv_perfctr_nmi(unsigned int msr)
 	counter = nmi_perfctr_msr_to_bit(msr);
 	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
 
-	return (!test_bit(counter, perfctr_nmi_owner));
+	return !test_bit(counter, perfctr_nmi_owner);
 }
 EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi_bit);
 
@@ -237,7 +237,7 @@ static unsigned int adjust_for_32bit_ctr(unsigned int hz)
 	 */
 	counter_val = (u64)cpu_khz * 1000;
 	do_div(counter_val, retval);
- 	if (counter_val > 0x7fffffffULL) {
+	if (counter_val > 0x7fffffffULL) {
 		u64 count = (u64)cpu_khz * 1000;
 		do_div(count, 0x7fffffffUL);
 		retval = count + 1;
@@ -251,7 +251,7 @@ static void write_watchdog_counter(unsigned int perfctr_msr,
 	u64 count = (u64)cpu_khz * 1000;
 
 	do_div(count, nmi_hz);
-	if(descr)
+	if (descr)
 		pr_debug("setting %s to -0x%08Lx\n", descr, count);
 	wrmsrl(perfctr_msr, 0 - count);
 }
@@ -262,7 +262,7 @@ static void write_watchdog_counter32(unsigned int perfctr_msr,
 	u64 count = (u64)cpu_khz * 1000;
 
 	do_div(count, nmi_hz);
-	if(descr)
+	if (descr)
 		pr_debug("setting %s to -0x%08Lx\n", descr, count);
 	wrmsr(perfctr_msr, (u32)(-count), 0);
 }
@@ -296,7 +296,7 @@ static int setup_k7_watchdog(unsigned nmi_hz)
 
 	/* setup the timer */
 	wrmsr(evntsel_msr, evntsel, 0);
-	write_watchdog_counter(perfctr_msr, "K7_PERFCTR0",nmi_hz);
+	write_watchdog_counter(perfctr_msr, "K7_PERFCTR0", nmi_hz);
 
 	/* initialize the wd struct before enabling */
 	wd->perfctr_msr = perfctr_msr;
@@ -387,7 +387,7 @@ static int setup_p6_watchdog(unsigned nmi_hz)
 	/* setup the timer */
 	wrmsr(evntsel_msr, evntsel, 0);
 	nmi_hz = adjust_for_32bit_ctr(nmi_hz);
-	write_watchdog_counter32(perfctr_msr, "P6_PERFCTR0",nmi_hz);
+	write_watchdog_counter32(perfctr_msr, "P6_PERFCTR0", nmi_hz);
 
 	/* initialize the wd struct before enabling */
 	wd->perfctr_msr = perfctr_msr;
@@ -415,7 +415,7 @@ static void __kprobes p6_rearm(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz)
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 
 	/* P6/ARCH_PERFMON has 32 bit counter write */
-	write_watchdog_counter32(wd->perfctr_msr, NULL,nmi_hz);
+	write_watchdog_counter32(wd->perfctr_msr, NULL, nmi_hz);
 }
 
 static const struct wd_ops p6_wd_ops = {
@@ -490,9 +490,9 @@ static int setup_p4_watchdog(unsigned nmi_hz)
 	if (smp_num_siblings == 2) {
 		unsigned int ebx, apicid;
 
-        	ebx = cpuid_ebx(1);
-	        apicid = (ebx >> 24) & 0xff;
-        	ht_num = apicid & 1;
+		ebx = cpuid_ebx(1);
+		apicid = (ebx >> 24) & 0xff;
+		ht_num = apicid & 1;
 	} else
 #endif
 		ht_num = 0;
@@ -544,7 +544,7 @@ static int setup_p4_watchdog(unsigned nmi_hz)
 	}
 
 	evntsel = P4_ESCR_EVENT_SELECT(0x3F)
-	 	| P4_ESCR_OS
+		| P4_ESCR_OS
 		| P4_ESCR_USR;
 
 	cccr_val |= P4_CCCR_THRESHOLD(15)
@@ -612,7 +612,7 @@ static void __kprobes p4_rearm(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz)
 {
 	unsigned dummy;
 	/*
- 	 * P4 quirks:
+	 * P4 quirks:
 	 * - An overflown perfctr will assert its interrupt
 	 *   until the OVF flag in its CCCR is cleared.
 	 * - LVTPC is masked on interrupt and must be
@@ -662,7 +662,8 @@ static int setup_intel_arch_watchdog(unsigned nmi_hz)
 	 * NOTE: Corresponding bit = 0 in ebx indicates event present.
 	 */
 	cpuid(10, &(eax.full), &ebx, &unused, &unused);
-	if ((eax.split.mask_length < (ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX+1)) ||
+	if ((eax.split.mask_length <
+			(ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX+1)) ||
 	    (ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT))
 		return 0;
 
diff --git a/arch/x86/kernel/cpu/proc.c b/arch/x86/kernel/cpu/proc.c
index d5e30397246b..1e904346bbf4 100644
--- a/arch/x86/kernel/cpu/proc.c
+++ b/arch/x86/kernel/cpu/proc.c
@@ -128,7 +128,7 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 			if (i < ARRAY_SIZE(x86_power_flags) &&
 			    x86_power_flags[i])
 				seq_printf(m, "%s%s",
-					   x86_power_flags[i][0]?" ":"",
+					   x86_power_flags[i][0] ? " " : "",
 					   x86_power_flags[i]);
 			else
 				seq_printf(m, " [%d]", i);
diff --git a/arch/x86/kernel/cpu/vmware.c b/arch/x86/kernel/cpu/vmware.c
index 284c399e3234..bc24f514ec93 100644
--- a/arch/x86/kernel/cpu/vmware.c
+++ b/arch/x86/kernel/cpu/vmware.c
@@ -49,17 +49,17 @@ static inline int __vmware_platform(void)
 
 static unsigned long __vmware_get_tsc_khz(void)
 {
-        uint64_t tsc_hz;
-        uint32_t eax, ebx, ecx, edx;
+	uint64_t tsc_hz;
+	uint32_t eax, ebx, ecx, edx;
 
-        VMWARE_PORT(GETHZ, eax, ebx, ecx, edx);
+	VMWARE_PORT(GETHZ, eax, ebx, ecx, edx);
 
-        if (ebx == UINT_MAX)
-                return 0;
-        tsc_hz = eax | (((uint64_t)ebx) << 32);
-        do_div(tsc_hz, 1000);
-        BUG_ON(tsc_hz >> 32);
-        return tsc_hz;
+	if (ebx == UINT_MAX)
+		return 0;
+	tsc_hz = eax | (((uint64_t)ebx) << 32);
+	do_div(tsc_hz, 1000);
+	BUG_ON(tsc_hz >> 32);
+	return tsc_hz;
 }
 
 /*
-- 
cgit v1.2.3


From 8045a4c293d36c61656a20d581b11f7f0cd7acd5 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Tue, 7 Jul 2009 19:30:25 +0200
Subject: x86/oprofile: Fix cast of counter value

When casting the counter value to a 64 bit value in 32 bit mode, sign
extension may lead to broken counter values. This patch fixes this by
casting to (u64) instead of (s64).

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_amd.c | 4 ++--
 arch/x86/oprofile/op_model_p4.c  | 6 +++---
 2 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index e95268eb9220..7ca8306aefae 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -111,7 +111,7 @@ static void op_amd_setup_ctrs(struct op_x86_model_spec const *model,
 		if (counter_config[i].enabled && msrs->counters[i].addr) {
 			reset_value[i] = counter_config[i].count;
 			wrmsrl(msrs->counters[i].addr,
-			       -(s64)counter_config[i].count);
+			       -(u64)counter_config[i].count);
 			rdmsrl(msrs->controls[i].addr, val);
 			val &= model->reserved;
 			val |= op_x86_get_ctrl(model, &counter_config[i]);
@@ -237,7 +237,7 @@ static int op_amd_check_ctrs(struct pt_regs * const regs,
 		if (val & OP_CTR_OVERFLOW)
 			continue;
 		oprofile_add_sample(regs, i);
-		wrmsrl(msrs->counters[i].addr, -(s64)reset_value[i]);
+		wrmsrl(msrs->counters[i].addr, -(u64)reset_value[i]);
 	}
 
 	op_amd_handle_ibs(regs, msrs);
diff --git a/arch/x86/oprofile/op_model_p4.c b/arch/x86/oprofile/op_model_p4.c
index f01e53b118fa..9db9e361182c 100644
--- a/arch/x86/oprofile/op_model_p4.c
+++ b/arch/x86/oprofile/op_model_p4.c
@@ -580,7 +580,7 @@ static void p4_setup_ctrs(struct op_x86_model_spec const *model,
 			reset_value[i] = counter_config[i].count;
 			pmc_setup_one_p4_counter(i);
 			wrmsrl(p4_counters[VIRT_CTR(stag, i)].counter_address,
-			       -(s64)counter_config[i].count);
+			       -(u64)counter_config[i].count);
 		} else {
 			reset_value[i] = 0;
 		}
@@ -625,11 +625,11 @@ static int p4_check_ctrs(struct pt_regs * const regs,
 		if (CCCR_OVF_P(low) || !(ctr & OP_CTR_OVERFLOW)) {
 			oprofile_add_sample(regs, i);
 			wrmsrl(p4_counters[real].counter_address,
-			       -(s64)reset_value[i]);
+			       -(u64)reset_value[i]);
 			CCCR_CLEAR_OVF(low);
 			wrmsr(p4_counters[real].cccr_address, low, high);
 			wrmsrl(p4_counters[real].counter_address,
-			       -(s64)reset_value[i]);
+			       -(u64)reset_value[i]);
 		}
 	}
 
-- 
cgit v1.2.3


From 44ab9a6b0e909145d42615493952fe986b1ce5c2 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Thu, 9 Jul 2009 18:33:02 +0200
Subject: x86/oprofile: Rework and simplify nmi_cpu_setup()

This patch removes the function nmi_save_registers(). Per-cpu code is
now executed only in the function nmi_cpu_setup().  Also, it renames
the per-cpu function nmi_restore_registers() to
nmi_cpu_restore_registers().

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c | 13 +++----------
 1 file changed, 3 insertions(+), 10 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index 93df76dd60f4..25da1e17815d 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -87,13 +87,6 @@ static void nmi_cpu_save_registers(struct op_msrs *msrs)
 	}
 }
 
-static void nmi_save_registers(void *dummy)
-{
-	int cpu = smp_processor_id();
-	struct op_msrs *msrs = &per_cpu(cpu_msrs, cpu);
-	nmi_cpu_save_registers(msrs);
-}
-
 static void free_msrs(void)
 {
 	int i;
@@ -137,6 +130,7 @@ static void nmi_cpu_setup(void *dummy)
 {
 	int cpu = smp_processor_id();
 	struct op_msrs *msrs = &per_cpu(cpu_msrs, cpu);
+	nmi_cpu_save_registers(msrs);
 	spin_lock(&oprofilefs_lock);
 	model->setup_ctrs(model, msrs);
 	spin_unlock(&oprofilefs_lock);
@@ -182,13 +176,12 @@ static int nmi_setup(void)
 		}
 
 	}
-	on_each_cpu(nmi_save_registers, NULL, 1);
 	on_each_cpu(nmi_cpu_setup, NULL, 1);
 	nmi_enabled = 1;
 	return 0;
 }
 
-static void nmi_restore_registers(struct op_msrs *msrs)
+static void nmi_cpu_restore_registers(struct op_msrs *msrs)
 {
 	struct op_msr *counters = msrs->counters;
 	struct op_msr *controls = msrs->controls;
@@ -220,7 +213,7 @@ static void nmi_cpu_shutdown(void *dummy)
 	apic_write(APIC_LVTERR, v | APIC_LVT_MASKED);
 	apic_write(APIC_LVTPC, per_cpu(saved_lvtpc, cpu));
 	apic_write(APIC_LVTERR, v);
-	nmi_restore_registers(msrs);
+	nmi_cpu_restore_registers(msrs);
 }
 
 static void nmi_shutdown(void)
-- 
cgit v1.2.3


From 6e63ea4b0b14ff5fb8a3ca704fcda7d28b95f079 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Tue, 7 Jul 2009 19:25:39 +0200
Subject: x86/oprofile: Whitespaces changes only

This patch fixes whitespace changes of code that will be touched in
follow-on patches.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c       | 12 ++++++------
 arch/x86/oprofile/op_model_amd.c  | 12 ++++++------
 arch/x86/oprofile/op_model_p4.c   |  8 ++++----
 arch/x86/oprofile/op_model_ppro.c |  8 ++++----
 4 files changed, 20 insertions(+), 20 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index 25da1e17815d..fca8dc94531e 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -516,12 +516,12 @@ int __init op_nmi_init(struct oprofile_operations *ops)
 	register_cpu_notifier(&oprofile_cpu_nb);
 #endif
 	/* default values, can be overwritten by model */
-	ops->create_files = nmi_create_files;
-	ops->setup = nmi_setup;
-	ops->shutdown = nmi_shutdown;
-	ops->start = nmi_start;
-	ops->stop = nmi_stop;
-	ops->cpu_type = cpu_type;
+	ops->create_files	= nmi_create_files;
+	ops->setup		= nmi_setup;
+	ops->shutdown		= nmi_shutdown;
+	ops->start		= nmi_start;
+	ops->stop		= nmi_stop;
+	ops->cpu_type		= cpu_type;
 
 	if (model->init)
 		ret = model->init(ops);
diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index 7ca8306aefae..f676f8825a3f 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -91,7 +91,7 @@ static void op_amd_setup_ctrs(struct op_x86_model_spec const *model,
 	int i;
 
 	/* clear all counters */
-	for (i = 0 ; i < NUM_CONTROLS; ++i) {
+	for (i = 0; i < NUM_CONTROLS; ++i) {
 		if (unlikely(!msrs->controls[i].addr))
 			continue;
 		rdmsrl(msrs->controls[i].addr, val);
@@ -229,7 +229,7 @@ static int op_amd_check_ctrs(struct pt_regs * const regs,
 	u64 val;
 	int i;
 
-	for (i = 0 ; i < NUM_COUNTERS; ++i) {
+	for (i = 0; i < NUM_COUNTERS; ++i) {
 		if (!reset_value[i])
 			continue;
 		rdmsrl(msrs->counters[i].addr, val);
@@ -250,7 +250,7 @@ static void op_amd_start(struct op_msrs const * const msrs)
 {
 	u64 val;
 	int i;
-	for (i = 0 ; i < NUM_COUNTERS ; ++i) {
+	for (i = 0; i < NUM_COUNTERS; ++i) {
 		if (reset_value[i]) {
 			rdmsrl(msrs->controls[i].addr, val);
 			val |= ARCH_PERFMON_EVENTSEL0_ENABLE;
@@ -270,7 +270,7 @@ static void op_amd_stop(struct op_msrs const * const msrs)
 	 * Subtle: stop on all counters to avoid race with setting our
 	 * pm callback
 	 */
-	for (i = 0 ; i < NUM_COUNTERS ; ++i) {
+	for (i = 0; i < NUM_COUNTERS; ++i) {
 		if (!reset_value[i])
 			continue;
 		rdmsrl(msrs->controls[i].addr, val);
@@ -285,11 +285,11 @@ static void op_amd_shutdown(struct op_msrs const * const msrs)
 {
 	int i;
 
-	for (i = 0 ; i < NUM_COUNTERS ; ++i) {
+	for (i = 0; i < NUM_COUNTERS; ++i) {
 		if (msrs->counters[i].addr)
 			release_perfctr_nmi(MSR_K7_PERFCTR0 + i);
 	}
-	for (i = 0 ; i < NUM_CONTROLS ; ++i) {
+	for (i = 0; i < NUM_CONTROLS; ++i) {
 		if (msrs->controls[i].addr)
 			release_evntsel_nmi(MSR_K7_EVNTSEL0 + i);
 	}
diff --git a/arch/x86/oprofile/op_model_p4.c b/arch/x86/oprofile/op_model_p4.c
index 9db9e361182c..5921b7fc724b 100644
--- a/arch/x86/oprofile/op_model_p4.c
+++ b/arch/x86/oprofile/op_model_p4.c
@@ -558,7 +558,7 @@ static void p4_setup_ctrs(struct op_x86_model_spec const *model,
 	}
 
 	/* clear the cccrs we will use */
-	for (i = 0 ; i < num_counters ; i++) {
+	for (i = 0; i < num_counters; i++) {
 		if (unlikely(!msrs->controls[i].addr))
 			continue;
 		rdmsr(p4_counters[VIRT_CTR(stag, i)].cccr_address, low, high);
@@ -575,7 +575,7 @@ static void p4_setup_ctrs(struct op_x86_model_spec const *model,
 	}
 
 	/* setup all counters */
-	for (i = 0 ; i < num_counters ; ++i) {
+	for (i = 0; i < num_counters; ++i) {
 		if (counter_config[i].enabled && msrs->controls[i].addr) {
 			reset_value[i] = counter_config[i].count;
 			pmc_setup_one_p4_counter(i);
@@ -678,7 +678,7 @@ static void p4_shutdown(struct op_msrs const * const msrs)
 {
 	int i;
 
-	for (i = 0 ; i < num_counters ; ++i) {
+	for (i = 0; i < num_counters; ++i) {
 		if (msrs->counters[i].addr)
 			release_perfctr_nmi(msrs->counters[i].addr);
 	}
@@ -687,7 +687,7 @@ static void p4_shutdown(struct op_msrs const * const msrs)
 	 * conjunction with the counter registers (hence the starting offset).
 	 * This saves a few bits.
 	 */
-	for (i = num_counters ; i < num_controls ; ++i) {
+	for (i = num_counters; i < num_controls; ++i) {
 		if (msrs->controls[i].addr)
 			release_evntsel_nmi(msrs->controls[i].addr);
 	}
diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c
index cd72d5c73b49..570d717c3308 100644
--- a/arch/x86/oprofile/op_model_ppro.c
+++ b/arch/x86/oprofile/op_model_ppro.c
@@ -81,7 +81,7 @@ static void ppro_setup_ctrs(struct op_x86_model_spec const *model,
 	}
 
 	/* clear all counters */
-	for (i = 0 ; i < num_counters; ++i) {
+	for (i = 0; i < num_counters; ++i) {
 		if (unlikely(!msrs->controls[i].addr))
 			continue;
 		rdmsrl(msrs->controls[i].addr, val);
@@ -125,7 +125,7 @@ static int ppro_check_ctrs(struct pt_regs * const regs,
 	if (unlikely(!reset_value))
 		goto out;
 
-	for (i = 0 ; i < num_counters; ++i) {
+	for (i = 0; i < num_counters; ++i) {
 		if (!reset_value[i])
 			continue;
 		rdmsrl(msrs->counters[i].addr, val);
@@ -188,11 +188,11 @@ static void ppro_shutdown(struct op_msrs const * const msrs)
 {
 	int i;
 
-	for (i = 0 ; i < num_counters ; ++i) {
+	for (i = 0; i < num_counters; ++i) {
 		if (msrs->counters[i].addr)
 			release_perfctr_nmi(MSR_P6_PERFCTR0 + i);
 	}
-	for (i = 0 ; i < num_counters ; ++i) {
+	for (i = 0; i < num_counters; ++i) {
 		if (msrs->controls[i].addr)
 			release_evntsel_nmi(MSR_P6_EVNTSEL0 + i);
 	}
-- 
cgit v1.2.3


From 6b2b171a774af256082635b53ac387b1613b7b4c Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Date: Mon, 8 Jun 2009 02:53:50 -0700
Subject: x86/acpi: acpi_parse_madt_ioapic_entries: remove redundant braces

We don't put braces around a single statement.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
---
 arch/x86/kernel/acpi/boot.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c
index 6b8ca3a0285d..ce31c1af854f 100644
--- a/arch/x86/kernel/acpi/boot.c
+++ b/arch/x86/kernel/acpi/boot.c
@@ -1179,9 +1179,8 @@ static int __init acpi_parse_madt_ioapic_entries(void)
 	 * If MPS is present, it will handle them,
 	 * otherwise the system will stay in PIC mode
 	 */
-	if (acpi_disabled || acpi_noirq) {
+	if (acpi_disabled || acpi_noirq)
 		return -ENODEV;
-	}
 
 	if (!cpu_has_apic)
 		return -ENODEV;
-- 
cgit v1.2.3


From 2f210deba9887dd9143b63b217506f1ac152e91c Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Date: Mon, 8 Jun 2009 02:55:22 -0700
Subject: x86/ioapic.c: ioapic_modify_irq is too large to inline

If ioapic_modify_irq() is marked inline, it gets inlined several times.
Un-inlining it saves around 200 bytes in .text for me.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
---
 arch/x86/kernel/apic/io_apic.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 90b5e6efa938..82271eb87bb0 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -556,9 +556,9 @@ static void __init replace_pin_at_irq_node(struct irq_cfg *cfg, int node,
 		add_pin_to_irq_node(cfg, node, newapic, newpin);
 }
 
-static inline void io_apic_modify_irq(struct irq_cfg *cfg,
-				int mask_and, int mask_or,
-				void (*final)(struct irq_pin_list *entry))
+static void io_apic_modify_irq(struct irq_cfg *cfg,
+			       int mask_and, int mask_or,
+			       void (*final)(struct irq_pin_list *entry))
 {
 	int pin;
 	struct irq_pin_list *entry;
-- 
cgit v1.2.3


From 890aeacf64c55a7ada7054a140d249ab13899f2d Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Date: Mon, 8 Jun 2009 02:57:43 -0700
Subject: x86/ioapic.c: unify __mask_IO_APIC_irq()

The main difference between 32 and 64-bit __mask_IO_APIC_irq() does a
readback from the I/O APIC to synchronize it.

If there's a hardware requirement to do a readback sync after updating
an APIC register, then it will be a hardware requrement regardless of
whether the kernel is compiled 32 or 64-bit.

Unify __mask_IO_APIC_irq() using the 64-bit version which always syncs
with io_apic_sync().

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
---
 arch/x86/kernel/apic/io_apic.c | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 82271eb87bb0..f8aa5461071b 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -580,7 +580,6 @@ static void __unmask_IO_APIC_irq(struct irq_cfg *cfg)
 	io_apic_modify_irq(cfg, ~IO_APIC_REDIR_MASKED, 0, NULL);
 }
 
-#ifdef CONFIG_X86_64
 static void io_apic_sync(struct irq_pin_list *entry)
 {
 	/*
@@ -596,12 +595,8 @@ static void __mask_IO_APIC_irq(struct irq_cfg *cfg)
 {
 	io_apic_modify_irq(cfg, ~0, IO_APIC_REDIR_MASKED, &io_apic_sync);
 }
-#else /* CONFIG_X86_32 */
-static void __mask_IO_APIC_irq(struct irq_cfg *cfg)
-{
-	io_apic_modify_irq(cfg, ~0, IO_APIC_REDIR_MASKED, NULL);
-}
 
+#ifdef CONFIG_X86_32
 static void __mask_and_edge_IO_APIC_irq(struct irq_cfg *cfg)
 {
 	io_apic_modify_irq(cfg, ~IO_APIC_REDIR_LEVEL_TRIGGER,
-- 
cgit v1.2.3


From 916a0fe739f151664f7f07b42543ae6fd4caec49 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Date: Mon, 8 Jun 2009 03:00:22 -0700
Subject: x86/ioapic.c: remove #ifdef for 82093AA workaround

While no 64-bit hardware will have a version 0x11 I/O APIC which needs
the level/edge bug workaround, that's not a particular reason to use
CONFIG_X86_32 to #ifdef the code out.  Most 32-bit machines will no
longer need the workaround either, so the test to see whether it is
necessary should be more fine-grained than "32-bit=yes, 64-bit=no".

(Also fix formatting of block comment.)

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
---
 arch/x86/kernel/apic/io_apic.c | 47 +++++++++++++++++-------------------------
 1 file changed, 19 insertions(+), 28 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index f8aa5461071b..1a3414442583 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -596,7 +596,6 @@ static void __mask_IO_APIC_irq(struct irq_cfg *cfg)
 	io_apic_modify_irq(cfg, ~0, IO_APIC_REDIR_MASKED, &io_apic_sync);
 }
 
-#ifdef CONFIG_X86_32
 static void __mask_and_edge_IO_APIC_irq(struct irq_cfg *cfg)
 {
 	io_apic_modify_irq(cfg, ~IO_APIC_REDIR_LEVEL_TRIGGER,
@@ -608,7 +607,6 @@ static void __unmask_and_level_IO_APIC_irq(struct irq_cfg *cfg)
 	io_apic_modify_irq(cfg, ~IO_APIC_REDIR_MASKED,
 			IO_APIC_REDIR_LEVEL_TRIGGER, NULL);
 }
-#endif /* CONFIG_X86_32 */
 
 static void mask_IO_APIC_irq_desc(struct irq_desc *desc)
 {
@@ -2510,11 +2508,8 @@ atomic_t irq_mis_count;
 static void ack_apic_level(unsigned int irq)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
-
-#ifdef CONFIG_X86_32
 	unsigned long v;
 	int i;
-#endif
 	struct irq_cfg *cfg;
 	int do_unmask_irq = 0;
 
@@ -2527,31 +2522,28 @@ static void ack_apic_level(unsigned int irq)
 	}
 #endif
 
-#ifdef CONFIG_X86_32
 	/*
-	* It appears there is an erratum which affects at least version 0x11
-	* of I/O APIC (that's the 82093AA and cores integrated into various
-	* chipsets).  Under certain conditions a level-triggered interrupt is
-	* erroneously delivered as edge-triggered one but the respective IRR
-	* bit gets set nevertheless.  As a result the I/O unit expects an EOI
-	* message but it will never arrive and further interrupts are blocked
-	* from the source.  The exact reason is so far unknown, but the
-	* phenomenon was observed when two consecutive interrupt requests
-	* from a given source get delivered to the same CPU and the source is
-	* temporarily disabled in between.
-	*
-	* A workaround is to simulate an EOI message manually.  We achieve it
-	* by setting the trigger mode to edge and then to level when the edge
-	* trigger mode gets detected in the TMR of a local APIC for a
-	* level-triggered interrupt.  We mask the source for the time of the
-	* operation to prevent an edge-triggered interrupt escaping meanwhile.
-	* The idea is from Manfred Spraul.  --macro
-	*/
+	 * It appears there is an erratum which affects at least version 0x11
+	 * of I/O APIC (that's the 82093AA and cores integrated into various
+	 * chipsets).  Under certain conditions a level-triggered interrupt is
+	 * erroneously delivered as edge-triggered one but the respective IRR
+	 * bit gets set nevertheless.  As a result the I/O unit expects an EOI
+	 * message but it will never arrive and further interrupts are blocked
+	 * from the source.  The exact reason is so far unknown, but the
+	 * phenomenon was observed when two consecutive interrupt requests
+	 * from a given source get delivered to the same CPU and the source is
+	 * temporarily disabled in between.
+	 *
+	 * A workaround is to simulate an EOI message manually.  We achieve it
+	 * by setting the trigger mode to edge and then to level when the edge
+	 * trigger mode gets detected in the TMR of a local APIC for a
+	 * level-triggered interrupt.  We mask the source for the time of the
+	 * operation to prevent an edge-triggered interrupt escaping meanwhile.
+	 * The idea is from Manfred Spraul.  --macro
+	 */
 	cfg = desc->chip_data;
 	i = cfg->vector;
-
 	v = apic_read(APIC_TMR + ((i & ~0x1f) >> 1));
-#endif
 
 	/*
 	 * We must acknowledge the irq before we move it or the acknowledge will
@@ -2593,7 +2585,7 @@ static void ack_apic_level(unsigned int irq)
 		unmask_IO_APIC_irq_desc(desc);
 	}
 
-#ifdef CONFIG_X86_32
+	/* Tail end of version 0x11 I/O APIC bug workaround */
 	if (!(v & (1 << (i & 0x1f)))) {
 		atomic_inc(&irq_mis_count);
 		spin_lock(&ioapic_lock);
@@ -2601,7 +2593,6 @@ static void ack_apic_level(unsigned int irq)
 		__unmask_and_level_IO_APIC_irq(cfg);
 		spin_unlock(&ioapic_lock);
 	}
-#endif
 }
 
 #ifdef CONFIG_INTR_REMAP
-- 
cgit v1.2.3


From 83c21bedf63ce92a2dd82ae2c7a96179b0aa4372 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Date: Mon, 8 Jun 2009 03:13:04 -0700
Subject: x86/ioapic.c: remove redundant declaration of irq_pin_list

The structure is defined immediately below, so there's no need
to forward declare it.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
---
 arch/x86/kernel/apic/io_apic.c | 2 --
 1 file changed, 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 1a3414442583..ec52e0c045c4 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -116,8 +116,6 @@ static int __init parse_noapic(char *str)
 }
 early_param("noapic", parse_noapic);
 
-struct irq_pin_list;
-
 /*
  * This is performance-critical, we want to do it O(1)
  *
-- 
cgit v1.2.3


From 8e13d697febc1ba17e70ed88789255c8bc25aa41 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Date: Mon, 8 Jun 2009 03:14:59 -0700
Subject: x86/ioapic.c: move lost comment to what seems like appropriate place

The comment got separated from its subject, so move it to what
appears to be the right place, and update to describe the current
structure.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
---
 arch/x86/kernel/apic/io_apic.c | 12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index ec52e0c045c4..a097a773bc76 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -116,13 +116,6 @@ static int __init parse_noapic(char *str)
 }
 early_param("noapic", parse_noapic);
 
-/*
- * This is performance-critical, we want to do it O(1)
- *
- * the indexing order of this array favors 1:1 mappings
- * between pins and IRQs.
- */
-
 struct irq_pin_list {
 	int apic, pin;
 	struct irq_pin_list *next;
@@ -137,6 +130,11 @@ static struct irq_pin_list *get_one_free_irq_2_pin(int node)
 	return pin;
 }
 
+/*
+ * This is performance-critical, we want to do it O(1)
+ *
+ * Most irqs are mapped 1:1 with pins.
+ */
 struct irq_cfg {
 	struct irq_pin_list *irq_2_pin;
 	cpumask_var_t domain;
-- 
cgit v1.2.3


From d8c52063ed85dda61b70bc05b90711478db5dc17 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Date: Mon, 8 Jun 2009 03:17:58 -0700
Subject: x86/ioapic.c: convert io_apic_level_ack_pending loop to normal for()
 loop

Convert the unconventional loop in io_apic_level_ack_pending() to
a conventional for() loop.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
---
 arch/x86/kernel/apic/io_apic.c | 8 +-------
 1 file changed, 1 insertion(+), 7 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index a097a773bc76..0d0401802d4f 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -410,13 +410,10 @@ static bool io_apic_level_ack_pending(struct irq_cfg *cfg)
 	unsigned long flags;
 
 	spin_lock_irqsave(&ioapic_lock, flags);
-	entry = cfg->irq_2_pin;
-	for (;;) {
+	for (entry = cfg->irq_2_pin; entry != NULL; entry = entry->next) {
 		unsigned int reg;
 		int pin;
 
-		if (!entry)
-			break;
 		pin = entry->pin;
 		reg = io_apic_read(entry->apic, 0x10 + pin*2);
 		/* Is the remote IRR bit set? */
@@ -424,9 +421,6 @@ static bool io_apic_level_ack_pending(struct irq_cfg *cfg)
 			spin_unlock_irqrestore(&ioapic_lock, flags);
 			return true;
 		}
-		if (!entry->next)
-			break;
-		entry = entry->next;
 	}
 	spin_unlock_irqrestore(&ioapic_lock, flags);
 
-- 
cgit v1.2.3


From 875e68ec32fc5495f3edf987aaae1c52306184b7 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Date: Mon, 8 Jun 2009 03:24:11 -0700
Subject: x86/ioapic.c: simplify add_pin_to_irq_node()

Rather than duplicating the same alloc/init code twice, restructure
the function to look for duplicates and then add an entry
if none is found.

This function is not performance critical; all but one of its callers
are __init functions, and the non-__init caller is for PCI device setup.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
---
 arch/x86/kernel/apic/io_apic.c | 28 ++++++++--------------------
 1 file changed, 8 insertions(+), 20 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 0d0401802d4f..d9e8f19088d4 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -490,34 +490,22 @@ static void ioapic_mask_entry(int apic, int pin)
  */
 static void add_pin_to_irq_node(struct irq_cfg *cfg, int node, int apic, int pin)
 {
-	struct irq_pin_list *entry;
+	struct irq_pin_list **entryp, *entry;
 
-	entry = cfg->irq_2_pin;
-	if (!entry) {
-		entry = get_one_free_irq_2_pin(node);
-		if (!entry) {
-			printk(KERN_ERR "can not alloc irq_2_pin to add %d - %d\n",
-					apic, pin);
-			return;
-		}
-		cfg->irq_2_pin = entry;
-		entry->apic = apic;
-		entry->pin = pin;
-		return;
-	}
-
-	while (entry->next) {
+	for (entryp = &cfg->irq_2_pin;
+	     *entryp != NULL;
+	     entryp = &(*entryp)->next) {
+		entry = *entryp;
 		/* not again, please */
 		if (entry->apic == apic && entry->pin == pin)
 			return;
-
-		entry = entry->next;
 	}
 
-	entry->next = get_one_free_irq_2_pin(node);
-	entry = entry->next;
+	entry = get_one_free_irq_2_pin(node);
 	entry->apic = apic;
 	entry->pin = pin;
+
+	*entryp = entry;
 }
 
 /*
-- 
cgit v1.2.3


From 535b64291a9d1ff8bc54642494a5fce27e1e1170 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Date: Mon, 8 Jun 2009 03:29:26 -0700
Subject: x86/ioapic.c: convert replace_pin_at_irq_node to conventional for()
 loop

Use a conventional for() loop in replace_pin_at_irq_node().

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
---
 arch/x86/kernel/apic/io_apic.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index d9e8f19088d4..9386976b675b 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -515,10 +515,10 @@ static void __init replace_pin_at_irq_node(struct irq_cfg *cfg, int node,
 				      int oldapic, int oldpin,
 				      int newapic, int newpin)
 {
-	struct irq_pin_list *entry = cfg->irq_2_pin;
+	struct irq_pin_list *entry;
 	int replaced = 0;
 
-	while (entry) {
+	for (entry = cfg->irq_2_pin; entry != NULL; entry = entry->next) {
 		if (entry->apic == oldapic && entry->pin == oldpin) {
 			entry->apic = newapic;
 			entry->pin = newpin;
@@ -526,7 +526,6 @@ static void __init replace_pin_at_irq_node(struct irq_cfg *cfg, int node,
 			/* every one is different, right? */
 			break;
 		}
-		entry = entry->next;
 	}
 
 	/* why? call replace before add? */
-- 
cgit v1.2.3


From 4eea6fff612f54380dd642b045bf03ac0613fe3e Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Date: Mon, 8 Jun 2009 03:32:15 -0700
Subject: x86/ioapic.c: clean up replace_pin_at_irq_node logic and comments

There's no need for a control variable in replace_pin_at_irq_node();
it can just return if it finds the old apic/pin to replace.

If the loop terminates, then it didn't find the old apic/pin, so it can
add the new ones.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
---
 arch/x86/kernel/apic/io_apic.c | 13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 9386976b675b..8245e62ed93f 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -512,25 +512,22 @@ static void add_pin_to_irq_node(struct irq_cfg *cfg, int node, int apic, int pin
  * Reroute an IRQ to a different pin.
  */
 static void __init replace_pin_at_irq_node(struct irq_cfg *cfg, int node,
-				      int oldapic, int oldpin,
-				      int newapic, int newpin)
+					   int oldapic, int oldpin,
+					   int newapic, int newpin)
 {
 	struct irq_pin_list *entry;
-	int replaced = 0;
 
 	for (entry = cfg->irq_2_pin; entry != NULL; entry = entry->next) {
 		if (entry->apic == oldapic && entry->pin == oldpin) {
 			entry->apic = newapic;
 			entry->pin = newpin;
-			replaced = 1;
 			/* every one is different, right? */
-			break;
+			return;
 		}
 	}
 
-	/* why? call replace before add? */
-	if (!replaced)
-		add_pin_to_irq_node(cfg, node, newapic, newpin);
+	/* old apic/pin didn't exist, so just add new ones */
+	add_pin_to_irq_node(cfg, node, newapic, newpin);
 }
 
 static void io_apic_modify_irq(struct irq_cfg *cfg,
-- 
cgit v1.2.3


From 638f2f8c52a92c15ebda9e50d84c1ab56fc42e42 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Date: Mon, 8 Jun 2009 03:37:52 -0700
Subject: x86/ioapic.c: convert __target_IO_APIC_irq to conventional for() loop

Use a normal for() loop in __target_IO_APIC_irq().

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
---
 arch/x86/kernel/apic/io_apic.c | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 8245e62ed93f..17883cd82592 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -2236,13 +2236,9 @@ static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq
 	struct irq_pin_list *entry;
 	u8 vector = cfg->vector;
 
-	entry = cfg->irq_2_pin;
-	for (;;) {
+	for (entry = cfg->irq_2_pin; entry != NULL; entry = entry->next) {
 		unsigned int reg;
 
-		if (!entry)
-			break;
-
 		apic = entry->apic;
 		pin = entry->pin;
 		/*
@@ -2255,9 +2251,6 @@ static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq
 		reg &= ~IO_APIC_REDIR_VECTOR_MASK;
 		reg |= vector;
 		io_apic_modify(apic, 0x10 + pin*2, reg);
-		if (!entry->next)
-			break;
-		entry = entry->next;
 	}
 }
 
-- 
cgit v1.2.3


From e25371d60cb06a44d7a32d7966ab9bfbeacb9390 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Date: Mon, 8 Jun 2009 03:49:01 -0700
Subject: x86/ioapic.c: unify ioapic_retrigger_irq()

The 32 and 64-bit versions of ioapic_retrigger_irq() are identical
except the 64-bit one takes vector_lock.  vector_lock is defined and
used on 32-bit too, so just use a common ioapic_retrigger_irq().

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
---
 arch/x86/kernel/apic/io_apic.c | 9 ---------
 1 file changed, 9 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 17883cd82592..cf51b0b58c56 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -2178,7 +2178,6 @@ static unsigned int startup_ioapic_irq(unsigned int irq)
 	return was_pending;
 }
 
-#ifdef CONFIG_X86_64
 static int ioapic_retrigger_irq(unsigned int irq)
 {
 
@@ -2191,14 +2190,6 @@ static int ioapic_retrigger_irq(unsigned int irq)
 
 	return 1;
 }
-#else
-static int ioapic_retrigger_irq(unsigned int irq)
-{
-	apic->send_IPI_self(irq_cfg(irq)->vector);
-
-	return 1;
-}
-#endif
 
 /*
  * Level and edge triggered IO-APIC interrupts need different handling,
-- 
cgit v1.2.3


From 254e0a6bff87ab8b22293c4bd1443507df698407 Mon Sep 17 00:00:00 2001
From: Akinobu Mita <akinobu.mita@gmail.com>
Date: Sun, 19 Jul 2009 00:08:54 +0900
Subject: x86: Use get_desc_base()

Use get_desc_base() to get the base address in desc_struct

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
LKML-Reference: <20090718150853.GA11294@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/doublefault_32.c | 4 +---
 arch/x86/kernel/step.c           | 9 ++++-----
 2 files changed, 5 insertions(+), 8 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/doublefault_32.c b/arch/x86/kernel/doublefault_32.c
index b4f14c6c09d9..37250fe490b1 100644
--- a/arch/x86/kernel/doublefault_32.c
+++ b/arch/x86/kernel/doublefault_32.c
@@ -27,9 +27,7 @@ static void doublefault_fn(void)
 
 	if (ptr_ok(gdt)) {
 		gdt += GDT_ENTRY_TSS << 3;
-		tss = *(u16 *)(gdt+2);
-		tss += *(u8 *)(gdt+4) << 16;
-		tss += *(u8 *)(gdt+7) << 24;
+		tss = get_desc_base((struct desc_struct *)gdt);
 		printk(KERN_EMERG "double fault, tss at %08lx\n", tss);
 
 		if (ptr_ok(tss)) {
diff --git a/arch/x86/kernel/step.c b/arch/x86/kernel/step.c
index e8b9863ef8c4..3149032ff107 100644
--- a/arch/x86/kernel/step.c
+++ b/arch/x86/kernel/step.c
@@ -4,6 +4,7 @@
 #include <linux/sched.h>
 #include <linux/mm.h>
 #include <linux/ptrace.h>
+#include <asm/desc.h>
 
 unsigned long convert_ip_to_linear(struct task_struct *child, struct pt_regs *regs)
 {
@@ -23,7 +24,7 @@ unsigned long convert_ip_to_linear(struct task_struct *child, struct pt_regs *re
 	 * and APM bios ones we just ignore here.
 	 */
 	if ((seg & SEGMENT_TI_MASK) == SEGMENT_LDT) {
-		u32 *desc;
+		struct desc_struct *desc;
 		unsigned long base;
 
 		seg &= ~7UL;
@@ -33,12 +34,10 @@ unsigned long convert_ip_to_linear(struct task_struct *child, struct pt_regs *re
 			addr = -1L; /* bogus selector, access would fault */
 		else {
 			desc = child->mm->context.ldt + seg;
-			base = ((desc[0] >> 16) |
-				((desc[1] & 0xff) << 16) |
-				(desc[1] & 0xff000000));
+			base = get_desc_base(desc);
 
 			/* 16-bit code segment? */
-			if (!((desc[1] >> 22) & 1))
+			if (!desc->d)
 				addr &= 0xffff;
 			addr += base;
 		}
-- 
cgit v1.2.3


From fde0312d01b60a3fd5dc56e69a9613defbbc7097 Mon Sep 17 00:00:00 2001
From: Akinobu Mita <akinobu.mita@gmail.com>
Date: Sun, 19 Jul 2009 00:09:56 +0900
Subject: x86: Remove unused patch_espfix_desc()

patch_espfix_desc() is not used after commit
dc4c2a0aed3b09f6e255bd5c3faa50fe6e0b2ded

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
LKML-Reference: <20090718150955.GB11294@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/include/asm/traps.h |  4 +---
 arch/x86/kernel/traps.c      | 21 ---------------------
 2 files changed, 1 insertion(+), 24 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/traps.h b/arch/x86/include/asm/traps.h
index bfd74c032fca..4da91ad69e0d 100644
--- a/arch/x86/include/asm/traps.h
+++ b/arch/x86/include/asm/traps.h
@@ -81,9 +81,7 @@ extern int panic_on_unrecovered_nmi;
 
 void math_error(void __user *);
 void math_emulate(struct math_emu_info *);
-#ifdef CONFIG_X86_32
-unsigned long patch_espfix_desc(unsigned long, unsigned long);
-#else
+#ifndef CONFIG_X86_32
 asmlinkage void smp_thermal_interrupt(void);
 asmlinkage void mce_threshold_interrupt(void);
 #endif
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index 5204332f475d..236794110207 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -786,27 +786,6 @@ do_spurious_interrupt_bug(struct pt_regs *regs, long error_code)
 #endif
 }
 
-#ifdef CONFIG_X86_32
-unsigned long patch_espfix_desc(unsigned long uesp, unsigned long kesp)
-{
-	struct desc_struct *gdt = get_cpu_gdt_table(smp_processor_id());
-	unsigned long base = (kesp - uesp) & -THREAD_SIZE;
-	unsigned long new_kesp = kesp - base;
-	unsigned long lim_pages = (new_kesp | (THREAD_SIZE - 1)) >> PAGE_SHIFT;
-	__u64 desc = *(__u64 *)&gdt[GDT_ENTRY_ESPFIX_SS];
-
-	/* Set up base for espfix segment */
-	desc &= 0x00f0ff0000000000ULL;
-	desc |=	((((__u64)base) << 16) & 0x000000ffffff0000ULL) |
-		((((__u64)base) << 32) & 0xff00000000000000ULL) |
-		((((__u64)lim_pages) << 32) & 0x000f000000000000ULL) |
-		(lim_pages & 0xffff);
-	*(__u64 *)&gdt[GDT_ENTRY_ESPFIX_SS] = desc;
-
-	return new_kesp;
-}
-#endif
-
 asmlinkage void __attribute__((weak)) smp_thermal_interrupt(void)
 {
 }
-- 
cgit v1.2.3


From 57594742a2b545f8f114cda34f15650be8ae976d Mon Sep 17 00:00:00 2001
From: Akinobu Mita <akinobu.mita@gmail.com>
Date: Sun, 19 Jul 2009 00:11:06 +0900
Subject: x86: Introduce set_desc_base() and set_desc_limit()

Rename set_base()/set_limit to set_desc_base()/set_desc_limit()
and rewrite them in C. These are naturally introduced by the
idea of get_desc_base()/get_desc_limit().

The conversion actually found the bug in apm_32.c:
bad_bios_desc is written at run-time, but it is defined const
variable.

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
LKML-Reference: <20090718151105.GC11294@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/include/asm/desc.h           | 13 +++++++++++++
 arch/x86/include/asm/stackprotector.h |  4 +---
 arch/x86/include/asm/system.h         | 27 ---------------------------
 arch/x86/kernel/apm_32.c              | 18 +++++++++---------
 drivers/pnp/pnpbios/bioscalls.c       | 21 +++++++++++----------
 5 files changed, 34 insertions(+), 49 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/desc.h b/arch/x86/include/asm/desc.h
index c993e9e0fed4..e8de2f6f5ca5 100644
--- a/arch/x86/include/asm/desc.h
+++ b/arch/x86/include/asm/desc.h
@@ -291,11 +291,24 @@ static inline unsigned long get_desc_base(const struct desc_struct *desc)
 	return desc->base0 | ((desc->base1) << 16) | ((desc->base2) << 24);
 }
 
+static inline void set_desc_base(struct desc_struct *desc, unsigned long base)
+{
+	desc->base0 = base & 0xffff;
+	desc->base1 = (base >> 16) & 0xff;
+	desc->base2 = (base >> 24) & 0xff;
+}
+
 static inline unsigned long get_desc_limit(const struct desc_struct *desc)
 {
 	return desc->limit0 | (desc->limit << 16);
 }
 
+static inline void set_desc_limit(struct desc_struct *desc, unsigned long limit)
+{
+	desc->limit0 = limit & 0xffff;
+	desc->limit = (limit >> 16) & 0xf;
+}
+
 static inline void _set_gate(int gate, unsigned type, void *addr,
 			     unsigned dpl, unsigned ist, unsigned seg)
 {
diff --git a/arch/x86/include/asm/stackprotector.h b/arch/x86/include/asm/stackprotector.h
index c2d742c6e15f..cdc5e0b126a7 100644
--- a/arch/x86/include/asm/stackprotector.h
+++ b/arch/x86/include/asm/stackprotector.h
@@ -90,9 +90,7 @@ static inline void setup_stack_canary_segment(int cpu)
 	struct desc_struct desc;
 
 	desc = gdt_table[GDT_ENTRY_STACK_CANARY];
-	desc.base0 = canary & 0xffff;
-	desc.base1 = (canary >> 16) & 0xff;
-	desc.base2 = (canary >> 24) & 0xff;
+	set_desc_base(&desc, canary);
 	write_gdt_entry(gdt_table, GDT_ENTRY_STACK_CANARY, &desc, DESCTYPE_S);
 #endif
 }
diff --git a/arch/x86/include/asm/system.h b/arch/x86/include/asm/system.h
index 643c59b4bc6e..75c49c782e20 100644
--- a/arch/x86/include/asm/system.h
+++ b/arch/x86/include/asm/system.h
@@ -150,33 +150,6 @@ do {									\
 #endif
 
 #ifdef __KERNEL__
-#define _set_base(addr, base) do { unsigned long __pr; \
-__asm__ __volatile__ ("movw %%dx,%1\n\t" \
-	"rorl $16,%%edx\n\t" \
-	"movb %%dl,%2\n\t" \
-	"movb %%dh,%3" \
-	:"=&d" (__pr) \
-	:"m" (*((addr)+2)), \
-	 "m" (*((addr)+4)), \
-	 "m" (*((addr)+7)), \
-	 "0" (base) \
-	); } while (0)
-
-#define _set_limit(addr, limit) do { unsigned long __lr; \
-__asm__ __volatile__ ("movw %%dx,%1\n\t" \
-	"rorl $16,%%edx\n\t" \
-	"movb %2,%%dh\n\t" \
-	"andb $0xf0,%%dh\n\t" \
-	"orb %%dh,%%dl\n\t" \
-	"movb %%dl,%2" \
-	:"=&d" (__lr) \
-	:"m" (*(addr)), \
-	 "m" (*((addr)+6)), \
-	 "0" (limit) \
-	); } while (0)
-
-#define set_base(ldt, base) _set_base(((char *)&(ldt)) , (base))
-#define set_limit(ldt, limit) _set_limit(((char *)&(ldt)) , ((limit)-1))
 
 extern void native_load_gs_index(unsigned);
 
diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c
index 79302e9a33a4..b5e841bd60d9 100644
--- a/arch/x86/kernel/apm_32.c
+++ b/arch/x86/kernel/apm_32.c
@@ -403,7 +403,7 @@ static DECLARE_WAIT_QUEUE_HEAD(apm_waitqueue);
 static DECLARE_WAIT_QUEUE_HEAD(apm_suspend_waitqueue);
 static struct apm_user *user_list;
 static DEFINE_SPINLOCK(user_list_lock);
-static const struct desc_struct	bad_bios_desc = { { { 0, 0x00409200 } } };
+static struct desc_struct bad_bios_desc = { { { 0, 0x00409200 } } };
 
 static const char driver_version[] = "1.16ac";	/* no spaces */
 
@@ -2337,8 +2337,8 @@ static int __init apm_init(void)
 	 * This is for buggy BIOS's that refer to (real mode) segment 0x40
 	 * even though they are called in protected mode.
 	 */
-	set_base(bad_bios_desc, __va((unsigned long)0x40 << 4));
-	_set_limit((char *)&bad_bios_desc, 4095 - (0x40 << 4));
+	set_desc_base(&bad_bios_desc, (unsigned long)__va(0x40UL << 4));
+	set_desc_limit(&bad_bios_desc, 4095 - (0x40 << 4));
 
 	/*
 	 * Set up the long jump entry point to the APM BIOS, which is called
@@ -2358,12 +2358,12 @@ static int __init apm_init(void)
 	 * code to that CPU.
 	 */
 	gdt = get_cpu_gdt_table(0);
-	set_base(gdt[APM_CS >> 3],
-		 __va((unsigned long)apm_info.bios.cseg << 4));
-	set_base(gdt[APM_CS_16 >> 3],
-		 __va((unsigned long)apm_info.bios.cseg_16 << 4));
-	set_base(gdt[APM_DS >> 3],
-		 __va((unsigned long)apm_info.bios.dseg << 4));
+	set_desc_base(&gdt[APM_CS >> 3],
+		 (unsigned long)__va((unsigned long)apm_info.bios.cseg << 4));
+	set_desc_base(&gdt[APM_CS_16 >> 3],
+		 (unsigned long)__va((unsigned long)apm_info.bios.cseg_16 << 4));
+	set_desc_base(&gdt[APM_DS >> 3],
+		 (unsigned long)__va((unsigned long)apm_info.bios.dseg << 4));
 
 	proc_create("apm", 0, NULL, &apm_file_ops);
 
diff --git a/drivers/pnp/pnpbios/bioscalls.c b/drivers/pnp/pnpbios/bioscalls.c
index 7e6b5a3b3281..45ad3e9cc362 100644
--- a/drivers/pnp/pnpbios/bioscalls.c
+++ b/drivers/pnp/pnpbios/bioscalls.c
@@ -55,9 +55,9 @@ __asm__(".text			\n"
 
 #define Q2_SET_SEL(cpu, selname, address, size) \
 do { \
-struct desc_struct *gdt = get_cpu_gdt_table((cpu)); \
-set_base(gdt[(selname) >> 3], (u32)(address)); \
-set_limit(gdt[(selname) >> 3], size); \
+	struct desc_struct *gdt = get_cpu_gdt_table((cpu)); \
+	set_desc_base(&gdt[(selname) >> 3], (u32)(address)); \
+	set_desc_limit(&gdt[(selname) >> 3], (size) - 1); \
 } while(0)
 
 static struct desc_struct bad_bios_desc;
@@ -479,16 +479,17 @@ void pnpbios_calls_init(union pnp_bios_install_struct *header)
 	bad_bios_desc.a = 0;
 	bad_bios_desc.b = 0x00409200;
 
-	set_base(bad_bios_desc, __va((unsigned long)0x40 << 4));
-	_set_limit((char *)&bad_bios_desc, 4095 - (0x40 << 4));
+	set_desc_base(&bad_bios_desc, (unsigned long)__va(0x40UL << 4));
+	set_desc_limit(&bad_bios_desc, 4095 - (0x40 << 4));
 	for_each_possible_cpu(i) {
 		struct desc_struct *gdt = get_cpu_gdt_table(i);
 		if (!gdt)
 			continue;
-		set_base(gdt[GDT_ENTRY_PNPBIOS_CS32], &pnp_bios_callfunc);
-		set_base(gdt[GDT_ENTRY_PNPBIOS_CS16],
-			 __va(header->fields.pm16cseg));
-		set_base(gdt[GDT_ENTRY_PNPBIOS_DS],
-			 __va(header->fields.pm16dseg));
+		set_desc_base(&gdt[GDT_ENTRY_PNPBIOS_CS32],
+			 (unsigned long)&pnp_bios_callfunc);
+		set_desc_base(&gdt[GDT_ENTRY_PNPBIOS_CS16],
+			 (unsigned long)__va(header->fields.pm16cseg));
+		set_desc_base(&gdt[GDT_ENTRY_PNPBIOS_DS],
+			 (unsigned long)__va(header->fields.pm16dseg));
 	}
 }
-- 
cgit v1.2.3


From 4d4036e0e7299c6cbb2d2421b4b30b7a409ce61a Mon Sep 17 00:00:00 2001
From: Jason Yeh <jason.yeh@amd.com>
Date: Wed, 8 Jul 2009 13:49:38 +0200
Subject: oprofile: Implement performance counter multiplexing

The number of hardware counters is limited. The multiplexing feature
enables OProfile to gather more events than counters are provided by
the hardware. This is realized by switching between events at an user
specified time interval.

A new file (/dev/oprofile/time_slice) is added for the user to specify
the timer interval in ms. If the number of events to profile is higher
than the number of hardware counters available, the patch will
schedule a work queue that switches the event counter and re-writes
the different sets of values into it. The switching mechanism needs to
be implemented for each architecture to support multiplexing. This
patch only implements AMD CPU support, but multiplexing can be easily
extended for other models and architectures.

There are follow-on patches that rework parts of this patch.

Signed-off-by: Jason Yeh <jason.yeh@amd.com>
Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/Kconfig                      |  12 +++
 arch/x86/oprofile/nmi_int.c       | 162 ++++++++++++++++++++++++++++++++++++--
 arch/x86/oprofile/op_counter.h    |   2 +-
 arch/x86/oprofile/op_model_amd.c  | 110 ++++++++++++++++++++++----
 arch/x86/oprofile/op_model_p4.c   |   4 +
 arch/x86/oprofile/op_model_ppro.c |   2 +
 arch/x86/oprofile/op_x86_model.h  |   7 ++
 drivers/oprofile/oprof.c          |  78 ++++++++++++++++++
 drivers/oprofile/oprof.h          |   2 +
 drivers/oprofile/oprofile_files.c |  43 ++++++++++
 drivers/oprofile/oprofile_stats.c |  10 +++
 include/linux/oprofile.h          |   3 +
 12 files changed, 415 insertions(+), 20 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/Kconfig b/arch/Kconfig
index 99193b160232..beea3ccebb5e 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -30,6 +30,18 @@ config OPROFILE_IBS
 
 	  If unsure, say N.
 
+config OPROFILE_EVENT_MULTIPLEX
+	bool "OProfile multiplexing support (EXPERIMENTAL)"
+	default n
+	depends on OPROFILE && X86
+	help
+	  The number of hardware counters is limited. The multiplexing
+	  feature enables OProfile to gather more events than counters
+	  are provided by the hardware. This is realized by switching
+	  between events at an user specified time interval.
+
+	  If unsure, say N.
+
 config HAVE_OPROFILE
 	bool
 
diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index fca8dc94531e..e54f6a0b35ac 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -1,11 +1,14 @@
 /**
  * @file nmi_int.c
  *
- * @remark Copyright 2002-2008 OProfile authors
+ * @remark Copyright 2002-2009 OProfile authors
  * @remark Read the file COPYING
  *
  * @author John Levon <levon@movementarian.org>
  * @author Robert Richter <robert.richter@amd.com>
+ * @author Barry Kasindorf <barry.kasindorf@amd.com>
+ * @author Jason Yeh <jason.yeh@amd.com>
+ * @author Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
  */
 
 #include <linux/init.h>
@@ -24,6 +27,12 @@
 #include "op_counter.h"
 #include "op_x86_model.h"
 
+
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+DEFINE_PER_CPU(int, switch_index);
+#endif
+
+
 static struct op_x86_model_spec const *model;
 static DEFINE_PER_CPU(struct op_msrs, cpu_msrs);
 static DEFINE_PER_CPU(unsigned long, saved_lvtpc);
@@ -31,6 +40,13 @@ static DEFINE_PER_CPU(unsigned long, saved_lvtpc);
 /* 0 == registered but off, 1 == registered and on */
 static int nmi_enabled = 0;
 
+
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+extern atomic_t multiplex_counter;
+#endif
+
+struct op_counter_config counter_config[OP_MAX_COUNTER];
+
 /* common functions */
 
 u64 op_x86_get_ctrl(struct op_x86_model_spec const *model,
@@ -95,6 +111,11 @@ static void free_msrs(void)
 		per_cpu(cpu_msrs, i).counters = NULL;
 		kfree(per_cpu(cpu_msrs, i).controls);
 		per_cpu(cpu_msrs, i).controls = NULL;
+
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+		kfree(per_cpu(cpu_msrs, i).multiplex);
+		per_cpu(cpu_msrs, i).multiplex = NULL;
+#endif
 	}
 }
 
@@ -103,6 +124,9 @@ static int allocate_msrs(void)
 	int success = 1;
 	size_t controls_size = sizeof(struct op_msr) * model->num_controls;
 	size_t counters_size = sizeof(struct op_msr) * model->num_counters;
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+	size_t multiplex_size = sizeof(struct op_msr) * model->num_virt_counters;
+#endif
 
 	int i;
 	for_each_possible_cpu(i) {
@@ -118,6 +142,14 @@ static int allocate_msrs(void)
 			success = 0;
 			break;
 		}
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+		per_cpu(cpu_msrs, i).multiplex =
+				kmalloc(multiplex_size, GFP_KERNEL);
+		if (!per_cpu(cpu_msrs, i).multiplex) {
+			success = 0;
+			break;
+		}
+#endif
 	}
 
 	if (!success)
@@ -126,6 +158,25 @@ static int allocate_msrs(void)
 	return success;
 }
 
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+
+static void nmi_setup_cpu_mux(struct op_msrs const * const msrs)
+{
+	int i;
+	struct op_msr *multiplex = msrs->multiplex;
+
+	for (i = 0; i < model->num_virt_counters; ++i) {
+		if (counter_config[i].enabled) {
+			multiplex[i].saved = -(u64)counter_config[i].count;
+		} else {
+			multiplex[i].addr  = 0;
+			multiplex[i].saved = 0;
+		}
+	}
+}
+
+#endif
+
 static void nmi_cpu_setup(void *dummy)
 {
 	int cpu = smp_processor_id();
@@ -133,6 +184,9 @@ static void nmi_cpu_setup(void *dummy)
 	nmi_cpu_save_registers(msrs);
 	spin_lock(&oprofilefs_lock);
 	model->setup_ctrs(model, msrs);
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+	nmi_setup_cpu_mux(msrs);
+#endif
 	spin_unlock(&oprofilefs_lock);
 	per_cpu(saved_lvtpc, cpu) = apic_read(APIC_LVTPC);
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
@@ -173,14 +227,52 @@ static int nmi_setup(void)
 			memcpy(per_cpu(cpu_msrs, cpu).controls,
 				per_cpu(cpu_msrs, 0).controls,
 				sizeof(struct op_msr) * model->num_controls);
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+			memcpy(per_cpu(cpu_msrs, cpu).multiplex,
+				per_cpu(cpu_msrs, 0).multiplex,
+				sizeof(struct op_msr) * model->num_virt_counters);
+#endif
 		}
-
 	}
 	on_each_cpu(nmi_cpu_setup, NULL, 1);
 	nmi_enabled = 1;
 	return 0;
 }
 
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+
+static void nmi_cpu_save_mpx_registers(struct op_msrs *msrs)
+{
+	unsigned int si = __get_cpu_var(switch_index);
+	struct op_msr *multiplex = msrs->multiplex;
+	unsigned int i;
+
+	for (i = 0; i < model->num_counters; ++i) {
+		int offset = i + si;
+		if (multiplex[offset].addr) {
+			rdmsrl(multiplex[offset].addr,
+			       multiplex[offset].saved);
+		}
+	}
+}
+
+static void nmi_cpu_restore_mpx_registers(struct op_msrs *msrs)
+{
+	unsigned int si = __get_cpu_var(switch_index);
+	struct op_msr *multiplex = msrs->multiplex;
+	unsigned int i;
+
+	for (i = 0; i < model->num_counters; ++i) {
+		int offset = i + si;
+		if (multiplex[offset].addr) {
+			wrmsrl(multiplex[offset].addr,
+			       multiplex[offset].saved);
+		}
+	}
+}
+
+#endif
+
 static void nmi_cpu_restore_registers(struct op_msrs *msrs)
 {
 	struct op_msr *counters = msrs->counters;
@@ -214,6 +306,9 @@ static void nmi_cpu_shutdown(void *dummy)
 	apic_write(APIC_LVTPC, per_cpu(saved_lvtpc, cpu));
 	apic_write(APIC_LVTERR, v);
 	nmi_cpu_restore_registers(msrs);
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+	__get_cpu_var(switch_index) = 0;
+#endif
 }
 
 static void nmi_shutdown(void)
@@ -252,16 +347,15 @@ static void nmi_stop(void)
 	on_each_cpu(nmi_cpu_stop, NULL, 1);
 }
 
-struct op_counter_config counter_config[OP_MAX_COUNTER];
-
 static int nmi_create_files(struct super_block *sb, struct dentry *root)
 {
 	unsigned int i;
 
-	for (i = 0; i < model->num_counters; ++i) {
+	for (i = 0; i < model->num_virt_counters; ++i) {
 		struct dentry *dir;
 		char buf[4];
 
+#ifndef CONFIG_OPROFILE_EVENT_MULTIPLEX
 		/* quick little hack to _not_ expose a counter if it is not
 		 * available for use.  This should protect userspace app.
 		 * NOTE:  assumes 1:1 mapping here (that counters are organized
@@ -269,6 +363,7 @@ static int nmi_create_files(struct super_block *sb, struct dentry *root)
 		 */
 		if (unlikely(!avail_to_resrv_perfctr_nmi_bit(i)))
 			continue;
+#endif /* CONFIG_OPROFILE_EVENT_MULTIPLEX */
 
 		snprintf(buf,  sizeof(buf), "%d", i);
 		dir = oprofilefs_mkdir(sb, root, buf);
@@ -283,6 +378,57 @@ static int nmi_create_files(struct super_block *sb, struct dentry *root)
 	return 0;
 }
 
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+
+static void nmi_cpu_switch(void *dummy)
+{
+	int cpu = smp_processor_id();
+	int si = per_cpu(switch_index, cpu);
+	struct op_msrs *msrs = &per_cpu(cpu_msrs, cpu);
+
+	nmi_cpu_stop(NULL);
+	nmi_cpu_save_mpx_registers(msrs);
+
+	/* move to next set */
+	si += model->num_counters;
+	if ((si > model->num_virt_counters) || (counter_config[si].count == 0))
+		per_cpu(switch_index, cpu) = 0;
+	else
+		per_cpu(switch_index, cpu) = si;
+
+	model->switch_ctrl(model, msrs);
+	nmi_cpu_restore_mpx_registers(msrs);
+
+	nmi_cpu_start(NULL);
+}
+
+
+/*
+ * Quick check to see if multiplexing is necessary.
+ * The check should be sufficient since counters are used
+ * in ordre.
+ */
+static int nmi_multiplex_on(void)
+{
+	return counter_config[model->num_counters].count ? 0 : -EINVAL;
+}
+
+static int nmi_switch_event(void)
+{
+	if (!model->switch_ctrl)
+		return -ENOSYS;		/* not implemented */
+	if (nmi_multiplex_on() < 0)
+		return -EINVAL;		/* not necessary */
+
+	on_each_cpu(nmi_cpu_switch, NULL, 1);
+
+	atomic_inc(&multiplex_counter);
+
+	return 0;
+}
+
+#endif
+
 #ifdef CONFIG_SMP
 static int oprofile_cpu_notifier(struct notifier_block *b, unsigned long action,
 				 void *data)
@@ -516,12 +662,18 @@ int __init op_nmi_init(struct oprofile_operations *ops)
 	register_cpu_notifier(&oprofile_cpu_nb);
 #endif
 	/* default values, can be overwritten by model */
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+	__raw_get_cpu_var(switch_index) = 0;
+#endif
 	ops->create_files	= nmi_create_files;
 	ops->setup		= nmi_setup;
 	ops->shutdown		= nmi_shutdown;
 	ops->start		= nmi_start;
 	ops->stop		= nmi_stop;
 	ops->cpu_type		= cpu_type;
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+	ops->switch_events	= nmi_switch_event;
+#endif
 
 	if (model->init)
 		ret = model->init(ops);
diff --git a/arch/x86/oprofile/op_counter.h b/arch/x86/oprofile/op_counter.h
index 91b6a116165e..e28398df0df2 100644
--- a/arch/x86/oprofile/op_counter.h
+++ b/arch/x86/oprofile/op_counter.h
@@ -10,7 +10,7 @@
 #ifndef OP_COUNTER_H
 #define OP_COUNTER_H
 
-#define OP_MAX_COUNTER 8
+#define OP_MAX_COUNTER 32
 
 /* Per-perfctr configuration as set via
  * oprofilefs.
diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index f676f8825a3f..fdbed3a0c877 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -9,12 +9,15 @@
  * @author Philippe Elie
  * @author Graydon Hoare
  * @author Robert Richter <robert.richter@amd.com>
- * @author Barry Kasindorf
+ * @author Barry Kasindorf <barry.kasindorf@amd.com>
+ * @author Jason Yeh <jason.yeh@amd.com>
+ * @author Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
  */
 
 #include <linux/oprofile.h>
 #include <linux/device.h>
 #include <linux/pci.h>
+#include <linux/percpu.h>
 
 #include <asm/ptrace.h>
 #include <asm/msr.h>
@@ -25,12 +28,23 @@
 
 #define NUM_COUNTERS 4
 #define NUM_CONTROLS 4
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+#define NUM_VIRT_COUNTERS 32
+#define NUM_VIRT_CONTROLS 32
+#else
+#define NUM_VIRT_COUNTERS NUM_COUNTERS
+#define NUM_VIRT_CONTROLS NUM_CONTROLS
+#endif
+
 #define OP_EVENT_MASK			0x0FFF
 #define OP_CTR_OVERFLOW			(1ULL<<31)
 
 #define MSR_AMD_EVENTSEL_RESERVED	((0xFFFFFCF0ULL<<32)|(1ULL<<21))
 
-static unsigned long reset_value[NUM_COUNTERS];
+static unsigned long reset_value[NUM_VIRT_COUNTERS];
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+DECLARE_PER_CPU(int, switch_index);
+#endif
 
 #ifdef CONFIG_OPROFILE_IBS
 
@@ -82,6 +96,16 @@ static void op_amd_fill_in_addresses(struct op_msrs * const msrs)
 		else
 			msrs->controls[i].addr = 0;
 	}
+
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+	for (i = 0; i < NUM_VIRT_COUNTERS; i++) {
+		int hw_counter = i % NUM_CONTROLS;
+		if (reserve_perfctr_nmi(MSR_K7_PERFCTR0 + i))
+			msrs->multiplex[i].addr = MSR_K7_PERFCTR0 + hw_counter;
+		else
+			msrs->multiplex[i].addr = 0;
+	}
+#endif
 }
 
 static void op_amd_setup_ctrs(struct op_x86_model_spec const *model,
@@ -90,6 +114,15 @@ static void op_amd_setup_ctrs(struct op_x86_model_spec const *model,
 	u64 val;
 	int i;
 
+	/* setup reset_value */
+	for (i = 0; i < NUM_VIRT_COUNTERS; ++i) {
+		if (counter_config[i].enabled) {
+			reset_value[i] = counter_config[i].count;
+		} else {
+			reset_value[i] = 0;
+		}
+	}
+
 	/* clear all counters */
 	for (i = 0; i < NUM_CONTROLS; ++i) {
 		if (unlikely(!msrs->controls[i].addr))
@@ -108,20 +141,49 @@ static void op_amd_setup_ctrs(struct op_x86_model_spec const *model,
 
 	/* enable active counters */
 	for (i = 0; i < NUM_COUNTERS; ++i) {
-		if (counter_config[i].enabled && msrs->counters[i].addr) {
-			reset_value[i] = counter_config[i].count;
-			wrmsrl(msrs->counters[i].addr,
-			       -(u64)counter_config[i].count);
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+		int offset = i + __get_cpu_var(switch_index);
+#else
+		int offset = i;
+#endif
+		if (counter_config[offset].enabled && msrs->counters[i].addr) {
+			/* setup counter registers */
+			wrmsrl(msrs->counters[i].addr, -(u64)reset_value[offset]);
+
+			/* setup control registers */
 			rdmsrl(msrs->controls[i].addr, val);
 			val &= model->reserved;
-			val |= op_x86_get_ctrl(model, &counter_config[i]);
+			val |= op_x86_get_ctrl(model, &counter_config[offset]);
+			wrmsrl(msrs->controls[i].addr, val);
+		}
+	}
+}
+
+
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+
+static void op_amd_switch_ctrl(struct op_x86_model_spec const *model,
+			       struct op_msrs const * const msrs)
+{
+	u64 val;
+	int i;
+
+	/* enable active counters */
+	for (i = 0; i < NUM_COUNTERS; ++i) {
+		int offset = i + __get_cpu_var(switch_index);
+		if (counter_config[offset].enabled) {
+			/* setup control registers */
+			rdmsrl(msrs->controls[i].addr, val);
+			val &= model->reserved;
+			val |= op_x86_get_ctrl(model, &counter_config[offset]);
 			wrmsrl(msrs->controls[i].addr, val);
-		} else {
-			reset_value[i] = 0;
 		}
 	}
 }
 
+#endif
+
+
 #ifdef CONFIG_OPROFILE_IBS
 
 static inline int
@@ -230,14 +292,19 @@ static int op_amd_check_ctrs(struct pt_regs * const regs,
 	int i;
 
 	for (i = 0; i < NUM_COUNTERS; ++i) {
-		if (!reset_value[i])
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+		int offset = i + __get_cpu_var(switch_index);
+#else
+		int offset = i;
+#endif
+		if (!reset_value[offset])
 			continue;
 		rdmsrl(msrs->counters[i].addr, val);
 		/* bit is clear if overflowed: */
 		if (val & OP_CTR_OVERFLOW)
 			continue;
-		oprofile_add_sample(regs, i);
-		wrmsrl(msrs->counters[i].addr, -(u64)reset_value[i]);
+		oprofile_add_sample(regs, offset);
+		wrmsrl(msrs->counters[i].addr, -(u64)reset_value[offset]);
 	}
 
 	op_amd_handle_ibs(regs, msrs);
@@ -250,8 +317,14 @@ static void op_amd_start(struct op_msrs const * const msrs)
 {
 	u64 val;
 	int i;
+
 	for (i = 0; i < NUM_COUNTERS; ++i) {
-		if (reset_value[i]) {
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+		int offset = i + __get_cpu_var(switch_index);
+#else
+		int offset = i;
+#endif
+		if (reset_value[offset]) {
 			rdmsrl(msrs->controls[i].addr, val);
 			val |= ARCH_PERFMON_EVENTSEL0_ENABLE;
 			wrmsrl(msrs->controls[i].addr, val);
@@ -271,7 +344,11 @@ static void op_amd_stop(struct op_msrs const * const msrs)
 	 * pm callback
 	 */
 	for (i = 0; i < NUM_COUNTERS; ++i) {
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+		if (!reset_value[i + per_cpu(switch_index, smp_processor_id())])
+#else
 		if (!reset_value[i])
+#endif
 			continue;
 		rdmsrl(msrs->controls[i].addr, val);
 		val &= ~ARCH_PERFMON_EVENTSEL0_ENABLE;
@@ -289,7 +366,7 @@ static void op_amd_shutdown(struct op_msrs const * const msrs)
 		if (msrs->counters[i].addr)
 			release_perfctr_nmi(MSR_K7_PERFCTR0 + i);
 	}
-	for (i = 0; i < NUM_CONTROLS; ++i) {
+	for (i = 0; i < NUM_COUNTERS; ++i) {
 		if (msrs->controls[i].addr)
 			release_evntsel_nmi(MSR_K7_EVNTSEL0 + i);
 	}
@@ -463,6 +540,8 @@ static void op_amd_exit(void) {}
 struct op_x86_model_spec const op_amd_spec = {
 	.num_counters		= NUM_COUNTERS,
 	.num_controls		= NUM_CONTROLS,
+	.num_virt_counters	= NUM_VIRT_COUNTERS,
+	.num_virt_controls	= NUM_VIRT_CONTROLS,
 	.reserved		= MSR_AMD_EVENTSEL_RESERVED,
 	.event_mask		= OP_EVENT_MASK,
 	.init			= op_amd_init,
@@ -473,4 +552,7 @@ struct op_x86_model_spec const op_amd_spec = {
 	.start			= &op_amd_start,
 	.stop			= &op_amd_stop,
 	.shutdown		= &op_amd_shutdown,
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+	.switch_ctrl		= &op_amd_switch_ctrl,
+#endif
 };
diff --git a/arch/x86/oprofile/op_model_p4.c b/arch/x86/oprofile/op_model_p4.c
index 5921b7fc724b..65b9237cde8b 100644
--- a/arch/x86/oprofile/op_model_p4.c
+++ b/arch/x86/oprofile/op_model_p4.c
@@ -698,6 +698,8 @@ static void p4_shutdown(struct op_msrs const * const msrs)
 struct op_x86_model_spec const op_p4_ht2_spec = {
 	.num_counters		= NUM_COUNTERS_HT2,
 	.num_controls		= NUM_CONTROLS_HT2,
+	.num_virt_counters	= NUM_COUNTERS_HT2,
+	.num_virt_controls	= NUM_CONTROLS_HT2,
 	.fill_in_addresses	= &p4_fill_in_addresses,
 	.setup_ctrs		= &p4_setup_ctrs,
 	.check_ctrs		= &p4_check_ctrs,
@@ -710,6 +712,8 @@ struct op_x86_model_spec const op_p4_ht2_spec = {
 struct op_x86_model_spec const op_p4_spec = {
 	.num_counters		= NUM_COUNTERS_NON_HT,
 	.num_controls		= NUM_CONTROLS_NON_HT,
+	.num_virt_counters	= NUM_COUNTERS_NON_HT,
+	.num_virt_controls	= NUM_CONTROLS_NON_HT,
 	.fill_in_addresses	= &p4_fill_in_addresses,
 	.setup_ctrs		= &p4_setup_ctrs,
 	.check_ctrs		= &p4_check_ctrs,
diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c
index 570d717c3308..098cbca5c0b0 100644
--- a/arch/x86/oprofile/op_model_ppro.c
+++ b/arch/x86/oprofile/op_model_ppro.c
@@ -206,6 +206,8 @@ static void ppro_shutdown(struct op_msrs const * const msrs)
 struct op_x86_model_spec const op_ppro_spec = {
 	.num_counters		= 2,
 	.num_controls		= 2,
+	.num_virt_counters	= 2,
+	.num_virt_controls	= 2,
 	.reserved		= MSR_PPRO_EVENTSEL_RESERVED,
 	.fill_in_addresses	= &ppro_fill_in_addresses,
 	.setup_ctrs		= &ppro_setup_ctrs,
diff --git a/arch/x86/oprofile/op_x86_model.h b/arch/x86/oprofile/op_x86_model.h
index 505489873b9d..0d07d23cb062 100644
--- a/arch/x86/oprofile/op_x86_model.h
+++ b/arch/x86/oprofile/op_x86_model.h
@@ -23,6 +23,7 @@ struct op_msr {
 struct op_msrs {
 	struct op_msr *counters;
 	struct op_msr *controls;
+	struct op_msr *multiplex;
 };
 
 struct pt_regs;
@@ -35,6 +36,8 @@ struct oprofile_operations;
 struct op_x86_model_spec {
 	unsigned int	num_counters;
 	unsigned int	num_controls;
+	unsigned int	num_virt_counters;
+	unsigned int	num_virt_controls;
 	u64		reserved;
 	u16		event_mask;
 	int		(*init)(struct oprofile_operations *ops);
@@ -47,6 +50,10 @@ struct op_x86_model_spec {
 	void		(*start)(struct op_msrs const * const msrs);
 	void		(*stop)(struct op_msrs const * const msrs);
 	void		(*shutdown)(struct op_msrs const * const msrs);
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+	void		(*switch_ctrl)(struct op_x86_model_spec const *model,
+				       struct op_msrs const * const msrs);
+#endif
 };
 
 struct op_counter_config;
diff --git a/drivers/oprofile/oprof.c b/drivers/oprofile/oprof.c
index 3cffce90f82a..7bc64af7cf99 100644
--- a/drivers/oprofile/oprof.c
+++ b/drivers/oprofile/oprof.c
@@ -12,6 +12,8 @@
 #include <linux/init.h>
 #include <linux/oprofile.h>
 #include <linux/moduleparam.h>
+#include <linux/workqueue.h>
+#include <linux/time.h>
 #include <asm/mutex.h>
 
 #include "oprof.h"
@@ -27,6 +29,15 @@ unsigned long oprofile_backtrace_depth;
 static unsigned long is_setup;
 static DEFINE_MUTEX(start_mutex);
 
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+
+static void switch_worker(struct work_struct *work);
+static DECLARE_DELAYED_WORK(switch_work, switch_worker);
+unsigned long timeout_jiffies;
+#define MULTIPLEXING_TIMER_DEFAULT 1
+
+#endif
+
 /* timer
    0 - use performance monitoring hardware if available
    1 - use the timer int mechanism regardless
@@ -87,6 +98,20 @@ out:
 	return err;
 }
 
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+
+static void start_switch_worker(void)
+{
+	schedule_delayed_work(&switch_work, timeout_jiffies);
+}
+
+static void switch_worker(struct work_struct *work)
+{
+	if (!oprofile_ops.switch_events())
+		start_switch_worker();
+}
+
+#endif
 
 /* Actually start profiling (echo 1>/dev/oprofile/enable) */
 int oprofile_start(void)
@@ -108,6 +133,11 @@ int oprofile_start(void)
 	if ((err = oprofile_ops.start()))
 		goto out;
 
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+	if (oprofile_ops.switch_events)
+		start_switch_worker();
+#endif
+
 	oprofile_started = 1;
 out:
 	mutex_unlock(&start_mutex);
@@ -123,6 +153,11 @@ void oprofile_stop(void)
 		goto out;
 	oprofile_ops.stop();
 	oprofile_started = 0;
+
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+	cancel_delayed_work_sync(&switch_work);
+#endif
+
 	/* wake up the daemon to read what remains */
 	wake_up_buffer_waiter();
 out:
@@ -155,6 +190,36 @@ post_sync:
 	mutex_unlock(&start_mutex);
 }
 
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+
+/* User inputs in ms, converts to jiffies */
+int oprofile_set_timeout(unsigned long val_msec)
+{
+	int err = 0;
+
+	mutex_lock(&start_mutex);
+
+	if (oprofile_started) {
+		err = -EBUSY;
+		goto out;
+	}
+
+	if (!oprofile_ops.switch_events) {
+		err = -EINVAL;
+		goto out;
+	}
+
+	timeout_jiffies = msecs_to_jiffies(val_msec);
+	if (timeout_jiffies == MAX_JIFFY_OFFSET)
+		timeout_jiffies = msecs_to_jiffies(MULTIPLEXING_TIMER_DEFAULT);
+
+out:
+	mutex_unlock(&start_mutex);
+	return err;
+
+}
+
+#endif
 
 int oprofile_set_backtrace(unsigned long val)
 {
@@ -179,10 +244,23 @@ out:
 	return err;
 }
 
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+
+static void __init oprofile_multiplexing_init(void)
+{
+	timeout_jiffies = msecs_to_jiffies(MULTIPLEXING_TIMER_DEFAULT);
+}
+
+#endif
+
 static int __init oprofile_init(void)
 {
 	int err;
 
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+	oprofile_multiplexing_init();
+#endif
+
 	err = oprofile_arch_init(&oprofile_ops);
 
 	if (err < 0 || timer) {
diff --git a/drivers/oprofile/oprof.h b/drivers/oprofile/oprof.h
index c288d3c24b50..ee38abcc74f3 100644
--- a/drivers/oprofile/oprof.h
+++ b/drivers/oprofile/oprof.h
@@ -27,6 +27,7 @@ extern unsigned long oprofile_buffer_watershed;
 extern struct oprofile_operations oprofile_ops;
 extern unsigned long oprofile_started;
 extern unsigned long oprofile_backtrace_depth;
+extern unsigned long timeout_jiffies;
 
 struct super_block;
 struct dentry;
@@ -35,5 +36,6 @@ void oprofile_create_files(struct super_block *sb, struct dentry *root);
 void oprofile_timer_init(struct oprofile_operations *ops);
 
 int oprofile_set_backtrace(unsigned long depth);
+int oprofile_set_timeout(unsigned long time);
 
 #endif /* OPROF_H */
diff --git a/drivers/oprofile/oprofile_files.c b/drivers/oprofile/oprofile_files.c
index 5d36ffc30dd5..468ec3e4f856 100644
--- a/drivers/oprofile/oprofile_files.c
+++ b/drivers/oprofile/oprofile_files.c
@@ -9,6 +9,7 @@
 
 #include <linux/fs.h>
 #include <linux/oprofile.h>
+#include <linux/jiffies.h>
 
 #include "event_buffer.h"
 #include "oprofile_stats.h"
@@ -22,6 +23,45 @@ unsigned long oprofile_buffer_size;
 unsigned long oprofile_cpu_buffer_size;
 unsigned long oprofile_buffer_watershed;
 
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+
+static ssize_t timeout_read(struct file *file, char __user *buf,
+		size_t count, loff_t *offset)
+{
+	return oprofilefs_ulong_to_user(jiffies_to_msecs(timeout_jiffies),
+				buf, count, offset);
+}
+
+
+static ssize_t timeout_write(struct file *file, char const __user *buf,
+		size_t count, loff_t *offset)
+{
+	unsigned long val;
+	int retval;
+
+	if (*offset)
+		return -EINVAL;
+
+	retval = oprofilefs_ulong_from_user(&val, buf, count);
+	if (retval)
+		return retval;
+
+	retval = oprofile_set_timeout(val);
+
+	if (retval)
+		return retval;
+	return count;
+}
+
+
+static const struct file_operations timeout_fops = {
+	.read		= timeout_read,
+	.write		= timeout_write,
+};
+
+#endif
+
+
 static ssize_t depth_read(struct file *file, char __user *buf, size_t count, loff_t *offset)
 {
 	return oprofilefs_ulong_to_user(oprofile_backtrace_depth, buf, count,
@@ -139,6 +179,9 @@ void oprofile_create_files(struct super_block *sb, struct dentry *root)
 	oprofilefs_create_file(sb, root, "cpu_type", &cpu_type_fops);
 	oprofilefs_create_file(sb, root, "backtrace_depth", &depth_fops);
 	oprofilefs_create_file(sb, root, "pointer_size", &pointer_size_fops);
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+	oprofilefs_create_file(sb, root, "time_slice", &timeout_fops);
+#endif
 	oprofile_create_stats_files(sb, root);
 	if (oprofile_ops.create_files)
 		oprofile_ops.create_files(sb, root);
diff --git a/drivers/oprofile/oprofile_stats.c b/drivers/oprofile/oprofile_stats.c
index 3c2270a8300c..77a57a6792f6 100644
--- a/drivers/oprofile/oprofile_stats.c
+++ b/drivers/oprofile/oprofile_stats.c
@@ -16,6 +16,9 @@
 #include "cpu_buffer.h"
 
 struct oprofile_stat_struct oprofile_stats;
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+atomic_t multiplex_counter;
+#endif
 
 void oprofile_reset_stats(void)
 {
@@ -34,6 +37,9 @@ void oprofile_reset_stats(void)
 	atomic_set(&oprofile_stats.sample_lost_no_mapping, 0);
 	atomic_set(&oprofile_stats.event_lost_overflow, 0);
 	atomic_set(&oprofile_stats.bt_lost_no_mapping, 0);
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+	atomic_set(&multiplex_counter, 0);
+#endif
 }
 
 
@@ -76,4 +82,8 @@ void oprofile_create_stats_files(struct super_block *sb, struct dentry *root)
 		&oprofile_stats.event_lost_overflow);
 	oprofilefs_create_ro_atomic(sb, dir, "bt_lost_no_mapping",
 		&oprofile_stats.bt_lost_no_mapping);
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+	oprofilefs_create_ro_atomic(sb, dir, "multiplex_counter",
+		&multiplex_counter);
+#endif
 }
diff --git a/include/linux/oprofile.h b/include/linux/oprofile.h
index d68d2ed94f15..5171639ecf0f 100644
--- a/include/linux/oprofile.h
+++ b/include/linux/oprofile.h
@@ -67,6 +67,9 @@ struct oprofile_operations {
 
 	/* Initiate a stack backtrace. Optional. */
 	void (*backtrace)(struct pt_regs * const regs, unsigned int depth);
+
+	/* Multiplex between different events. Optional. */
+	int (*switch_events)(void);
 	/* CPU identification string. */
 	char * cpu_type;
 };
-- 
cgit v1.2.3


From 5e766e3e433fa2d5d2fdfd8e2432804c91393387 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Wed, 8 Jul 2009 14:54:17 +0200
Subject: x86/oprofile: Fix usage of NUM_CONTROLS/NUM_COUNTERS macros

Use the corresponding macros when iterating over counter and control
registers. Since NUM_CONTROLS and NUM_COUNTERS are equal for AMD cpus
the fix is more a cosmetical change.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_amd.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index fdbed3a0c877..dcfd4505cacc 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -99,7 +99,7 @@ static void op_amd_fill_in_addresses(struct op_msrs * const msrs)
 
 #ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
 	for (i = 0; i < NUM_VIRT_COUNTERS; i++) {
-		int hw_counter = i % NUM_CONTROLS;
+		int hw_counter = i % NUM_COUNTERS;
 		if (reserve_perfctr_nmi(MSR_K7_PERFCTR0 + i))
 			msrs->multiplex[i].addr = MSR_K7_PERFCTR0 + hw_counter;
 		else
@@ -366,7 +366,7 @@ static void op_amd_shutdown(struct op_msrs const * const msrs)
 		if (msrs->counters[i].addr)
 			release_perfctr_nmi(MSR_K7_PERFCTR0 + i);
 	}
-	for (i = 0; i < NUM_COUNTERS; ++i) {
+	for (i = 0; i < NUM_CONTROLS; ++i) {
 		if (msrs->controls[i].addr)
 			release_evntsel_nmi(MSR_K7_EVNTSEL0 + i);
 	}
-- 
cgit v1.2.3


From 82a225283fb0d9438549595d9e6f3ecc42b42ad6 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Thu, 9 Jul 2009 16:29:34 +0200
Subject: x86/oprofile: Use per_cpu() instead of __get_cpu_var()

__get_cpu_var() calls smp_processor_id(). When the cpu id is already
known, instead use per_cpu() to avoid generating the id again.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index e54f6a0b35ac..8cd4658370be 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -294,7 +294,7 @@ static void nmi_cpu_shutdown(void *dummy)
 {
 	unsigned int v;
 	int cpu = smp_processor_id();
-	struct op_msrs *msrs = &__get_cpu_var(cpu_msrs);
+	struct op_msrs *msrs = &per_cpu(cpu_msrs, cpu);
 
 	/* restoring APIC_LVTPC can trigger an apic error because the delivery
 	 * mode and vector nr combination can be illegal. That's by design: on
@@ -307,7 +307,7 @@ static void nmi_cpu_shutdown(void *dummy)
 	apic_write(APIC_LVTERR, v);
 	nmi_cpu_restore_registers(msrs);
 #ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-	__get_cpu_var(switch_index) = 0;
+	per_cpu(switch_index, cpu) = 0;
 #endif
 }
 
-- 
cgit v1.2.3


From 6bfccd099c2841e1c42530f1b6d2553bfa13be3a Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Thu, 9 Jul 2009 19:23:50 +0200
Subject: x86/oprofile: Fix initialization of switch_index

Variable switch_index must be initialized for each cpu. This patch
fixes the initialization by moving it to the per-cpu init function
nmi_cpu_setup().

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c | 16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index 8cd4658370be..b211d335e075 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -160,7 +160,7 @@ static int allocate_msrs(void)
 
 #ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
 
-static void nmi_setup_cpu_mux(struct op_msrs const * const msrs)
+static void nmi_cpu_setup_mux(int cpu, struct op_msrs const * const msrs)
 {
 	int i;
 	struct op_msr *multiplex = msrs->multiplex;
@@ -173,8 +173,15 @@ static void nmi_setup_cpu_mux(struct op_msrs const * const msrs)
 			multiplex[i].saved = 0;
 		}
 	}
+
+	per_cpu(switch_index, cpu) = 0;
 }
 
+#else
+
+static inline void
+nmi_cpu_setup_mux(int cpu, struct op_msrs const * const msrs) { }
+
 #endif
 
 static void nmi_cpu_setup(void *dummy)
@@ -184,9 +191,7 @@ static void nmi_cpu_setup(void *dummy)
 	nmi_cpu_save_registers(msrs);
 	spin_lock(&oprofilefs_lock);
 	model->setup_ctrs(model, msrs);
-#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-	nmi_setup_cpu_mux(msrs);
-#endif
+	nmi_cpu_setup_mux(cpu, msrs);
 	spin_unlock(&oprofilefs_lock);
 	per_cpu(saved_lvtpc, cpu) = apic_read(APIC_LVTPC);
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
@@ -662,9 +667,6 @@ int __init op_nmi_init(struct oprofile_operations *ops)
 	register_cpu_notifier(&oprofile_cpu_nb);
 #endif
 	/* default values, can be overwritten by model */
-#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-	__raw_get_cpu_var(switch_index) = 0;
-#endif
 	ops->create_files	= nmi_create_files;
 	ops->setup		= nmi_setup;
 	ops->shutdown		= nmi_shutdown;
-- 
cgit v1.2.3


From d8471ad3ab613a1ba7abd3aad46659de39a2871c Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Thu, 16 Jul 2009 13:04:43 +0200
Subject: oprofile: Introduce op_x86_phys_to_virt()

This new function translates physical to virtual counter numbers.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c      | 43 +++++++++++----------
 arch/x86/oprofile/op_model_amd.c | 80 ++++++++++++++++------------------------
 arch/x86/oprofile/op_x86_model.h |  1 +
 3 files changed, 55 insertions(+), 69 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index b211d335e075..02b57b8d0e61 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -27,12 +27,6 @@
 #include "op_counter.h"
 #include "op_x86_model.h"
 
-
-#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-DEFINE_PER_CPU(int, switch_index);
-#endif
-
-
 static struct op_x86_model_spec const *model;
 static DEFINE_PER_CPU(struct op_msrs, cpu_msrs);
 static DEFINE_PER_CPU(unsigned long, saved_lvtpc);
@@ -103,6 +97,21 @@ static void nmi_cpu_save_registers(struct op_msrs *msrs)
 	}
 }
 
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+
+static DEFINE_PER_CPU(int, switch_index);
+
+inline int op_x86_phys_to_virt(int phys)
+{
+	return __get_cpu_var(switch_index) + phys;
+}
+
+#else
+
+inline int op_x86_phys_to_virt(int phys) { return phys; }
+
+#endif
+
 static void free_msrs(void)
 {
 	int i;
@@ -248,31 +257,25 @@ static int nmi_setup(void)
 
 static void nmi_cpu_save_mpx_registers(struct op_msrs *msrs)
 {
-	unsigned int si = __get_cpu_var(switch_index);
 	struct op_msr *multiplex = msrs->multiplex;
-	unsigned int i;
+	int i;
 
 	for (i = 0; i < model->num_counters; ++i) {
-		int offset = i + si;
-		if (multiplex[offset].addr) {
-			rdmsrl(multiplex[offset].addr,
-			       multiplex[offset].saved);
-		}
+		int virt = op_x86_phys_to_virt(i);
+		if (multiplex[virt].addr)
+			rdmsrl(multiplex[virt].addr, multiplex[virt].saved);
 	}
 }
 
 static void nmi_cpu_restore_mpx_registers(struct op_msrs *msrs)
 {
-	unsigned int si = __get_cpu_var(switch_index);
 	struct op_msr *multiplex = msrs->multiplex;
-	unsigned int i;
+	int i;
 
 	for (i = 0; i < model->num_counters; ++i) {
-		int offset = i + si;
-		if (multiplex[offset].addr) {
-			wrmsrl(multiplex[offset].addr,
-			       multiplex[offset].saved);
-		}
+		int virt = op_x86_phys_to_virt(i);
+		if (multiplex[virt].addr)
+			wrmsrl(multiplex[virt].addr, multiplex[virt].saved);
 	}
 }
 
diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index dcfd4505cacc..67f830d12e0e 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -42,9 +42,6 @@
 #define MSR_AMD_EVENTSEL_RESERVED	((0xFFFFFCF0ULL<<32)|(1ULL<<21))
 
 static unsigned long reset_value[NUM_VIRT_COUNTERS];
-#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-DECLARE_PER_CPU(int, switch_index);
-#endif
 
 #ifdef CONFIG_OPROFILE_IBS
 
@@ -141,21 +138,20 @@ static void op_amd_setup_ctrs(struct op_x86_model_spec const *model,
 
 	/* enable active counters */
 	for (i = 0; i < NUM_COUNTERS; ++i) {
-#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-		int offset = i + __get_cpu_var(switch_index);
-#else
-		int offset = i;
-#endif
-		if (counter_config[offset].enabled && msrs->counters[i].addr) {
-			/* setup counter registers */
-			wrmsrl(msrs->counters[i].addr, -(u64)reset_value[offset]);
-
-			/* setup control registers */
-			rdmsrl(msrs->controls[i].addr, val);
-			val &= model->reserved;
-			val |= op_x86_get_ctrl(model, &counter_config[offset]);
-			wrmsrl(msrs->controls[i].addr, val);
-		}
+		int virt = op_x86_phys_to_virt(i);
+		if (!counter_config[virt].enabled)
+			continue;
+		if (!msrs->counters[i].addr)
+			continue;
+
+		/* setup counter registers */
+		wrmsrl(msrs->counters[i].addr, -(u64)reset_value[virt]);
+
+		/* setup control registers */
+		rdmsrl(msrs->controls[i].addr, val);
+		val &= model->reserved;
+		val |= op_x86_get_ctrl(model, &counter_config[virt]);
+		wrmsrl(msrs->controls[i].addr, val);
 	}
 }
 
@@ -170,14 +166,13 @@ static void op_amd_switch_ctrl(struct op_x86_model_spec const *model,
 
 	/* enable active counters */
 	for (i = 0; i < NUM_COUNTERS; ++i) {
-		int offset = i + __get_cpu_var(switch_index);
-		if (counter_config[offset].enabled) {
-			/* setup control registers */
-			rdmsrl(msrs->controls[i].addr, val);
-			val &= model->reserved;
-			val |= op_x86_get_ctrl(model, &counter_config[offset]);
-			wrmsrl(msrs->controls[i].addr, val);
-		}
+		int virt = op_x86_phys_to_virt(i);
+		if (!counter_config[virt].enabled)
+			continue;
+		rdmsrl(msrs->controls[i].addr, val);
+		val &= model->reserved;
+		val |= op_x86_get_ctrl(model, &counter_config[virt]);
+		wrmsrl(msrs->controls[i].addr, val);
 	}
 }
 
@@ -292,19 +287,15 @@ static int op_amd_check_ctrs(struct pt_regs * const regs,
 	int i;
 
 	for (i = 0; i < NUM_COUNTERS; ++i) {
-#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-		int offset = i + __get_cpu_var(switch_index);
-#else
-		int offset = i;
-#endif
-		if (!reset_value[offset])
+		int virt = op_x86_phys_to_virt(i);
+		if (!reset_value[virt])
 			continue;
 		rdmsrl(msrs->counters[i].addr, val);
 		/* bit is clear if overflowed: */
 		if (val & OP_CTR_OVERFLOW)
 			continue;
-		oprofile_add_sample(regs, offset);
-		wrmsrl(msrs->counters[i].addr, -(u64)reset_value[offset]);
+		oprofile_add_sample(regs, virt);
+		wrmsrl(msrs->counters[i].addr, -(u64)reset_value[virt]);
 	}
 
 	op_amd_handle_ibs(regs, msrs);
@@ -319,16 +310,11 @@ static void op_amd_start(struct op_msrs const * const msrs)
 	int i;
 
 	for (i = 0; i < NUM_COUNTERS; ++i) {
-#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-		int offset = i + __get_cpu_var(switch_index);
-#else
-		int offset = i;
-#endif
-		if (reset_value[offset]) {
-			rdmsrl(msrs->controls[i].addr, val);
-			val |= ARCH_PERFMON_EVENTSEL0_ENABLE;
-			wrmsrl(msrs->controls[i].addr, val);
-		}
+		if (!reset_value[op_x86_phys_to_virt(i)])
+			continue;
+		rdmsrl(msrs->controls[i].addr, val);
+		val |= ARCH_PERFMON_EVENTSEL0_ENABLE;
+		wrmsrl(msrs->controls[i].addr, val);
 	}
 
 	op_amd_start_ibs();
@@ -344,11 +330,7 @@ static void op_amd_stop(struct op_msrs const * const msrs)
 	 * pm callback
 	 */
 	for (i = 0; i < NUM_COUNTERS; ++i) {
-#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-		if (!reset_value[i + per_cpu(switch_index, smp_processor_id())])
-#else
-		if (!reset_value[i])
-#endif
+		if (!reset_value[op_x86_phys_to_virt(i)])
 			continue;
 		rdmsrl(msrs->controls[i].addr, val);
 		val &= ~ARCH_PERFMON_EVENTSEL0_ENABLE;
diff --git a/arch/x86/oprofile/op_x86_model.h b/arch/x86/oprofile/op_x86_model.h
index 0d07d23cb062..e874dc3565ae 100644
--- a/arch/x86/oprofile/op_x86_model.h
+++ b/arch/x86/oprofile/op_x86_model.h
@@ -60,6 +60,7 @@ struct op_counter_config;
 
 extern u64 op_x86_get_ctrl(struct op_x86_model_spec const *model,
 			   struct op_counter_config *counter_config);
+extern int op_x86_phys_to_virt(int phys);
 
 extern struct op_x86_model_spec const op_ppro_spec;
 extern struct op_x86_model_spec const op_p4_spec;
-- 
cgit v1.2.3


From 7e7478c6bc0e011d2854b21f190cc3a1dba89905 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Thu, 16 Jul 2009 13:09:53 +0200
Subject: oprofile: Grouping multiplexing code in op_model_amd.c

This patch moves some multiplexing code to the new function
op_mux_fill_in_addresses(). Also, the whole multiplexing code is now
at a single location.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_amd.c | 75 ++++++++++++++++++++++------------------
 1 file changed, 41 insertions(+), 34 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index 67f830d12e0e..644980f03924 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -74,6 +74,45 @@ static struct op_ibs_config ibs_config;
 
 #endif
 
+#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
+
+static void op_mux_fill_in_addresses(struct op_msrs * const msrs)
+{
+	int i;
+
+	for (i = 0; i < NUM_VIRT_COUNTERS; i++) {
+		int hw_counter = i % NUM_COUNTERS;
+		if (reserve_perfctr_nmi(MSR_K7_PERFCTR0 + i))
+			msrs->multiplex[i].addr = MSR_K7_PERFCTR0 + hw_counter;
+		else
+			msrs->multiplex[i].addr = 0;
+	}
+}
+
+static void op_mux_switch_ctrl(struct op_x86_model_spec const *model,
+			       struct op_msrs const * const msrs)
+{
+	u64 val;
+	int i;
+
+	/* enable active counters */
+	for (i = 0; i < NUM_COUNTERS; ++i) {
+		int virt = op_x86_phys_to_virt(i);
+		if (!counter_config[virt].enabled)
+			continue;
+		rdmsrl(msrs->controls[i].addr, val);
+		val &= model->reserved;
+		val |= op_x86_get_ctrl(model, &counter_config[virt]);
+		wrmsrl(msrs->controls[i].addr, val);
+	}
+}
+
+#else
+
+static inline void op_mux_fill_in_addresses(struct op_msrs * const msrs) { }
+
+#endif
+
 /* functions for op_amd_spec */
 
 static void op_amd_fill_in_addresses(struct op_msrs * const msrs)
@@ -94,15 +133,7 @@ static void op_amd_fill_in_addresses(struct op_msrs * const msrs)
 			msrs->controls[i].addr = 0;
 	}
 
-#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-	for (i = 0; i < NUM_VIRT_COUNTERS; i++) {
-		int hw_counter = i % NUM_COUNTERS;
-		if (reserve_perfctr_nmi(MSR_K7_PERFCTR0 + i))
-			msrs->multiplex[i].addr = MSR_K7_PERFCTR0 + hw_counter;
-		else
-			msrs->multiplex[i].addr = 0;
-	}
-#endif
+	op_mux_fill_in_addresses(msrs);
 }
 
 static void op_amd_setup_ctrs(struct op_x86_model_spec const *model,
@@ -155,30 +186,6 @@ static void op_amd_setup_ctrs(struct op_x86_model_spec const *model,
 	}
 }
 
-
-#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-
-static void op_amd_switch_ctrl(struct op_x86_model_spec const *model,
-			       struct op_msrs const * const msrs)
-{
-	u64 val;
-	int i;
-
-	/* enable active counters */
-	for (i = 0; i < NUM_COUNTERS; ++i) {
-		int virt = op_x86_phys_to_virt(i);
-		if (!counter_config[virt].enabled)
-			continue;
-		rdmsrl(msrs->controls[i].addr, val);
-		val &= model->reserved;
-		val |= op_x86_get_ctrl(model, &counter_config[virt]);
-		wrmsrl(msrs->controls[i].addr, val);
-	}
-}
-
-#endif
-
-
 #ifdef CONFIG_OPROFILE_IBS
 
 static inline int
@@ -535,6 +542,6 @@ struct op_x86_model_spec const op_amd_spec = {
 	.stop			= &op_amd_stop,
 	.shutdown		= &op_amd_shutdown,
 #ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-	.switch_ctrl		= &op_amd_switch_ctrl,
+	.switch_ctrl		= &op_mux_switch_ctrl,
 #endif
 };
-- 
cgit v1.2.3


From 6ab82f958a5dca591a6ea17a3ca6f2aca06f4f2f Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Thu, 9 Jul 2009 14:40:04 +0200
Subject: x86/oprofile: Implement multiplexing setup/shutdown functions

This patch implements nmi_setup_mux() and nmi_shutdown_mux() functions
to setup/shutdown multiplexing. Multiplexing code in nmi_int.c is now
much more separated.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c | 76 ++++++++++++++++++++++++---------------------
 1 file changed, 40 insertions(+), 36 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index 02b57b8d0e61..674fa37d1502 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -106,9 +106,35 @@ inline int op_x86_phys_to_virt(int phys)
 	return __get_cpu_var(switch_index) + phys;
 }
 
+static void nmi_shutdown_mux(void)
+{
+	int i;
+	for_each_possible_cpu(i) {
+		kfree(per_cpu(cpu_msrs, i).multiplex);
+		per_cpu(cpu_msrs, i).multiplex = NULL;
+		per_cpu(switch_index, i) = 0;
+	}
+}
+
+static int nmi_setup_mux(void)
+{
+	size_t multiplex_size =
+		sizeof(struct op_msr) * model->num_virt_counters;
+	int i;
+	for_each_possible_cpu(i) {
+		per_cpu(cpu_msrs, i).multiplex =
+			kmalloc(multiplex_size, GFP_KERNEL);
+		if (!per_cpu(cpu_msrs, i).multiplex)
+			return 0;
+	}
+	return 1;
+}
+
 #else
 
 inline int op_x86_phys_to_virt(int phys) { return phys; }
+static inline void nmi_shutdown_mux(void) { }
+static inline int nmi_setup_mux(void) { return 1; }
 
 #endif
 
@@ -120,51 +146,27 @@ static void free_msrs(void)
 		per_cpu(cpu_msrs, i).counters = NULL;
 		kfree(per_cpu(cpu_msrs, i).controls);
 		per_cpu(cpu_msrs, i).controls = NULL;
-
-#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-		kfree(per_cpu(cpu_msrs, i).multiplex);
-		per_cpu(cpu_msrs, i).multiplex = NULL;
-#endif
 	}
 }
 
 static int allocate_msrs(void)
 {
-	int success = 1;
 	size_t controls_size = sizeof(struct op_msr) * model->num_controls;
 	size_t counters_size = sizeof(struct op_msr) * model->num_counters;
-#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-	size_t multiplex_size = sizeof(struct op_msr) * model->num_virt_counters;
-#endif
 
 	int i;
 	for_each_possible_cpu(i) {
 		per_cpu(cpu_msrs, i).counters = kmalloc(counters_size,
-								GFP_KERNEL);
-		if (!per_cpu(cpu_msrs, i).counters) {
-			success = 0;
-			break;
-		}
+							GFP_KERNEL);
+		if (!per_cpu(cpu_msrs, i).counters)
+			return 0;
 		per_cpu(cpu_msrs, i).controls = kmalloc(controls_size,
-								GFP_KERNEL);
-		if (!per_cpu(cpu_msrs, i).controls) {
-			success = 0;
-			break;
-		}
-#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-		per_cpu(cpu_msrs, i).multiplex =
-				kmalloc(multiplex_size, GFP_KERNEL);
-		if (!per_cpu(cpu_msrs, i).multiplex) {
-			success = 0;
-			break;
-		}
-#endif
+							GFP_KERNEL);
+		if (!per_cpu(cpu_msrs, i).controls)
+			return 0;
 	}
 
-	if (!success)
-		free_msrs();
-
-	return success;
+	return 1;
 }
 
 #ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
@@ -218,11 +220,15 @@ static int nmi_setup(void)
 	int cpu;
 
 	if (!allocate_msrs())
-		return -ENOMEM;
+		err = -ENOMEM;
+	else if (!nmi_setup_mux())
+		err = -ENOMEM;
+	else
+		err = register_die_notifier(&profile_exceptions_nb);
 
-	err = register_die_notifier(&profile_exceptions_nb);
 	if (err) {
 		free_msrs();
+		nmi_shutdown_mux();
 		return err;
 	}
 
@@ -314,9 +320,6 @@ static void nmi_cpu_shutdown(void *dummy)
 	apic_write(APIC_LVTPC, per_cpu(saved_lvtpc, cpu));
 	apic_write(APIC_LVTERR, v);
 	nmi_cpu_restore_registers(msrs);
-#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-	per_cpu(switch_index, cpu) = 0;
-#endif
 }
 
 static void nmi_shutdown(void)
@@ -326,6 +329,7 @@ static void nmi_shutdown(void)
 	nmi_enabled = 0;
 	on_each_cpu(nmi_cpu_shutdown, NULL, 1);
 	unregister_die_notifier(&profile_exceptions_nb);
+	nmi_shutdown_mux();
 	msrs = &get_cpu_var(cpu_msrs);
 	model->shutdown(msrs);
 	free_msrs();
-- 
cgit v1.2.3


From 48fb4b46712c7d3e8adc79826311abd9ccbf7f1d Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Thu, 9 Jul 2009 14:38:49 +0200
Subject: x86/oprofile: Moving nmi_setup_cpu_mux() in nmi_int.c

This patch moves some code in nmi_int.c to get a single separate
multiplexing code section.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c | 45 +++++++++++++++++++--------------------------
 1 file changed, 19 insertions(+), 26 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index 674fa37d1502..b1edfc922e7f 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -130,11 +130,30 @@ static int nmi_setup_mux(void)
 	return 1;
 }
 
+static void nmi_cpu_setup_mux(int cpu, struct op_msrs const * const msrs)
+{
+	int i;
+	struct op_msr *multiplex = msrs->multiplex;
+
+	for (i = 0; i < model->num_virt_counters; ++i) {
+		if (counter_config[i].enabled) {
+			multiplex[i].saved = -(u64)counter_config[i].count;
+		} else {
+			multiplex[i].addr  = 0;
+			multiplex[i].saved = 0;
+		}
+	}
+
+	per_cpu(switch_index, cpu) = 0;
+}
+
 #else
 
 inline int op_x86_phys_to_virt(int phys) { return phys; }
 static inline void nmi_shutdown_mux(void) { }
 static inline int nmi_setup_mux(void) { return 1; }
+static inline void
+nmi_cpu_setup_mux(int cpu, struct op_msrs const * const msrs) { }
 
 #endif
 
@@ -169,32 +188,6 @@ static int allocate_msrs(void)
 	return 1;
 }
 
-#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-
-static void nmi_cpu_setup_mux(int cpu, struct op_msrs const * const msrs)
-{
-	int i;
-	struct op_msr *multiplex = msrs->multiplex;
-
-	for (i = 0; i < model->num_virt_counters; ++i) {
-		if (counter_config[i].enabled) {
-			multiplex[i].saved = -(u64)counter_config[i].count;
-		} else {
-			multiplex[i].addr  = 0;
-			multiplex[i].saved = 0;
-		}
-	}
-
-	per_cpu(switch_index, cpu) = 0;
-}
-
-#else
-
-static inline void
-nmi_cpu_setup_mux(int cpu, struct op_msrs const * const msrs) { }
-
-#endif
-
 static void nmi_cpu_setup(void *dummy)
 {
 	int cpu = smp_processor_id();
-- 
cgit v1.2.3


From d0f585dd20010f8479e56b5c6f391ef18e26877e Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Thu, 9 Jul 2009 14:38:49 +0200
Subject: x86/oprofile: Moving nmi_cpu_save/restore_mpx_registers() in
 nmi_int.c

This patch moves some code in nmi_int.c to get a single separate
multiplexing code section.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c | 52 +++++++++++++++++++++------------------------
 1 file changed, 24 insertions(+), 28 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index b1edfc922e7f..f38c5cf0fdbb 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -147,6 +147,30 @@ static void nmi_cpu_setup_mux(int cpu, struct op_msrs const * const msrs)
 	per_cpu(switch_index, cpu) = 0;
 }
 
+static void nmi_cpu_save_mpx_registers(struct op_msrs *msrs)
+{
+	struct op_msr *multiplex = msrs->multiplex;
+	int i;
+
+	for (i = 0; i < model->num_counters; ++i) {
+		int virt = op_x86_phys_to_virt(i);
+		if (multiplex[virt].addr)
+			rdmsrl(multiplex[virt].addr, multiplex[virt].saved);
+	}
+}
+
+static void nmi_cpu_restore_mpx_registers(struct op_msrs *msrs)
+{
+	struct op_msr *multiplex = msrs->multiplex;
+	int i;
+
+	for (i = 0; i < model->num_counters; ++i) {
+		int virt = op_x86_phys_to_virt(i);
+		if (multiplex[virt].addr)
+			wrmsrl(multiplex[virt].addr, multiplex[virt].saved);
+	}
+}
+
 #else
 
 inline int op_x86_phys_to_virt(int phys) { return phys; }
@@ -252,34 +276,6 @@ static int nmi_setup(void)
 	return 0;
 }
 
-#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-
-static void nmi_cpu_save_mpx_registers(struct op_msrs *msrs)
-{
-	struct op_msr *multiplex = msrs->multiplex;
-	int i;
-
-	for (i = 0; i < model->num_counters; ++i) {
-		int virt = op_x86_phys_to_virt(i);
-		if (multiplex[virt].addr)
-			rdmsrl(multiplex[virt].addr, multiplex[virt].saved);
-	}
-}
-
-static void nmi_cpu_restore_mpx_registers(struct op_msrs *msrs)
-{
-	struct op_msr *multiplex = msrs->multiplex;
-	int i;
-
-	for (i = 0; i < model->num_counters; ++i) {
-		int virt = op_x86_phys_to_virt(i);
-		if (multiplex[virt].addr)
-			wrmsrl(multiplex[virt].addr, multiplex[virt].saved);
-	}
-}
-
-#endif
-
 static void nmi_cpu_restore_registers(struct op_msrs *msrs)
 {
 	struct op_msr *counters = msrs->counters;
-- 
cgit v1.2.3


From b28d1b923ab52d535c0719155dccf3b3d98bab9f Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Thu, 9 Jul 2009 14:38:49 +0200
Subject: x86/oprofile: Moving nmi_cpu_switch() in nmi_int.c

This patch moves some code in nmi_int.c to get a single separate
multiplexing code section.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c | 144 +++++++++++++++++++++-----------------------
 1 file changed, 70 insertions(+), 74 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index f38c5cf0fdbb..998c7dca31e7 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -97,6 +97,29 @@ static void nmi_cpu_save_registers(struct op_msrs *msrs)
 	}
 }
 
+static void nmi_cpu_start(void *dummy)
+{
+	struct op_msrs const *msrs = &__get_cpu_var(cpu_msrs);
+	model->start(msrs);
+}
+
+static int nmi_start(void)
+{
+	on_each_cpu(nmi_cpu_start, NULL, 1);
+	return 0;
+}
+
+static void nmi_cpu_stop(void *dummy)
+{
+	struct op_msrs const *msrs = &__get_cpu_var(cpu_msrs);
+	model->stop(msrs);
+}
+
+static void nmi_stop(void)
+{
+	on_each_cpu(nmi_cpu_stop, NULL, 1);
+}
+
 #ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
 
 static DEFINE_PER_CPU(int, switch_index);
@@ -171,6 +194,53 @@ static void nmi_cpu_restore_mpx_registers(struct op_msrs *msrs)
 	}
 }
 
+static void nmi_cpu_switch(void *dummy)
+{
+	int cpu = smp_processor_id();
+	int si = per_cpu(switch_index, cpu);
+	struct op_msrs *msrs = &per_cpu(cpu_msrs, cpu);
+
+	nmi_cpu_stop(NULL);
+	nmi_cpu_save_mpx_registers(msrs);
+
+	/* move to next set */
+	si += model->num_counters;
+	if ((si > model->num_virt_counters) || (counter_config[si].count == 0))
+		per_cpu(switch_index, cpu) = 0;
+	else
+		per_cpu(switch_index, cpu) = si;
+
+	model->switch_ctrl(model, msrs);
+	nmi_cpu_restore_mpx_registers(msrs);
+
+	nmi_cpu_start(NULL);
+}
+
+
+/*
+ * Quick check to see if multiplexing is necessary.
+ * The check should be sufficient since counters are used
+ * in ordre.
+ */
+static int nmi_multiplex_on(void)
+{
+	return counter_config[model->num_counters].count ? 0 : -EINVAL;
+}
+
+static int nmi_switch_event(void)
+{
+	if (!model->switch_ctrl)
+		return -ENOSYS;		/* not implemented */
+	if (nmi_multiplex_on() < 0)
+		return -EINVAL;		/* not necessary */
+
+	on_each_cpu(nmi_cpu_switch, NULL, 1);
+
+	atomic_inc(&multiplex_counter);
+
+	return 0;
+}
+
 #else
 
 inline int op_x86_phys_to_virt(int phys) { return phys; }
@@ -325,29 +395,6 @@ static void nmi_shutdown(void)
 	put_cpu_var(cpu_msrs);
 }
 
-static void nmi_cpu_start(void *dummy)
-{
-	struct op_msrs const *msrs = &__get_cpu_var(cpu_msrs);
-	model->start(msrs);
-}
-
-static int nmi_start(void)
-{
-	on_each_cpu(nmi_cpu_start, NULL, 1);
-	return 0;
-}
-
-static void nmi_cpu_stop(void *dummy)
-{
-	struct op_msrs const *msrs = &__get_cpu_var(cpu_msrs);
-	model->stop(msrs);
-}
-
-static void nmi_stop(void)
-{
-	on_each_cpu(nmi_cpu_stop, NULL, 1);
-}
-
 static int nmi_create_files(struct super_block *sb, struct dentry *root)
 {
 	unsigned int i;
@@ -379,57 +426,6 @@ static int nmi_create_files(struct super_block *sb, struct dentry *root)
 	return 0;
 }
 
-#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-
-static void nmi_cpu_switch(void *dummy)
-{
-	int cpu = smp_processor_id();
-	int si = per_cpu(switch_index, cpu);
-	struct op_msrs *msrs = &per_cpu(cpu_msrs, cpu);
-
-	nmi_cpu_stop(NULL);
-	nmi_cpu_save_mpx_registers(msrs);
-
-	/* move to next set */
-	si += model->num_counters;
-	if ((si > model->num_virt_counters) || (counter_config[si].count == 0))
-		per_cpu(switch_index, cpu) = 0;
-	else
-		per_cpu(switch_index, cpu) = si;
-
-	model->switch_ctrl(model, msrs);
-	nmi_cpu_restore_mpx_registers(msrs);
-
-	nmi_cpu_start(NULL);
-}
-
-
-/*
- * Quick check to see if multiplexing is necessary.
- * The check should be sufficient since counters are used
- * in ordre.
- */
-static int nmi_multiplex_on(void)
-{
-	return counter_config[model->num_counters].count ? 0 : -EINVAL;
-}
-
-static int nmi_switch_event(void)
-{
-	if (!model->switch_ctrl)
-		return -ENOSYS;		/* not implemented */
-	if (nmi_multiplex_on() < 0)
-		return -EINVAL;		/* not necessary */
-
-	on_each_cpu(nmi_cpu_switch, NULL, 1);
-
-	atomic_inc(&multiplex_counter);
-
-	return 0;
-}
-
-#endif
-
 #ifdef CONFIG_SMP
 static int oprofile_cpu_notifier(struct notifier_block *b, unsigned long action,
 				 void *data)
-- 
cgit v1.2.3


From 259a83a8abdb9d2664819ec80ad12ebaeb251e32 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Thu, 9 Jul 2009 15:12:35 +0200
Subject: x86/oprofile: Remove const qualifier from struct op_x86_model_spec

This patch removes the const qualifier from struct
op_x86_model_spec to make model parameters changable.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c       | 4 ++--
 arch/x86/oprofile/op_model_amd.c  | 2 +-
 arch/x86/oprofile/op_model_p4.c   | 4 ++--
 arch/x86/oprofile/op_model_ppro.c | 2 +-
 arch/x86/oprofile/op_x86_model.h  | 8 ++++----
 5 files changed, 10 insertions(+), 10 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index 998c7dca31e7..826f391b4229 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -27,7 +27,7 @@
 #include "op_counter.h"
 #include "op_x86_model.h"
 
-static struct op_x86_model_spec const *model;
+static struct op_x86_model_spec *model;
 static DEFINE_PER_CPU(struct op_msrs, cpu_msrs);
 static DEFINE_PER_CPU(unsigned long, saved_lvtpc);
 
@@ -542,7 +542,7 @@ module_param_call(cpu_type, force_cpu_type, NULL, NULL, 0);
 static int __init ppro_init(char **cpu_type)
 {
 	__u8 cpu_model = boot_cpu_data.x86_model;
-	struct op_x86_model_spec const *spec = &op_ppro_spec;	/* default */
+	struct op_x86_model_spec *spec = &op_ppro_spec;	/* default */
 
 	if (force_arch_perfmon && cpu_has_arch_perfmon)
 		return 0;
diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index 644980f03924..39604b429d69 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -526,7 +526,7 @@ static void op_amd_exit(void) {}
 
 #endif /* CONFIG_OPROFILE_IBS */
 
-struct op_x86_model_spec const op_amd_spec = {
+struct op_x86_model_spec op_amd_spec = {
 	.num_counters		= NUM_COUNTERS,
 	.num_controls		= NUM_CONTROLS,
 	.num_virt_counters	= NUM_VIRT_COUNTERS,
diff --git a/arch/x86/oprofile/op_model_p4.c b/arch/x86/oprofile/op_model_p4.c
index 65b9237cde8b..40df028d0d91 100644
--- a/arch/x86/oprofile/op_model_p4.c
+++ b/arch/x86/oprofile/op_model_p4.c
@@ -695,7 +695,7 @@ static void p4_shutdown(struct op_msrs const * const msrs)
 
 
 #ifdef CONFIG_SMP
-struct op_x86_model_spec const op_p4_ht2_spec = {
+struct op_x86_model_spec op_p4_ht2_spec = {
 	.num_counters		= NUM_COUNTERS_HT2,
 	.num_controls		= NUM_CONTROLS_HT2,
 	.num_virt_counters	= NUM_COUNTERS_HT2,
@@ -709,7 +709,7 @@ struct op_x86_model_spec const op_p4_ht2_spec = {
 };
 #endif
 
-struct op_x86_model_spec const op_p4_spec = {
+struct op_x86_model_spec op_p4_spec = {
 	.num_counters		= NUM_COUNTERS_NON_HT,
 	.num_controls		= NUM_CONTROLS_NON_HT,
 	.num_virt_counters	= NUM_COUNTERS_NON_HT,
diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c
index 098cbca5c0b0..659f3b6f86fb 100644
--- a/arch/x86/oprofile/op_model_ppro.c
+++ b/arch/x86/oprofile/op_model_ppro.c
@@ -203,7 +203,7 @@ static void ppro_shutdown(struct op_msrs const * const msrs)
 }
 
 
-struct op_x86_model_spec const op_ppro_spec = {
+struct op_x86_model_spec op_ppro_spec = {
 	.num_counters		= 2,
 	.num_controls		= 2,
 	.num_virt_counters	= 2,
diff --git a/arch/x86/oprofile/op_x86_model.h b/arch/x86/oprofile/op_x86_model.h
index e874dc3565ae..0c886fa0369c 100644
--- a/arch/x86/oprofile/op_x86_model.h
+++ b/arch/x86/oprofile/op_x86_model.h
@@ -62,10 +62,10 @@ extern u64 op_x86_get_ctrl(struct op_x86_model_spec const *model,
 			   struct op_counter_config *counter_config);
 extern int op_x86_phys_to_virt(int phys);
 
-extern struct op_x86_model_spec const op_ppro_spec;
-extern struct op_x86_model_spec const op_p4_spec;
-extern struct op_x86_model_spec const op_p4_ht2_spec;
-extern struct op_x86_model_spec const op_amd_spec;
+extern struct op_x86_model_spec op_ppro_spec;
+extern struct op_x86_model_spec op_p4_spec;
+extern struct op_x86_model_spec op_p4_ht2_spec;
+extern struct op_x86_model_spec op_amd_spec;
 extern struct op_x86_model_spec op_arch_perfmon_spec;
 
 #endif /* OP_X86_MODEL_H */
-- 
cgit v1.2.3


From 2904a527575344a804fdd82b1f8d09a8731d8d49 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Thu, 9 Jul 2009 12:33:41 +0200
Subject: x86/oprofile: Remove unused num_virt_controls from struct
 op_x86_model_spec

The member num_virt_controls of struct op_x86_model_spec is not
used. This patch removes it.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_amd.c  | 1 -
 arch/x86/oprofile/op_model_p4.c   | 2 --
 arch/x86/oprofile/op_model_ppro.c | 1 -
 arch/x86/oprofile/op_x86_model.h  | 1 -
 4 files changed, 5 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index 39604b429d69..dce69b5979e6 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -530,7 +530,6 @@ struct op_x86_model_spec op_amd_spec = {
 	.num_counters		= NUM_COUNTERS,
 	.num_controls		= NUM_CONTROLS,
 	.num_virt_counters	= NUM_VIRT_COUNTERS,
-	.num_virt_controls	= NUM_VIRT_CONTROLS,
 	.reserved		= MSR_AMD_EVENTSEL_RESERVED,
 	.event_mask		= OP_EVENT_MASK,
 	.init			= op_amd_init,
diff --git a/arch/x86/oprofile/op_model_p4.c b/arch/x86/oprofile/op_model_p4.c
index 40df028d0d91..0a4f2deb9e8f 100644
--- a/arch/x86/oprofile/op_model_p4.c
+++ b/arch/x86/oprofile/op_model_p4.c
@@ -699,7 +699,6 @@ struct op_x86_model_spec op_p4_ht2_spec = {
 	.num_counters		= NUM_COUNTERS_HT2,
 	.num_controls		= NUM_CONTROLS_HT2,
 	.num_virt_counters	= NUM_COUNTERS_HT2,
-	.num_virt_controls	= NUM_CONTROLS_HT2,
 	.fill_in_addresses	= &p4_fill_in_addresses,
 	.setup_ctrs		= &p4_setup_ctrs,
 	.check_ctrs		= &p4_check_ctrs,
@@ -713,7 +712,6 @@ struct op_x86_model_spec op_p4_spec = {
 	.num_counters		= NUM_COUNTERS_NON_HT,
 	.num_controls		= NUM_CONTROLS_NON_HT,
 	.num_virt_counters	= NUM_COUNTERS_NON_HT,
-	.num_virt_controls	= NUM_CONTROLS_NON_HT,
 	.fill_in_addresses	= &p4_fill_in_addresses,
 	.setup_ctrs		= &p4_setup_ctrs,
 	.check_ctrs		= &p4_check_ctrs,
diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c
index 659f3b6f86fb..753a02ab215b 100644
--- a/arch/x86/oprofile/op_model_ppro.c
+++ b/arch/x86/oprofile/op_model_ppro.c
@@ -207,7 +207,6 @@ struct op_x86_model_spec op_ppro_spec = {
 	.num_counters		= 2,
 	.num_controls		= 2,
 	.num_virt_counters	= 2,
-	.num_virt_controls	= 2,
 	.reserved		= MSR_PPRO_EVENTSEL_RESERVED,
 	.fill_in_addresses	= &ppro_fill_in_addresses,
 	.setup_ctrs		= &ppro_setup_ctrs,
diff --git a/arch/x86/oprofile/op_x86_model.h b/arch/x86/oprofile/op_x86_model.h
index 0c886fa0369c..4e2e7c2c519f 100644
--- a/arch/x86/oprofile/op_x86_model.h
+++ b/arch/x86/oprofile/op_x86_model.h
@@ -37,7 +37,6 @@ struct op_x86_model_spec {
 	unsigned int	num_counters;
 	unsigned int	num_controls;
 	unsigned int	num_virt_counters;
-	unsigned int	num_virt_controls;
 	u64		reserved;
 	u16		event_mask;
 	int		(*init)(struct oprofile_operations *ops);
-- 
cgit v1.2.3


From 52471c67ee2fa5ed6f700ef57bf27833c63b2192 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Mon, 6 Jul 2009 14:43:55 +0200
Subject: x86/oprofile: Modify initialization of num_virt_counters

Models that do not yet support counter multiplexing have to setup
num_virt_counters. This patch implements the setup from num_counters
if num_virt_counters is not set. Thus, num_virt_counters must be setup
only for multiplexing support.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c       | 3 +++
 arch/x86/oprofile/op_model_p4.c   | 2 --
 arch/x86/oprofile/op_model_ppro.c | 1 -
 3 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index 826f391b4229..82ee29517f16 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -674,6 +674,9 @@ int __init op_nmi_init(struct oprofile_operations *ops)
 	if (ret)
 		return ret;
 
+	if (!model->num_virt_counters)
+		model->num_virt_counters = model->num_counters;
+
 	init_sysfs();
 	using_nmi = 1;
 	printk(KERN_INFO "oprofile: using NMI interrupt.\n");
diff --git a/arch/x86/oprofile/op_model_p4.c b/arch/x86/oprofile/op_model_p4.c
index 0a4f2deb9e8f..ac6b354becdf 100644
--- a/arch/x86/oprofile/op_model_p4.c
+++ b/arch/x86/oprofile/op_model_p4.c
@@ -698,7 +698,6 @@ static void p4_shutdown(struct op_msrs const * const msrs)
 struct op_x86_model_spec op_p4_ht2_spec = {
 	.num_counters		= NUM_COUNTERS_HT2,
 	.num_controls		= NUM_CONTROLS_HT2,
-	.num_virt_counters	= NUM_COUNTERS_HT2,
 	.fill_in_addresses	= &p4_fill_in_addresses,
 	.setup_ctrs		= &p4_setup_ctrs,
 	.check_ctrs		= &p4_check_ctrs,
@@ -711,7 +710,6 @@ struct op_x86_model_spec op_p4_ht2_spec = {
 struct op_x86_model_spec op_p4_spec = {
 	.num_counters		= NUM_COUNTERS_NON_HT,
 	.num_controls		= NUM_CONTROLS_NON_HT,
-	.num_virt_counters	= NUM_COUNTERS_NON_HT,
 	.fill_in_addresses	= &p4_fill_in_addresses,
 	.setup_ctrs		= &p4_setup_ctrs,
 	.check_ctrs		= &p4_check_ctrs,
diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c
index 753a02ab215b..4899215999de 100644
--- a/arch/x86/oprofile/op_model_ppro.c
+++ b/arch/x86/oprofile/op_model_ppro.c
@@ -206,7 +206,6 @@ static void ppro_shutdown(struct op_msrs const * const msrs)
 struct op_x86_model_spec op_ppro_spec = {
 	.num_counters		= 2,
 	.num_controls		= 2,
-	.num_virt_counters	= 2,
 	.reserved		= MSR_PPRO_EVENTSEL_RESERVED,
 	.fill_in_addresses	= &ppro_fill_in_addresses,
 	.setup_ctrs		= &ppro_setup_ctrs,
-- 
cgit v1.2.3


From 39e97f40c3a5e71de0532368deaa683e09b74ba2 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Thu, 9 Jul 2009 15:11:45 +0200
Subject: x86/oprofile: Add function has_mux() to check multiplexing support

The check is used to prevent running multiplexing code for models not
supporting multiplexing. Before, the code was running but without
effect.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c | 19 ++++++++++++++++++-
 1 file changed, 18 insertions(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index 82ee29517f16..dca7240aeb26 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -124,6 +124,11 @@ static void nmi_stop(void)
 
 static DEFINE_PER_CPU(int, switch_index);
 
+static inline int has_mux(void)
+{
+	return !!model->switch_ctrl;
+}
+
 inline int op_x86_phys_to_virt(int phys)
 {
 	return __get_cpu_var(switch_index) + phys;
@@ -132,6 +137,10 @@ inline int op_x86_phys_to_virt(int phys)
 static void nmi_shutdown_mux(void)
 {
 	int i;
+
+	if (!has_mux())
+		return;
+
 	for_each_possible_cpu(i) {
 		kfree(per_cpu(cpu_msrs, i).multiplex);
 		per_cpu(cpu_msrs, i).multiplex = NULL;
@@ -144,12 +153,17 @@ static int nmi_setup_mux(void)
 	size_t multiplex_size =
 		sizeof(struct op_msr) * model->num_virt_counters;
 	int i;
+
+	if (!has_mux())
+		return 1;
+
 	for_each_possible_cpu(i) {
 		per_cpu(cpu_msrs, i).multiplex =
 			kmalloc(multiplex_size, GFP_KERNEL);
 		if (!per_cpu(cpu_msrs, i).multiplex)
 			return 0;
 	}
+
 	return 1;
 }
 
@@ -158,6 +172,9 @@ static void nmi_cpu_setup_mux(int cpu, struct op_msrs const * const msrs)
 	int i;
 	struct op_msr *multiplex = msrs->multiplex;
 
+	if (!has_mux())
+		return;
+
 	for (i = 0; i < model->num_virt_counters; ++i) {
 		if (counter_config[i].enabled) {
 			multiplex[i].saved = -(u64)counter_config[i].count;
@@ -229,7 +246,7 @@ static int nmi_multiplex_on(void)
 
 static int nmi_switch_event(void)
 {
-	if (!model->switch_ctrl)
+	if (!has_mux())
 		return -ENOSYS;		/* not implemented */
 	if (nmi_multiplex_on() < 0)
 		return -EINVAL;		/* not necessary */
-- 
cgit v1.2.3


From 5280514471c2803776701c43c027038decac1103 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Thu, 9 Jul 2009 16:02:44 +0200
Subject: x86/oprofile: Enable multiplexing only if the model supports it

This patch checks if the model supports multiplexing. Only then
multiplexing will be enabled. The code is added to the common x86
initialization.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index dca7240aeb26..f0fb44725d80 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -258,6 +258,12 @@ static int nmi_switch_event(void)
 	return 0;
 }
 
+static inline void mux_init(struct oprofile_operations *ops)
+{
+	if (has_mux())
+		ops->switch_events = nmi_switch_event;
+}
+
 #else
 
 inline int op_x86_phys_to_virt(int phys) { return phys; }
@@ -265,6 +271,7 @@ static inline void nmi_shutdown_mux(void) { }
 static inline int nmi_setup_mux(void) { return 1; }
 static inline void
 nmi_cpu_setup_mux(int cpu, struct op_msrs const * const msrs) { }
+static inline void mux_init(struct oprofile_operations *ops) { }
 
 #endif
 
@@ -682,9 +689,6 @@ int __init op_nmi_init(struct oprofile_operations *ops)
 	ops->start		= nmi_start;
 	ops->stop		= nmi_stop;
 	ops->cpu_type		= cpu_type;
-#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-	ops->switch_events	= nmi_switch_event;
-#endif
 
 	if (model->init)
 		ret = model->init(ops);
@@ -694,6 +698,8 @@ int __init op_nmi_init(struct oprofile_operations *ops)
 	if (!model->num_virt_counters)
 		model->num_virt_counters = model->num_counters;
 
+	mux_init(ops);
+
 	init_sysfs();
 	using_nmi = 1;
 	printk(KERN_INFO "oprofile: using NMI interrupt.\n");
-- 
cgit v1.2.3


From 4d015f79e972cea1761cfee8872b1c0992ccd8b2 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Thu, 9 Jul 2009 21:42:51 +0200
Subject: x86/oprofile: Implement mux_clone()

To setup a counter for all cpus, its structure is cloned from cpu
0. This patch implements mux_clone() to do this part for multiplexing
data.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c | 37 +++++++++++++++++++++++--------------
 1 file changed, 23 insertions(+), 14 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index f0fb44725d80..da6d2ab31c6c 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -264,6 +264,16 @@ static inline void mux_init(struct oprofile_operations *ops)
 		ops->switch_events = nmi_switch_event;
 }
 
+static void mux_clone(int cpu)
+{
+	if (!has_mux())
+		return;
+
+	memcpy(per_cpu(cpu_msrs, cpu).multiplex,
+	       per_cpu(cpu_msrs, 0).multiplex,
+	       sizeof(struct op_msr) * model->num_virt_counters);
+}
+
 #else
 
 inline int op_x86_phys_to_virt(int phys) { return phys; }
@@ -272,6 +282,7 @@ static inline int nmi_setup_mux(void) { return 1; }
 static inline void
 nmi_cpu_setup_mux(int cpu, struct op_msrs const * const msrs) { }
 static inline void mux_init(struct oprofile_operations *ops) { }
+static void mux_clone(int cpu) { }
 
 #endif
 
@@ -350,20 +361,18 @@ static int nmi_setup(void)
 	/* Assume saved/restored counters are the same on all CPUs */
 	model->fill_in_addresses(&per_cpu(cpu_msrs, 0));
 	for_each_possible_cpu(cpu) {
-		if (cpu != 0) {
-			memcpy(per_cpu(cpu_msrs, cpu).counters,
-				per_cpu(cpu_msrs, 0).counters,
-				sizeof(struct op_msr) * model->num_counters);
-
-			memcpy(per_cpu(cpu_msrs, cpu).controls,
-				per_cpu(cpu_msrs, 0).controls,
-				sizeof(struct op_msr) * model->num_controls);
-#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-			memcpy(per_cpu(cpu_msrs, cpu).multiplex,
-				per_cpu(cpu_msrs, 0).multiplex,
-				sizeof(struct op_msr) * model->num_virt_counters);
-#endif
-		}
+		if (!cpu)
+			continue;
+
+		memcpy(per_cpu(cpu_msrs, cpu).counters,
+		       per_cpu(cpu_msrs, 0).counters,
+		       sizeof(struct op_msr) * model->num_counters);
+
+		memcpy(per_cpu(cpu_msrs, cpu).controls,
+		       per_cpu(cpu_msrs, 0).controls,
+		       sizeof(struct op_msr) * model->num_controls);
+
+		mux_clone(cpu);
 	}
 	on_each_cpu(nmi_cpu_setup, NULL, 1);
 	nmi_enabled = 1;
-- 
cgit v1.2.3


From 1b294f5960cd89e49eeb3e797860c552b03f2272 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Thu, 9 Jul 2009 14:56:25 +0200
Subject: oprofile: Adding switch counter to oprofile statistic variables

This patch moves the multiplexing switch counter from x86 code to
common oprofile statistic variables. Now the value will be available
and usable for all architectures. The initialization and
incrementation also moved to common code.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c       | 7 -------
 drivers/oprofile/oprof.c          | 7 +++++--
 drivers/oprofile/oprofile_stats.c | 9 ++-------
 drivers/oprofile/oprofile_stats.h | 1 +
 4 files changed, 8 insertions(+), 16 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index da6d2ab31c6c..7b3362f9abdb 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -34,11 +34,6 @@ static DEFINE_PER_CPU(unsigned long, saved_lvtpc);
 /* 0 == registered but off, 1 == registered and on */
 static int nmi_enabled = 0;
 
-
-#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-extern atomic_t multiplex_counter;
-#endif
-
 struct op_counter_config counter_config[OP_MAX_COUNTER];
 
 /* common functions */
@@ -253,8 +248,6 @@ static int nmi_switch_event(void)
 
 	on_each_cpu(nmi_cpu_switch, NULL, 1);
 
-	atomic_inc(&multiplex_counter);
-
 	return 0;
 }
 
diff --git a/drivers/oprofile/oprof.c b/drivers/oprofile/oprof.c
index a48294a8ebe8..dc8a0428260d 100644
--- a/drivers/oprofile/oprof.c
+++ b/drivers/oprofile/oprof.c
@@ -107,8 +107,11 @@ static void stop_switch_worker(void)
 
 static void switch_worker(struct work_struct *work)
 {
-	if (!oprofile_ops.switch_events())
-		start_switch_worker();
+	if (oprofile_ops.switch_events())
+		return;
+
+	atomic_inc(&oprofile_stats.multiplex_counter);
+	start_switch_worker();
 }
 
 /* User inputs in ms, converts to jiffies */
diff --git a/drivers/oprofile/oprofile_stats.c b/drivers/oprofile/oprofile_stats.c
index 77a57a6792f6..61689e814d46 100644
--- a/drivers/oprofile/oprofile_stats.c
+++ b/drivers/oprofile/oprofile_stats.c
@@ -16,9 +16,6 @@
 #include "cpu_buffer.h"
 
 struct oprofile_stat_struct oprofile_stats;
-#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-atomic_t multiplex_counter;
-#endif
 
 void oprofile_reset_stats(void)
 {
@@ -37,9 +34,7 @@ void oprofile_reset_stats(void)
 	atomic_set(&oprofile_stats.sample_lost_no_mapping, 0);
 	atomic_set(&oprofile_stats.event_lost_overflow, 0);
 	atomic_set(&oprofile_stats.bt_lost_no_mapping, 0);
-#ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
-	atomic_set(&multiplex_counter, 0);
-#endif
+	atomic_set(&oprofile_stats.multiplex_counter, 0);
 }
 
 
@@ -84,6 +79,6 @@ void oprofile_create_stats_files(struct super_block *sb, struct dentry *root)
 		&oprofile_stats.bt_lost_no_mapping);
 #ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX
 	oprofilefs_create_ro_atomic(sb, dir, "multiplex_counter",
-		&multiplex_counter);
+		&oprofile_stats.multiplex_counter);
 #endif
 }
diff --git a/drivers/oprofile/oprofile_stats.h b/drivers/oprofile/oprofile_stats.h
index 3da0d08dc1f9..0b54e46c3c14 100644
--- a/drivers/oprofile/oprofile_stats.h
+++ b/drivers/oprofile/oprofile_stats.h
@@ -17,6 +17,7 @@ struct oprofile_stat_struct {
 	atomic_t sample_lost_no_mapping;
 	atomic_t bt_lost_no_mapping;
 	atomic_t event_lost_overflow;
+	atomic_t multiplex_counter;
 };
 
 extern struct oprofile_stat_struct oprofile_stats;
-- 
cgit v1.2.3


From 61d149d5248ad7428801cdede0f5fcc2b90cd61c Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Fri, 10 Jul 2009 15:47:17 +0200
Subject: x86/oprofile: Implement op_x86_virt_to_phys()

This patch implements a common x86 function to convert virtual counter
numbers to physical.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c      | 6 ++++++
 arch/x86/oprofile/op_model_amd.c | 2 +-
 arch/x86/oprofile/op_x86_model.h | 1 +
 3 files changed, 8 insertions(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index 7b3362f9abdb..5856e61cb098 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -129,6 +129,11 @@ inline int op_x86_phys_to_virt(int phys)
 	return __get_cpu_var(switch_index) + phys;
 }
 
+inline int op_x86_virt_to_phys(int virt)
+{
+	return virt % model->num_counters;
+}
+
 static void nmi_shutdown_mux(void)
 {
 	int i;
@@ -270,6 +275,7 @@ static void mux_clone(int cpu)
 #else
 
 inline int op_x86_phys_to_virt(int phys) { return phys; }
+inline int op_x86_virt_to_phys(int virt) { return virt; }
 static inline void nmi_shutdown_mux(void) { }
 static inline int nmi_setup_mux(void) { return 1; }
 static inline void
diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index dce69b5979e6..1ea19829d985 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -81,7 +81,7 @@ static void op_mux_fill_in_addresses(struct op_msrs * const msrs)
 	int i;
 
 	for (i = 0; i < NUM_VIRT_COUNTERS; i++) {
-		int hw_counter = i % NUM_COUNTERS;
+		int hw_counter = op_x86_virt_to_phys(i);
 		if (reserve_perfctr_nmi(MSR_K7_PERFCTR0 + i))
 			msrs->multiplex[i].addr = MSR_K7_PERFCTR0 + hw_counter;
 		else
diff --git a/arch/x86/oprofile/op_x86_model.h b/arch/x86/oprofile/op_x86_model.h
index 4e2e7c2c519f..b83776180c7f 100644
--- a/arch/x86/oprofile/op_x86_model.h
+++ b/arch/x86/oprofile/op_x86_model.h
@@ -60,6 +60,7 @@ struct op_counter_config;
 extern u64 op_x86_get_ctrl(struct op_x86_model_spec const *model,
 			   struct op_counter_config *counter_config);
 extern int op_x86_phys_to_virt(int phys);
+extern int op_x86_virt_to_phys(int virt);
 
 extern struct op_x86_model_spec op_ppro_spec;
 extern struct op_x86_model_spec op_p4_spec;
-- 
cgit v1.2.3


From 11be1a7b54283021777f409aa983ce125945e67c Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Fri, 10 Jul 2009 18:15:21 +0200
Subject: x86/oprofile: Add counter reservation check for virtual counters

This patch adds a check for the availability of a counter. A virtual
counter is used only if its physical counter is not reserved.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/nmi_int.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index 5856e61cb098..cb88b1a0bd5f 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -435,15 +435,13 @@ static int nmi_create_files(struct super_block *sb, struct dentry *root)
 		struct dentry *dir;
 		char buf[4];
 
-#ifndef CONFIG_OPROFILE_EVENT_MULTIPLEX
 		/* quick little hack to _not_ expose a counter if it is not
 		 * available for use.  This should protect userspace app.
 		 * NOTE:  assumes 1:1 mapping here (that counters are organized
 		 *        sequentially in their struct assignment).
 		 */
-		if (unlikely(!avail_to_resrv_perfctr_nmi_bit(i)))
+		if (!avail_to_resrv_perfctr_nmi_bit(op_x86_virt_to_phys(i)))
 			continue;
-#endif /* CONFIG_OPROFILE_EVENT_MULTIPLEX */
 
 		snprintf(buf,  sizeof(buf), "%d", i);
 		dir = oprofilefs_mkdir(sb, root, buf);
-- 
cgit v1.2.3


From c550091edd6fac2ed9dac1b30d986b6c58b216fa Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Thu, 16 Jul 2009 13:11:16 +0200
Subject: x86/oprofile: Small coding style fixes

Some small coding style fixes.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_amd.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index 1ea19829d985..827beecb67a4 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -144,11 +144,10 @@ static void op_amd_setup_ctrs(struct op_x86_model_spec const *model,
 
 	/* setup reset_value */
 	for (i = 0; i < NUM_VIRT_COUNTERS; ++i) {
-		if (counter_config[i].enabled) {
+		if (counter_config[i].enabled)
 			reset_value[i] = counter_config[i].count;
-		} else {
+		else
 			reset_value[i] = 0;
-		}
 	}
 
 	/* clear all counters */
-- 
cgit v1.2.3


From 3885123da8335dc6b67387e5e626acbffc56f664 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Fri, 10 Jul 2009 10:04:50 +0900
Subject: swiotlb: remove unused swiotlb_alloc_boot()

Nobody uses swiotlb_alloc_boot().

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Acked-by: Becky Bruce <beckyb@kernel.crashing.org>
---
 arch/x86/kernel/pci-swiotlb.c | 5 -----
 include/linux/swiotlb.h       | 2 --
 lib/swiotlb.c                 | 7 +------
 3 files changed, 1 insertion(+), 13 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/pci-swiotlb.c b/arch/x86/kernel/pci-swiotlb.c
index 6af96ee44200..0ac7cd524788 100644
--- a/arch/x86/kernel/pci-swiotlb.c
+++ b/arch/x86/kernel/pci-swiotlb.c
@@ -13,11 +13,6 @@
 
 int swiotlb __read_mostly;
 
-void * __init swiotlb_alloc_boot(size_t size, unsigned long nslabs)
-{
-	return alloc_bootmem_low_pages(size);
-}
-
 void *swiotlb_alloc(unsigned order, unsigned long nslabs)
 {
 	return (void *)__get_free_pages(GFP_DMA | __GFP_NOWARN, order);
diff --git a/include/linux/swiotlb.h b/include/linux/swiotlb.h
index cb1a6631b8f4..94db70444c17 100644
--- a/include/linux/swiotlb.h
+++ b/include/linux/swiotlb.h
@@ -14,7 +14,6 @@ struct scatterlist;
  */
 #define IO_TLB_SEGSIZE	128
 
-
 /*
  * log of the size of each IO TLB slab.  The number of slabs is command line
  * controllable.
@@ -24,7 +23,6 @@ struct scatterlist;
 extern void
 swiotlb_init(void);
 
-extern void *swiotlb_alloc_boot(size_t bytes, unsigned long nslabs);
 extern void *swiotlb_alloc(unsigned order, unsigned long nslabs);
 
 extern dma_addr_t swiotlb_phys_to_bus(struct device *hwdev,
diff --git a/lib/swiotlb.c b/lib/swiotlb.c
index bffe6d7ef9d9..9edfdd442edc 100644
--- a/lib/swiotlb.c
+++ b/lib/swiotlb.c
@@ -114,11 +114,6 @@ setup_io_tlb_npages(char *str)
 __setup("swiotlb=", setup_io_tlb_npages);
 /* make io_tlb_overflow tunable too? */
 
-void * __weak __init swiotlb_alloc_boot(size_t size, unsigned long nslabs)
-{
-	return alloc_bootmem_low_pages(size);
-}
-
 void * __weak swiotlb_alloc(unsigned order, unsigned long nslabs)
 {
 	return (void *)__get_free_pages(GFP_DMA | __GFP_NOWARN, order);
@@ -189,7 +184,7 @@ swiotlb_init_with_default_size(size_t default_size)
 	/*
 	 * Get IO TLB memory from the low pages
 	 */
-	io_tlb_start = swiotlb_alloc_boot(bytes, io_tlb_nslabs);
+	io_tlb_start = alloc_bootmem_low_pages(bytes);
 	if (!io_tlb_start)
 		panic("Cannot allocate SWIOTLB buffer");
 	io_tlb_end = io_tlb_start + bytes;
-- 
cgit v1.2.3


From bb52196be37ce154ddc50b1f39496146d181cbe7 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Fri, 10 Jul 2009 10:04:51 +0900
Subject: swiotlb: remove unused swiotlb_alloc()

Nobody uses swiotlb_alloc().

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Acked-by: Becky Bruce <beckyb@kernel.crashing.org>
---
 arch/x86/kernel/pci-swiotlb.c | 5 -----
 include/linux/swiotlb.h       | 2 --
 lib/swiotlb.c                 | 8 ++------
 3 files changed, 2 insertions(+), 13 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/pci-swiotlb.c b/arch/x86/kernel/pci-swiotlb.c
index 0ac7cd524788..ea675cfe76fe 100644
--- a/arch/x86/kernel/pci-swiotlb.c
+++ b/arch/x86/kernel/pci-swiotlb.c
@@ -13,11 +13,6 @@
 
 int swiotlb __read_mostly;
 
-void *swiotlb_alloc(unsigned order, unsigned long nslabs)
-{
-	return (void *)__get_free_pages(GFP_DMA | __GFP_NOWARN, order);
-}
-
 dma_addr_t swiotlb_phys_to_bus(struct device *hwdev, phys_addr_t paddr)
 {
 	return paddr;
diff --git a/include/linux/swiotlb.h b/include/linux/swiotlb.h
index 94db70444c17..6bc50944040f 100644
--- a/include/linux/swiotlb.h
+++ b/include/linux/swiotlb.h
@@ -23,8 +23,6 @@ struct scatterlist;
 extern void
 swiotlb_init(void);
 
-extern void *swiotlb_alloc(unsigned order, unsigned long nslabs);
-
 extern dma_addr_t swiotlb_phys_to_bus(struct device *hwdev,
 				      phys_addr_t address);
 extern phys_addr_t swiotlb_bus_to_phys(struct device *hwdev,
diff --git a/lib/swiotlb.c b/lib/swiotlb.c
index 9edfdd442edc..3c4c21cdf43d 100644
--- a/lib/swiotlb.c
+++ b/lib/swiotlb.c
@@ -114,11 +114,6 @@ setup_io_tlb_npages(char *str)
 __setup("swiotlb=", setup_io_tlb_npages);
 /* make io_tlb_overflow tunable too? */
 
-void * __weak swiotlb_alloc(unsigned order, unsigned long nslabs)
-{
-	return (void *)__get_free_pages(GFP_DMA | __GFP_NOWARN, order);
-}
-
 dma_addr_t __weak swiotlb_phys_to_bus(struct device *hwdev, phys_addr_t paddr)
 {
 	return paddr;
@@ -240,7 +235,8 @@ swiotlb_late_init_with_default_size(size_t default_size)
 	bytes = io_tlb_nslabs << IO_TLB_SHIFT;
 
 	while ((SLABS_PER_PAGE << order) > IO_TLB_MIN_SLABS) {
-		io_tlb_start = swiotlb_alloc(order, io_tlb_nslabs);
+		io_tlb_start = (void *)__get_free_pages(GFP_DMA | __GFP_NOWARN,
+							order);
 		if (io_tlb_start)
 			break;
 		order--;
-- 
cgit v1.2.3


From cf56e3f2e8a8d5b7bc719980869b0e7985c256f3 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Fri, 10 Jul 2009 10:04:52 +0900
Subject: swiotlb: remove swiotlb_arch_range_needs_mapping

Nobody uses swiotlb_arch_range_needs_mapping().

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Acked-by: Becky Bruce <beckyb@kernel.crashing.org>
---
 arch/x86/kernel/pci-swiotlb.c |  5 -----
 include/linux/swiotlb.h       |  2 --
 lib/swiotlb.c                 | 15 ++-------------
 3 files changed, 2 insertions(+), 20 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/pci-swiotlb.c b/arch/x86/kernel/pci-swiotlb.c
index ea675cfe76fe..165bd7f93bb1 100644
--- a/arch/x86/kernel/pci-swiotlb.c
+++ b/arch/x86/kernel/pci-swiotlb.c
@@ -23,11 +23,6 @@ phys_addr_t swiotlb_bus_to_phys(struct device *hwdev, dma_addr_t baddr)
 	return baddr;
 }
 
-int __weak swiotlb_arch_range_needs_mapping(phys_addr_t paddr, size_t size)
-{
-	return 0;
-}
-
 static void *x86_swiotlb_alloc_coherent(struct device *hwdev, size_t size,
 					dma_addr_t *dma_handle, gfp_t flags)
 {
diff --git a/include/linux/swiotlb.h b/include/linux/swiotlb.h
index 6bc50944040f..a977da24f17c 100644
--- a/include/linux/swiotlb.h
+++ b/include/linux/swiotlb.h
@@ -28,8 +28,6 @@ extern dma_addr_t swiotlb_phys_to_bus(struct device *hwdev,
 extern phys_addr_t swiotlb_bus_to_phys(struct device *hwdev,
 				       dma_addr_t address);
 
-extern int swiotlb_arch_range_needs_mapping(phys_addr_t paddr, size_t size);
-
 extern void
 *swiotlb_alloc_coherent(struct device *hwdev, size_t size,
 			dma_addr_t *dma_handle, gfp_t flags);
diff --git a/lib/swiotlb.c b/lib/swiotlb.c
index 3c4c21cdf43d..dc1cd1f5369e 100644
--- a/lib/swiotlb.c
+++ b/lib/swiotlb.c
@@ -141,11 +141,6 @@ int __weak swiotlb_arch_address_needs_mapping(struct device *hwdev,
 	return !is_buffer_dma_capable(dma_get_mask(hwdev), addr, size);
 }
 
-int __weak swiotlb_arch_range_needs_mapping(phys_addr_t paddr, size_t size)
-{
-	return 0;
-}
-
 static void swiotlb_print_info(unsigned long bytes)
 {
 	phys_addr_t pstart, pend;
@@ -312,11 +307,6 @@ address_needs_mapping(struct device *hwdev, dma_addr_t addr, size_t size)
 	return swiotlb_arch_address_needs_mapping(hwdev, addr, size);
 }
 
-static inline int range_needs_mapping(phys_addr_t paddr, size_t size)
-{
-	return swiotlb_force || swiotlb_arch_range_needs_mapping(paddr, size);
-}
-
 static int is_swiotlb_buffer(char *addr)
 {
 	return addr >= io_tlb_start && addr < io_tlb_end;
@@ -646,8 +636,7 @@ dma_addr_t swiotlb_map_page(struct device *dev, struct page *page,
 	 * we can safely return the device addr and not worry about bounce
 	 * buffering it.
 	 */
-	if (!address_needs_mapping(dev, dev_addr, size) &&
-	    !range_needs_mapping(phys, size))
+	if (!address_needs_mapping(dev, dev_addr, size) && !swiotlb_force)
 		return dev_addr;
 
 	/*
@@ -810,7 +799,7 @@ swiotlb_map_sg_attrs(struct device *hwdev, struct scatterlist *sgl, int nelems,
 		phys_addr_t paddr = sg_phys(sg);
 		dma_addr_t dev_addr = swiotlb_phys_to_bus(hwdev, paddr);
 
-		if (range_needs_mapping(paddr, sg->length) ||
+		if (swiotlb_force ||
 		    address_needs_mapping(hwdev, dev_addr, sg->length)) {
 			void *map = map_single(hwdev, sg_phys(sg),
 					       sg->length, dir);
-- 
cgit v1.2.3


From 99becaca86d184a4433e9fde879ff97303d7669f Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Fri, 10 Jul 2009 10:04:54 +0900
Subject: x86: add dma_capable() to replace is_buffer_dma_capable()

dma_capable() eventually replaces is_buffer_dma_capable(), which tells
if a memory area is dma-capable or not. The problem of
is_buffer_dma_capable() is that it doesn't take a pointer to struct
device so it doesn't work for POWERPC.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
---
 arch/x86/include/asm/dma-mapping.h | 8 ++++++++
 1 file changed, 8 insertions(+)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/dma-mapping.h b/arch/x86/include/asm/dma-mapping.h
index 1c3f9435f1c9..adac59c8f69b 100644
--- a/arch/x86/include/asm/dma-mapping.h
+++ b/arch/x86/include/asm/dma-mapping.h
@@ -55,6 +55,14 @@ extern int dma_set_mask(struct device *dev, u64 mask);
 extern void *dma_generic_alloc_coherent(struct device *dev, size_t size,
 					dma_addr_t *dma_addr, gfp_t flag);
 
+static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
+{
+	if (!dev->dma_mask)
+		return 0;
+
+	return addr + size <= *dev->dma_mask;
+}
+
 static inline void
 dma_cache_sync(struct device *dev, void *vaddr, size_t size,
 	enum dma_data_direction dir)
-- 
cgit v1.2.3


From a4c2baa6e148adfb27beaf16b6fb6d465b5b3acb Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Fri, 10 Jul 2009 10:04:55 +0900
Subject: x86: replace is_buffer_dma_capable() with dma_capable

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
---
 arch/x86/kernel/pci-dma.c     | 2 +-
 arch/x86/kernel/pci-gart_64.c | 5 ++---
 arch/x86/kernel/pci-nommu.c   | 2 +-
 3 files changed, 4 insertions(+), 5 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c
index 1a041bcf506b..3c945c0b3501 100644
--- a/arch/x86/kernel/pci-dma.c
+++ b/arch/x86/kernel/pci-dma.c
@@ -147,7 +147,7 @@ again:
 		return NULL;
 
 	addr = page_to_phys(page);
-	if (!is_buffer_dma_capable(dma_mask, addr, size)) {
+	if (addr + size > dma_mask) {
 		__free_pages(page, get_order(size));
 
 		if (dma_mask < DMA_BIT_MASK(32) && !(flag & GFP_DMA)) {
diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c
index d2e56b8f48e7..98a827ee9ed7 100644
--- a/arch/x86/kernel/pci-gart_64.c
+++ b/arch/x86/kernel/pci-gart_64.c
@@ -190,14 +190,13 @@ static void iommu_full(struct device *dev, size_t size, int dir)
 static inline int
 need_iommu(struct device *dev, unsigned long addr, size_t size)
 {
-	return force_iommu ||
-		!is_buffer_dma_capable(*dev->dma_mask, addr, size);
+	return force_iommu || !dma_capable(dev, addr, size);
 }
 
 static inline int
 nonforced_iommu(struct device *dev, unsigned long addr, size_t size)
 {
-	return !is_buffer_dma_capable(*dev->dma_mask, addr, size);
+	return !dma_capable(dev, addr, size);
 }
 
 /* Map a single continuous physical area into the IOMMU.
diff --git a/arch/x86/kernel/pci-nommu.c b/arch/x86/kernel/pci-nommu.c
index 71d412a09f30..c0a4222bf62b 100644
--- a/arch/x86/kernel/pci-nommu.c
+++ b/arch/x86/kernel/pci-nommu.c
@@ -14,7 +14,7 @@
 static int
 check_addr(char *name, struct device *hwdev, dma_addr_t bus, size_t size)
 {
-	if (hwdev && !is_buffer_dma_capable(*hwdev->dma_mask, bus, size)) {
+	if (hwdev && !dma_capable(hwdev, bus, size)) {
 		if (*hwdev->dma_mask >= DMA_BIT_MASK(32))
 			printk(KERN_ERR
 			    "nommu_%s: overflow %Lx+%zu of device mask %Lx\n",
-- 
cgit v1.2.3


From 8d4f5339d1ee4027c07e6b2a1cfa9dc41b0d383b Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Fri, 10 Jul 2009 10:05:01 +0900
Subject: x86, IA64, powerpc: add phys_to_dma() and dma_to_phys()

This adds two functions, phys_to_dma() and dma_to_phys() to x86, IA64
and powerpc. swiotlb uses them. phys_to_dma() converts a physical
address to a dma address. dma_to_phys() does the opposite.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Acked-by: Becky Bruce <beckyb@kernel.crashing.org>
---
 arch/ia64/include/asm/dma-mapping.h    | 10 ++++++++++
 arch/powerpc/include/asm/dma-mapping.h | 10 ++++++++++
 arch/x86/include/asm/dma-mapping.h     | 10 ++++++++++
 3 files changed, 30 insertions(+)

(limited to 'arch/x86')

diff --git a/arch/ia64/include/asm/dma-mapping.h b/arch/ia64/include/asm/dma-mapping.h
index 88d0f860394d..f91829de329f 100644
--- a/arch/ia64/include/asm/dma-mapping.h
+++ b/arch/ia64/include/asm/dma-mapping.h
@@ -77,6 +77,16 @@ static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
 	return addr + size <= *dev->dma_mask;
 }
 
+static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr)
+{
+	return paddr;
+}
+
+static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr)
+{
+	return daddr;
+}
+
 extern int dma_get_cache_alignment(void);
 
 static inline void
diff --git a/arch/powerpc/include/asm/dma-mapping.h b/arch/powerpc/include/asm/dma-mapping.h
index 6ff1f8581d79..0c34371ec49c 100644
--- a/arch/powerpc/include/asm/dma-mapping.h
+++ b/arch/powerpc/include/asm/dma-mapping.h
@@ -437,6 +437,16 @@ static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
 	return addr + size <= *dev->dma_mask;
 }
 
+static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr)
+{
+	return paddr + get_dma_direct_offset(dev);
+}
+
+static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr)
+{
+	return daddr - get_dma_direct_offset(dev);
+}
+
 #define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
 #define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
 #ifdef CONFIG_NOT_COHERENT_CACHE
diff --git a/arch/x86/include/asm/dma-mapping.h b/arch/x86/include/asm/dma-mapping.h
index adac59c8f69b..0ee770d23d0e 100644
--- a/arch/x86/include/asm/dma-mapping.h
+++ b/arch/x86/include/asm/dma-mapping.h
@@ -63,6 +63,16 @@ static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
 	return addr + size <= *dev->dma_mask;
 }
 
+static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr)
+{
+	return paddr;
+}
+
+static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr)
+{
+	return daddr;
+}
+
 static inline void
 dma_cache_sync(struct device *dev, void *vaddr, size_t size,
 	enum dma_data_direction dir)
-- 
cgit v1.2.3


From b683d42693c4e92b838117f5c6f7b90bfa1525c9 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Fri, 10 Jul 2009 10:05:04 +0900
Subject: x86: remove unused swiotlb_phys_to_bus() and swiotlb_bus_to_phys()

phys_to_dma() and dma_to_phys() are used instead of
swiotlb_phys_to_bus() and swiotlb_bus_to_phys().

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
---
 arch/x86/kernel/pci-swiotlb.c | 10 ----------
 1 file changed, 10 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/pci-swiotlb.c b/arch/x86/kernel/pci-swiotlb.c
index 165bd7f93bb1..e8a35016115f 100644
--- a/arch/x86/kernel/pci-swiotlb.c
+++ b/arch/x86/kernel/pci-swiotlb.c
@@ -13,16 +13,6 @@
 
 int swiotlb __read_mostly;
 
-dma_addr_t swiotlb_phys_to_bus(struct device *hwdev, phys_addr_t paddr)
-{
-	return paddr;
-}
-
-phys_addr_t swiotlb_bus_to_phys(struct device *hwdev, dma_addr_t baddr)
-{
-	return baddr;
-}
-
 static void *x86_swiotlb_alloc_coherent(struct device *hwdev, size_t size,
 					dma_addr_t *dma_handle, gfp_t flags)
 {
-- 
cgit v1.2.3


From c1dc0b9c0c8979ce4d411caadff5c0d79dee58bc Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Sun, 2 Aug 2009 11:28:21 +0200
Subject: debug lockups: Improve lockup detection

When debugging a recent lockup bug i found various deficiencies
in how our current lockup detection helpers work:

 - SysRq-L is not very efficient as it uses a workqueue, hence
   it cannot punch through hard lockups and cannot see through
   most soft lockups either.

 - The SysRq-L code depends on the NMI watchdog - which is off
   by default.

 - We dont print backtraces from the RCU code's built-in
   'RCU state machine is stuck' debug code. This debug
   code tends to be one of the first (and only) mechanisms
   that show that a lockup has occured.

This patch changes the code so taht we:

 - Trigger the NMI backtrace code from SysRq-L instead of using
   a workqueue (which cannot punch through hard lockups)

 - Trigger print-all-CPU-backtraces from the RCU lockup detection
   code

Also decouple the backtrace printing code from the NMI watchdog:

 - Dont use variable size cpumasks (it might not be initialized
   and they are a bit more fragile anyway)

 - Trigger an NMI immediately via an IPI, instead of waiting
   for the NMI tick to occur. This is a lot faster and can
   produce more relevant backtraces. It will also work if the
   NMI watchdog is disabled.

 - Dont print the 'dazed and confused' message when we print
   a backtrace from the NMI

 - Do a show_regs() plus a dump_stack() to get maximum info
   out of the dump. Worst-case we get two stacktraces - which
   is not a big deal. Sometimes, if register content is
   corrupted, the precise stack walker in show_regs() wont
   give us a full backtrace - in this case dump_stack() will
   do it.

Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic/nmi.c | 18 ++++++++++++------
 drivers/char/sysrq.c       |  8 ++------
 kernel/rcutree.c           |  7 ++++++-
 3 files changed, 20 insertions(+), 13 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apic/nmi.c b/arch/x86/kernel/apic/nmi.c
index b3025b43b63a..1bb1ac20e9ec 100644
--- a/arch/x86/kernel/apic/nmi.c
+++ b/arch/x86/kernel/apic/nmi.c
@@ -39,7 +39,7 @@
 int unknown_nmi_panic;
 int nmi_watchdog_enabled;
 
-static cpumask_var_t backtrace_mask;
+static cpumask_t backtrace_mask __read_mostly;
 
 /* nmi_active:
  * >0: the lapic NMI watchdog is active, but can be disabled
@@ -138,7 +138,6 @@ int __init check_nmi_watchdog(void)
 	if (!prev_nmi_count)
 		goto error;
 
-	alloc_cpumask_var(&backtrace_mask, GFP_KERNEL|__GFP_ZERO);
 	printk(KERN_INFO "Testing NMI watchdog ... ");
 
 #ifdef CONFIG_SMP
@@ -415,14 +414,17 @@ nmi_watchdog_tick(struct pt_regs *regs, unsigned reason)
 	}
 
 	/* We can be called before check_nmi_watchdog, hence NULL check. */
-	if (backtrace_mask != NULL && cpumask_test_cpu(cpu, backtrace_mask)) {
+	if (cpumask_test_cpu(cpu, &backtrace_mask)) {
 		static DEFINE_SPINLOCK(lock);	/* Serialise the printks */
 
 		spin_lock(&lock);
 		printk(KERN_WARNING "NMI backtrace for cpu %d\n", cpu);
+		show_regs(regs);
 		dump_stack();
 		spin_unlock(&lock);
-		cpumask_clear_cpu(cpu, backtrace_mask);
+		cpumask_clear_cpu(cpu, &backtrace_mask);
+
+		rc = 1;
 	}
 
 	/* Could check oops_in_progress here too, but it's safer not to */
@@ -556,10 +558,14 @@ void __trigger_all_cpu_backtrace(void)
 {
 	int i;
 
-	cpumask_copy(backtrace_mask, cpu_online_mask);
+	cpumask_copy(&backtrace_mask, cpu_online_mask);
+
+	printk(KERN_INFO "sending NMI to all CPUs:\n");
+	apic->send_IPI_all(NMI_VECTOR);
+
 	/* Wait for up to 10 seconds for all CPUs to do the backtrace */
 	for (i = 0; i < 10 * 1000; i++) {
-		if (cpumask_empty(backtrace_mask))
+		if (cpumask_empty(&backtrace_mask))
 			break;
 		mdelay(1);
 	}
diff --git a/drivers/char/sysrq.c b/drivers/char/sysrq.c
index 5d7a02f63e1c..165f307f30e8 100644
--- a/drivers/char/sysrq.c
+++ b/drivers/char/sysrq.c
@@ -24,6 +24,7 @@
 #include <linux/sysrq.h>
 #include <linux/kbd_kern.h>
 #include <linux/proc_fs.h>
+#include <linux/nmi.h>
 #include <linux/quotaops.h>
 #include <linux/perf_counter.h>
 #include <linux/kernel.h>
@@ -222,12 +223,7 @@ static DECLARE_WORK(sysrq_showallcpus, sysrq_showregs_othercpus);
 
 static void sysrq_handle_showallcpus(int key, struct tty_struct *tty)
 {
-	struct pt_regs *regs = get_irq_regs();
-	if (regs) {
-		printk(KERN_INFO "CPU%d:\n", smp_processor_id());
-		show_regs(regs);
-	}
-	schedule_work(&sysrq_showallcpus);
+	trigger_all_cpu_backtrace();
 }
 
 static struct sysrq_key_op sysrq_showallcpus_op = {
diff --git a/kernel/rcutree.c b/kernel/rcutree.c
index 7717b95c2027..9c5fa9fc57ec 100644
--- a/kernel/rcutree.c
+++ b/kernel/rcutree.c
@@ -35,6 +35,7 @@
 #include <linux/rcupdate.h>
 #include <linux/interrupt.h>
 #include <linux/sched.h>
+#include <linux/nmi.h>
 #include <asm/atomic.h>
 #include <linux/bitops.h>
 #include <linux/module.h>
@@ -469,6 +470,8 @@ static void print_other_cpu_stall(struct rcu_state *rsp)
 	}
 	printk(" (detected by %d, t=%ld jiffies)\n",
 	       smp_processor_id(), (long)(jiffies - rsp->gp_start));
+	trigger_all_cpu_backtrace();
+
 	force_quiescent_state(rsp, 0);  /* Kick them all. */
 }
 
@@ -479,12 +482,14 @@ static void print_cpu_stall(struct rcu_state *rsp)
 
 	printk(KERN_ERR "INFO: RCU detected CPU %d stall (t=%lu jiffies)\n",
 			smp_processor_id(), jiffies - rsp->gp_start);
-	dump_stack();
+	trigger_all_cpu_backtrace();
+
 	spin_lock_irqsave(&rnp->lock, flags);
 	if ((long)(jiffies - rsp->jiffies_stall) >= 0)
 		rsp->jiffies_stall =
 			jiffies + RCU_SECONDS_TILL_STALL_RECHECK;
 	spin_unlock_irqrestore(&rnp->lock, flags);
+
 	set_need_resched();  /* kick ourselves to get things going. */
 }
 
-- 
cgit v1.2.3


From 25f6e89bedd29cc49bfa0d55497e91a671b9ae6e Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 30 Jul 2009 23:21:18 +0200
Subject: x86: Remove superfluous NULL pointer check in destroy_irq()

This takes care of the following entry from Dan's list:

  arch/x86/kernel/apic/io_apic.c +3241 destroy_irq(11) warning: variable derefenced before check 'desc'

Reported-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Eugene Teo <eteo@redhat.com>
Cc: Julia Lawall <julia@diku.dk>
LKML-Reference: <200907302321.19086.bzolnier@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic/io_apic.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index cf51b0b58c56..7e92a9212fd7 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -3185,8 +3185,7 @@ void destroy_irq(unsigned int irq)
 	cfg = desc->chip_data;
 	dynamic_irq_cleanup(irq);
 	/* connect back irq_cfg */
-	if (desc)
-		desc->chip_data = cfg;
+	desc->chip_data = cfg;
 
 	free_irte(irq);
 	spin_lock_irqsave(&vector_lock, flags);
-- 
cgit v1.2.3


From 47cab6a722d44c71c4f8224017ef548522243cf4 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 3 Aug 2009 09:31:54 +0200
Subject: debug lockups: Improve lockup detection, fix generic arch fallback

As Andrew noted, my previous patch ("debug lockups: Improve lockup
detection") broke/removed SysRq-L support from architecture that do
not provide a __trigger_all_cpu_backtrace implementation.

Restore a fallback path and clean up the SysRq-L machinery a bit:

 - Rename the arch method to arch_trigger_all_cpu_backtrace()

 - Simplify the define

 - Document the method a bit - in the hope of more architectures
   adding support for it.

[ The patch touches Sparc code for the rename. ]

Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: "David S. Miller" <davem@davemloft.net>
LKML-Reference: <20090802140809.7ec4bb6b.akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/sparc/include/asm/irq_64.h |  4 ++--
 arch/sparc/kernel/process_64.c  |  4 ++--
 arch/x86/include/asm/nmi.h      |  4 ++--
 arch/x86/kernel/apic/nmi.c      |  2 +-
 drivers/char/sysrq.c            | 15 ++++++++++++++-
 include/linux/nmi.h             | 19 +++++++++++++++++--
 6 files changed, 38 insertions(+), 10 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/sparc/include/asm/irq_64.h b/arch/sparc/include/asm/irq_64.h
index 1934f2cbf513..a0b443cb3c1f 100644
--- a/arch/sparc/include/asm/irq_64.h
+++ b/arch/sparc/include/asm/irq_64.h
@@ -89,8 +89,8 @@ static inline unsigned long get_softint(void)
 	return retval;
 }
 
-void __trigger_all_cpu_backtrace(void);
-#define trigger_all_cpu_backtrace() __trigger_all_cpu_backtrace()
+void arch_trigger_all_cpu_backtrace(void);
+#define arch_trigger_all_cpu_backtrace arch_trigger_all_cpu_backtrace
 
 extern void *hardirq_stack[NR_CPUS];
 extern void *softirq_stack[NR_CPUS];
diff --git a/arch/sparc/kernel/process_64.c b/arch/sparc/kernel/process_64.c
index 4041f94e7724..18d67854a1b8 100644
--- a/arch/sparc/kernel/process_64.c
+++ b/arch/sparc/kernel/process_64.c
@@ -251,7 +251,7 @@ static void __global_reg_poll(struct global_reg_snapshot *gp)
 	}
 }
 
-void __trigger_all_cpu_backtrace(void)
+void arch_trigger_all_cpu_backtrace(void)
 {
 	struct thread_info *tp = current_thread_info();
 	struct pt_regs *regs = get_irq_regs();
@@ -304,7 +304,7 @@ void __trigger_all_cpu_backtrace(void)
 
 static void sysrq_handle_globreg(int key, struct tty_struct *tty)
 {
-	__trigger_all_cpu_backtrace();
+	arch_trigger_all_cpu_backtrace();
 }
 
 static struct sysrq_key_op sparc_globalreg_op = {
diff --git a/arch/x86/include/asm/nmi.h b/arch/x86/include/asm/nmi.h
index c86e5ed4af51..e63cf7d441e1 100644
--- a/arch/x86/include/asm/nmi.h
+++ b/arch/x86/include/asm/nmi.h
@@ -45,8 +45,8 @@ extern int proc_nmi_enabled(struct ctl_table *, int , struct file *,
 			void __user *, size_t *, loff_t *);
 extern int unknown_nmi_panic;
 
-void __trigger_all_cpu_backtrace(void);
-#define trigger_all_cpu_backtrace() __trigger_all_cpu_backtrace()
+void arch_trigger_all_cpu_backtrace(void);
+#define arch_trigger_all_cpu_backtrace arch_trigger_all_cpu_backtrace
 
 static inline void localise_nmi_watchdog(void)
 {
diff --git a/arch/x86/kernel/apic/nmi.c b/arch/x86/kernel/apic/nmi.c
index 1bb1ac20e9ec..db7220220d09 100644
--- a/arch/x86/kernel/apic/nmi.c
+++ b/arch/x86/kernel/apic/nmi.c
@@ -554,7 +554,7 @@ int do_nmi_callback(struct pt_regs *regs, int cpu)
 	return 0;
 }
 
-void __trigger_all_cpu_backtrace(void)
+void arch_trigger_all_cpu_backtrace(void)
 {
 	int i;
 
diff --git a/drivers/char/sysrq.c b/drivers/char/sysrq.c
index 165f307f30e8..50eecfe1d724 100644
--- a/drivers/char/sysrq.c
+++ b/drivers/char/sysrq.c
@@ -223,7 +223,20 @@ static DECLARE_WORK(sysrq_showallcpus, sysrq_showregs_othercpus);
 
 static void sysrq_handle_showallcpus(int key, struct tty_struct *tty)
 {
-	trigger_all_cpu_backtrace();
+	/*
+	 * Fall back to the workqueue based printing if the
+	 * backtrace printing did not succeed or the
+	 * architecture has no support for it:
+	 */
+	if (!trigger_all_cpu_backtrace()) {
+		struct pt_regs *regs = get_irq_regs();
+
+		if (regs) {
+			printk(KERN_INFO "CPU%d:\n", smp_processor_id());
+			show_regs(regs);
+		}
+		schedule_work(&sysrq_showallcpus);
+	}
 }
 
 static struct sysrq_key_op sysrq_showallcpus_op = {
diff --git a/include/linux/nmi.h b/include/linux/nmi.h
index 29af2d5df097..b752e807adde 100644
--- a/include/linux/nmi.h
+++ b/include/linux/nmi.h
@@ -28,8 +28,23 @@ static inline void acpi_nmi_disable(void) { }
 static inline void acpi_nmi_enable(void) { }
 #endif
 
-#ifndef trigger_all_cpu_backtrace
-#define trigger_all_cpu_backtrace() do { } while (0)
+/*
+ * Create trigger_all_cpu_backtrace() out of the arch-provided
+ * base function. Return whether such support was available,
+ * to allow calling code to fall back to some other mechanism:
+ */
+#ifdef arch_trigger_all_cpu_backtrace
+static inline bool trigger_all_cpu_backtrace(void)
+{
+	arch_trigger_all_cpu_backtrace();
+
+	return true;
+}
+#else
+static inline bool trigger_all_cpu_backtrace(void)
+{
+	return false;
+}
 #endif
 
 #endif
-- 
cgit v1.2.3


From ed8d9adf357ec331603fa1049510399812cea7e5 Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Mon, 3 Aug 2009 14:08:48 +0900
Subject: x86, percpu: Add 'percpu_read_stable()' interface for cacheable
 accesses

This is very useful for some common things like 'get_current()' and
'get_thread_info()', which can be used multiple times in a function, and
where the result is cacheable.

tj: Added the magical undocumented "P" modifier to UP __percpu_arg()
    to force gcc to dereference the pointer value passed in via the
    "p" input constraint.  Without this, percpu_read_stable() returns
    the address of the percpu variable.  Also added comment explaining
    the difference between percpu_read() and percpu_read_stable().

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/include/asm/current.h     |  2 +-
 arch/x86/include/asm/percpu.h      | 26 +++++++++++++++++++-------
 arch/x86/include/asm/thread_info.h |  2 +-
 3 files changed, 21 insertions(+), 9 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/current.h b/arch/x86/include/asm/current.h
index c68c361697e1..4d447b732d82 100644
--- a/arch/x86/include/asm/current.h
+++ b/arch/x86/include/asm/current.h
@@ -11,7 +11,7 @@ DECLARE_PER_CPU(struct task_struct *, current_task);
 
 static __always_inline struct task_struct *get_current(void)
 {
-	return percpu_read(current_task);
+	return percpu_read_stable(current_task);
 }
 
 #define current get_current()
diff --git a/arch/x86/include/asm/percpu.h b/arch/x86/include/asm/percpu.h
index 103f1ddb0d85..04eacefcfd26 100644
--- a/arch/x86/include/asm/percpu.h
+++ b/arch/x86/include/asm/percpu.h
@@ -49,7 +49,7 @@
 #define __percpu_arg(x)		"%%"__stringify(__percpu_seg)":%P" #x
 #define __my_cpu_offset		percpu_read(this_cpu_off)
 #else
-#define __percpu_arg(x)		"%" #x
+#define __percpu_arg(x)		"%P" #x
 #endif
 
 /*
@@ -104,36 +104,48 @@ do {							\
 	}						\
 } while (0)
 
-#define percpu_from_op(op, var)				\
+#define percpu_from_op(op, var, constraint)		\
 ({							\
 	typeof(var) ret__;				\
 	switch (sizeof(var)) {				\
 	case 1:						\
 		asm(op "b "__percpu_arg(1)",%0"		\
 		    : "=q" (ret__)			\
-		    : "m" (var));			\
+		    : constraint);			\
 		break;					\
 	case 2:						\
 		asm(op "w "__percpu_arg(1)",%0"		\
 		    : "=r" (ret__)			\
-		    : "m" (var));			\
+		    : constraint);			\
 		break;					\
 	case 4:						\
 		asm(op "l "__percpu_arg(1)",%0"		\
 		    : "=r" (ret__)			\
-		    : "m" (var));			\
+		    : constraint);			\
 		break;					\
 	case 8:						\
 		asm(op "q "__percpu_arg(1)",%0"		\
 		    : "=r" (ret__)			\
-		    : "m" (var));			\
+		    : constraint);			\
 		break;					\
 	default: __bad_percpu_size();			\
 	}						\
 	ret__;						\
 })
 
-#define percpu_read(var)	percpu_from_op("mov", per_cpu__##var)
+/*
+ * percpu_read() makes gcc load the percpu variable every time it is
+ * accessed while percpu_read_stable() allows the value to be cached.
+ * percpu_read_stable() is more efficient and can be used if its value
+ * is guaranteed to be valid across cpus.  The current users include
+ * get_current() and get_thread_info() both of which are actually
+ * per-thread variables implemented as per-cpu variables and thus
+ * stable for the duration of the respective task.
+ */
+#define percpu_read(var)	percpu_from_op("mov", per_cpu__##var,	\
+					       "m" (per_cpu__##var))
+#define percpu_read_stable(var)	percpu_from_op("mov", per_cpu__##var,	\
+					       "p" (&per_cpu__##var))
 #define percpu_write(var, val)	percpu_to_op("mov", per_cpu__##var, val)
 #define percpu_add(var, val)	percpu_to_op("add", per_cpu__##var, val)
 #define percpu_sub(var, val)	percpu_to_op("sub", per_cpu__##var, val)
diff --git a/arch/x86/include/asm/thread_info.h b/arch/x86/include/asm/thread_info.h
index fad7d40b75f8..a1bb5a114bf2 100644
--- a/arch/x86/include/asm/thread_info.h
+++ b/arch/x86/include/asm/thread_info.h
@@ -213,7 +213,7 @@ DECLARE_PER_CPU(unsigned long, kernel_stack);
 static inline struct thread_info *current_thread_info(void)
 {
 	struct thread_info *ti;
-	ti = (void *)(percpu_read(kernel_stack) +
+	ti = (void *)(percpu_read_stable(kernel_stack) +
 		      KERNEL_STACK_OFFSET - THREAD_SIZE);
 	return ti;
 }
-- 
cgit v1.2.3


From 3e352aa8ee2bd48f1a19c7742810b3a4a7ba605e Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Mon, 3 Aug 2009 14:10:11 +0900
Subject: x86, percpu: Fix DECLARE/DEFINE_PER_CPU_PAGE_ALIGNED()

DECLARE/DEFINE_PER_CPU_PAGE_ALIGNED() put percpu variables in
.page_aligned section without adding any alignment restrictions.
Currently, this doesn't cause any problem because all users of the
macros have explicit page alignment and page-sized but it's much safer
to enforce page alignment from the macros.  After all, it's what they
claim to do.

Add __aligned(PAGE_SIZE) to DECLARE/DEFINE_PER_CPU_PAGE_ALIGNED() and
drop explicit alignment from it users.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/kernel/cpu/common.c | 3 +--
 include/linux/percpu-defs.h  | 8 +++++---
 2 files changed, 6 insertions(+), 5 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index f1961c07af9a..12493c5485e5 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -1008,8 +1008,7 @@ static const unsigned int exception_stack_sizes[N_EXCEPTION_STACKS] = {
 };
 
 static DEFINE_PER_CPU_PAGE_ALIGNED(char, exception_stacks
-	[(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ + DEBUG_STKSZ])
-	__aligned(PAGE_SIZE);
+	[(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ + DEBUG_STKSZ]);
 
 /* May not be marked __init: used by software suspend */
 void syscall_init(void)
diff --git a/include/linux/percpu-defs.h b/include/linux/percpu-defs.h
index 68438e18fff4..afd5f8b7061f 100644
--- a/include/linux/percpu-defs.h
+++ b/include/linux/percpu-defs.h
@@ -69,11 +69,13 @@
 /*
  * Declaration/definition used for per-CPU variables that must be page aligned.
  */
-#define DECLARE_PER_CPU_PAGE_ALIGNED(type, name)				\
-	DECLARE_PER_CPU_SECTION(type, name, ".page_aligned")
+#define DECLARE_PER_CPU_PAGE_ALIGNED(type, name)			\
+	DECLARE_PER_CPU_SECTION(type, name, ".page_aligned")		\
+	__aligned(PAGE_SIZE)
 
 #define DEFINE_PER_CPU_PAGE_ALIGNED(type, name)				\
-	DEFINE_PER_CPU_SECTION(type, name, ".page_aligned")
+	DEFINE_PER_CPU_SECTION(type, name, ".page_aligned")		\
+	__aligned(PAGE_SIZE)
 
 /*
  * Intermodule exports for per-CPU variables.
-- 
cgit v1.2.3


From bdf977b37418cdf8a2252504779a7e12a09b7575 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Mon, 3 Aug 2009 14:12:19 +0900
Subject: x86, percpu: Collect hot percpu variables into one cacheline

On x86_64, percpu variables current_task and kernel_stack are used for
get_current() and current_thread_info() respectively and thus are
often used close to each other.  Move definition of current_task to
kernel/cpu/common.c right above kernel_stack definition and align it
to cacheline so that they always fall into the same cacheline.  Two
percpu variables defined there together - irq_stack_ptr and irq_count
- are also pretty hot and will benefit from sharing the cacheline.

For consistency, current_task definition for x86_32 is also moved to
kernel/cpu/common.c.

Putting current_task and kernel_stack into the same cacheline was
suggested by Linus Torvalds.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/kernel/cpu/common.c | 15 +++++++++++++--
 arch/x86/kernel/process_32.c |  3 ---
 arch/x86/kernel/process_64.c |  3 ---
 3 files changed, 13 insertions(+), 8 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 12493c5485e5..1bd88ed978bd 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -987,13 +987,21 @@ struct desc_ptr idt_descr = { 256 * 16 - 1, (unsigned long) idt_table };
 DEFINE_PER_CPU_FIRST(union irq_stack_union,
 		     irq_stack_union) __aligned(PAGE_SIZE);
 
-DEFINE_PER_CPU(char *, irq_stack_ptr) =
-	init_per_cpu_var(irq_stack_union.irq_stack) + IRQ_STACK_SIZE - 64;
+/*
+ * The following four percpu variables are hot.  Align current_task to
+ * cacheline size such that all four fall in the same cacheline.
+ */
+DEFINE_PER_CPU(struct task_struct *, current_task) ____cacheline_aligned =
+	&init_task;
+EXPORT_PER_CPU_SYMBOL(current_task);
 
 DEFINE_PER_CPU(unsigned long, kernel_stack) =
 	(unsigned long)&init_thread_union - KERNEL_STACK_OFFSET + THREAD_SIZE;
 EXPORT_PER_CPU_SYMBOL(kernel_stack);
 
+DEFINE_PER_CPU(char *, irq_stack_ptr) =
+	init_per_cpu_var(irq_stack_union.irq_stack) + IRQ_STACK_SIZE - 64;
+
 DEFINE_PER_CPU(unsigned int, irq_count) = -1;
 
 /*
@@ -1041,6 +1049,9 @@ DEFINE_PER_CPU(struct orig_ist, orig_ist);
 
 #else	/* CONFIG_X86_64 */
 
+DEFINE_PER_CPU(struct task_struct *, current_task) = &init_task;
+EXPORT_PER_CPU_SYMBOL(current_task);
+
 #ifdef CONFIG_CC_STACKPROTECTOR
 DEFINE_PER_CPU(unsigned long, stack_canary);
 #endif
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 59f4524984af..daa4107be3b4 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -61,9 +61,6 @@
 
 asmlinkage void ret_from_fork(void) __asm__("ret_from_fork");
 
-DEFINE_PER_CPU(struct task_struct *, current_task) = &init_task;
-EXPORT_PER_CPU_SYMBOL(current_task);
-
 /*
  * Return saved PC of a blocked thread.
  */
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index ebefb5407b9d..c4c675d5ba1a 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -55,9 +55,6 @@
 
 asmlinkage extern void ret_from_fork(void);
 
-DEFINE_PER_CPU(struct task_struct *, current_task) = &init_task;
-EXPORT_PER_CPU_SYMBOL(current_task);
-
 DEFINE_PER_CPU(unsigned long, old_rsp);
 static DEFINE_PER_CPU(unsigned char, is_idle);
 
-- 
cgit v1.2.3


From c7bd0414d681706a32105895cae20fb9090db52e Mon Sep 17 00:00:00 2001
From: Frans Pop <elendil@planet.nl>
Date: Thu, 23 Jul 2009 20:56:27 +0200
Subject: x86: Simplify the Makefile in a minor way through use of cc-ifversion

Signed-off-by: Frans Pop <elendil@planet.nl>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Reviewed-by: WANG Cong <xiyou.wangcong@gmail.com>
LKML-Reference: <200907232056.28635.elendil@planet.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/Makefile | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/Makefile b/arch/x86/Makefile
index 1b68659c41b4..1f3851a626d3 100644
--- a/arch/x86/Makefile
+++ b/arch/x86/Makefile
@@ -32,8 +32,8 @@ ifeq ($(CONFIG_X86_32),y)
 
         # Disable unit-at-a-time mode on pre-gcc-4.0 compilers, it makes gcc use
         # a lot more stack due to the lack of sharing of stacklots:
-        KBUILD_CFLAGS += $(shell if [ $(call cc-version) -lt 0400 ] ; then \
-                echo $(call cc-option,-fno-unit-at-a-time); fi ;)
+        KBUILD_CFLAGS += $(call cc-ifversion, -lt, 0400, \
+				$(call cc-option,-fno-unit-at-a-time))
 
         # CPU-specific tuning. Anything which can be shared with UML should go here.
         include $(srctree)/arch/x86/Makefile_32.cpu
-- 
cgit v1.2.3


From 54a0bf3c2cad3fd118ea725f26a493aece6ea01d Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Tue, 4 Aug 2009 15:52:38 +0200
Subject: Revert "x86: oprofile/op_model_amd.c set return values for
 op_amd_handle_ibs()"

This reverts commit 21e70878215f620fe99ea7d7c74bc641aeec932f.

Instead Andrew's patch will be applied he posted at the same time.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_amd.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index 827beecb67a4..37d19c768d5f 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -195,7 +195,7 @@ op_amd_handle_ibs(struct pt_regs * const regs,
 	struct op_entry entry;
 
 	if (!has_ibs)
-		return 0;
+		return 1;
 
 	if (ibs_config.fetch_enabled) {
 		rdmsrl(MSR_AMD64_IBSFETCHCTL, ctl);
@@ -277,10 +277,7 @@ static void op_amd_stop_ibs(void)
 #else
 
 static inline int op_amd_handle_ibs(struct pt_regs * const regs,
-				    struct op_msrs const * const msrs)
-{
-	return 0;
-}
+				    struct op_msrs const * const msrs) { }
 static inline void op_amd_start_ibs(void) { }
 static inline void op_amd_stop_ibs(void) { }
 
-- 
cgit v1.2.3


From 4680e64a88c4ce2c4e736dade99233e3def13fa7 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@linux-foundation.org>
Date: Tue, 23 Jun 2009 12:36:08 -0700
Subject: arch/x86/oprofile/op_model_amd.c: fix op_amd_handle_ibs() return type

arch/x86/oprofile/op_model_amd.c: In function 'op_amd_handle_ibs':
arch/x86/oprofile/op_model_amd.c:217: warning: no return statement in function returning non-void

Fix this by making op_amd_handle_ibs() return void.

Cc: Robert Richter <robert.richter@amd.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_amd.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index 37d19c768d5f..39686c29f03a 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -187,7 +187,7 @@ static void op_amd_setup_ctrs(struct op_x86_model_spec const *model,
 
 #ifdef CONFIG_OPROFILE_IBS
 
-static inline int
+static inline void
 op_amd_handle_ibs(struct pt_regs * const regs,
 		  struct op_msrs const * const msrs)
 {
@@ -195,7 +195,7 @@ op_amd_handle_ibs(struct pt_regs * const regs,
 	struct op_entry entry;
 
 	if (!has_ibs)
-		return 1;
+		return;
 
 	if (ibs_config.fetch_enabled) {
 		rdmsrl(MSR_AMD64_IBSFETCHCTL, ctl);
@@ -241,8 +241,6 @@ op_amd_handle_ibs(struct pt_regs * const regs,
 			wrmsrl(MSR_AMD64_IBSOPCTL, ctl);
 		}
 	}
-
-	return 1;
 }
 
 static inline void op_amd_start_ibs(void)
@@ -276,7 +274,7 @@ static void op_amd_stop_ibs(void)
 
 #else
 
-static inline int op_amd_handle_ibs(struct pt_regs * const regs,
+static inline void op_amd_handle_ibs(struct pt_regs * const regs,
 				    struct op_msrs const * const msrs) { }
 static inline void op_amd_start_ibs(void) { }
 static inline void op_amd_stop_ibs(void) { }
-- 
cgit v1.2.3


From 2977fb3ffc8493a2f4f0a362e8660a6cde9f1bb9 Mon Sep 17 00:00:00 2001
From: Cyrill Gorcunov <gorcunov@openvz.org>
Date: Sat, 1 Aug 2009 11:47:59 +0400
Subject: x86, ioapic: Introduce for_each_irq_pin() helper

This allow us to save a few lines of code.

Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
Cc: yinghai@kernel.org
LKML-Reference: <20090801075435.597863129@openvz.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic/io_apic.c | 43 +++++++++++++++---------------------------
 1 file changed, 15 insertions(+), 28 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 7e92a9212fd7..ffd8fdfcbe4c 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -66,6 +66,8 @@
 #include <asm/apic.h>
 
 #define __apicdebuginit(type) static type __init
+#define for_each_irq_pin(entry, head) \
+	for (entry = head; entry; entry = entry->next)
 
 /*
  *      Is the SiS APIC rmw bug present ?
@@ -410,7 +412,7 @@ static bool io_apic_level_ack_pending(struct irq_cfg *cfg)
 	unsigned long flags;
 
 	spin_lock_irqsave(&ioapic_lock, flags);
-	for (entry = cfg->irq_2_pin; entry != NULL; entry = entry->next) {
+	for_each_irq_pin(entry, cfg->irq_2_pin) {
 		unsigned int reg;
 		int pin;
 
@@ -490,22 +492,21 @@ static void ioapic_mask_entry(int apic, int pin)
  */
 static void add_pin_to_irq_node(struct irq_cfg *cfg, int node, int apic, int pin)
 {
-	struct irq_pin_list **entryp, *entry;
+	struct irq_pin_list **last, *entry;
 
-	for (entryp = &cfg->irq_2_pin;
-	     *entryp != NULL;
-	     entryp = &(*entryp)->next) {
-		entry = *entryp;
-		/* not again, please */
+	/* don't allow duplicates */
+	last = &cfg->irq_2_pin;
+	for_each_irq_pin(entry, cfg->irq_2_pin) {
 		if (entry->apic == apic && entry->pin == pin)
 			return;
+		last = &entry->next;
 	}
 
 	entry = get_one_free_irq_2_pin(node);
 	entry->apic = apic;
 	entry->pin = pin;
 
-	*entryp = entry;
+	*last = entry;
 }
 
 /*
@@ -517,7 +518,7 @@ static void __init replace_pin_at_irq_node(struct irq_cfg *cfg, int node,
 {
 	struct irq_pin_list *entry;
 
-	for (entry = cfg->irq_2_pin; entry != NULL; entry = entry->next) {
+	for_each_irq_pin(entry, cfg->irq_2_pin) {
 		if (entry->apic == oldapic && entry->pin == oldpin) {
 			entry->apic = newapic;
 			entry->pin = newpin;
@@ -537,7 +538,7 @@ static void io_apic_modify_irq(struct irq_cfg *cfg,
 	int pin;
 	struct irq_pin_list *entry;
 
-	for (entry = cfg->irq_2_pin; entry != NULL; entry = entry->next) {
+	for_each_irq_pin(entry, cfg->irq_2_pin) {
 		unsigned int reg;
 		pin = entry->pin;
 		reg = io_apic_read(entry->apic, 0x10 + pin * 2);
@@ -1669,12 +1670,8 @@ __apicdebuginit(void) print_IO_APIC(void)
 		if (!entry)
 			continue;
 		printk(KERN_DEBUG "IRQ%d ", irq);
-		for (;;) {
+		for_each_irq_pin(entry, cfg->irq_2_pin)
 			printk("-> %d:%d", entry->apic, entry->pin);
-			if (!entry->next)
-				break;
-			entry = entry->next;
-		}
 		printk("\n");
 	}
 
@@ -2227,7 +2224,7 @@ static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq
 	struct irq_pin_list *entry;
 	u8 vector = cfg->vector;
 
-	for (entry = cfg->irq_2_pin; entry != NULL; entry = entry->next) {
+	for_each_irq_pin(entry, cfg->irq_2_pin) {
 		unsigned int reg;
 
 		apic = entry->apic;
@@ -2556,20 +2553,10 @@ static void ack_apic_level(unsigned int irq)
 #ifdef CONFIG_INTR_REMAP
 static void __eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg)
 {
-	int apic, pin;
 	struct irq_pin_list *entry;
 
-	entry = cfg->irq_2_pin;
-	for (;;) {
-
-		if (!entry)
-			break;
-
-		apic = entry->apic;
-		pin = entry->pin;
-		io_apic_eoi(apic, pin);
-		entry = entry->next;
-	}
+	for_each_irq_pin(entry, cfg->irq_2_pin)
+		io_apic_eoi(entry->apic, entry->pin);
 }
 
 static void
-- 
cgit v1.2.3


From a7428cd2ef77734465e36bceb43290e37e2a97c6 Mon Sep 17 00:00:00 2001
From: Cyrill Gorcunov <gorcunov@openvz.org>
Date: Sat, 1 Aug 2009 11:48:00 +0400
Subject: x86, ioapic: Throw BUG instead of NULL dereference

Instead of plain NULL deref we better throw error
message with a backtrace. Actually we need more
gracious error handling here. Meanwhile leave it
as is.

Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
Cc: yinghai@kernel.org
LKML-Reference: <20090801075435.769301745@openvz.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic/io_apic.c | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index ffd8fdfcbe4c..2a145d3a8375 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -503,6 +503,10 @@ static void add_pin_to_irq_node(struct irq_cfg *cfg, int node, int apic, int pin
 	}
 
 	entry = get_one_free_irq_2_pin(node);
+	if (!entry) {
+		printk(KERN_ERR "can not alloc irq_pin_list\n");
+		BUG_ON(1);
+	}
 	entry->apic = apic;
 	entry->pin = pin;
 
-- 
cgit v1.2.3


From 9910887af84e33ba98fd6792029470ae80166208 Mon Sep 17 00:00:00 2001
From: Cyrill Gorcunov <gorcunov@gmail.com>
Date: Thu, 23 Jul 2009 00:52:59 +0400
Subject: x86, apic: Drop redundant bit assignment

cpu_has_apic has already investigated boot_cpu_data
X86_FEATURE_APIC bit for being clear if condition is
triggered.

So there is no need to clear this bit second time.

Signed-off-by: Cyrill Gorcuno v <gorcunov@openvz.org>
Cc: "Maciej W. Rozycki" <macro@linux-mips.org>
LKML-Reference: <20090722205259.GE15805@lenovo>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic/apic.c | 1 -
 1 file changed, 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c
index 0a1c2830ec66..0b021c56e822 100644
--- a/arch/x86/kernel/apic/apic.c
+++ b/arch/x86/kernel/apic/apic.c
@@ -1651,7 +1651,6 @@ int __init APIC_init_uniprocessor(void)
 	    APIC_INTEGRATED(apic_version[boot_cpu_physical_apicid])) {
 		pr_err("BIOS bug, local APIC 0x%x not detected!...\n",
 			boot_cpu_physical_apicid);
-		clear_cpu_cap(&boot_cpu_data, X86_FEATURE_APIC);
 		return -1;
 	}
 #endif
-- 
cgit v1.2.3


From ce69a784504222c3ab6f1b3c357d09ec5772127a Mon Sep 17 00:00:00 2001
From: Gleb Natapov <gleb@redhat.com>
Date: Mon, 20 Jul 2009 15:24:17 +0300
Subject: x86/apic: Enable x2APIC without interrupt remapping under KVM

KVM would like to provide x2APIC interface to a guest without emulating
interrupt remapping device. The reason KVM prefers guest to use x2APIC
is that x2APIC interface is better virtualizable and provides better
performance than mmio xAPIC interface:

 - msr exits are faster than mmio (no page table walk, emulation)
 - no need to read back ICR to look at the busy bit
 - one 64 bit ICR write instead of two 32 bit writes
 - shared code with the Hyper-V paravirt interface

Included patch changes x2APIC enabling logic to enable it even if IR
initialization failed, but kernel runs under KVM and no apic id is
greater than 255 (if there is one spec requires BIOS to move to x2apic
mode before starting an OS).

-v2: fix build
-v3: fix bug causing compiler warning

Signed-off-by: Gleb Natapov <gleb@redhat.com>
Acked-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Sheng Yang <sheng@linux.intel.com>
Cc: "avi@redhat.com" <avi@redhat.com>
LKML-Reference: <20090720122417.GR5638@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/include/asm/apic.h     |  7 ++++
 arch/x86/kernel/apic/apic.c     | 83 +++++++++++++++++++++++------------------
 arch/x86/kernel/apic/probe_64.c |  6 +--
 3 files changed, 56 insertions(+), 40 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/apic.h b/arch/x86/include/asm/apic.h
index bb7d47925847..586b7adb8e53 100644
--- a/arch/x86/include/asm/apic.h
+++ b/arch/x86/include/asm/apic.h
@@ -183,6 +183,10 @@ static inline int x2apic_enabled(void)
 }
 
 #define x2apic_supported()	(cpu_has_x2apic)
+static inline void x2apic_force_phys(void)
+{
+	x2apic_phys = 1;
+}
 #else
 static inline void check_x2apic(void)
 {
@@ -194,6 +198,9 @@ static inline int x2apic_enabled(void)
 {
 	return 0;
 }
+static inline void x2apic_force_phys(void)
+{
+}
 
 #define	x2apic_preenabled 0
 #define	x2apic_supported()	0
diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c
index 0b021c56e822..de039fcdd053 100644
--- a/arch/x86/kernel/apic/apic.c
+++ b/arch/x86/kernel/apic/apic.c
@@ -49,6 +49,7 @@
 #include <asm/mtrr.h>
 #include <asm/smp.h>
 #include <asm/mce.h>
+#include <asm/kvm_para.h>
 
 unsigned int num_processors;
 
@@ -1361,52 +1362,76 @@ void enable_x2apic(void)
 }
 #endif /* CONFIG_X86_X2APIC */
 
-void __init enable_IR_x2apic(void)
+int __init enable_IR(void)
 {
 #ifdef CONFIG_INTR_REMAP
 	int ret;
-	unsigned long flags;
-	struct IO_APIC_route_entry **ioapic_entries = NULL;
 
 	ret = dmar_table_init();
 	if (ret) {
 		pr_debug("dmar_table_init() failed with %d:\n", ret);
-		goto ir_failed;
+		return 0;
 	}
 
 	if (!intr_remapping_supported()) {
 		pr_debug("intr-remapping not supported\n");
-		goto ir_failed;
+		return 0;
 	}
 
-
 	if (!x2apic_preenabled && skip_ioapic_setup) {
 		pr_info("Skipped enabling intr-remap because of skipping "
 			"io-apic setup\n");
-		return;
+		return 0;
 	}
 
+	if (enable_intr_remapping(x2apic_supported()))
+		return 0;
+
+	pr_info("Enabled Interrupt-remapping\n");
+
+	return 1;
+
+#endif
+	return 0;
+}
+
+void __init enable_IR_x2apic(void)
+{
+	unsigned long flags;
+	struct IO_APIC_route_entry **ioapic_entries = NULL;
+	int ret, x2apic_enabled = 0;
+
 	ioapic_entries = alloc_ioapic_entries();
 	if (!ioapic_entries) {
-		pr_info("Allocate ioapic_entries failed: %d\n", ret);
-		goto end;
+		pr_err("Allocate ioapic_entries failed\n");
+		goto out;
 	}
 
 	ret = save_IO_APIC_setup(ioapic_entries);
 	if (ret) {
 		pr_info("Saving IO-APIC state failed: %d\n", ret);
-		goto end;
+		goto out;
 	}
 
 	local_irq_save(flags);
-	mask_IO_APIC_setup(ioapic_entries);
 	mask_8259A();
+	mask_IO_APIC_setup(ioapic_entries);
 
-	ret = enable_intr_remapping(x2apic_supported());
-	if (ret)
-		goto end_restore;
+	ret = enable_IR();
+	if (!ret) {
+		/* IR is required if there is APIC ID > 255 even when running
+		 * under KVM
+		 */
+		if (max_physical_apicid > 255 || !kvm_para_available())
+			goto nox2apic;
+		/*
+		 * without IR all CPUs can be addressed by IOAPIC/MSI
+		 * only in physical mode
+		 */
+		x2apic_force_phys();
+	}
 
-	pr_info("Enabled Interrupt-remapping\n");
+	x2apic_enabled = 1;
 
 	if (x2apic_supported() && !x2apic_mode) {
 		x2apic_mode = 1;
@@ -1414,41 +1439,25 @@ void __init enable_IR_x2apic(void)
 		pr_info("Enabled x2apic\n");
 	}
 
-end_restore:
-	if (ret)
-		/*
-		 * IR enabling failed
-		 */
+nox2apic:
+	if (!ret) /* IR enabling failed */
 		restore_IO_APIC_setup(ioapic_entries);
-
 	unmask_8259A();
 	local_irq_restore(flags);
 
-end:
+out:
 	if (ioapic_entries)
 		free_ioapic_entries(ioapic_entries);
 
-	if (!ret)
+	if (x2apic_enabled)
 		return;
 
-ir_failed:
 	if (x2apic_preenabled)
-		panic("x2apic enabled by bios. But IR enabling failed");
+		panic("x2apic: enabled by BIOS but kernel init failed.");
 	else if (cpu_has_x2apic)
-		pr_info("Not enabling x2apic,Intr-remapping\n");
-#else
-	if (!cpu_has_x2apic)
-		return;
-
-	if (x2apic_preenabled)
-		panic("x2apic enabled prior OS handover,"
-		      " enable CONFIG_X86_X2APIC, CONFIG_INTR_REMAP");
-#endif
-
-	return;
+		pr_info("Not enabling x2apic, Intr-remapping init failed.\n");
 }
 
-
 #ifdef CONFIG_X86_64
 /*
  * Detect and enable local APICs on non-SMP boards.
diff --git a/arch/x86/kernel/apic/probe_64.c b/arch/x86/kernel/apic/probe_64.c
index bc3e880f9b82..f3b1037076e4 100644
--- a/arch/x86/kernel/apic/probe_64.c
+++ b/arch/x86/kernel/apic/probe_64.c
@@ -50,11 +50,11 @@ static struct apic *apic_probe[] __initdata = {
 void __init default_setup_apic_routing(void)
 {
 #ifdef CONFIG_X86_X2APIC
-	if (x2apic_mode && (apic != &apic_x2apic_phys &&
+	if (x2apic_mode
 #ifdef CONFIG_X86_UV
-		       apic != &apic_x2apic_uv_x &&
+		       && apic != &apic_x2apic_uv_x
 #endif
-		       apic != &apic_x2apic_cluster)) {
+		       ) {
 		if (x2apic_phys)
 			apic = &apic_x2apic_phys;
 		else
-- 
cgit v1.2.3


From 07868b086cca784f4b532fc2ab574ec3a73b468a Mon Sep 17 00:00:00 2001
From: Frederic Weisbecker <fweisbec@gmail.com>
Date: Wed, 29 Jul 2009 03:33:51 +0200
Subject: tracing/function-graph-tracer: Drop the useless nmi protection

The function graph tracer used to have a protection against NMI
while entering a function entry tracing. But this is useless now,
this tracer is reentrant and the ring buffer supports the NMI tracing.
We can then drop this protection.

Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
---
 arch/x86/kernel/ftrace.c | 4 ----
 1 file changed, 4 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c
index d94e1ea3b9fe..8e9663413b7f 100644
--- a/arch/x86/kernel/ftrace.c
+++ b/arch/x86/kernel/ftrace.c
@@ -417,10 +417,6 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr,
 	unsigned long return_hooker = (unsigned long)
 				&return_to_handler;
 
-	/* Nmi's are currently unsupported */
-	if (unlikely(in_nmi()))
-		return;
-
 	if (unlikely(atomic_read(&current->tracing_graph_pause)))
 		return;
 
-- 
cgit v1.2.3


From f3d1915a8623b9248572d3ee44e19a80b7a3520b Mon Sep 17 00:00:00 2001
From: Cyrill Gorcunov <gorcunov@gmail.com>
Date: Thu, 6 Aug 2009 00:09:31 +0400
Subject: x86, ioapic: Panic on irq-pin binding only if needed

Though the most time we are to panic on irq-pin allocation
fails, for PCI interrupts it's not the case and we could
continue operate even if irq-pin allocation failed.

Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <20090805200931.GB5319@lenovo>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic/io_apic.c | 23 ++++++++++++++++++-----
 1 file changed, 18 insertions(+), 5 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 2a145d3a8375..2999f3dd5889 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -490,7 +490,8 @@ static void ioapic_mask_entry(int apic, int pin)
  * shared ISA-space IRQs, so we have to support them. We are super
  * fast in the common case, and fast for shared ISA-space IRQs.
  */
-static void add_pin_to_irq_node(struct irq_cfg *cfg, int node, int apic, int pin)
+static int
+add_pin_to_irq_node_nopanic(struct irq_cfg *cfg, int node, int apic, int pin)
 {
 	struct irq_pin_list **last, *entry;
 
@@ -498,19 +499,27 @@ static void add_pin_to_irq_node(struct irq_cfg *cfg, int node, int apic, int pin
 	last = &cfg->irq_2_pin;
 	for_each_irq_pin(entry, cfg->irq_2_pin) {
 		if (entry->apic == apic && entry->pin == pin)
-			return;
+			return 0;
 		last = &entry->next;
 	}
 
 	entry = get_one_free_irq_2_pin(node);
 	if (!entry) {
-		printk(KERN_ERR "can not alloc irq_pin_list\n");
-		BUG_ON(1);
+		printk(KERN_ERR "can not alloc irq_pin_list (%d,%d,%d)\n",
+				node, apic, pin);
+		return -ENOMEM;
 	}
 	entry->apic = apic;
 	entry->pin = pin;
 
 	*last = entry;
+	return 0;
+}
+
+static void add_pin_to_irq_node(struct irq_cfg *cfg, int node, int apic, int pin)
+{
+	if (add_pin_to_irq_node_nopanic(cfg, node, apic, pin))
+		panic("IO-APIC: failed to add irq-pin. Can not proceed\n");
 }
 
 /*
@@ -3843,7 +3852,11 @@ static int __io_apic_set_pci_routing(struct device *dev, int irq,
 	 */
 	if (irq >= NR_IRQS_LEGACY) {
 		cfg = desc->chip_data;
-		add_pin_to_irq_node(cfg, node, ioapic, pin);
+		if (add_pin_to_irq_node_nopanic(cfg, node, ioapic, pin)) {
+			printk(KERN_INFO "can not add pin %d for irq %d\n",
+				pin, irq);
+			return 0;
+		}
 	}
 
 	setup_IO_APIC_irq(ioapic, pin, irq, desc, trigger, polarity);
-- 
cgit v1.2.3


From 1e5de18278e6862f4198412b5059a03770fa816a Mon Sep 17 00:00:00 2001
From: Akinobu Mita <akinobu.mita@gmail.com>
Date: Sun, 19 Jul 2009 00:12:20 +0900
Subject: x86: Introduce GDT_ENTRY_INIT()

GDT_ENTRY_INIT is static initializer of desc_struct.

We already have similar macro GDT_ENTRY() but it's static
initializer for u64 and it cannot be used for desc_struct.

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
LKML-Reference: <20090718151219.GD11294@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/include/asm/desc_defs.h      |  6 ++++++
 arch/x86/include/asm/lguest.h         |  5 +++--
 arch/x86/include/asm/stackprotector.h |  2 +-
 arch/x86/kernel/apm_32.c              |  2 +-
 arch/x86/kernel/cpu/common.c          | 40 +++++++++++++++++------------------
 drivers/pnp/pnpbios/bioscalls.c       |  5 +----
 6 files changed, 32 insertions(+), 28 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/desc_defs.h b/arch/x86/include/asm/desc_defs.h
index a6adefa28b94..9d6684849fd9 100644
--- a/arch/x86/include/asm/desc_defs.h
+++ b/arch/x86/include/asm/desc_defs.h
@@ -34,6 +34,12 @@ struct desc_struct {
 	};
 } __attribute__((packed));
 
+#define GDT_ENTRY_INIT(flags, base, limit) { { { \
+		.a = ((limit) & 0xffff) | (((base) & 0xffff) << 16), \
+		.b = (((base) & 0xff0000) >> 16) | (((flags) & 0xf0ff) << 8) | \
+			((limit) & 0xf0000) | ((base) & 0xff000000), \
+	} } }
+
 enum {
 	GATE_INTERRUPT = 0xE,
 	GATE_TRAP = 0xF,
diff --git a/arch/x86/include/asm/lguest.h b/arch/x86/include/asm/lguest.h
index 313389cd50d2..94cd69858b14 100644
--- a/arch/x86/include/asm/lguest.h
+++ b/arch/x86/include/asm/lguest.h
@@ -91,8 +91,9 @@ static inline void lguest_set_ts(void)
 }
 
 /* Full 4G segment descriptors, suitable for CS and DS. */
-#define FULL_EXEC_SEGMENT ((struct desc_struct){ { {0x0000ffff, 0x00cf9b00} } })
-#define FULL_SEGMENT ((struct desc_struct){ { {0x0000ffff, 0x00cf9300} } })
+#define FULL_EXEC_SEGMENT \
+	((struct desc_struct)GDT_ENTRY_INIT(0xc09b, 0, 0xfffff))
+#define FULL_SEGMENT ((struct desc_struct)GDT_ENTRY_INIT(0xc093, 0, 0xfffff))
 
 #endif /* __ASSEMBLY__ */
 
diff --git a/arch/x86/include/asm/stackprotector.h b/arch/x86/include/asm/stackprotector.h
index cdc5e0b126a7..44efdff3975d 100644
--- a/arch/x86/include/asm/stackprotector.h
+++ b/arch/x86/include/asm/stackprotector.h
@@ -48,7 +48,7 @@
  * head_32 for boot CPU and setup_per_cpu_areas() for others.
  */
 #define GDT_STACK_CANARY_INIT						\
-	[GDT_ENTRY_STACK_CANARY] = { { { 0x00000018, 0x00409000 } } },
+	[GDT_ENTRY_STACK_CANARY] = GDT_ENTRY_INIT(0x4090, 0, 0x18),
 
 /*
  * Initialize the stackprotector canary value.
diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c
index b5e841bd60d9..febb2dab254f 100644
--- a/arch/x86/kernel/apm_32.c
+++ b/arch/x86/kernel/apm_32.c
@@ -403,7 +403,7 @@ static DECLARE_WAIT_QUEUE_HEAD(apm_waitqueue);
 static DECLARE_WAIT_QUEUE_HEAD(apm_suspend_waitqueue);
 static struct apm_user *user_list;
 static DEFINE_SPINLOCK(user_list_lock);
-static struct desc_struct bad_bios_desc = { { { 0, 0x00409200 } } };
+static struct desc_struct bad_bios_desc = GDT_ENTRY_INIT(0x4092, 0, 0);
 
 static const char driver_version[] = "1.16ac";	/* no spaces */
 
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index f1961c07af9a..8c9bc287f8fb 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -71,45 +71,45 @@ DEFINE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page) = { .gdt = {
 	 * TLS descriptors are currently at a different place compared to i386.
 	 * Hopefully nobody expects them at a fixed place (Wine?)
 	 */
-	[GDT_ENTRY_KERNEL32_CS]		= { { { 0x0000ffff, 0x00cf9b00 } } },
-	[GDT_ENTRY_KERNEL_CS]		= { { { 0x0000ffff, 0x00af9b00 } } },
-	[GDT_ENTRY_KERNEL_DS]		= { { { 0x0000ffff, 0x00cf9300 } } },
-	[GDT_ENTRY_DEFAULT_USER32_CS]	= { { { 0x0000ffff, 0x00cffb00 } } },
-	[GDT_ENTRY_DEFAULT_USER_DS]	= { { { 0x0000ffff, 0x00cff300 } } },
-	[GDT_ENTRY_DEFAULT_USER_CS]	= { { { 0x0000ffff, 0x00affb00 } } },
+	[GDT_ENTRY_KERNEL32_CS]		= GDT_ENTRY_INIT(0xc09b, 0, 0xfffff),
+	[GDT_ENTRY_KERNEL_CS]		= GDT_ENTRY_INIT(0xa09b, 0, 0xfffff),
+	[GDT_ENTRY_KERNEL_DS]		= GDT_ENTRY_INIT(0xc093, 0, 0xfffff),
+	[GDT_ENTRY_DEFAULT_USER32_CS]	= GDT_ENTRY_INIT(0xc0fb, 0, 0xfffff),
+	[GDT_ENTRY_DEFAULT_USER_DS]	= GDT_ENTRY_INIT(0xc0f3, 0, 0xfffff),
+	[GDT_ENTRY_DEFAULT_USER_CS]	= GDT_ENTRY_INIT(0xa0fb, 0, 0xfffff),
 #else
-	[GDT_ENTRY_KERNEL_CS]		= { { { 0x0000ffff, 0x00cf9a00 } } },
-	[GDT_ENTRY_KERNEL_DS]		= { { { 0x0000ffff, 0x00cf9200 } } },
-	[GDT_ENTRY_DEFAULT_USER_CS]	= { { { 0x0000ffff, 0x00cffa00 } } },
-	[GDT_ENTRY_DEFAULT_USER_DS]	= { { { 0x0000ffff, 0x00cff200 } } },
+	[GDT_ENTRY_KERNEL_CS]		= GDT_ENTRY_INIT(0xc09a, 0, 0xfffff),
+	[GDT_ENTRY_KERNEL_DS]		= GDT_ENTRY_INIT(0xc092, 0, 0xfffff),
+	[GDT_ENTRY_DEFAULT_USER_CS]	= GDT_ENTRY_INIT(0xc0fa, 0, 0xfffff),
+	[GDT_ENTRY_DEFAULT_USER_DS]	= GDT_ENTRY_INIT(0xc0f2, 0, 0xfffff),
 	/*
 	 * Segments used for calling PnP BIOS have byte granularity.
 	 * They code segments and data segments have fixed 64k limits,
 	 * the transfer segment sizes are set at run time.
 	 */
 	/* 32-bit code */
-	[GDT_ENTRY_PNPBIOS_CS32]	= { { { 0x0000ffff, 0x00409a00 } } },
+	[GDT_ENTRY_PNPBIOS_CS32]	= GDT_ENTRY_INIT(0x409a, 0, 0xffff),
 	/* 16-bit code */
-	[GDT_ENTRY_PNPBIOS_CS16]	= { { { 0x0000ffff, 0x00009a00 } } },
+	[GDT_ENTRY_PNPBIOS_CS16]	= GDT_ENTRY_INIT(0x009a, 0, 0xffff),
 	/* 16-bit data */
-	[GDT_ENTRY_PNPBIOS_DS]		= { { { 0x0000ffff, 0x00009200 } } },
+	[GDT_ENTRY_PNPBIOS_DS]		= GDT_ENTRY_INIT(0x0092, 0, 0xffff),
 	/* 16-bit data */
-	[GDT_ENTRY_PNPBIOS_TS1]		= { { { 0x00000000, 0x00009200 } } },
+	[GDT_ENTRY_PNPBIOS_TS1]		= GDT_ENTRY_INIT(0x0092, 0, 0),
 	/* 16-bit data */
-	[GDT_ENTRY_PNPBIOS_TS2]		= { { { 0x00000000, 0x00009200 } } },
+	[GDT_ENTRY_PNPBIOS_TS2]		= GDT_ENTRY_INIT(0x0092, 0, 0),
 	/*
 	 * The APM segments have byte granularity and their bases
 	 * are set at run time.  All have 64k limits.
 	 */
 	/* 32-bit code */
-	[GDT_ENTRY_APMBIOS_BASE]	= { { { 0x0000ffff, 0x00409a00 } } },
+	[GDT_ENTRY_APMBIOS_BASE]	= GDT_ENTRY_INIT(0x409a, 0, 0xffff),
 	/* 16-bit code */
-	[GDT_ENTRY_APMBIOS_BASE+1]	= { { { 0x0000ffff, 0x00009a00 } } },
+	[GDT_ENTRY_APMBIOS_BASE+1]	= GDT_ENTRY_INIT(0x009a, 0, 0xffff),
 	/* data */
-	[GDT_ENTRY_APMBIOS_BASE+2]	= { { { 0x0000ffff, 0x00409200 } } },
+	[GDT_ENTRY_APMBIOS_BASE+2]	= GDT_ENTRY_INIT(0x409a, 0, 0xffff),
 
-	[GDT_ENTRY_ESPFIX_SS]		= { { { 0x0000ffff, 0x00cf9200 } } },
-	[GDT_ENTRY_PERCPU]		= { { { 0x0000ffff, 0x00cf9200 } } },
+	[GDT_ENTRY_ESPFIX_SS]		= GDT_ENTRY_INIT(0xc092, 0, 0xfffff),
+	[GDT_ENTRY_PERCPU]		= GDT_ENTRY_INIT(0xc092, 0, 0xfffff),
 	GDT_STACK_CANARY_INIT
 #endif
 } };
diff --git a/drivers/pnp/pnpbios/bioscalls.c b/drivers/pnp/pnpbios/bioscalls.c
index 45ad3e9cc362..bd035e3d3550 100644
--- a/drivers/pnp/pnpbios/bioscalls.c
+++ b/drivers/pnp/pnpbios/bioscalls.c
@@ -60,7 +60,7 @@ do { \
 	set_desc_limit(&gdt[(selname) >> 3], (size) - 1); \
 } while(0)
 
-static struct desc_struct bad_bios_desc;
+static struct desc_struct bad_bios_desc = GDT_ENTRY_INIT(0x4092, 0, 0);
 
 /*
  * At some point we want to use this stack frame pointer to unwind
@@ -476,9 +476,6 @@ void pnpbios_calls_init(union pnp_bios_install_struct *header)
 	pnp_bios_callpoint.offset = header->fields.pm16offset;
 	pnp_bios_callpoint.segment = PNP_CS16;
 
-	bad_bios_desc.a = 0;
-	bad_bios_desc.b = 0x00409200;
-
 	set_desc_base(&bad_bios_desc, (unsigned long)__va(0x40UL << 4));
 	set_desc_limit(&bad_bios_desc, 4095 - (0x40 << 4));
 	for_each_possible_cpu(i) {
-- 
cgit v1.2.3


From 72c4d8530244264317a662de9a55cc47e6c8e9df Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 3 Aug 2009 08:47:07 +0200
Subject: x86: Introduce GDT_ENTRY_INIT(), fix APM

This crash:

[    0.891983] calling  cache_sysfs_init+0x0/0x1ee @ 1
[    0.897251] initcall cache_sysfs_init+0x0/0x1ee returned 0 after 405 usecs
[    0.904019] calling  mce_init_device+0x0/0x242 @ 1
[    0.909124] initcall mce_init_device+0x0/0x242 returned 0 after 347 usecs
[    0.915815] calling  apm_init+0x0/0x38d @ 1
[    0.919967] apm: BIOS version 1.2 Flags 0x07 (Driver version 1.16ac)
[    0.926813] general protection fault: 0000 [#1]
[    0.927269] last sysfs file:
[    0.927269] Modules linked in:
[    0.927269]
[    0.927269] Pid: 271, comm: kapmd Not tainted (2.6.31-rc3-00100-gd520da1-dirty #311) System Product Name
[    0.927269] EIP: 00c0:[<000082b2>] EFLAGS: 00010002 CPU: 0
[    0.927269] EIP is at 0x82b2
[    0.927269] EAX: 0000530e EBX: 00000000 ECX: 00000102 EDX: 00000000
[    0.927269] ESI: 00000000 EDI: f6a4bf44 EBP: 67890000 ESP: f6a4beec
[    0.927269]  DS: 00c8 ES: 0000 FS: 0000 GS: 0000 SS: 0068
[    0.927269] Process kapmd (pid: 271, ti=f6a4a000 task=f7142280 task.ti=f6a4a000)
[    0.927269] Stack:
[    0.927269]  0000828d 02160000 00b88092 f6a4bf3c c102a63d 00000060 f6a4bf3c f6a4bf44
[    0.927269] <0> 0000007b 0000007b 00000000 00000000 00000000 00000000 560aae9e 00000000
[    0.927269] <0> 00000200 f705fd74 00000000 c102af70 f6a4bf60 c102a6ec 0000530e 00000000
[    0.927269] Call Trace:
[    0.927269]  [<c102a63d>] ? __apm_bios_call_simple+0x7d/0x110
[    0.927269]  [<c102af70>] ? apm+0x0/0x6a0
[    0.927269]  [<c102a6ec>] ? apm_bios_call_simple+0x1c/0x50
[    0.927269]  [<c102b3f5>] ? apm+0x485/0x6a0
[    0.927269]  [<c1038e7a>] ? finish_task_switch+0x2a/0xb0
[    0.927269]  [<c164a69e>] ? schedule+0x31e/0x480
[    0.927269]  [<c102af70>] ? apm+0x0/0x6a0
[    0.927269]  [<c102af70>] ? apm+0x0/0x6a0
[    0.927269]  [<c1052654>] ? kthread+0x74/0x80
[    0.927269]  [<c10525e0>] ? kthread+0x0/0x80
[    0.927269]  [<c101d627>] ? kernel_thread_helper+0x7/0x10
[    0.927269] Code:  Bad EIP value.
[    0.927269] EIP: [<000082b2>] 0x82b2 SS:ESP 0068:f6a4beec
[    0.927269] ---[ end trace a7919e7f17c0a725 ]---
[    0.927269] Kernel panic - not syncing: Fatal exception
[    0.927269] Pid: 271, comm: kapmd Tainted: G      D    2.6.31-rc3-00100-gd520da1-dirty #311

Is caused by an incorrect GDT_ENTRY_INIT() conversion in the apm
code, as noticed by hpa.

Reported-by: Ingo Molnar <mingo@elte.hu>
Noticed-by: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
LKML-Reference: <20090808094905.GA2954@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/common.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 8c9bc287f8fb..b5764a269645 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -106,7 +106,7 @@ DEFINE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page) = { .gdt = {
 	/* 16-bit code */
 	[GDT_ENTRY_APMBIOS_BASE+1]	= GDT_ENTRY_INIT(0x009a, 0, 0xffff),
 	/* data */
-	[GDT_ENTRY_APMBIOS_BASE+2]	= GDT_ENTRY_INIT(0x409a, 0, 0xffff),
+	[GDT_ENTRY_APMBIOS_BASE+2]	= GDT_ENTRY_INIT(0x4092, 0, 0xffff),
 
 	[GDT_ENTRY_ESPFIX_SS]		= GDT_ENTRY_INIT(0xc092, 0, 0xfffff),
 	[GDT_ENTRY_PERCPU]		= GDT_ENTRY_INIT(0xc092, 0, 0xfffff),
-- 
cgit v1.2.3


From 4c711576b90cc36c13b94816a953a8de6a53d03c Mon Sep 17 00:00:00 2001
From: Amerigo Wang <amwang@redhat.com>
Date: Thu, 6 Aug 2009 15:58:12 -0700
Subject: x86, 32-bit: Use generic sys_pipe()

As suggested by Al, it's better to use the generic sys_pipe() for ia32.

Signed-off-by: WANG Cong <amwang@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/ia32/ia32entry.S |  2 +-
 arch/x86/ia32/sys_ia32.c  | 14 --------------
 2 files changed, 1 insertion(+), 15 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S
index e590261ba059..ba331bfd1112 100644
--- a/arch/x86/ia32/ia32entry.S
+++ b/arch/x86/ia32/ia32entry.S
@@ -537,7 +537,7 @@ ia32_sys_call_table:
 	.quad sys_mkdir
 	.quad sys_rmdir		/* 40 */
 	.quad sys_dup
-	.quad sys32_pipe
+	.quad sys_pipe
 	.quad compat_sys_times
 	.quad quiet_ni_syscall			/* old prof syscall holder */
 	.quad sys_brk		/* 45 */
diff --git a/arch/x86/ia32/sys_ia32.c b/arch/x86/ia32/sys_ia32.c
index 085a8c35f149..9f5527198825 100644
--- a/arch/x86/ia32/sys_ia32.c
+++ b/arch/x86/ia32/sys_ia32.c
@@ -189,20 +189,6 @@ asmlinkage long sys32_mprotect(unsigned long start, size_t len,
 	return sys_mprotect(start, len, prot);
 }
 
-asmlinkage long sys32_pipe(int __user *fd)
-{
-	int retval;
-	int fds[2];
-
-	retval = do_pipe_flags(fds, 0);
-	if (retval)
-		goto out;
-	if (copy_to_user(fd, fds, sizeof(fds)))
-		retval = -EFAULT;
-out:
-	return retval;
-}
-
 asmlinkage long sys32_rt_sigaction(int sig, struct sigaction32 __user *act,
 				   struct sigaction32 __user *oact,
 				   unsigned int sigsetsize)
-- 
cgit v1.2.3


From 30dd568c912602b7dbd609a45d053e01b13422bb Mon Sep 17 00:00:00 2001
From: Markus Metzger <markus.t.metzger@intel.com>
Date: Tue, 21 Jul 2009 15:56:48 +0200
Subject: x86, perf_counter, bts: Add BTS support to perfcounters

Implement a performance counter with:

    attr.type           = PERF_TYPE_HARDWARE
    attr.config         = PERF_COUNT_HW_BRANCH_INSTRUCTIONS
    attr.sample_period  = 1

Using branch trace store (BTS) on x86 hardware, if available.

The from and to address for each branch can be sampled using:

    PERF_SAMPLE_IP      for the from address
    PERF_SAMPLE_ADDR    for the to address

[ v2: address review feedback, fix bugs ]

Signed-off-by: Markus Metzger <markus.t.metzger@intel.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/include/asm/perf_counter.h |  10 ++
 arch/x86/kernel/cpu/perf_counter.c  | 325 +++++++++++++++++++++++++++++++++++-
 include/linux/perf_counter.h        |   2 +
 kernel/perf_counter.c               |  10 +-
 4 files changed, 340 insertions(+), 7 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/perf_counter.h b/arch/x86/include/asm/perf_counter.h
index fa64e401589d..e7b7c938ae27 100644
--- a/arch/x86/include/asm/perf_counter.h
+++ b/arch/x86/include/asm/perf_counter.h
@@ -84,6 +84,16 @@ union cpuid10_edx {
 #define MSR_ARCH_PERFMON_FIXED_CTR2			0x30b
 #define X86_PMC_IDX_FIXED_BUS_CYCLES			(X86_PMC_IDX_FIXED + 2)
 
+/*
+ * We model BTS tracing as another fixed-mode PMC.
+ *
+ * We choose a value in the middle of the fixed counter range, since lower
+ * values are used by actual fixed counters and higher values are used
+ * to indicate other overflow conditions in the PERF_GLOBAL_STATUS msr.
+ */
+#define X86_PMC_IDX_FIXED_BTS				(X86_PMC_IDX_FIXED + 16)
+
+
 #ifdef CONFIG_PERF_COUNTERS
 extern void init_hw_perf_counters(void);
 extern void perf_counters_lapic_init(void);
diff --git a/arch/x86/kernel/cpu/perf_counter.c b/arch/x86/kernel/cpu/perf_counter.c
index a7aa8f900954..b237c181aa41 100644
--- a/arch/x86/kernel/cpu/perf_counter.c
+++ b/arch/x86/kernel/cpu/perf_counter.c
@@ -6,6 +6,7 @@
  *  Copyright (C) 2009 Jaswinder Singh Rajput
  *  Copyright (C) 2009 Advanced Micro Devices, Inc., Robert Richter
  *  Copyright (C) 2008-2009 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com>
+ *  Copyright (C) 2009 Intel Corporation, <markus.t.metzger@intel.com>
  *
  *  For licencing details see kernel-base/COPYING
  */
@@ -20,6 +21,7 @@
 #include <linux/sched.h>
 #include <linux/uaccess.h>
 #include <linux/highmem.h>
+#include <linux/cpu.h>
 
 #include <asm/apic.h>
 #include <asm/stacktrace.h>
@@ -27,12 +29,52 @@
 
 static u64 perf_counter_mask __read_mostly;
 
+/* The maximal number of PEBS counters: */
+#define MAX_PEBS_COUNTERS	4
+
+/* The size of a BTS record in bytes: */
+#define BTS_RECORD_SIZE		24
+
+/* The size of a per-cpu BTS buffer in bytes: */
+#define BTS_BUFFER_SIZE		(BTS_RECORD_SIZE * 1024)
+
+/* The BTS overflow threshold in bytes from the end of the buffer: */
+#define BTS_OVFL_TH		(BTS_RECORD_SIZE * 64)
+
+
+/*
+ * Bits in the debugctlmsr controlling branch tracing.
+ */
+#define X86_DEBUGCTL_TR			(1 << 6)
+#define X86_DEBUGCTL_BTS		(1 << 7)
+#define X86_DEBUGCTL_BTINT		(1 << 8)
+#define X86_DEBUGCTL_BTS_OFF_OS		(1 << 9)
+#define X86_DEBUGCTL_BTS_OFF_USR	(1 << 10)
+
+/*
+ * A debug store configuration.
+ *
+ * We only support architectures that use 64bit fields.
+ */
+struct debug_store {
+	u64	bts_buffer_base;
+	u64	bts_index;
+	u64	bts_absolute_maximum;
+	u64	bts_interrupt_threshold;
+	u64	pebs_buffer_base;
+	u64	pebs_index;
+	u64	pebs_absolute_maximum;
+	u64	pebs_interrupt_threshold;
+	u64	pebs_counter_reset[MAX_PEBS_COUNTERS];
+};
+
 struct cpu_hw_counters {
 	struct perf_counter	*counters[X86_PMC_IDX_MAX];
 	unsigned long		used_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
 	unsigned long		active_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
 	unsigned long		interrupts;
 	int			enabled;
+	struct debug_store	*ds;
 };
 
 /*
@@ -57,6 +99,8 @@ struct x86_pmu {
 	u64		counter_mask;
 	u64		max_period;
 	u64		intel_ctrl;
+	void		(*enable_bts)(u64 config);
+	void		(*disable_bts)(void);
 };
 
 static struct x86_pmu x86_pmu __read_mostly;
@@ -576,6 +620,9 @@ x86_perf_counter_update(struct perf_counter *counter,
 	u64 prev_raw_count, new_raw_count;
 	s64 delta;
 
+	if (idx == X86_PMC_IDX_FIXED_BTS)
+		return 0;
+
 	/*
 	 * Careful: an NMI might modify the previous counter value.
 	 *
@@ -659,10 +706,109 @@ static void release_pmc_hardware(void)
 		enable_lapic_nmi_watchdog();
 }
 
+static inline bool bts_available(void)
+{
+	return x86_pmu.enable_bts != NULL;
+}
+
+static inline void init_debug_store_on_cpu(int cpu)
+{
+	struct debug_store *ds = per_cpu(cpu_hw_counters, cpu).ds;
+
+	if (!ds)
+		return;
+
+	wrmsr_on_cpu(cpu, MSR_IA32_DS_AREA,
+		     (u32)((u64)(long)ds), (u32)((u64)(long)ds >> 32));
+}
+
+static inline void fini_debug_store_on_cpu(int cpu)
+{
+	if (!per_cpu(cpu_hw_counters, cpu).ds)
+		return;
+
+	wrmsr_on_cpu(cpu, MSR_IA32_DS_AREA, 0, 0);
+}
+
+static void release_bts_hardware(void)
+{
+	int cpu;
+
+	if (!bts_available())
+		return;
+
+	get_online_cpus();
+
+	for_each_online_cpu(cpu)
+		fini_debug_store_on_cpu(cpu);
+
+	for_each_possible_cpu(cpu) {
+		struct debug_store *ds = per_cpu(cpu_hw_counters, cpu).ds;
+
+		if (!ds)
+			continue;
+
+		per_cpu(cpu_hw_counters, cpu).ds = NULL;
+
+		kfree((void *)(long)ds->bts_buffer_base);
+		kfree(ds);
+	}
+
+	put_online_cpus();
+}
+
+static int reserve_bts_hardware(void)
+{
+	int cpu, err = 0;
+
+	if (!bts_available())
+		return -EOPNOTSUPP;
+
+	get_online_cpus();
+
+	for_each_possible_cpu(cpu) {
+		struct debug_store *ds;
+		void *buffer;
+
+		err = -ENOMEM;
+		buffer = kzalloc(BTS_BUFFER_SIZE, GFP_KERNEL);
+		if (unlikely(!buffer))
+			break;
+
+		ds = kzalloc(sizeof(*ds), GFP_KERNEL);
+		if (unlikely(!ds)) {
+			kfree(buffer);
+			break;
+		}
+
+		ds->bts_buffer_base = (u64)(long)buffer;
+		ds->bts_index = ds->bts_buffer_base;
+		ds->bts_absolute_maximum =
+			ds->bts_buffer_base + BTS_BUFFER_SIZE;
+		ds->bts_interrupt_threshold =
+			ds->bts_absolute_maximum - BTS_OVFL_TH;
+
+		per_cpu(cpu_hw_counters, cpu).ds = ds;
+		err = 0;
+	}
+
+	if (err)
+		release_bts_hardware();
+	else {
+		for_each_online_cpu(cpu)
+			init_debug_store_on_cpu(cpu);
+	}
+
+	put_online_cpus();
+
+	return err;
+}
+
 static void hw_perf_counter_destroy(struct perf_counter *counter)
 {
 	if (atomic_dec_and_mutex_lock(&active_counters, &pmc_reserve_mutex)) {
 		release_pmc_hardware();
+		release_bts_hardware();
 		mutex_unlock(&pmc_reserve_mutex);
 	}
 }
@@ -705,6 +851,42 @@ set_ext_hw_attr(struct hw_perf_counter *hwc, struct perf_counter_attr *attr)
 	return 0;
 }
 
+static void intel_pmu_enable_bts(u64 config)
+{
+	unsigned long debugctlmsr;
+
+	debugctlmsr = get_debugctlmsr();
+
+	debugctlmsr |= X86_DEBUGCTL_TR;
+	debugctlmsr |= X86_DEBUGCTL_BTS;
+	debugctlmsr |= X86_DEBUGCTL_BTINT;
+
+	if (!(config & ARCH_PERFMON_EVENTSEL_OS))
+		debugctlmsr |= X86_DEBUGCTL_BTS_OFF_OS;
+
+	if (!(config & ARCH_PERFMON_EVENTSEL_USR))
+		debugctlmsr |= X86_DEBUGCTL_BTS_OFF_USR;
+
+	update_debugctlmsr(debugctlmsr);
+}
+
+static void intel_pmu_disable_bts(void)
+{
+	struct cpu_hw_counters *cpuc = &__get_cpu_var(cpu_hw_counters);
+	unsigned long debugctlmsr;
+
+	if (!cpuc->ds)
+		return;
+
+	debugctlmsr = get_debugctlmsr();
+
+	debugctlmsr &=
+		~(X86_DEBUGCTL_TR | X86_DEBUGCTL_BTS | X86_DEBUGCTL_BTINT |
+		  X86_DEBUGCTL_BTS_OFF_OS | X86_DEBUGCTL_BTS_OFF_USR);
+
+	update_debugctlmsr(debugctlmsr);
+}
+
 /*
  * Setup the hardware configuration for a given attr_type
  */
@@ -721,9 +903,13 @@ static int __hw_perf_counter_init(struct perf_counter *counter)
 	err = 0;
 	if (!atomic_inc_not_zero(&active_counters)) {
 		mutex_lock(&pmc_reserve_mutex);
-		if (atomic_read(&active_counters) == 0 && !reserve_pmc_hardware())
-			err = -EBUSY;
-		else
+		if (atomic_read(&active_counters) == 0) {
+			if (!reserve_pmc_hardware())
+				err = -EBUSY;
+			else
+				reserve_bts_hardware();
+		}
+		if (!err)
 			atomic_inc(&active_counters);
 		mutex_unlock(&pmc_reserve_mutex);
 	}
@@ -801,7 +987,18 @@ static void p6_pmu_disable_all(void)
 
 static void intel_pmu_disable_all(void)
 {
+	struct cpu_hw_counters *cpuc = &__get_cpu_var(cpu_hw_counters);
+
+	if (!cpuc->enabled)
+		return;
+
+	cpuc->enabled = 0;
+	barrier();
+
 	wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0);
+
+	if (test_bit(X86_PMC_IDX_FIXED_BTS, cpuc->active_mask))
+		intel_pmu_disable_bts();
 }
 
 static void amd_pmu_disable_all(void)
@@ -859,7 +1056,25 @@ static void p6_pmu_enable_all(void)
 
 static void intel_pmu_enable_all(void)
 {
+	struct cpu_hw_counters *cpuc = &__get_cpu_var(cpu_hw_counters);
+
+	if (cpuc->enabled)
+		return;
+
+	cpuc->enabled = 1;
+	barrier();
+
 	wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, x86_pmu.intel_ctrl);
+
+	if (test_bit(X86_PMC_IDX_FIXED_BTS, cpuc->active_mask)) {
+		struct perf_counter *counter =
+			cpuc->counters[X86_PMC_IDX_FIXED_BTS];
+
+		if (WARN_ON_ONCE(!counter))
+			return;
+
+		intel_pmu_enable_bts(counter->hw.config);
+	}
 }
 
 static void amd_pmu_enable_all(void)
@@ -946,6 +1161,11 @@ p6_pmu_disable_counter(struct hw_perf_counter *hwc, int idx)
 static inline void
 intel_pmu_disable_counter(struct hw_perf_counter *hwc, int idx)
 {
+	if (unlikely(idx == X86_PMC_IDX_FIXED_BTS)) {
+		intel_pmu_disable_bts();
+		return;
+	}
+
 	if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) {
 		intel_pmu_disable_fixed(hwc, idx);
 		return;
@@ -974,6 +1194,9 @@ x86_perf_counter_set_period(struct perf_counter *counter,
 	s64 period = hwc->sample_period;
 	int err, ret = 0;
 
+	if (idx == X86_PMC_IDX_FIXED_BTS)
+		return 0;
+
 	/*
 	 * If we are way outside a reasoable range then just skip forward:
 	 */
@@ -1056,6 +1279,14 @@ static void p6_pmu_enable_counter(struct hw_perf_counter *hwc, int idx)
 
 static void intel_pmu_enable_counter(struct hw_perf_counter *hwc, int idx)
 {
+	if (unlikely(idx == X86_PMC_IDX_FIXED_BTS)) {
+		if (!__get_cpu_var(cpu_hw_counters).enabled)
+			return;
+
+		intel_pmu_enable_bts(hwc->config);
+		return;
+	}
+
 	if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) {
 		intel_pmu_enable_fixed(hwc, idx);
 		return;
@@ -1077,11 +1308,16 @@ fixed_mode_idx(struct perf_counter *counter, struct hw_perf_counter *hwc)
 {
 	unsigned int event;
 
+	event = hwc->config & ARCH_PERFMON_EVENT_MASK;
+
+	if (unlikely((event ==
+		      x86_pmu.event_map(PERF_COUNT_HW_BRANCH_INSTRUCTIONS)) &&
+		     (hwc->sample_period == 1)))
+		return X86_PMC_IDX_FIXED_BTS;
+
 	if (!x86_pmu.num_counters_fixed)
 		return -1;
 
-	event = hwc->config & ARCH_PERFMON_EVENT_MASK;
-
 	if (unlikely(event == x86_pmu.event_map(PERF_COUNT_HW_INSTRUCTIONS)))
 		return X86_PMC_IDX_FIXED_INSTRUCTIONS;
 	if (unlikely(event == x86_pmu.event_map(PERF_COUNT_HW_CPU_CYCLES)))
@@ -1102,7 +1338,25 @@ static int x86_pmu_enable(struct perf_counter *counter)
 	int idx;
 
 	idx = fixed_mode_idx(counter, hwc);
-	if (idx >= 0) {
+	if (idx == X86_PMC_IDX_FIXED_BTS) {
+		/*
+		 * Try to use BTS for branch tracing. If that is not
+		 * available, try to get a generic counter.
+		 */
+		if (unlikely(!cpuc->ds))
+			goto try_generic;
+
+		/*
+		 * Try to get the fixed counter, if that is already taken
+		 * then try to get a generic counter:
+		 */
+		if (test_and_set_bit(idx, cpuc->used_mask))
+			goto try_generic;
+
+		hwc->config_base	= 0;
+		hwc->counter_base	= 0;
+		hwc->idx		= idx;
+	} else if (idx >= 0) {
 		/*
 		 * Try to get the fixed counter, if that is already taken
 		 * then try to get a generic counter:
@@ -1213,6 +1467,45 @@ void perf_counter_print_debug(void)
 	local_irq_restore(flags);
 }
 
+static void intel_pmu_drain_bts_buffer(struct cpu_hw_counters *cpuc,
+				       struct perf_sample_data *data)
+{
+	struct debug_store *ds = cpuc->ds;
+	struct bts_record {
+		u64	from;
+		u64	to;
+		u64	flags;
+	};
+	struct perf_counter *counter = cpuc->counters[X86_PMC_IDX_FIXED_BTS];
+	unsigned long orig_ip = data->regs->ip;
+	u64 at;
+
+	if (!counter)
+		return;
+
+	if (!ds)
+		return;
+
+	for (at = ds->bts_buffer_base;
+	     at < ds->bts_index;
+	     at += sizeof(struct bts_record)) {
+		struct bts_record *rec = (struct bts_record *)(long)at;
+
+		data->regs->ip	= rec->from;
+		data->addr	= rec->to;
+
+		perf_counter_output(counter, 1, data);
+	}
+
+	ds->bts_index = ds->bts_buffer_base;
+
+	data->regs->ip	= orig_ip;
+	data->addr	= 0;
+
+	/* There's new data available. */
+	counter->pending_kill = POLL_IN;
+}
+
 static void x86_pmu_disable(struct perf_counter *counter)
 {
 	struct cpu_hw_counters *cpuc = &__get_cpu_var(cpu_hw_counters);
@@ -1237,6 +1530,15 @@ static void x86_pmu_disable(struct perf_counter *counter)
 	 * that we are disabling:
 	 */
 	x86_perf_counter_update(counter, hwc, idx);
+
+	/* Drain the remaining BTS records. */
+	if (unlikely(idx == X86_PMC_IDX_FIXED_BTS)) {
+		struct perf_sample_data data;
+		struct pt_regs regs;
+
+		data.regs = &regs;
+		intel_pmu_drain_bts_buffer(cpuc, &data);
+	}
 	cpuc->counters[idx] = NULL;
 	clear_bit(idx, cpuc->used_mask);
 
@@ -1264,6 +1566,7 @@ static int intel_pmu_save_and_restart(struct perf_counter *counter)
 
 static void intel_pmu_reset(void)
 {
+	struct debug_store *ds = __get_cpu_var(cpu_hw_counters).ds;
 	unsigned long flags;
 	int idx;
 
@@ -1281,6 +1584,8 @@ static void intel_pmu_reset(void)
 	for (idx = 0; idx < x86_pmu.num_counters_fixed; idx++) {
 		checking_wrmsrl(MSR_ARCH_PERFMON_FIXED_CTR0 + idx, 0ull);
 	}
+	if (ds)
+		ds->bts_index = ds->bts_buffer_base;
 
 	local_irq_restore(flags);
 }
@@ -1346,6 +1651,7 @@ static int intel_pmu_handle_irq(struct pt_regs *regs)
 	cpuc = &__get_cpu_var(cpu_hw_counters);
 
 	perf_disable();
+	intel_pmu_drain_bts_buffer(cpuc, &data);
 	status = intel_pmu_get_status();
 	if (!status) {
 		perf_enable();
@@ -1547,6 +1853,8 @@ static struct x86_pmu intel_pmu = {
 	 * the generic counter period:
 	 */
 	.max_period		= (1ULL << 31) - 1,
+	.enable_bts		= intel_pmu_enable_bts,
+	.disable_bts		= intel_pmu_disable_bts,
 };
 
 static struct x86_pmu amd_pmu = {
@@ -1936,3 +2244,8 @@ struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
 
 	return entry;
 }
+
+void hw_perf_counter_setup_online(int cpu)
+{
+	init_debug_store_on_cpu(cpu);
+}
diff --git a/include/linux/perf_counter.h b/include/linux/perf_counter.h
index 2aabe43c1d04..0a6f3209c9de 100644
--- a/include/linux/perf_counter.h
+++ b/include/linux/perf_counter.h
@@ -692,6 +692,8 @@ struct perf_sample_data {
 
 extern int perf_counter_overflow(struct perf_counter *counter, int nmi,
 				 struct perf_sample_data *data);
+extern void perf_counter_output(struct perf_counter *counter, int nmi,
+				struct perf_sample_data *data);
 
 /*
  * Return 1 for a software counter, 0 for a hardware counter
diff --git a/kernel/perf_counter.c b/kernel/perf_counter.c
index 546e62d62941..bf8110b35c51 100644
--- a/kernel/perf_counter.c
+++ b/kernel/perf_counter.c
@@ -88,6 +88,7 @@ void __weak hw_perf_disable(void)		{ barrier(); }
 void __weak hw_perf_enable(void)		{ barrier(); }
 
 void __weak hw_perf_counter_setup(int cpu)	{ barrier(); }
+void __weak hw_perf_counter_setup_online(int cpu)	{ barrier(); }
 
 int __weak
 hw_perf_group_sched_in(struct perf_counter *group_leader,
@@ -2630,7 +2631,7 @@ static u32 perf_counter_tid(struct perf_counter *counter, struct task_struct *p)
 	return task_pid_nr_ns(p, counter->ns);
 }
 
-static void perf_counter_output(struct perf_counter *counter, int nmi,
+void perf_counter_output(struct perf_counter *counter, int nmi,
 				struct perf_sample_data *data)
 {
 	int ret;
@@ -4566,6 +4567,11 @@ perf_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu)
 		perf_counter_init_cpu(cpu);
 		break;
 
+	case CPU_ONLINE:
+	case CPU_ONLINE_FROZEN:
+		hw_perf_counter_setup_online(cpu);
+		break;
+
 	case CPU_DOWN_PREPARE:
 	case CPU_DOWN_PREPARE_FROZEN:
 		perf_counter_exit_cpu(cpu);
@@ -4590,6 +4596,8 @@ void __init perf_counter_init(void)
 {
 	perf_cpu_notify(&perf_cpu_nb, (unsigned long)CPU_UP_PREPARE,
 			(void *)(long)smp_processor_id());
+	perf_cpu_notify(&perf_cpu_nb, (unsigned long)CPU_ONLINE,
+			(void *)(long)smp_processor_id());
 	register_cpu_notifier(&perf_cpu_nb);
 }
 
-- 
cgit v1.2.3


From 9f51e24ee8b5a1595b6a5ac0c2be278a16488e75 Mon Sep 17 00:00:00 2001
From: Marcin Slusarz <marcin.slusarz@gmail.com>
Date: Sun, 9 Aug 2009 21:54:00 +0200
Subject: x86: Use printk_once()

Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
LKML-Reference: <1249847649-11631-6-git-send-email-marcin.slusarz@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/irq_32.c | 5 ++---
 arch/x86/kvm/x86.c       | 7 +------
 2 files changed, 3 insertions(+), 9 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c
index 3b09634a5153..7d35d0fe2329 100644
--- a/arch/x86/kernel/irq_32.c
+++ b/arch/x86/kernel/irq_32.c
@@ -218,7 +218,6 @@ bool handle_irq(unsigned irq, struct pt_regs *regs)
 void fixup_irqs(void)
 {
 	unsigned int irq;
-	static int warned;
 	struct irq_desc *desc;
 
 	for_each_irq_desc(irq, desc) {
@@ -236,8 +235,8 @@ void fixup_irqs(void)
 		}
 		if (desc->chip->set_affinity)
 			desc->chip->set_affinity(irq, affinity);
-		else if (desc->action && !(warned++))
-			printk("Cannot set affinity for irq %i\n", irq);
+		else if (desc->action)
+			printk_once("Cannot set affinity for irq %i\n", irq);
 	}
 
 #if 0
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index fe5474aec41a..0572c90f0c84 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -2261,12 +2261,7 @@ static int emulator_cmpxchg_emulated(unsigned long addr,
 				     unsigned int bytes,
 				     struct kvm_vcpu *vcpu)
 {
-	static int reported;
-
-	if (!reported) {
-		reported = 1;
-		printk(KERN_WARNING "kvm: emulating exchange as write\n");
-	}
+	printk_once(KERN_WARNING "kvm: emulating exchange as write\n");
 #ifndef CONFIG_X86_64
 	/* guests cmpxchg8b have to be emulated atomically */
 	if (bytes == 8) {
-- 
cgit v1.2.3


From a8ad568dd8ca122aa8048ea067d3599820d1c1b4 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 10 Aug 2009 11:53:10 +0900
Subject: dma-ops: Remove flush_write_buffers() in dma-mapping-common.h

This moves flush_write_buffers() in
asm-generic/dma-mapping-common.h to
arch/x86/kernel/pci-nommu.c.

The purpose of this patch is that, we can avoid defining NULL
flush_write_buffers() on IA64 and SPARC.

dma-mapping-common.h is used by X86 and IA64 (and SPARC soon)
but only X86 with CONFIG_X86_OOSTORE or CONFIG_X86_PPRO_FENCE
actually uses flush_write_buffers(). CONFIG_X86_OOSTORE or
CONFIG_X86_PPRO_FENCE is usable with only kernel/pci-nommu.c
(that is, not usable with other X86 IOMMU implementations such
as SWIOTLB, VT-d, etc) so we can safely move
flush_write_buffers() in asm-generic/dma-mapping-common.h to
arch/x86/kernel/pci-nommu.c.

The further discussion is:

  http://lkml.org/lkml/2009/6/28/104

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: davem@davemloft.net
Cc: tony.luck@intel.com
Cc: fenghua.yu@intel.com
LKML-Reference: <1249872797-1314-2-git-send-email-fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/pci-nommu.c              | 27 ++++++++++++++++++++++-----
 include/asm-generic/dma-mapping-common.h |  6 ------
 2 files changed, 22 insertions(+), 11 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/pci-nommu.c b/arch/x86/kernel/pci-nommu.c
index c0a4222bf62b..a3933d4330cd 100644
--- a/arch/x86/kernel/pci-nommu.c
+++ b/arch/x86/kernel/pci-nommu.c
@@ -79,12 +79,29 @@ static void nommu_free_coherent(struct device *dev, size_t size, void *vaddr,
 	free_pages((unsigned long)vaddr, get_order(size));
 }
 
+static void nommu_sync_single_for_device(struct device *dev,
+			dma_addr_t addr, size_t size,
+			enum dma_data_direction dir)
+{
+	flush_write_buffers();
+}
+
+
+static void nommu_sync_sg_for_device(struct device *dev,
+			struct scatterlist *sg, int nelems,
+			enum dma_data_direction dir)
+{
+	flush_write_buffers();
+}
+
 struct dma_map_ops nommu_dma_ops = {
-	.alloc_coherent	= dma_generic_alloc_coherent,
-	.free_coherent	= nommu_free_coherent,
-	.map_sg		= nommu_map_sg,
-	.map_page	= nommu_map_page,
-	.is_phys	= 1,
+	.alloc_coherent		= dma_generic_alloc_coherent,
+	.free_coherent		= nommu_free_coherent,
+	.map_sg			= nommu_map_sg,
+	.map_page		= nommu_map_page,
+	.sync_single_for_device = nommu_sync_single_for_device,
+	.sync_sg_for_device	= nommu_sync_sg_for_device,
+	.is_phys		= 1,
 };
 
 void __init no_iommu_init(void)
diff --git a/include/asm-generic/dma-mapping-common.h b/include/asm-generic/dma-mapping-common.h
index 5406a601185c..e694263445f7 100644
--- a/include/asm-generic/dma-mapping-common.h
+++ b/include/asm-generic/dma-mapping-common.h
@@ -103,7 +103,6 @@ static inline void dma_sync_single_for_cpu(struct device *dev, dma_addr_t addr,
 	if (ops->sync_single_for_cpu)
 		ops->sync_single_for_cpu(dev, addr, size, dir);
 	debug_dma_sync_single_for_cpu(dev, addr, size, dir);
-	flush_write_buffers();
 }
 
 static inline void dma_sync_single_for_device(struct device *dev,
@@ -116,7 +115,6 @@ static inline void dma_sync_single_for_device(struct device *dev,
 	if (ops->sync_single_for_device)
 		ops->sync_single_for_device(dev, addr, size, dir);
 	debug_dma_sync_single_for_device(dev, addr, size, dir);
-	flush_write_buffers();
 }
 
 static inline void dma_sync_single_range_for_cpu(struct device *dev,
@@ -132,7 +130,6 @@ static inline void dma_sync_single_range_for_cpu(struct device *dev,
 		ops->sync_single_range_for_cpu(dev, addr, offset, size, dir);
 		debug_dma_sync_single_range_for_cpu(dev, addr, offset, size, dir);
 
-		flush_write_buffers();
 	} else
 		dma_sync_single_for_cpu(dev, addr, size, dir);
 }
@@ -150,7 +147,6 @@ static inline void dma_sync_single_range_for_device(struct device *dev,
 		ops->sync_single_range_for_device(dev, addr, offset, size, dir);
 		debug_dma_sync_single_range_for_device(dev, addr, offset, size, dir);
 
-		flush_write_buffers();
 	} else
 		dma_sync_single_for_device(dev, addr, size, dir);
 }
@@ -165,7 +161,6 @@ dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg,
 	if (ops->sync_sg_for_cpu)
 		ops->sync_sg_for_cpu(dev, sg, nelems, dir);
 	debug_dma_sync_sg_for_cpu(dev, sg, nelems, dir);
-	flush_write_buffers();
 }
 
 static inline void
@@ -179,7 +174,6 @@ dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg,
 		ops->sync_sg_for_device(dev, sg, nelems, dir);
 	debug_dma_sync_sg_for_device(dev, sg, nelems, dir);
 
-	flush_write_buffers();
 }
 
 #define dma_map_single(d, a, s, r) dma_map_single_attrs(d, a, s, r, NULL)
-- 
cgit v1.2.3


From c7425314c755d5f94da7c978205c85a7c6201212 Mon Sep 17 00:00:00 2001
From: Akinobu Mita <akinobu.mita@gmail.com>
Date: Sun, 9 Aug 2009 17:03:52 +0900
Subject: x86: Introduce GDT_ENTRY_INIT(), initialize bad_bios_desc statically

Fully initialize bad_bios_desc statically instead of doing some
fields statically and some dynamically.

Suggested-by: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
LKML-Reference: <20090809080350.GA4765@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apm_32.c        | 19 +++++++++----------
 drivers/pnp/pnpbios/bioscalls.c |  5 ++---
 2 files changed, 11 insertions(+), 13 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c
index febb2dab254f..39a4462ef8a7 100644
--- a/arch/x86/kernel/apm_32.c
+++ b/arch/x86/kernel/apm_32.c
@@ -403,7 +403,15 @@ static DECLARE_WAIT_QUEUE_HEAD(apm_waitqueue);
 static DECLARE_WAIT_QUEUE_HEAD(apm_suspend_waitqueue);
 static struct apm_user *user_list;
 static DEFINE_SPINLOCK(user_list_lock);
-static struct desc_struct bad_bios_desc = GDT_ENTRY_INIT(0x4092, 0, 0);
+
+/*
+ * Set up a segment that references the real mode segment 0x40
+ * that extends up to the end of page zero (that we have reserved).
+ * This is for buggy BIOS's that refer to (real mode) segment 0x40
+ * even though they are called in protected mode.
+ */
+static struct desc_struct bad_bios_desc = GDT_ENTRY_INIT(0x4092,
+			(unsigned long)__va(0x400UL), PAGE_SIZE - 0x400 - 1);
 
 static const char driver_version[] = "1.16ac";	/* no spaces */
 
@@ -2331,15 +2339,6 @@ static int __init apm_init(void)
 	}
 	pm_flags |= PM_APM;
 
-	/*
-	 * Set up a segment that references the real mode segment 0x40
-	 * that extends up to the end of page zero (that we have reserved).
-	 * This is for buggy BIOS's that refer to (real mode) segment 0x40
-	 * even though they are called in protected mode.
-	 */
-	set_desc_base(&bad_bios_desc, (unsigned long)__va(0x40UL << 4));
-	set_desc_limit(&bad_bios_desc, 4095 - (0x40 << 4));
-
 	/*
 	 * Set up the long jump entry point to the APM BIOS, which is called
 	 * from inline assembly.
diff --git a/drivers/pnp/pnpbios/bioscalls.c b/drivers/pnp/pnpbios/bioscalls.c
index bd035e3d3550..fc83783c3a96 100644
--- a/drivers/pnp/pnpbios/bioscalls.c
+++ b/drivers/pnp/pnpbios/bioscalls.c
@@ -60,7 +60,8 @@ do { \
 	set_desc_limit(&gdt[(selname) >> 3], (size) - 1); \
 } while(0)
 
-static struct desc_struct bad_bios_desc = GDT_ENTRY_INIT(0x4092, 0, 0);
+static struct desc_struct bad_bios_desc = GDT_ENTRY_INIT(0x4092,
+			(unsigned long)__va(0x400UL), PAGE_SIZE - 0x400 - 1);
 
 /*
  * At some point we want to use this stack frame pointer to unwind
@@ -476,8 +477,6 @@ void pnpbios_calls_init(union pnp_bios_install_struct *header)
 	pnp_bios_callpoint.offset = header->fields.pm16offset;
 	pnp_bios_callpoint.segment = PNP_CS16;
 
-	set_desc_base(&bad_bios_desc, (unsigned long)__va(0x40UL << 4));
-	set_desc_limit(&bad_bios_desc, 4095 - (0x40 << 4));
 	for_each_possible_cpu(i) {
 		struct desc_struct *gdt = get_cpu_gdt_table(i);
 		if (!gdt)
-- 
cgit v1.2.3


From eeac19a7efa150231e4a6bb110d6f27500bcc8ce Mon Sep 17 00:00:00 2001
From: Jason Baron <jbaron@redhat.com>
Date: Mon, 10 Aug 2009 16:52:13 -0400
Subject: tracing: Map syscall name to number

Add a new function to support translating a syscall name to number at
runtime.
This allows the syscall event tracer to map syscall names to number.

Signed-off-by: Jason Baron <jbaron@redhat.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: Jiaying Zhang <jiayingz@google.com>
Cc: Martin Bligh <mbligh@google.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
---
 arch/x86/kernel/ftrace.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c
index 8e9663413b7f..afb31d72618d 100644
--- a/arch/x86/kernel/ftrace.c
+++ b/arch/x86/kernel/ftrace.c
@@ -500,6 +500,22 @@ struct syscall_metadata *syscall_nr_to_meta(int nr)
 	return syscalls_metadata[nr];
 }
 
+int syscall_name_to_nr(char *name)
+{
+	int i;
+
+	if (!syscalls_metadata)
+		return -1;
+
+	for (i = 0; i < FTRACE_SYSCALL_MAX; i++) {
+		if (syscalls_metadata[i]) {
+			if (!strcmp(syscalls_metadata[i]->name, name))
+				return i;
+		}
+	}
+	return -1;
+}
+
 void arch_init_ftrace_syscalls(void)
 {
 	int i;
-- 
cgit v1.2.3


From 066e0378c23f0a3db730893f6a041e4a3922a385 Mon Sep 17 00:00:00 2001
From: Jason Baron <jbaron@redhat.com>
Date: Mon, 10 Aug 2009 16:52:23 -0400
Subject: tracing: Call arch_init_ftrace_syscalls at boot

Call arch_init_ftrace_syscalls at boot, so we can determine early the
set of syscalls for the syscall trace events.

Signed-off-by: Jason Baron <jbaron@redhat.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: Jiaying Zhang <jiayingz@google.com>
Cc: Martin Bligh <mbligh@google.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
---
 arch/x86/kernel/ftrace.c      | 15 ++++-----------
 include/trace/syscall.h       |  1 -
 kernel/trace/trace_syscalls.c |  1 -
 3 files changed, 4 insertions(+), 13 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c
index afb31d72618d..0d93d409b8d2 100644
--- a/arch/x86/kernel/ftrace.c
+++ b/arch/x86/kernel/ftrace.c
@@ -516,31 +516,24 @@ int syscall_name_to_nr(char *name)
 	return -1;
 }
 
-void arch_init_ftrace_syscalls(void)
+static int __init arch_init_ftrace_syscalls(void)
 {
 	int i;
 	struct syscall_metadata *meta;
 	unsigned long **psys_syscall_table = &sys_call_table;
-	static atomic_t refs;
-
-	if (atomic_inc_return(&refs) != 1)
-		goto end;
 
 	syscalls_metadata = kzalloc(sizeof(*syscalls_metadata) *
 					FTRACE_SYSCALL_MAX, GFP_KERNEL);
 	if (!syscalls_metadata) {
 		WARN_ON(1);
-		return;
+		return -ENOMEM;
 	}
 
 	for (i = 0; i < FTRACE_SYSCALL_MAX; i++) {
 		meta = find_syscall_meta(psys_syscall_table[i]);
 		syscalls_metadata[i] = meta;
 	}
-	return;
-
-	/* Paranoid: avoid overflow */
-end:
-	atomic_dec(&refs);
+	return 0;
 }
+arch_initcall(arch_init_ftrace_syscalls);
 #endif
diff --git a/include/trace/syscall.h b/include/trace/syscall.h
index 8cfe515cbc47..c55fcce4fbb2 100644
--- a/include/trace/syscall.h
+++ b/include/trace/syscall.h
@@ -19,7 +19,6 @@ struct syscall_metadata {
 };
 
 #ifdef CONFIG_FTRACE_SYSCALLS
-extern void arch_init_ftrace_syscalls(void);
 extern struct syscall_metadata *syscall_nr_to_meta(int nr);
 extern void start_ftrace_syscalls(void);
 extern void stop_ftrace_syscalls(void);
diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c
index 5e579645ac86..08aed439feaf 100644
--- a/kernel/trace/trace_syscalls.c
+++ b/kernel/trace/trace_syscalls.c
@@ -106,7 +106,6 @@ void start_ftrace_syscalls(void)
 	if (++refcount != 1)
 		goto unlock;
 
-	arch_init_ftrace_syscalls();
 	read_lock_irqsave(&tasklist_lock, flags);
 
 	do_each_thread(g, t) {
-- 
cgit v1.2.3


From a871bd33a6c0bc86fb47cd02ea2650dd43d3d95f Mon Sep 17 00:00:00 2001
From: Jason Baron <jbaron@redhat.com>
Date: Mon, 10 Aug 2009 16:52:31 -0400
Subject: tracing: Add syscall tracepoints

add two tracepoints in syscall exit and entry path, conditioned on
TIF_SYSCALL_FTRACE. Supports the syscall trace event code.

Signed-off-by: Jason Baron <jbaron@redhat.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: Jiaying Zhang <jiayingz@google.com>
Cc: Martin Bligh <mbligh@google.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
---
 arch/x86/kernel/ptrace.c |  7 +++++--
 include/trace/syscall.h  | 20 ++++++++++++++++++++
 kernel/tracepoint.c      | 38 ++++++++++++++++++++++++++++++++++++++
 3 files changed, 63 insertions(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index 09ecbde91c13..34dd6f15185d 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -37,6 +37,9 @@
 
 #include <trace/syscall.h>
 
+DEFINE_TRACE(syscall_enter);
+DEFINE_TRACE(syscall_exit);
+
 #include "tls.h"
 
 enum x86_regset {
@@ -1498,7 +1501,7 @@ asmregparm long syscall_trace_enter(struct pt_regs *regs)
 		ret = -1L;
 
 	if (unlikely(test_thread_flag(TIF_SYSCALL_FTRACE)))
-		ftrace_syscall_enter(regs);
+		trace_syscall_enter(regs, regs->orig_ax);
 
 	if (unlikely(current->audit_context)) {
 		if (IS_IA32)
@@ -1524,7 +1527,7 @@ asmregparm void syscall_trace_leave(struct pt_regs *regs)
 		audit_syscall_exit(AUDITSC_RESULT(regs->ax), regs->ax);
 
 	if (unlikely(test_thread_flag(TIF_SYSCALL_FTRACE)))
-		ftrace_syscall_exit(regs);
+		trace_syscall_exit(regs, regs->ax);
 
 	if (test_thread_flag(TIF_SYSCALL_TRACE))
 		tracehook_report_syscall_exit(regs, 0);
diff --git a/include/trace/syscall.h b/include/trace/syscall.h
index c55fcce4fbb2..3951d774de18 100644
--- a/include/trace/syscall.h
+++ b/include/trace/syscall.h
@@ -1,8 +1,28 @@
 #ifndef _TRACE_SYSCALL_H
 #define _TRACE_SYSCALL_H
 
+#include <linux/tracepoint.h>
+
 #include <asm/ptrace.h>
 
+
+extern void syscall_regfunc(void);
+extern void syscall_unregfunc(void);
+
+DECLARE_TRACE_WITH_CALLBACK(syscall_enter,
+	TP_PROTO(struct pt_regs *regs, long id),
+	TP_ARGS(regs, id),
+	syscall_regfunc,
+	syscall_unregfunc
+);
+
+DECLARE_TRACE_WITH_CALLBACK(syscall_exit,
+	TP_PROTO(struct pt_regs *regs, long ret),
+	TP_ARGS(regs, ret),
+	syscall_regfunc,
+	syscall_unregfunc
+);
+
 /*
  * A syscall entry in the ftrace syscalls array.
  *
diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c
index 1ef5d3a601c7..070a42bb8920 100644
--- a/kernel/tracepoint.c
+++ b/kernel/tracepoint.c
@@ -24,6 +24,7 @@
 #include <linux/tracepoint.h>
 #include <linux/err.h>
 #include <linux/slab.h>
+#include <linux/sched.h>
 
 extern struct tracepoint __start___tracepoints[];
 extern struct tracepoint __stop___tracepoints[];
@@ -577,3 +578,40 @@ static int init_tracepoints(void)
 __initcall(init_tracepoints);
 
 #endif /* CONFIG_MODULES */
+
+static DEFINE_MUTEX(regfunc_mutex);
+static int sys_tracepoint_refcount;
+
+void syscall_regfunc(void)
+{
+	unsigned long flags;
+	struct task_struct *g, *t;
+
+	mutex_lock(&regfunc_mutex);
+	if (!sys_tracepoint_refcount) {
+		read_lock_irqsave(&tasklist_lock, flags);
+		do_each_thread(g, t) {
+			set_tsk_thread_flag(t, TIF_SYSCALL_FTRACE);
+		} while_each_thread(g, t);
+		read_unlock_irqrestore(&tasklist_lock, flags);
+	}
+	sys_tracepoint_refcount++;
+	mutex_unlock(&regfunc_mutex);
+}
+
+void syscall_unregfunc(void)
+{
+	unsigned long flags;
+	struct task_struct *g, *t;
+
+	mutex_lock(&regfunc_mutex);
+	sys_tracepoint_refcount--;
+	if (!sys_tracepoint_refcount) {
+		read_lock_irqsave(&tasklist_lock, flags);
+		do_each_thread(g, t) {
+			clear_tsk_thread_flag(t, TIF_SYSCALL_FTRACE);
+		} while_each_thread(g, t);
+		read_unlock_irqrestore(&tasklist_lock, flags);
+	}
+	mutex_unlock(&regfunc_mutex);
+}
-- 
cgit v1.2.3


From 9daa77e2e9a6b8b859660d5e24d0f8cd77c2af39 Mon Sep 17 00:00:00 2001
From: Jason Baron <jbaron@redhat.com>
Date: Mon, 10 Aug 2009 16:52:35 -0400
Subject: tracing: Update FTRACE_SYSCALL_MAX

update FTRACE_SYSCALL_MAX to the current number of syscalls

FTRACE_SYSCALL_MAX is a temporary solution to get the number of
syscalls supported by the arch until we find a more dynamic way
to get this number.

Signed-off-by: Jason Baron <jbaron@redhat.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: Jiaying Zhang <jiayingz@google.com>
Cc: Martin Bligh <mbligh@google.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
---
 arch/x86/include/asm/ftrace.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/ftrace.h b/arch/x86/include/asm/ftrace.h
index bd2c6511c887..71136545187a 100644
--- a/arch/x86/include/asm/ftrace.h
+++ b/arch/x86/include/asm/ftrace.h
@@ -30,9 +30,9 @@
 
 /* FIXME: I don't want to stay hardcoded */
 #ifdef CONFIG_X86_64
-# define FTRACE_SYSCALL_MAX     296
+# define FTRACE_SYSCALL_MAX     299
 #else
-# define FTRACE_SYSCALL_MAX     333
+# define FTRACE_SYSCALL_MAX     337
 #endif
 
 #ifdef CONFIG_FUNCTION_TRACER
-- 
cgit v1.2.3


From 64c12e0444fcc6b75eb49144ba46d43dbdc6bc8f Mon Sep 17 00:00:00 2001
From: Jason Baron <jbaron@redhat.com>
Date: Mon, 10 Aug 2009 16:52:53 -0400
Subject: tracing: Add individual syscalls tracepoint id support

The current state of syscalls tracepoints generates only one event id
for every syscall events.

This patch associates an id with each syscall trace event, so that we
can identify each syscall trace event using the 'perf' tool.

Signed-off-by: Jason Baron <jbaron@redhat.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: Jiaying Zhang <jiayingz@google.com>
Cc: Martin Bligh <mbligh@google.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
---
 arch/x86/kernel/ftrace.c      | 10 ++++++++++
 include/linux/syscalls.h      | 22 ++++++++++++++++++----
 include/trace/syscall.h       |  8 ++++++++
 kernel/trace/trace.h          |  6 ------
 kernel/trace/trace_syscalls.c | 26 ++++++++++++++++----------
 5 files changed, 52 insertions(+), 20 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c
index 0d93d409b8d2..3cff1214e176 100644
--- a/arch/x86/kernel/ftrace.c
+++ b/arch/x86/kernel/ftrace.c
@@ -516,6 +516,16 @@ int syscall_name_to_nr(char *name)
 	return -1;
 }
 
+void set_syscall_enter_id(int num, int id)
+{
+	syscalls_metadata[num]->enter_id = id;
+}
+
+void set_syscall_exit_id(int num, int id)
+{
+	syscalls_metadata[num]->exit_id = id;
+}
+
 static int __init arch_init_ftrace_syscalls(void)
 {
 	int i;
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 5e5b4d33a31c..ce4b01c658eb 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -116,13 +116,20 @@ struct perf_counter_attr;
 
 #define SYSCALL_TRACE_ENTER_EVENT(sname)				\
 	static struct ftrace_event_call event_enter_##sname;		\
+	struct trace_event enter_syscall_print_##sname = {		\
+		.trace                  = print_syscall_enter,		\
+	};								\
 	static int init_enter_##sname(void)				\
 	{								\
-		int num;						\
+		int num, id;						\
 		num = syscall_name_to_nr("sys"#sname);			\
 		if (num < 0)						\
 			return -ENOSYS;					\
-		register_ftrace_event(&event_syscall_enter);		\
+		id = register_ftrace_event(&enter_syscall_print_##sname);\
+		if (!id)						\
+			return -ENODEV;					\
+		event_enter_##sname.id = id;				\
+		set_syscall_enter_id(num, id);				\
 		INIT_LIST_HEAD(&event_enter_##sname.fields);		\
 		init_preds(&event_enter_##sname);			\
 		return 0;						\
@@ -142,13 +149,20 @@ struct perf_counter_attr;
 
 #define SYSCALL_TRACE_EXIT_EVENT(sname)					\
 	static struct ftrace_event_call event_exit_##sname;		\
+	struct trace_event exit_syscall_print_##sname = {		\
+		.trace                  = print_syscall_exit,		\
+	};								\
 	static int init_exit_##sname(void)				\
 	{								\
-		int num;						\
+		int num, id;						\
 		num = syscall_name_to_nr("sys"#sname);			\
 		if (num < 0)						\
 			return -ENOSYS;					\
-		register_ftrace_event(&event_syscall_exit);		\
+		id = register_ftrace_event(&exit_syscall_print_##sname);\
+		if (!id)						\
+			return -ENODEV;					\
+		event_exit_##sname.id = id;				\
+		set_syscall_exit_id(num, id);				\
 		INIT_LIST_HEAD(&event_exit_##sname.fields);		\
 		init_preds(&event_exit_##sname);			\
 		return 0;						\
diff --git a/include/trace/syscall.h b/include/trace/syscall.h
index 73fb8b4a9955..df628404241a 100644
--- a/include/trace/syscall.h
+++ b/include/trace/syscall.h
@@ -32,23 +32,31 @@ DECLARE_TRACE_WITH_CALLBACK(syscall_exit,
  * @nb_args: number of parameters it takes
  * @types: list of types as strings
  * @args: list of args as strings (args[i] matches types[i])
+ * @enter_id: associated ftrace enter event id
+ * @exit_id: associated ftrace exit event id
  */
 struct syscall_metadata {
 	const char	*name;
 	int		nb_args;
 	const char	**types;
 	const char	**args;
+	int		enter_id;
+	int		exit_id;
 };
 
 #ifdef CONFIG_FTRACE_SYSCALLS
 extern struct syscall_metadata *syscall_nr_to_meta(int nr);
 extern int syscall_name_to_nr(char *name);
+void set_syscall_enter_id(int num, int id);
+void set_syscall_exit_id(int num, int id);
 extern struct trace_event event_syscall_enter;
 extern struct trace_event event_syscall_exit;
 extern int reg_event_syscall_enter(void *ptr);
 extern void unreg_event_syscall_enter(void *ptr);
 extern int reg_event_syscall_exit(void *ptr);
 extern void unreg_event_syscall_exit(void *ptr);
+enum print_line_t print_syscall_enter(struct trace_iterator *iter, int flags);
+enum print_line_t print_syscall_exit(struct trace_iterator *iter, int flags);
 #endif
 
 #endif /* _TRACE_SYSCALL_H */
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index d682357e4b1f..300ef788c976 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -34,8 +34,6 @@ enum trace_type {
 	TRACE_GRAPH_ENT,
 	TRACE_USER_STACK,
 	TRACE_HW_BRANCHES,
-	TRACE_SYSCALL_ENTER,
-	TRACE_SYSCALL_EXIT,
 	TRACE_KMEM_ALLOC,
 	TRACE_KMEM_FREE,
 	TRACE_POWER,
@@ -319,10 +317,6 @@ extern void __ftrace_bad_type(void);
 			  TRACE_KMEM_ALLOC);	\
 		IF_ASSIGN(var, ent, struct kmemtrace_free_entry,	\
 			  TRACE_KMEM_FREE);	\
-		IF_ASSIGN(var, ent, struct syscall_trace_enter,		\
-			  TRACE_SYSCALL_ENTER);				\
-		IF_ASSIGN(var, ent, struct syscall_trace_exit,		\
-			  TRACE_SYSCALL_EXIT);				\
 		__ftrace_bad_type();					\
 	} while (0)
 
diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c
index c7ae25ee95d8..e58a9c11ba85 100644
--- a/kernel/trace/trace_syscalls.c
+++ b/kernel/trace/trace_syscalls.c
@@ -36,14 +36,18 @@ print_syscall_enter(struct trace_iterator *iter, int flags)
 	struct syscall_metadata *entry;
 	int i, ret, syscall;
 
-	trace_assign_type(trace, ent);
-
+	trace = (typeof(trace))ent;
 	syscall = trace->nr;
-
 	entry = syscall_nr_to_meta(syscall);
+
 	if (!entry)
 		goto end;
 
+	if (entry->enter_id != ent->type) {
+		WARN_ON_ONCE(1);
+		goto end;
+	}
+
 	ret = trace_seq_printf(s, "%s(", entry->name);
 	if (!ret)
 		return TRACE_TYPE_PARTIAL_LINE;
@@ -78,16 +82,20 @@ print_syscall_exit(struct trace_iterator *iter, int flags)
 	struct syscall_metadata *entry;
 	int ret;
 
-	trace_assign_type(trace, ent);
-
+	trace = (typeof(trace))ent;
 	syscall = trace->nr;
-
 	entry = syscall_nr_to_meta(syscall);
+
 	if (!entry) {
 		trace_seq_printf(s, "\n");
 		return TRACE_TYPE_HANDLED;
 	}
 
+	if (entry->exit_id != ent->type) {
+		WARN_ON_ONCE(1);
+		return TRACE_TYPE_UNHANDLED;
+	}
+
 	ret = trace_seq_printf(s, "%s -> 0x%lx\n", entry->name,
 				trace->ret);
 	if (!ret)
@@ -114,7 +122,7 @@ void ftrace_syscall_enter(struct pt_regs *regs, long id)
 
 	size = sizeof(*entry) + sizeof(unsigned long) * sys_data->nb_args;
 
-	event = trace_current_buffer_lock_reserve(TRACE_SYSCALL_ENTER, size,
+	event = trace_current_buffer_lock_reserve(sys_data->enter_id, size,
 							0, 0);
 	if (!event)
 		return;
@@ -142,7 +150,7 @@ void ftrace_syscall_exit(struct pt_regs *regs, long ret)
 	if (!sys_data)
 		return;
 
-	event = trace_current_buffer_lock_reserve(TRACE_SYSCALL_EXIT,
+	event = trace_current_buffer_lock_reserve(sys_data->exit_id,
 				sizeof(*entry), 0, 0);
 	if (!event)
 		return;
@@ -239,10 +247,8 @@ void unreg_event_syscall_exit(void *ptr)
 
 struct trace_event event_syscall_enter = {
 	.trace			= print_syscall_enter,
-	.type			= TRACE_SYSCALL_ENTER
 };
 
 struct trace_event event_syscall_exit = {
 	.trace			= print_syscall_exit,
-	.type			= TRACE_SYSCALL_EXIT
 };
-- 
cgit v1.2.3


From 0ac676fb50f5f8a22e5e80afc40bf38e31b77c00 Mon Sep 17 00:00:00 2001
From: Jason Baron <jbaron@redhat.com>
Date: Mon, 10 Aug 2009 16:53:11 -0400
Subject: tracing: Convert x86_64 mmap and uname to use DEFINE_SYSCALL

A number of syscalls are not using 'DEFINE_SYSCALL'. I'm not sure why.
Convert x86_64 uname and mmap to use DEFINE_SYSCALL.

Signed-off-by: Jason Baron <jbaron@redhat.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: Jiaying Zhang <jiayingz@google.com>
Cc: Martin Bligh <mbligh@google.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
---
 arch/x86/kernel/sys_x86_64.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/sys_x86_64.c b/arch/x86/kernel/sys_x86_64.c
index 6bc211accf08..45e00eb09c3a 100644
--- a/arch/x86/kernel/sys_x86_64.c
+++ b/arch/x86/kernel/sys_x86_64.c
@@ -18,9 +18,9 @@
 #include <asm/ia32.h>
 #include <asm/syscalls.h>
 
-asmlinkage long sys_mmap(unsigned long addr, unsigned long len,
-		unsigned long prot, unsigned long flags,
-		unsigned long fd, unsigned long off)
+SYSCALL_DEFINE6(mmap, unsigned long, addr, unsigned long, len,
+		unsigned long, prot, unsigned long, flags,
+		unsigned long, fd, unsigned long, off)
 {
 	long error;
 	struct file *file;
@@ -226,7 +226,7 @@ bottomup:
 }
 
 
-asmlinkage long sys_uname(struct new_utsname __user *name)
+SYSCALL_DEFINE1(uname, struct new_utsname __user *, name)
 {
 	int err;
 	down_read(&uts_sem);
-- 
cgit v1.2.3


From b7f42ab2e237f08a5bbcefa17473e80eb05e725c Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yinghai@kernel.org>
Date: Mon, 17 Aug 2009 11:19:40 -0700
Subject: x86, apic: Move dmar_table_init() out of enable_IR()

On an x2apic system, we got:

[    1.818072] ------------[ cut here ]------------
[    1.820376] WARNING: at kernel/lockdep.c:2461 lockdep_trace_alloc+0xa5/0xe9()
[    1.835282] Hardware name: ASSY,
[    1.839006] Modules linked in:
[    1.841253] Pid: 1, comm: swapper Not tainted 2.6.31-rc5-tip-03926-g39aaa80-dirty #510
[    1.858056] Call Trace:
[    1.859913]  [<ffffffff810d13aa>] ? lockdep_trace_alloc+0xa5/0xe9
[    1.876270]  [<ffffffff81093f37>] warn_slowpath_common+0x8d/0xd0
[    1.879132]  [<ffffffff81093fa1>] warn_slowpath_null+0x27/0x3d
[    1.896823]  [<ffffffff810d13aa>] lockdep_trace_alloc+0xa5/0xe9
[    1.900659]  [<ffffffff810cf5a0>] ? lock_release_holdtime+0x2f/0x199
[    1.917188]  [<ffffffff81167a3c>] kmem_cache_alloc_notrace+0x42/0x111
[    1.922320]  [<ffffffff8106fe8c>] ? reserve_memtype+0x152/0x518
[    1.938137]  [<ffffffff8106f8b1>] ? pat_pagerange_is_ram+0x4a/0x91
[    1.941730]  [<ffffffff8106fe8c>] reserve_memtype+0x152/0x518
[    1.958115]  [<ffffffff8106ce62>] __ioremap_caller+0x1dd/0x30f
[    1.975507]  [<ffffffff81ce2c5c>] ? acpi_os_map_memory+0x2a/0x47
[    1.978987]  [<ffffffff8106d0fd>] ioremap_nocache+0x2a/0x40
[    2.031400]  [<ffffffff810d0364>] ? trace_hardirqs_off+0x20/0x36
[    2.036096]  [<ffffffff81ce2c5c>] acpi_os_map_memory+0x2a/0x47
[    2.046263]  [<ffffffff815cd642>] acpi_tb_verify_table+0x3d/0x85
[    2.050349]  [<ffffffff81d34af7>] ? _spin_unlock_irqrestore+0x50/0x76
[    2.067327]  [<ffffffff815ccad6>] acpi_get_table_with_size+0x64/0xd9
[    2.070860]  [<ffffffff81d34af7>] ? _spin_unlock_irqrestore+0x50/0x76
[    2.088000]  [<ffffffff825c88d5>] dmar_table_detect+0x33/0x70
[    2.092047]  [<ffffffff825c8a01>] dmar_table_init+0x43/0x428
[    2.106854]  [<ffffffff825a7537>] enable_IR+0x1c/0x8d
[    2.110256]  [<ffffffff825a7624>] enable_IR_x2apic+0x7c/0x19e
[    2.127139]  [<ffffffff825a4876>] native_smp_prepare_cpus+0x139/0x3b8
[    2.145175]  [<ffffffff8259678d>] kernel_init+0x71/0x1da
[    2.148913]  [<ffffffff8104305a>] child_rip+0xa/0x20
[    2.152349]  [<ffffffff810429fc>] ? restore_args+0x0/0x30
[    2.167931]  [<ffffffff8259671c>] ? kernel_init+0x0/0x1da
[    2.171671]  [<ffffffff81043050>] ? child_rip+0x0/0x20
[    2.187607] ---[ end trace a7919e7f17c0a725 ]---

Venkatesh Pallipadi said:

| Looks like the problem started with this commit
|
| commit ce69a784504222c3ab6f1b3c357d09ec5772127a
| Author: Gleb Natapov <gleb@redhat.com>
| Date:   Mon Jul 20 15:24:17 2009 +0300
|
| x86/apic: Enable x2APIC without interrupt remapping under KVM
|
| Before this commit, dmar_table_init() was getting called
| with interrupts enabled and after this commit, it is getting
| called with interrupts disabled.

so try to move out dmar_table_init out of that function.

Analyzed-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Gleb Natapov <gleb@redhat.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: "Pallipadi, Venkatesh" <venkatesh.pallipadi@intel.com>
LKML-Reference: <4A899F3C.2050104@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic/apic.c | 22 +++++++++++++---------
 1 file changed, 13 insertions(+), 9 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c
index de039fcdd053..3fc3a6c428da 100644
--- a/arch/x86/kernel/apic/apic.c
+++ b/arch/x86/kernel/apic/apic.c
@@ -1365,14 +1365,6 @@ void enable_x2apic(void)
 int __init enable_IR(void)
 {
 #ifdef CONFIG_INTR_REMAP
-	int ret;
-
-	ret = dmar_table_init();
-	if (ret) {
-		pr_debug("dmar_table_init() failed with %d:\n", ret);
-		return 0;
-	}
-
 	if (!intr_remapping_supported()) {
 		pr_debug("intr-remapping not supported\n");
 		return 0;
@@ -1400,6 +1392,14 @@ void __init enable_IR_x2apic(void)
 	unsigned long flags;
 	struct IO_APIC_route_entry **ioapic_entries = NULL;
 	int ret, x2apic_enabled = 0;
+	int dmar_table_init_ret = 0;
+
+#ifdef CONFIG_INTR_REMAP
+	dmar_table_init_ret = dmar_table_init();
+	if (dmar_table_init_ret)
+		pr_debug("dmar_table_init() failed with %d:\n",
+				dmar_table_init_ret);
+#endif
 
 	ioapic_entries = alloc_ioapic_entries();
 	if (!ioapic_entries) {
@@ -1417,7 +1417,11 @@ void __init enable_IR_x2apic(void)
 	mask_8259A();
 	mask_IO_APIC_setup(ioapic_entries);
 
-	ret = enable_IR();
+	if (dmar_table_init_ret)
+		ret = 0;
+	else
+		ret = enable_IR();
+
 	if (!ret) {
 		/* IR is required if there is APIC ID > 255 even when running
 		 * under KVM
-- 
cgit v1.2.3


From 8b5a10fc6fd02289ea03480f93382b1a99006142 Mon Sep 17 00:00:00 2001
From: Jan Beulich <JBeulich@novell.com>
Date: Wed, 19 Aug 2009 08:40:48 +0100
Subject: x86: properly annotate alternatives.c

Some of the NOPs tables aren't used on 64-bits, quite some code and
data is needed post-init for module loading only, and a couple of
functions aren't used outside that file (i.e. can be static, and don't
need to be exported).

The change to __INITDATA/__INITRODATA is needed to avoid an assembler
warning.

Signed-off-by: Jan Beulich <jbeulich@novell.com>
LKML-Reference: <4A8BC8A00200007800010823@vpn.id2.novell.com>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/include/asm/alternative.h |  7 -----
 arch/x86/kernel/alternative.c      | 56 ++++++++++++++++++++++----------------
 include/linux/init.h               | 12 ++++++--
 3 files changed, 42 insertions(+), 33 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/alternative.h b/arch/x86/include/asm/alternative.h
index 1a37bcdc8606..c240efc74e00 100644
--- a/arch/x86/include/asm/alternative.h
+++ b/arch/x86/include/asm/alternative.h
@@ -73,8 +73,6 @@ static inline void alternatives_smp_module_del(struct module *mod) {}
 static inline void alternatives_smp_switch(int smp) {}
 #endif	/* CONFIG_SMP */
 
-const unsigned char *const *find_nop_table(void);
-
 /* alternative assembly primitive: */
 #define ALTERNATIVE(oldinstr, newinstr, feature)			\
 									\
@@ -144,8 +142,6 @@ static inline void apply_paravirt(struct paravirt_patch_site *start,
 #define __parainstructions_end	NULL
 #endif
 
-extern void add_nops(void *insns, unsigned int len);
-
 /*
  * Clear and restore the kernel write-protection flag on the local CPU.
  * Allows the kernel to edit read-only pages.
@@ -161,10 +157,7 @@ extern void add_nops(void *insns, unsigned int len);
  * Intel's errata.
  * On the local CPU you need to be protected again NMI or MCE handlers seeing an
  * inconsistent instruction while you patch.
- * The _early version expects the memory to already be RW.
  */
-
 extern void *text_poke(void *addr, const void *opcode, size_t len);
-extern void *text_poke_early(void *addr, const void *opcode, size_t len);
 
 #endif /* _ASM_X86_ALTERNATIVE_H */
diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c
index f57658702571..486935143e02 100644
--- a/arch/x86/kernel/alternative.c
+++ b/arch/x86/kernel/alternative.c
@@ -2,6 +2,7 @@
 #include <linux/sched.h>
 #include <linux/mutex.h>
 #include <linux/list.h>
+#include <linux/stringify.h>
 #include <linux/kprobes.h>
 #include <linux/mm.h>
 #include <linux/vmalloc.h>
@@ -32,7 +33,7 @@ __setup("smp-alt-boot", bootonly);
 #define smp_alt_once 1
 #endif
 
-static int debug_alternative;
+static int __initdata_or_module debug_alternative;
 
 static int __init debug_alt(char *str)
 {
@@ -51,7 +52,7 @@ static int __init setup_noreplace_smp(char *str)
 __setup("noreplace-smp", setup_noreplace_smp);
 
 #ifdef CONFIG_PARAVIRT
-static int noreplace_paravirt = 0;
+static int __initdata_or_module noreplace_paravirt = 0;
 
 static int __init setup_noreplace_paravirt(char *str)
 {
@@ -64,16 +65,17 @@ __setup("noreplace-paravirt", setup_noreplace_paravirt);
 #define DPRINTK(fmt, args...) if (debug_alternative) \
 	printk(KERN_DEBUG fmt, args)
 
-#ifdef GENERIC_NOP1
+#if defined(GENERIC_NOP1) && !defined(CONFIG_X86_64)
 /* Use inline assembly to define this because the nops are defined
    as inline assembly strings in the include files and we cannot
    get them easily into strings. */
-asm("\t.section .rodata, \"a\"\nintelnops: "
+asm("\t" __stringify(__INITRODATA_OR_MODULE) "\nintelnops: "
 	GENERIC_NOP1 GENERIC_NOP2 GENERIC_NOP3 GENERIC_NOP4 GENERIC_NOP5 GENERIC_NOP6
 	GENERIC_NOP7 GENERIC_NOP8
     "\t.previous");
 extern const unsigned char intelnops[];
-static const unsigned char *const intel_nops[ASM_NOP_MAX+1] = {
+static const unsigned char *const __initconst_or_module
+intel_nops[ASM_NOP_MAX+1] = {
 	NULL,
 	intelnops,
 	intelnops + 1,
@@ -87,12 +89,13 @@ static const unsigned char *const intel_nops[ASM_NOP_MAX+1] = {
 #endif
 
 #ifdef K8_NOP1
-asm("\t.section .rodata, \"a\"\nk8nops: "
+asm("\t" __stringify(__INITRODATA_OR_MODULE) "\nk8nops: "
 	K8_NOP1 K8_NOP2 K8_NOP3 K8_NOP4 K8_NOP5 K8_NOP6
 	K8_NOP7 K8_NOP8
     "\t.previous");
 extern const unsigned char k8nops[];
-static const unsigned char *const k8_nops[ASM_NOP_MAX+1] = {
+static const unsigned char *const __initconst_or_module
+k8_nops[ASM_NOP_MAX+1] = {
 	NULL,
 	k8nops,
 	k8nops + 1,
@@ -105,13 +108,14 @@ static const unsigned char *const k8_nops[ASM_NOP_MAX+1] = {
 };
 #endif
 
-#ifdef K7_NOP1
-asm("\t.section .rodata, \"a\"\nk7nops: "
+#if defined(K7_NOP1) && !defined(CONFIG_X86_64)
+asm("\t" __stringify(__INITRODATA_OR_MODULE) "\nk7nops: "
 	K7_NOP1 K7_NOP2 K7_NOP3 K7_NOP4 K7_NOP5 K7_NOP6
 	K7_NOP7 K7_NOP8
     "\t.previous");
 extern const unsigned char k7nops[];
-static const unsigned char *const k7_nops[ASM_NOP_MAX+1] = {
+static const unsigned char *const __initconst_or_module
+k7_nops[ASM_NOP_MAX+1] = {
 	NULL,
 	k7nops,
 	k7nops + 1,
@@ -125,12 +129,13 @@ static const unsigned char *const k7_nops[ASM_NOP_MAX+1] = {
 #endif
 
 #ifdef P6_NOP1
-asm("\t.section .rodata, \"a\"\np6nops: "
+asm("\t" __stringify(__INITRODATA_OR_MODULE) "\np6nops: "
 	P6_NOP1 P6_NOP2 P6_NOP3 P6_NOP4 P6_NOP5 P6_NOP6
 	P6_NOP7 P6_NOP8
     "\t.previous");
 extern const unsigned char p6nops[];
-static const unsigned char *const p6_nops[ASM_NOP_MAX+1] = {
+static const unsigned char *const __initconst_or_module
+p6_nops[ASM_NOP_MAX+1] = {
 	NULL,
 	p6nops,
 	p6nops + 1,
@@ -146,7 +151,7 @@ static const unsigned char *const p6_nops[ASM_NOP_MAX+1] = {
 #ifdef CONFIG_X86_64
 
 extern char __vsyscall_0;
-const unsigned char *const *find_nop_table(void)
+static const unsigned char *const *__init_or_module find_nop_table(void)
 {
 	if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL &&
 	    boot_cpu_has(X86_FEATURE_NOPL))
@@ -157,7 +162,7 @@ const unsigned char *const *find_nop_table(void)
 
 #else /* CONFIG_X86_64 */
 
-const unsigned char *const *find_nop_table(void)
+static const unsigned char *const *__init_or_module find_nop_table(void)
 {
 	if (boot_cpu_has(X86_FEATURE_K8))
 		return k8_nops;
@@ -172,7 +177,7 @@ const unsigned char *const *find_nop_table(void)
 #endif /* CONFIG_X86_64 */
 
 /* Use this to add nops to a buffer, then text_poke the whole buffer. */
-void add_nops(void *insns, unsigned int len)
+static void __init_or_module add_nops(void *insns, unsigned int len)
 {
 	const unsigned char *const *noptable = find_nop_table();
 
@@ -185,10 +190,10 @@ void add_nops(void *insns, unsigned int len)
 		len -= noplen;
 	}
 }
-EXPORT_SYMBOL_GPL(add_nops);
 
 extern struct alt_instr __alt_instructions[], __alt_instructions_end[];
 extern u8 *__smp_locks[], *__smp_locks_end[];
+static void *text_poke_early(void *addr, const void *opcode, size_t len);
 
 /* Replace instructions with better alternatives for this CPU type.
    This runs before SMP is initialized to avoid SMP problems with
@@ -196,7 +201,8 @@ extern u8 *__smp_locks[], *__smp_locks_end[];
    APs have less capabilities than the boot processor are not handled.
    Tough. Make sure you disable such features by hand. */
 
-void apply_alternatives(struct alt_instr *start, struct alt_instr *end)
+void __init_or_module apply_alternatives(struct alt_instr *start,
+					 struct alt_instr *end)
 {
 	struct alt_instr *a;
 	char insnbuf[MAX_PATCH_LEN];
@@ -279,9 +285,10 @@ static LIST_HEAD(smp_alt_modules);
 static DEFINE_MUTEX(smp_alt);
 static int smp_mode = 1;	/* protected by smp_alt */
 
-void alternatives_smp_module_add(struct module *mod, char *name,
-				 void *locks, void *locks_end,
-				 void *text,  void *text_end)
+void __init_or_module alternatives_smp_module_add(struct module *mod,
+						  char *name,
+						  void *locks, void *locks_end,
+						  void *text,  void *text_end)
 {
 	struct smp_alt_module *smp;
 
@@ -317,7 +324,7 @@ void alternatives_smp_module_add(struct module *mod, char *name,
 	mutex_unlock(&smp_alt);
 }
 
-void alternatives_smp_module_del(struct module *mod)
+void __init_or_module alternatives_smp_module_del(struct module *mod)
 {
 	struct smp_alt_module *item;
 
@@ -386,8 +393,8 @@ void alternatives_smp_switch(int smp)
 #endif
 
 #ifdef CONFIG_PARAVIRT
-void apply_paravirt(struct paravirt_patch_site *start,
-		    struct paravirt_patch_site *end)
+void __init_or_module apply_paravirt(struct paravirt_patch_site *start,
+				     struct paravirt_patch_site *end)
 {
 	struct paravirt_patch_site *p;
 	char insnbuf[MAX_PATCH_LEN];
@@ -485,7 +492,8 @@ void __init alternative_instructions(void)
  * instructions. And on the local CPU you need to be protected again NMI or MCE
  * handlers seeing an inconsistent instruction while you patch.
  */
-void *text_poke_early(void *addr, const void *opcode, size_t len)
+static void *__init_or_module text_poke_early(void *addr, const void *opcode,
+					      size_t len)
 {
 	unsigned long flags;
 	local_irq_save(flags);
diff --git a/include/linux/init.h b/include/linux/init.h
index 13b633ed695e..400adbb45414 100644
--- a/include/linux/init.h
+++ b/include/linux/init.h
@@ -103,8 +103,8 @@
 #define __INIT		.section	".init.text","ax"
 #define __FINIT		.previous
 
-#define __INITDATA	.section	".init.data","aw"
-#define __INITRODATA	.section	".init.rodata","a"
+#define __INITDATA	.section	".init.data","aw",%progbits
+#define __INITRODATA	.section	".init.rodata","a",%progbits
 #define __FINITDATA	.previous
 
 #define __DEVINIT        .section	".devinit.text", "ax"
@@ -305,9 +305,17 @@ void __init parse_early_options(char *cmdline);
 #ifdef CONFIG_MODULES
 #define __init_or_module
 #define __initdata_or_module
+#define __initconst_or_module
+#define __INIT_OR_MODULE	.text
+#define __INITDATA_OR_MODULE	.data
+#define __INITRODATA_OR_MODULE	.section ".rodata","a",%progbits
 #else
 #define __init_or_module __init
 #define __initdata_or_module __initdata
+#define __initconst_or_module __initconst
+#define __INIT_OR_MODULE __INIT
+#define __INITDATA_OR_MODULE __INITDATA
+#define __INITRODATA_OR_MODULE __INITRODATA
 #endif /*CONFIG_MODULES*/
 
 /* Functions marked as __devexit may be discarded at kernel link time, depending
-- 
cgit v1.2.3


From 366d19e181be873c70f4aafca3931d77d781ccd7 Mon Sep 17 00:00:00 2001
From: Tobias Doerffel <tobias.doerffel@gmail.com>
Date: Fri, 21 Aug 2009 23:06:23 +0200
Subject: x86: add specific support for Intel Atom architecture

Add another option when selecting CPU family so the kernel can be
optimized for Intel Atom CPUs. If GCC supports tuning options for
Intel Atom they will be used.

Signed-off-by: Tobias Doerffel <tobias.doerffel@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
LKML-Reference: <1251018457-19157-1-git-send-email-tobias.doerffel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/Kconfig.cpu          | 19 ++++++++++++++-----
 arch/x86/Makefile             |  2 ++
 arch/x86/Makefile_32.cpu      |  2 ++
 arch/x86/include/asm/module.h |  2 ++
 4 files changed, 20 insertions(+), 5 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/Kconfig.cpu b/arch/x86/Kconfig.cpu
index 8130334329c0..527519b8a9f9 100644
--- a/arch/x86/Kconfig.cpu
+++ b/arch/x86/Kconfig.cpu
@@ -262,6 +262,15 @@ config MCORE2
 	  family in /proc/cpuinfo. Newer ones have 6 and older ones 15
 	  (not a typo)
 
+config MATOM
+	bool "Intel Atom"
+	---help---
+
+	  Select this for the Intel Atom platform. Intel Atom CPUs have an
+	  in-order pipelining architecture and thus can benefit from
+	  accordingly optimized code. Use a recent GCC with specific Atom
+	  support in order to fully benefit from selecting this option.
+
 config GENERIC_CPU
 	bool "Generic-x86-64"
 	depends on X86_64
@@ -295,7 +304,7 @@ config X86_CPU
 config X86_L1_CACHE_BYTES
 	int
 	default "128" if MPSC
-	default "64" if GENERIC_CPU || MK8 || MCORE2 || X86_32
+	default "64" if GENERIC_CPU || MK8 || MCORE2 || MATOM || X86_32
 
 config X86_INTERNODE_CACHE_BYTES
 	int
@@ -310,7 +319,7 @@ config X86_L1_CACHE_SHIFT
 	default "7" if MPENTIUM4 || MPSC
 	default "4" if X86_ELAN || M486 || M386 || MGEODEGX1
 	default "5" if MWINCHIP3D || MWINCHIPC6 || MCRUSOE || MEFFICEON || MCYRIXIII || MK6 || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || M586 || MVIAC3_2 || MGEODE_LX
-	default "6" if MK7 || MK8 || MPENTIUMM || MCORE2 || MVIAC7 || X86_GENERIC || GENERIC_CPU
+	default "6" if MK7 || MK8 || MPENTIUMM || MCORE2 || MATOM || MVIAC7 || X86_GENERIC || GENERIC_CPU
 
 config X86_XADD
 	def_bool y
@@ -359,7 +368,7 @@ config X86_INTEL_USERCOPY
 
 config X86_USE_PPRO_CHECKSUM
 	def_bool y
-	depends on MWINCHIP3D || MWINCHIPC6 || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MK8 || MVIAC3_2 || MEFFICEON || MGEODE_LX || MCORE2
+	depends on MWINCHIP3D || MWINCHIPC6 || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MK8 || MVIAC3_2 || MEFFICEON || MGEODE_LX || MCORE2 || MATOM
 
 config X86_USE_3DNOW
 	def_bool y
@@ -387,7 +396,7 @@ config X86_P6_NOP
 
 config X86_TSC
 	def_bool y
-	depends on ((MWINCHIP3D || MCRUSOE || MEFFICEON || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || MK8 || MVIAC3_2 || MVIAC7 || MGEODEGX1 || MGEODE_LX || MCORE2) && !X86_NUMAQ) || X86_64
+	depends on ((MWINCHIP3D || MCRUSOE || MEFFICEON || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || MK8 || MVIAC3_2 || MVIAC7 || MGEODEGX1 || MGEODE_LX || MCORE2 || MATOM) && !X86_NUMAQ) || X86_64
 
 config X86_CMPXCHG64
 	def_bool y
@@ -397,7 +406,7 @@ config X86_CMPXCHG64
 # generates cmov.
 config X86_CMOV
 	def_bool y
-	depends on (MK8 || MK7 || MCORE2 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MVIAC3_2 || MVIAC7 || MCRUSOE || MEFFICEON || X86_64)
+	depends on (MK8 || MK7 || MCORE2 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MVIAC3_2 || MVIAC7 || MCRUSOE || MEFFICEON || X86_64 || MATOM)
 
 config X86_MINIMUM_CPU_FAMILY
 	int
diff --git a/arch/x86/Makefile b/arch/x86/Makefile
index 1b68659c41b4..8a4c24c96d08 100644
--- a/arch/x86/Makefile
+++ b/arch/x86/Makefile
@@ -55,6 +55,8 @@ else
 
         cflags-$(CONFIG_MCORE2) += \
                 $(call cc-option,-march=core2,$(call cc-option,-mtune=generic))
+	cflags-$(CONFIG_MATOM) += $(call cc-option,-march=atom) \
+		$(call cc-option,-mtune=atom,$(call cc-option,-mtune=generic))
         cflags-$(CONFIG_GENERIC_CPU) += $(call cc-option,-mtune=generic)
         KBUILD_CFLAGS += $(cflags-y)
 
diff --git a/arch/x86/Makefile_32.cpu b/arch/x86/Makefile_32.cpu
index 80177ec052f0..30e9a264f69d 100644
--- a/arch/x86/Makefile_32.cpu
+++ b/arch/x86/Makefile_32.cpu
@@ -33,6 +33,8 @@ cflags-$(CONFIG_MCYRIXIII)	+= $(call cc-option,-march=c3,-march=i486) $(align)-f
 cflags-$(CONFIG_MVIAC3_2)	+= $(call cc-option,-march=c3-2,-march=i686)
 cflags-$(CONFIG_MVIAC7)		+= -march=i686
 cflags-$(CONFIG_MCORE2)		+= -march=i686 $(call tune,core2)
+cflags-$(CONFIG_MATOM)		+= $(call cc-option,-march=atom,$(call cc-option,-march=core2,-march=i686)) \
+	$(call cc-option,-mtune=atom,$(call cc-option,-mtune=generic))
 
 # AMD Elan support
 cflags-$(CONFIG_X86_ELAN)	+= -march=i486
diff --git a/arch/x86/include/asm/module.h b/arch/x86/include/asm/module.h
index 47d62743c4d5..e959c4afab59 100644
--- a/arch/x86/include/asm/module.h
+++ b/arch/x86/include/asm/module.h
@@ -28,6 +28,8 @@ struct mod_arch_specific {};
 #define MODULE_PROC_FAMILY "586MMX "
 #elif defined CONFIG_MCORE2
 #define MODULE_PROC_FAMILY "CORE2 "
+#elif defined CONFIG_MATOM
+#define MODULE_PROC_FAMILY "ATOM "
 #elif defined CONFIG_M686
 #define MODULE_PROC_FAMILY "686 "
 #elif defined CONFIG_MPENTIUMII
-- 
cgit v1.2.3


From 10f02d1168585edf66229bb2ec90a42f32667a78 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Sun, 23 Aug 2009 23:17:27 +0400
Subject: x86: uv: Clean up uv_ptc_init(), use proc_create()

create_proc_entry() is getting duhprecated.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: cpw@sgi.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/tlb_uv.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/tlb_uv.c b/arch/x86/kernel/tlb_uv.c
index 77b9689f8edb..503c1f2e8835 100644
--- a/arch/x86/kernel/tlb_uv.c
+++ b/arch/x86/kernel/tlb_uv.c
@@ -640,13 +640,13 @@ static int __init uv_ptc_init(void)
 	if (!is_uv_system())
 		return 0;
 
-	proc_uv_ptc = create_proc_entry(UV_PTC_BASENAME, 0444, NULL);
+	proc_uv_ptc = proc_create(UV_PTC_BASENAME, 0444, NULL,
+				  &proc_uv_ptc_operations);
 	if (!proc_uv_ptc) {
 		printk(KERN_ERR "unable to create %s proc entry\n",
 		       UV_PTC_BASENAME);
 		return -EINVAL;
 	}
-	proc_uv_ptc->proc_fops = &proc_uv_ptc_operations;
 	return 0;
 }
 
-- 
cgit v1.2.3


From 005155b1f626d2b2d7932e4afdf4fead168c6888 Mon Sep 17 00:00:00 2001
From: Roel Kluin <roel.kluin@gmail.com>
Date: Tue, 25 Aug 2009 15:35:12 +0200
Subject: x86: Fix x86_model test in es7000_apic_is_cluster()

For the x86_model to be greater than 6 or less than 12 is
logically always true.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: <stable@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic/es7000_32.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apic/es7000_32.c b/arch/x86/kernel/apic/es7000_32.c
index 69328ac8de9c..420f95da7bf6 100644
--- a/arch/x86/kernel/apic/es7000_32.c
+++ b/arch/x86/kernel/apic/es7000_32.c
@@ -167,7 +167,7 @@ static int es7000_apic_is_cluster(void)
 {
 	/* MPENTIUMIII */
 	if (boot_cpu_data.x86 == 6 &&
-	    (boot_cpu_data.x86_model >= 7 || boot_cpu_data.x86_model <= 11))
+	    (boot_cpu_data.x86_model >= 7 && boot_cpu_data.x86_model <= 11))
 		return 1;
 
 	return 0;
-- 
cgit v1.2.3


From 667000011927b4fcc359beac4a2447889db6d349 Mon Sep 17 00:00:00 2001
From: Josh Stone <jistone@redhat.com>
Date: Mon, 24 Aug 2009 14:43:11 -0700
Subject: tracing: Rename FTRACE_SYSCALLS for tracepoints

s/HAVE_FTRACE_SYSCALLS/HAVE_SYSCALL_TRACEPOINTS/g
s/TIF_SYSCALL_FTRACE/TIF_SYSCALL_TRACEPOINT/g

The syscall enter/exit tracing is no longer specific to just ftrace, so
they now have names that reflect their tie to tracepoints instead.

Signed-off-by: Josh Stone <jistone@redhat.com>
Cc: Jason Baron <jbaron@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: Jiaying Zhang <jiayingz@google.com>
Cc: Martin Bligh <mbligh@google.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
LKML-Reference: <1251150194-1713-2-git-send-email-jistone@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
---
 arch/s390/Kconfig                   |  2 +-
 arch/s390/defconfig                 |  2 +-
 arch/s390/include/asm/thread_info.h |  4 ++--
 arch/s390/kernel/entry.S            |  2 +-
 arch/s390/kernel/entry64.S          |  2 +-
 arch/s390/kernel/ptrace.c           |  4 ++--
 arch/x86/Kconfig                    |  2 +-
 arch/x86/configs/i386_defconfig     |  2 +-
 arch/x86/configs/x86_64_defconfig   |  2 +-
 arch/x86/include/asm/thread_info.h  | 13 +++++++------
 arch/x86/kernel/ptrace.c            |  4 ++--
 kernel/trace/Kconfig                |  4 ++--
 kernel/tracepoint.c                 |  4 ++--
 13 files changed, 24 insertions(+), 23 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index 2ae5d72f47ed..7238ef4c7a6b 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -84,7 +84,7 @@ config S390
 	select HAVE_FUNCTION_TRACER
 	select HAVE_FUNCTION_TRACE_MCOUNT_TEST
 	select HAVE_FTRACE_MCOUNT_RECORD
-	select HAVE_FTRACE_SYSCALLS
+	select HAVE_SYSCALL_TRACEPOINTS
 	select HAVE_DYNAMIC_FTRACE
 	select HAVE_FUNCTION_GRAPH_TRACER
 	select HAVE_DEFAULT_NO_SPIN_MUTEXES
diff --git a/arch/s390/defconfig b/arch/s390/defconfig
index fcba206529f3..4e91a2573cc4 100644
--- a/arch/s390/defconfig
+++ b/arch/s390/defconfig
@@ -900,7 +900,7 @@ CONFIG_HAVE_FUNCTION_GRAPH_TRACER=y
 CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST=y
 CONFIG_HAVE_DYNAMIC_FTRACE=y
 CONFIG_HAVE_FTRACE_MCOUNT_RECORD=y
-CONFIG_HAVE_FTRACE_SYSCALLS=y
+CONFIG_HAVE_SYSCALL_TRACEPOINTS=y
 CONFIG_TRACING_SUPPORT=y
 CONFIG_FTRACE=y
 # CONFIG_FUNCTION_TRACER is not set
diff --git a/arch/s390/include/asm/thread_info.h b/arch/s390/include/asm/thread_info.h
index ba1cab9fc1f9..07eb61b2fb3a 100644
--- a/arch/s390/include/asm/thread_info.h
+++ b/arch/s390/include/asm/thread_info.h
@@ -92,7 +92,7 @@ static inline struct thread_info *current_thread_info(void)
 #define TIF_SYSCALL_TRACE	8	/* syscall trace active */
 #define TIF_SYSCALL_AUDIT	9	/* syscall auditing active */
 #define TIF_SECCOMP		10	/* secure computing */
-#define TIF_SYSCALL_FTRACE	11	/* ftrace syscall instrumentation */
+#define TIF_SYSCALL_TRACEPOINT	11	/* syscall tracepoint instrumentation */
 #define TIF_USEDFPU		16	/* FPU was used by this task this quantum (SMP) */
 #define TIF_POLLING_NRFLAG	17	/* true if poll_idle() is polling 
 					   TIF_NEED_RESCHED */
@@ -111,7 +111,7 @@ static inline struct thread_info *current_thread_info(void)
 #define _TIF_SYSCALL_TRACE	(1<<TIF_SYSCALL_TRACE)
 #define _TIF_SYSCALL_AUDIT	(1<<TIF_SYSCALL_AUDIT)
 #define _TIF_SECCOMP		(1<<TIF_SECCOMP)
-#define _TIF_SYSCALL_FTRACE	(1<<TIF_SYSCALL_FTRACE)
+#define _TIF_SYSCALL_TRACEPOINT	(1<<TIF_SYSCALL_TRACEPOINT)
 #define _TIF_USEDFPU		(1<<TIF_USEDFPU)
 #define _TIF_POLLING_NRFLAG	(1<<TIF_POLLING_NRFLAG)
 #define _TIF_31BIT		(1<<TIF_31BIT)
diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S
index c4c80a22bc1f..5d40fce878a5 100644
--- a/arch/s390/kernel/entry.S
+++ b/arch/s390/kernel/entry.S
@@ -54,7 +54,7 @@ _TIF_WORK_SVC = (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED | \
 _TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED | \
 		 _TIF_MCCK_PENDING)
 _TIF_SYSCALL = (_TIF_SYSCALL_TRACE>>8 | _TIF_SYSCALL_AUDIT>>8 | \
-		_TIF_SECCOMP>>8 | _TIF_SYSCALL_FTRACE>>8)
+		_TIF_SECCOMP>>8 | _TIF_SYSCALL_TRACEPOINT>>8)
 
 STACK_SHIFT = PAGE_SHIFT + THREAD_ORDER
 STACK_SIZE  = 1 << STACK_SHIFT
diff --git a/arch/s390/kernel/entry64.S b/arch/s390/kernel/entry64.S
index f6618e9e15ef..3ceb53c9c493 100644
--- a/arch/s390/kernel/entry64.S
+++ b/arch/s390/kernel/entry64.S
@@ -57,7 +57,7 @@ _TIF_WORK_SVC = (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED | \
 _TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED | \
 		 _TIF_MCCK_PENDING)
 _TIF_SYSCALL = (_TIF_SYSCALL_TRACE>>8 | _TIF_SYSCALL_AUDIT>>8 | \
-		_TIF_SECCOMP>>8 | _TIF_SYSCALL_FTRACE>>8)
+		_TIF_SECCOMP>>8 | _TIF_SYSCALL_TRACEPOINT>>8)
 
 #define BASED(name) name-system_call(%r13)
 
diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c
index c5e87d891ca0..9d3dcfa79ea2 100644
--- a/arch/s390/kernel/ptrace.c
+++ b/arch/s390/kernel/ptrace.c
@@ -664,7 +664,7 @@ asmlinkage long do_syscall_trace_enter(struct pt_regs *regs)
 		ret = -1;
 	}
 
-	if (unlikely(test_thread_flag(TIF_SYSCALL_FTRACE)))
+	if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT)))
 		trace_syscall_enter(regs, regs->gprs[2]);
 
 	if (unlikely(current->audit_context))
@@ -682,7 +682,7 @@ asmlinkage void do_syscall_trace_exit(struct pt_regs *regs)
 		audit_syscall_exit(AUDITSC_RESULT(regs->gprs[2]),
 				   regs->gprs[2]);
 
-	if (unlikely(test_thread_flag(TIF_SYSCALL_FTRACE)))
+	if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT)))
 		trace_syscall_exit(regs, regs->gprs[2]);
 
 	if (test_thread_flag(TIF_SYSCALL_TRACE))
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 738bdc6b0f8b..d59cbf758f34 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -37,7 +37,7 @@ config X86
 	select HAVE_FUNCTION_GRAPH_FP_TEST
 	select HAVE_FUNCTION_TRACE_MCOUNT_TEST
 	select HAVE_FTRACE_NMI_ENTER if DYNAMIC_FTRACE
-	select HAVE_FTRACE_SYSCALLS
+	select HAVE_SYSCALL_TRACEPOINTS
 	select HAVE_KVM
 	select HAVE_ARCH_KGDB
 	select HAVE_ARCH_TRACEHOOK
diff --git a/arch/x86/configs/i386_defconfig b/arch/x86/configs/i386_defconfig
index edb992ebef92..d28fad19654a 100644
--- a/arch/x86/configs/i386_defconfig
+++ b/arch/x86/configs/i386_defconfig
@@ -2355,7 +2355,7 @@ CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST=y
 CONFIG_HAVE_DYNAMIC_FTRACE=y
 CONFIG_HAVE_FTRACE_MCOUNT_RECORD=y
 CONFIG_HAVE_HW_BRANCH_TRACER=y
-CONFIG_HAVE_FTRACE_SYSCALLS=y
+CONFIG_HAVE_SYSCALL_TRACEPOINTS=y
 CONFIG_RING_BUFFER=y
 CONFIG_TRACING=y
 CONFIG_TRACING_SUPPORT=y
diff --git a/arch/x86/configs/x86_64_defconfig b/arch/x86/configs/x86_64_defconfig
index cee1dd2e69b2..6c86acd847a4 100644
--- a/arch/x86/configs/x86_64_defconfig
+++ b/arch/x86/configs/x86_64_defconfig
@@ -2329,7 +2329,7 @@ CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST=y
 CONFIG_HAVE_DYNAMIC_FTRACE=y
 CONFIG_HAVE_FTRACE_MCOUNT_RECORD=y
 CONFIG_HAVE_HW_BRANCH_TRACER=y
-CONFIG_HAVE_FTRACE_SYSCALLS=y
+CONFIG_HAVE_SYSCALL_TRACEPOINTS=y
 CONFIG_RING_BUFFER=y
 CONFIG_TRACING=y
 CONFIG_TRACING_SUPPORT=y
diff --git a/arch/x86/include/asm/thread_info.h b/arch/x86/include/asm/thread_info.h
index fad7d40b75f8..6f7786aea4fc 100644
--- a/arch/x86/include/asm/thread_info.h
+++ b/arch/x86/include/asm/thread_info.h
@@ -95,7 +95,7 @@ struct thread_info {
 #define TIF_DEBUGCTLMSR		25	/* uses thread_struct.debugctlmsr */
 #define TIF_DS_AREA_MSR		26      /* uses thread_struct.ds_area_msr */
 #define TIF_LAZY_MMU_UPDATES	27	/* task is updating the mmu lazily */
-#define TIF_SYSCALL_FTRACE	28	/* for ftrace syscall instrumentation */
+#define TIF_SYSCALL_TRACEPOINT	28	/* syscall tracepoint instrumentation */
 
 #define _TIF_SYSCALL_TRACE	(1 << TIF_SYSCALL_TRACE)
 #define _TIF_NOTIFY_RESUME	(1 << TIF_NOTIFY_RESUME)
@@ -118,17 +118,17 @@ struct thread_info {
 #define _TIF_DEBUGCTLMSR	(1 << TIF_DEBUGCTLMSR)
 #define _TIF_DS_AREA_MSR	(1 << TIF_DS_AREA_MSR)
 #define _TIF_LAZY_MMU_UPDATES	(1 << TIF_LAZY_MMU_UPDATES)
-#define _TIF_SYSCALL_FTRACE	(1 << TIF_SYSCALL_FTRACE)
+#define _TIF_SYSCALL_TRACEPOINT	(1 << TIF_SYSCALL_TRACEPOINT)
 
 /* work to do in syscall_trace_enter() */
 #define _TIF_WORK_SYSCALL_ENTRY	\
-	(_TIF_SYSCALL_TRACE | _TIF_SYSCALL_EMU | _TIF_SYSCALL_FTRACE |	\
-	 _TIF_SYSCALL_AUDIT | _TIF_SECCOMP | _TIF_SINGLESTEP)
+	(_TIF_SYSCALL_TRACE | _TIF_SYSCALL_EMU | _TIF_SYSCALL_AUDIT |	\
+	 _TIF_SECCOMP | _TIF_SINGLESTEP | _TIF_SYSCALL_TRACEPOINT)
 
 /* work to do in syscall_trace_leave() */
 #define _TIF_WORK_SYSCALL_EXIT	\
 	(_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT | _TIF_SINGLESTEP |	\
-	 _TIF_SYSCALL_FTRACE)
+	 _TIF_SYSCALL_TRACEPOINT)
 
 /* work to do on interrupt/exception return */
 #define _TIF_WORK_MASK							\
@@ -137,7 +137,8 @@ struct thread_info {
 	   _TIF_SINGLESTEP|_TIF_SECCOMP|_TIF_SYSCALL_EMU))
 
 /* work to do on any return to user space */
-#define _TIF_ALLWORK_MASK ((0x0000FFFF & ~_TIF_SECCOMP) | _TIF_SYSCALL_FTRACE)
+#define _TIF_ALLWORK_MASK						\
+	((0x0000FFFF & ~_TIF_SECCOMP) | _TIF_SYSCALL_TRACEPOINT)
 
 /* Only used for 64 bit */
 #define _TIF_DO_NOTIFY_MASK						\
diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index 34dd6f15185d..a909afef44f4 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -1500,7 +1500,7 @@ asmregparm long syscall_trace_enter(struct pt_regs *regs)
 	    tracehook_report_syscall_entry(regs))
 		ret = -1L;
 
-	if (unlikely(test_thread_flag(TIF_SYSCALL_FTRACE)))
+	if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT)))
 		trace_syscall_enter(regs, regs->orig_ax);
 
 	if (unlikely(current->audit_context)) {
@@ -1526,7 +1526,7 @@ asmregparm void syscall_trace_leave(struct pt_regs *regs)
 	if (unlikely(current->audit_context))
 		audit_syscall_exit(AUDITSC_RESULT(regs->ax), regs->ax);
 
-	if (unlikely(test_thread_flag(TIF_SYSCALL_FTRACE)))
+	if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT)))
 		trace_syscall_exit(regs, regs->ax);
 
 	if (test_thread_flag(TIF_SYSCALL_TRACE))
diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
index 019f380fd764..06be85a7ef8c 100644
--- a/kernel/trace/Kconfig
+++ b/kernel/trace/Kconfig
@@ -41,7 +41,7 @@ config HAVE_FTRACE_MCOUNT_RECORD
 config HAVE_HW_BRANCH_TRACER
 	bool
 
-config HAVE_FTRACE_SYSCALLS
+config HAVE_SYSCALL_TRACEPOINTS
 	bool
 
 config TRACER_MAX_TRACE
@@ -211,7 +211,7 @@ config ENABLE_DEFAULT_TRACERS
 
 config FTRACE_SYSCALLS
 	bool "Trace syscalls"
-	depends on HAVE_FTRACE_SYSCALLS
+	depends on HAVE_SYSCALL_TRACEPOINTS
 	select GENERIC_TRACER
 	select KALLSYMS
 	help
diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c
index 06f165a44083..be86b9a01a09 100644
--- a/kernel/tracepoint.c
+++ b/kernel/tracepoint.c
@@ -590,7 +590,7 @@ void syscall_regfunc(void)
 	if (!sys_tracepoint_refcount) {
 		read_lock_irqsave(&tasklist_lock, flags);
 		do_each_thread(g, t) {
-			set_tsk_thread_flag(t, TIF_SYSCALL_FTRACE);
+			set_tsk_thread_flag(t, TIF_SYSCALL_TRACEPOINT);
 		} while_each_thread(g, t);
 		read_unlock_irqrestore(&tasklist_lock, flags);
 	}
@@ -608,7 +608,7 @@ void syscall_unregfunc(void)
 	if (!sys_tracepoint_refcount) {
 		read_lock_irqsave(&tasklist_lock, flags);
 		do_each_thread(g, t) {
-			clear_tsk_thread_flag(t, TIF_SYSCALL_FTRACE);
+			clear_tsk_thread_flag(t, TIF_SYSCALL_TRACEPOINT);
 		} while_each_thread(g, t);
 		read_unlock_irqrestore(&tasklist_lock, flags);
 	}
-- 
cgit v1.2.3


From 97419875865859fd2403e66266c02ce028e2f5ab Mon Sep 17 00:00:00 2001
From: Josh Stone <jistone@redhat.com>
Date: Mon, 24 Aug 2009 14:43:13 -0700
Subject: tracing: Move tracepoint callbacks from declaration to definition

It's not strictly correct for the tracepoint reg/unreg callbacks to
occur when a client is hooking up, because the actual tracepoint may not
be present yet.  This happens to be fine for syscall, since that's in
the core kernel, but it would cause problems for tracepoints defined in
a module that hasn't been loaded yet.  It also means the reg/unreg has
to be EXPORTed for any modules to use the tracepoint (as in SystemTap).

This patch removes DECLARE_TRACE_WITH_CALLBACK, and instead introduces
DEFINE_TRACE_FN which stores the callbacks in struct tracepoint.  The
callbacks are used now when the active state of the tracepoint changes
in set_tracepoint & disable_tracepoint.

This also introduces TRACE_EVENT_FN, so ftrace events can also provide
registration callbacks if needed.

Signed-off-by: Josh Stone <jistone@redhat.com>
Cc: Jason Baron <jbaron@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: Jiaying Zhang <jiayingz@google.com>
Cc: Martin Bligh <mbligh@google.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
LKML-Reference: <1251150194-1713-4-git-send-email-jistone@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
---
 arch/s390/kernel/ptrace.c    |  4 ++--
 arch/x86/kernel/ptrace.c     |  4 ++--
 include/linux/tracepoint.h   | 46 +++++++++++++++++---------------------------
 include/trace/define_trace.h |  5 +++++
 include/trace/ftrace.h       |  9 +++++++++
 include/trace/syscall.h      | 12 ++++--------
 kernel/tracepoint.c          | 14 +++++++++-----
 7 files changed, 49 insertions(+), 45 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c
index 9d3dcfa79ea2..c05b44b80c23 100644
--- a/arch/s390/kernel/ptrace.c
+++ b/arch/s390/kernel/ptrace.c
@@ -51,8 +51,8 @@
 #include "compat_ptrace.h"
 #endif
 
-DEFINE_TRACE(syscall_enter);
-DEFINE_TRACE(syscall_exit);
+DEFINE_TRACE_FN(syscall_enter, syscall_regfunc, syscall_unregfunc);
+DEFINE_TRACE_FN(syscall_exit, syscall_regfunc, syscall_unregfunc);
 
 enum s390_regset {
 	REGSET_GENERAL,
diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index a909afef44f4..31e9b97ec4d6 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -37,8 +37,8 @@
 
 #include <trace/syscall.h>
 
-DEFINE_TRACE(syscall_enter);
-DEFINE_TRACE(syscall_exit);
+DEFINE_TRACE_FN(syscall_enter, syscall_regfunc, syscall_unregfunc);
+DEFINE_TRACE_FN(syscall_exit, syscall_regfunc, syscall_unregfunc);
 
 #include "tls.h"
 
diff --git a/include/linux/tracepoint.h b/include/linux/tracepoint.h
index 5984ed04c03b..846a4ae501eb 100644
--- a/include/linux/tracepoint.h
+++ b/include/linux/tracepoint.h
@@ -23,6 +23,8 @@ struct tracepoint;
 struct tracepoint {
 	const char *name;		/* Tracepoint name */
 	int state;			/* State. */
+	void (*regfunc)(void);
+	void (*unregfunc)(void);
 	void **funcs;
 } __attribute__((aligned(32)));		/*
 					 * Aligned on 32 bytes because it is
@@ -60,10 +62,8 @@ struct tracepoint {
  * Make sure the alignment of the structure in the __tracepoints section will
  * not add unwanted padding between the beginning of the section and the
  * structure. Force alignment to the same alignment as the section start.
- * An optional set of (un)registration functions can be passed to perform any
- * additional (un)registration work.
  */
-#define DECLARE_TRACE_WITH_CALLBACK(name, proto, args, reg, unreg)	\
+#define DECLARE_TRACE(name, proto, args)				\
 	extern struct tracepoint __tracepoint_##name;			\
 	static inline void trace_##name(proto)				\
 	{								\
@@ -73,36 +73,23 @@ struct tracepoint {
 	}								\
 	static inline int register_trace_##name(void (*probe)(proto))	\
 	{								\
-		int ret;						\
-		void (*func)(void) = reg;				\
-									\
-		ret = tracepoint_probe_register(#name, (void *)probe);	\
-		if (func && !ret)					\
-			func();						\
-		return ret;						\
+		return tracepoint_probe_register(#name, (void *)probe);	\
 	}								\
 	static inline int unregister_trace_##name(void (*probe)(proto))	\
 	{								\
-		int ret;						\
-		void (*func)(void) = unreg;				\
-									\
-		ret = tracepoint_probe_unregister(#name, (void *)probe);\
-		if (func && !ret)					\
-			func();						\
-		return ret;						\
+		return tracepoint_probe_unregister(#name, (void *)probe);\
 	}
 
 
-#define DECLARE_TRACE(name, proto, args)				 \
-	DECLARE_TRACE_WITH_CALLBACK(name, TP_PROTO(proto), TP_ARGS(args),\
-					NULL, NULL);
-
-#define DEFINE_TRACE(name)						\
+#define DEFINE_TRACE_FN(name, reg, unreg)				\
 	static const char __tpstrtab_##name[]				\
 	__attribute__((section("__tracepoints_strings"))) = #name;	\
 	struct tracepoint __tracepoint_##name				\
 	__attribute__((section("__tracepoints"), aligned(32))) =	\
-		{ __tpstrtab_##name, 0, NULL }
+		{ __tpstrtab_##name, 0, reg, unreg, NULL }
+
+#define DEFINE_TRACE(name)						\
+	DEFINE_TRACE_FN(name, NULL, NULL);
 
 #define EXPORT_TRACEPOINT_SYMBOL_GPL(name)				\
 	EXPORT_SYMBOL_GPL(__tracepoint_##name)
@@ -113,7 +100,7 @@ extern void tracepoint_update_probe_range(struct tracepoint *begin,
 	struct tracepoint *end);
 
 #else /* !CONFIG_TRACEPOINTS */
-#define DECLARE_TRACE_WITH_CALLBACK(name, proto, args, reg, unreg)	\
+#define DECLARE_TRACE(name, proto, args)				\
 	static inline void _do_trace_##name(struct tracepoint *tp, proto) \
 	{ }								\
 	static inline void trace_##name(proto)				\
@@ -127,10 +114,7 @@ extern void tracepoint_update_probe_range(struct tracepoint *begin,
 		return -ENOSYS;						\
 	}
 
-#define DECLARE_TRACE(name, proto, args)				 \
-	DECLARE_TRACE_WITH_CALLBACK(name, TP_PROTO(proto), TP_ARGS(args),\
-					NULL, NULL);
-
+#define DEFINE_TRACE_FN(name, reg, unreg)
 #define DEFINE_TRACE(name)
 #define EXPORT_TRACEPOINT_SYMBOL_GPL(name)
 #define EXPORT_TRACEPOINT_SYMBOL(name)
@@ -282,10 +266,16 @@ static inline void tracepoint_synchronize_unregister(void)
  * can also by used by generic instrumentation like SystemTap), and
  * it is also used to expose a structured trace record in
  * /sys/kernel/debug/tracing/events/.
+ *
+ * A set of (un)registration functions can be passed to the variant
+ * TRACE_EVENT_FN to perform any (un)registration work.
  */
 
 #define TRACE_EVENT(name, proto, args, struct, assign, print)	\
 	DECLARE_TRACE(name, PARAMS(proto), PARAMS(args))
+#define TRACE_EVENT_FN(name, proto, args, struct,		\
+		assign, print, reg, unreg)			\
+	DECLARE_TRACE(name, PARAMS(proto), PARAMS(args))
 #endif
 
 #endif
diff --git a/include/trace/define_trace.h b/include/trace/define_trace.h
index f7a7ae1e8f90..2a969850736d 100644
--- a/include/trace/define_trace.h
+++ b/include/trace/define_trace.h
@@ -26,6 +26,11 @@
 #define TRACE_EVENT(name, proto, args, tstruct, assign, print)	\
 	DEFINE_TRACE(name)
 
+#undef TRACE_EVENT_FN
+#define TRACE_EVENT_FN(name, proto, args, tstruct,		\
+		assign, print, reg, unreg)			\
+	DEFINE_TRACE_FN(name, reg, unreg)
+
 #undef DECLARE_TRACE
 #define DECLARE_TRACE(name, proto, args)	\
 	DEFINE_TRACE(name)
diff --git a/include/trace/ftrace.h b/include/trace/ftrace.h
index 127400255e4c..3a0b44bdabf7 100644
--- a/include/trace/ftrace.h
+++ b/include/trace/ftrace.h
@@ -42,6 +42,15 @@
 	};							\
 	static struct ftrace_event_call event_##name
 
+/* Callbacks are meaningless to ftrace. */
+#undef TRACE_EVENT_FN
+#define TRACE_EVENT_FN(name, proto, args, tstruct,		\
+		assign, print, reg, unreg)			\
+	TRACE_EVENT(name, TP_PROTO(proto), TP_ARGS(args),	\
+		TP_STRUCT__entry(tstruct),			\
+		TP_fast_assign(assign),				\
+		TP_printk(print))
+
 #include TRACE_INCLUDE(TRACE_INCLUDE_FILE)
 
 
diff --git a/include/trace/syscall.h b/include/trace/syscall.h
index 5dcb7e3a544c..4e1943001854 100644
--- a/include/trace/syscall.h
+++ b/include/trace/syscall.h
@@ -13,18 +13,14 @@
 extern void syscall_regfunc(void);
 extern void syscall_unregfunc(void);
 
-DECLARE_TRACE_WITH_CALLBACK(syscall_enter,
+DECLARE_TRACE(syscall_enter,
 	TP_PROTO(struct pt_regs *regs, long id),
-	TP_ARGS(regs, id),
-	syscall_regfunc,
-	syscall_unregfunc
+	TP_ARGS(regs, id)
 );
 
-DECLARE_TRACE_WITH_CALLBACK(syscall_exit,
+DECLARE_TRACE(syscall_exit,
 	TP_PROTO(struct pt_regs *regs, long ret),
-	TP_ARGS(regs, ret),
-	syscall_regfunc,
-	syscall_unregfunc
+	TP_ARGS(regs, ret)
 );
 
 #endif
diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c
index 9e0a36f0e2a9..1a6a453b7efb 100644
--- a/kernel/tracepoint.c
+++ b/kernel/tracepoint.c
@@ -243,6 +243,11 @@ static void set_tracepoint(struct tracepoint_entry **entry,
 {
 	WARN_ON(strcmp((*entry)->name, elem->name) != 0);
 
+	if (elem->regfunc && !elem->state && active)
+		elem->regfunc();
+	else if (elem->unregfunc && elem->state && !active)
+		elem->unregfunc();
+
 	/*
 	 * rcu_assign_pointer has a smp_wmb() which makes sure that the new
 	 * probe callbacks array is consistent before setting a pointer to it.
@@ -262,6 +267,9 @@ static void set_tracepoint(struct tracepoint_entry **entry,
  */
 static void disable_tracepoint(struct tracepoint *elem)
 {
+	if (elem->unregfunc && elem->state)
+		elem->unregfunc();
+
 	elem->state = 0;
 	rcu_assign_pointer(elem->funcs, NULL);
 }
@@ -578,7 +586,7 @@ __initcall(init_tracepoints);
 
 #ifdef CONFIG_HAVE_SYSCALL_TRACEPOINTS
 
-static DEFINE_MUTEX(regfunc_mutex);
+/* NB: reg/unreg are called while guarded with the tracepoints_mutex */
 static int sys_tracepoint_refcount;
 
 void syscall_regfunc(void)
@@ -586,7 +594,6 @@ void syscall_regfunc(void)
 	unsigned long flags;
 	struct task_struct *g, *t;
 
-	mutex_lock(&regfunc_mutex);
 	if (!sys_tracepoint_refcount) {
 		read_lock_irqsave(&tasklist_lock, flags);
 		do_each_thread(g, t) {
@@ -595,7 +602,6 @@ void syscall_regfunc(void)
 		read_unlock_irqrestore(&tasklist_lock, flags);
 	}
 	sys_tracepoint_refcount++;
-	mutex_unlock(&regfunc_mutex);
 }
 
 void syscall_unregfunc(void)
@@ -603,7 +609,6 @@ void syscall_unregfunc(void)
 	unsigned long flags;
 	struct task_struct *g, *t;
 
-	mutex_lock(&regfunc_mutex);
 	sys_tracepoint_refcount--;
 	if (!sys_tracepoint_refcount) {
 		read_lock_irqsave(&tasklist_lock, flags);
@@ -612,6 +617,5 @@ void syscall_unregfunc(void)
 		} while_each_thread(g, t);
 		read_unlock_irqrestore(&tasklist_lock, flags);
 	}
-	mutex_unlock(&regfunc_mutex);
 }
 #endif
-- 
cgit v1.2.3


From 1c569f0264ea629c10bbab471dd0626ce4d3f19f Mon Sep 17 00:00:00 2001
From: Josh Stone <jistone@redhat.com>
Date: Mon, 24 Aug 2009 14:43:14 -0700
Subject: tracing: Create generic syscall TRACE_EVENTs

This converts the syscall_enter/exit tracepoints into TRACE_EVENTs, so
you can have generic ftrace events that capture all system calls with
arguments and return values.  These generic events are also renamed to
sys_enter/exit, so they're more closely aligned to the specific
sys_enter_foo events.

Signed-off-by: Josh Stone <jistone@redhat.com>
Cc: Jason Baron <jbaron@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: Jiaying Zhang <jiayingz@google.com>
Cc: Martin Bligh <mbligh@google.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
LKML-Reference: <1251150194-1713-5-git-send-email-jistone@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
---
 arch/s390/kernel/ptrace.c       |  8 ++---
 arch/x86/kernel/ptrace.c        | 12 +++----
 include/trace/events/syscalls.h | 70 +++++++++++++++++++++++++++++++++++++++++
 include/trace/syscall.h         | 17 ----------
 kernel/trace/trace_syscalls.c   | 17 +++++-----
 5 files changed, 88 insertions(+), 36 deletions(-)
 create mode 100644 include/trace/events/syscalls.h

(limited to 'arch/x86')

diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c
index c05b44b80c23..f3ddd7ac06c5 100644
--- a/arch/s390/kernel/ptrace.c
+++ b/arch/s390/kernel/ptrace.c
@@ -51,8 +51,8 @@
 #include "compat_ptrace.h"
 #endif
 
-DEFINE_TRACE_FN(syscall_enter, syscall_regfunc, syscall_unregfunc);
-DEFINE_TRACE_FN(syscall_exit, syscall_regfunc, syscall_unregfunc);
+#define CREATE_TRACE_POINTS
+#include <trace/events/syscalls.h>
 
 enum s390_regset {
 	REGSET_GENERAL,
@@ -665,7 +665,7 @@ asmlinkage long do_syscall_trace_enter(struct pt_regs *regs)
 	}
 
 	if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT)))
-		trace_syscall_enter(regs, regs->gprs[2]);
+		trace_sys_enter(regs, regs->gprs[2]);
 
 	if (unlikely(current->audit_context))
 		audit_syscall_entry(is_compat_task() ?
@@ -683,7 +683,7 @@ asmlinkage void do_syscall_trace_exit(struct pt_regs *regs)
 				   regs->gprs[2]);
 
 	if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT)))
-		trace_syscall_exit(regs, regs->gprs[2]);
+		trace_sys_exit(regs, regs->gprs[2]);
 
 	if (test_thread_flag(TIF_SYSCALL_TRACE))
 		tracehook_report_syscall_exit(regs, 0);
diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index 31e9b97ec4d6..8d7d5c9c1be3 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -35,13 +35,11 @@
 #include <asm/proto.h>
 #include <asm/ds.h>
 
-#include <trace/syscall.h>
-
-DEFINE_TRACE_FN(syscall_enter, syscall_regfunc, syscall_unregfunc);
-DEFINE_TRACE_FN(syscall_exit, syscall_regfunc, syscall_unregfunc);
-
 #include "tls.h"
 
+#define CREATE_TRACE_POINTS
+#include <trace/events/syscalls.h>
+
 enum x86_regset {
 	REGSET_GENERAL,
 	REGSET_FP,
@@ -1501,7 +1499,7 @@ asmregparm long syscall_trace_enter(struct pt_regs *regs)
 		ret = -1L;
 
 	if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT)))
-		trace_syscall_enter(regs, regs->orig_ax);
+		trace_sys_enter(regs, regs->orig_ax);
 
 	if (unlikely(current->audit_context)) {
 		if (IS_IA32)
@@ -1527,7 +1525,7 @@ asmregparm void syscall_trace_leave(struct pt_regs *regs)
 		audit_syscall_exit(AUDITSC_RESULT(regs->ax), regs->ax);
 
 	if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT)))
-		trace_syscall_exit(regs, regs->ax);
+		trace_sys_exit(regs, regs->ax);
 
 	if (test_thread_flag(TIF_SYSCALL_TRACE))
 		tracehook_report_syscall_exit(regs, 0);
diff --git a/include/trace/events/syscalls.h b/include/trace/events/syscalls.h
new file mode 100644
index 000000000000..397dff2dbd5a
--- /dev/null
+++ b/include/trace/events/syscalls.h
@@ -0,0 +1,70 @@
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM syscalls
+
+#if !defined(_TRACE_EVENTS_SYSCALLS_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_EVENTS_SYSCALLS_H
+
+#include <linux/tracepoint.h>
+
+#include <asm/ptrace.h>
+#include <asm/syscall.h>
+
+
+#ifdef CONFIG_HAVE_SYSCALL_TRACEPOINTS
+
+extern void syscall_regfunc(void);
+extern void syscall_unregfunc(void);
+
+TRACE_EVENT_FN(sys_enter,
+
+	TP_PROTO(struct pt_regs *regs, long id),
+
+	TP_ARGS(regs, id),
+
+	TP_STRUCT__entry(
+		__field(	long,		id		)
+		__array(	unsigned long,	args,	6	)
+	),
+
+	TP_fast_assign(
+		__entry->id	= id;
+		syscall_get_arguments(current, regs, 0, 6, __entry->args);
+	),
+
+	TP_printk("NR %ld (%lx, %lx, %lx, %lx, %lx, %lx)",
+		  __entry->id,
+		  __entry->args[0], __entry->args[1], __entry->args[2],
+		  __entry->args[3], __entry->args[4], __entry->args[5]),
+
+	syscall_regfunc, syscall_unregfunc
+);
+
+TRACE_EVENT_FN(sys_exit,
+
+	TP_PROTO(struct pt_regs *regs, long ret),
+
+	TP_ARGS(regs, ret),
+
+	TP_STRUCT__entry(
+		__field(	long,	id	)
+		__field(	long,	ret	)
+	),
+
+	TP_fast_assign(
+		__entry->id	= syscall_get_nr(current, regs);
+		__entry->ret	= ret;
+	),
+
+	TP_printk("NR %ld = %ld",
+		  __entry->id, __entry->ret),
+
+	syscall_regfunc, syscall_unregfunc
+);
+
+#endif /* CONFIG_HAVE_SYSCALL_TRACEPOINTS */
+
+#endif /* _TRACE_EVENTS_SYSCALLS_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
+
diff --git a/include/trace/syscall.h b/include/trace/syscall.h
index 4e1943001854..5dc283ba5ae0 100644
--- a/include/trace/syscall.h
+++ b/include/trace/syscall.h
@@ -8,23 +8,6 @@
 #include <asm/ptrace.h>
 
 
-#ifdef CONFIG_HAVE_SYSCALL_TRACEPOINTS
-
-extern void syscall_regfunc(void);
-extern void syscall_unregfunc(void);
-
-DECLARE_TRACE(syscall_enter,
-	TP_PROTO(struct pt_regs *regs, long id),
-	TP_ARGS(regs, id)
-);
-
-DECLARE_TRACE(syscall_exit,
-	TP_PROTO(struct pt_regs *regs, long ret),
-	TP_ARGS(regs, ret)
-);
-
-#endif
-
 /*
  * A syscall entry in the ftrace syscalls array.
  *
diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c
index 46c1b977a2cb..2698fe401ebd 100644
--- a/kernel/trace/trace_syscalls.c
+++ b/kernel/trace/trace_syscalls.c
@@ -1,4 +1,5 @@
 #include <trace/syscall.h>
+#include <trace/events/syscalls.h>
 #include <linux/kernel.h>
 #include <linux/ftrace.h>
 #include <linux/perf_counter.h>
@@ -286,7 +287,7 @@ int reg_event_syscall_enter(void *ptr)
 		return -ENOSYS;
 	mutex_lock(&syscall_trace_lock);
 	if (!sys_refcount_enter)
-		ret = register_trace_syscall_enter(ftrace_syscall_enter);
+		ret = register_trace_sys_enter(ftrace_syscall_enter);
 	if (ret) {
 		pr_info("event trace: Could not activate"
 				"syscall entry trace point");
@@ -311,7 +312,7 @@ void unreg_event_syscall_enter(void *ptr)
 	sys_refcount_enter--;
 	clear_bit(num, enabled_enter_syscalls);
 	if (!sys_refcount_enter)
-		unregister_trace_syscall_enter(ftrace_syscall_enter);
+		unregister_trace_sys_enter(ftrace_syscall_enter);
 	mutex_unlock(&syscall_trace_lock);
 }
 
@@ -327,7 +328,7 @@ int reg_event_syscall_exit(void *ptr)
 		return -ENOSYS;
 	mutex_lock(&syscall_trace_lock);
 	if (!sys_refcount_exit)
-		ret = register_trace_syscall_exit(ftrace_syscall_exit);
+		ret = register_trace_sys_exit(ftrace_syscall_exit);
 	if (ret) {
 		pr_info("event trace: Could not activate"
 				"syscall exit trace point");
@@ -352,7 +353,7 @@ void unreg_event_syscall_exit(void *ptr)
 	sys_refcount_exit--;
 	clear_bit(num, enabled_exit_syscalls);
 	if (!sys_refcount_exit)
-		unregister_trace_syscall_exit(ftrace_syscall_exit);
+		unregister_trace_sys_exit(ftrace_syscall_exit);
 	mutex_unlock(&syscall_trace_lock);
 }
 
@@ -418,7 +419,7 @@ int reg_prof_syscall_enter(char *name)
 
 	mutex_lock(&syscall_trace_lock);
 	if (!sys_prof_refcount_enter)
-		ret = register_trace_syscall_enter(prof_syscall_enter);
+		ret = register_trace_sys_enter(prof_syscall_enter);
 	if (ret) {
 		pr_info("event trace: Could not activate"
 				"syscall entry trace point");
@@ -442,7 +443,7 @@ void unreg_prof_syscall_enter(char *name)
 	sys_prof_refcount_enter--;
 	clear_bit(num, enabled_prof_enter_syscalls);
 	if (!sys_prof_refcount_enter)
-		unregister_trace_syscall_enter(prof_syscall_enter);
+		unregister_trace_sys_enter(prof_syscall_enter);
 	mutex_unlock(&syscall_trace_lock);
 }
 
@@ -479,7 +480,7 @@ int reg_prof_syscall_exit(char *name)
 
 	mutex_lock(&syscall_trace_lock);
 	if (!sys_prof_refcount_exit)
-		ret = register_trace_syscall_exit(prof_syscall_exit);
+		ret = register_trace_sys_exit(prof_syscall_exit);
 	if (ret) {
 		pr_info("event trace: Could not activate"
 				"syscall entry trace point");
@@ -503,7 +504,7 @@ void unreg_prof_syscall_exit(char *name)
 	sys_prof_refcount_exit--;
 	clear_bit(num, enabled_prof_exit_syscalls);
 	if (!sys_prof_refcount_exit)
-		unregister_trace_syscall_exit(prof_syscall_exit);
+		unregister_trace_sys_exit(prof_syscall_exit);
 	mutex_unlock(&syscall_trace_lock);
 }
 
-- 
cgit v1.2.3


From ab94fcf528d127fcb490175512a8910f37e5b346 Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Tue, 25 Aug 2009 16:47:16 -0700
Subject: x86: allow "=rm" in native_save_fl()

This is a partial revert of f1f029c7bfbf4ee1918b90a431ab823bed812504.

"=rm" is allowed in this context, because "pop" is explicitly defined
to adjust the stack pointer *before* it evaluates its effective
address, if it has one.  Thus, we do end up writing to the correct
address even if we use an on-stack memory argument.

The original reporter for f1f029c7bfbf4ee1918b90a431ab823bed812504 was
apparently using a broken x86 simulator.

[ Impact: performance ]

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Cc: Gabe Black <spamforgabe@umich.edu>
---
 arch/x86/include/asm/irqflags.h | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/irqflags.h b/arch/x86/include/asm/irqflags.h
index c6ccbe7e81ad..9e2b952f810a 100644
--- a/arch/x86/include/asm/irqflags.h
+++ b/arch/x86/include/asm/irqflags.h
@@ -13,14 +13,13 @@ static inline unsigned long native_save_fl(void)
 	unsigned long flags;
 
 	/*
-	 * Note: this needs to be "=r" not "=rm", because we have the
-	 * stack offset from what gcc expects at the time the "pop" is
-	 * executed, and so a memory reference with respect to the stack
-	 * would end up using the wrong address.
+	 * "=rm" is safe here, because "pop" adjusts the stack before
+	 * it evaluates its effective address -- this is part of the
+	 * documented behavior of the "pop" instruction.
 	 */
 	asm volatile("# __raw_save_flags\n\t"
 		     "pushf ; pop %0"
-		     : "=r" (flags)
+		     : "=rm" (flags)
 		     : /* no input */
 		     : "memory");
 
-- 
cgit v1.2.3


From 8f3e1df48baf728bbb0f242c9dff9c9d7108218a Mon Sep 17 00:00:00 2001
From: Cyrill Gorcunov <gorcunov@openvz.org>
Date: Mon, 24 Aug 2009 21:53:36 +0400
Subject: x86, ioapic: Define IO_APIC_DEFAULT_PHYS_BASE constant

We already have APIC_DEFAULT_PHYS_BASE so just to be
consistent.

Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
LKML-Reference: <20090824175550.927946757@openvz.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/include/asm/apicdef.h |  3 ++-
 arch/x86/kernel/mpparse.c      | 10 +++++-----
 2 files changed, 7 insertions(+), 6 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/apicdef.h b/arch/x86/include/asm/apicdef.h
index 7ddb36ab933b..7386bfa4f4bc 100644
--- a/arch/x86/include/asm/apicdef.h
+++ b/arch/x86/include/asm/apicdef.h
@@ -8,7 +8,8 @@
  * Ingo Molnar <mingo@redhat.com>, 1999, 2000
  */
 
-#define	APIC_DEFAULT_PHYS_BASE	0xfee00000
+#define IO_APIC_DEFAULT_PHYS_BASE	0xfec00000
+#define	APIC_DEFAULT_PHYS_BASE		0xfee00000
 
 #define	APIC_ID		0x20
 
diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c
index 651c93b28862..fcd513bf2846 100644
--- a/arch/x86/kernel/mpparse.c
+++ b/arch/x86/kernel/mpparse.c
@@ -482,11 +482,11 @@ static void __init construct_ioapic_table(int mpc_default_type)
 		MP_bus_info(&bus);
 	}
 
-	ioapic.type = MP_IOAPIC;
-	ioapic.apicid = 2;
-	ioapic.apicver = mpc_default_type > 4 ? 0x10 : 0x01;
-	ioapic.flags = MPC_APIC_USABLE;
-	ioapic.apicaddr = 0xFEC00000;
+	ioapic.type	= MP_IOAPIC;
+	ioapic.apicid	= 2;
+	ioapic.apicver	= mpc_default_type > 4 ? 0x10 : 0x01;
+	ioapic.flags	= MPC_APIC_USABLE;
+	ioapic.apicaddr	= IO_APIC_DEFAULT_PHYS_BASE;
 	MP_ioapic_info(&ioapic);
 
 	/*
-- 
cgit v1.2.3


From ffc438366c2660a6a811b94ba33229bf217f8254 Mon Sep 17 00:00:00 2001
From: Cyrill Gorcunov <gorcunov@openvz.org>
Date: Mon, 24 Aug 2009 21:53:39 +0400
Subject: x86, ioapic: Get rid of needless check and simplify
 ioapic_setup_resources()

alloc_bootmem() already panics on allocation failure. There is
no need to check the result.

Also there is a way to unbind global variable from its body and
use it as a parameter which allow us to simplify
ioapic_init_mappings as well -- "for" cycle already uses
nr_ioapics as a conditional variable and there is no need to
check if ioapic_setup_resources was returning NULL again.

Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <20090824175551.493629148@openvz.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic/io_apic.c | 26 +++++++++++---------------
 1 file changed, 11 insertions(+), 15 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 2999f3dd5889..d836b4d347e6 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -4053,7 +4053,7 @@ void __init setup_ioapic_dest(void)
 
 static struct resource *ioapic_resources;
 
-static struct resource * __init ioapic_setup_resources(void)
+static struct resource * __init ioapic_setup_resources(int nr_ioapics)
 {
 	unsigned long n;
 	struct resource *res;
@@ -4069,15 +4069,13 @@ static struct resource * __init ioapic_setup_resources(void)
 	mem = alloc_bootmem(n);
 	res = (void *)mem;
 
-	if (mem != NULL) {
-		mem += sizeof(struct resource) * nr_ioapics;
+	mem += sizeof(struct resource) * nr_ioapics;
 
-		for (i = 0; i < nr_ioapics; i++) {
-			res[i].name = mem;
-			res[i].flags = IORESOURCE_MEM | IORESOURCE_BUSY;
-			sprintf(mem,  "IOAPIC %u", i);
-			mem += IOAPIC_RESOURCE_NAME_SIZE;
-		}
+	for (i = 0; i < nr_ioapics; i++) {
+		res[i].name = mem;
+		res[i].flags = IORESOURCE_MEM | IORESOURCE_BUSY;
+		sprintf(mem,  "IOAPIC %u", i);
+		mem += IOAPIC_RESOURCE_NAME_SIZE;
 	}
 
 	ioapic_resources = res;
@@ -4091,7 +4089,7 @@ void __init ioapic_init_mappings(void)
 	struct resource *ioapic_res;
 	int i;
 
-	ioapic_res = ioapic_setup_resources();
+	ioapic_res = ioapic_setup_resources(nr_ioapics);
 	for (i = 0; i < nr_ioapics; i++) {
 		if (smp_found_config) {
 			ioapic_phys = mp_ioapics[i].apicaddr;
@@ -4120,11 +4118,9 @@ fake_ioapic_page:
 			    __fix_to_virt(idx), ioapic_phys);
 		idx++;
 
-		if (ioapic_res != NULL) {
-			ioapic_res->start = ioapic_phys;
-			ioapic_res->end = ioapic_phys + (4 * 1024) - 1;
-			ioapic_res++;
-		}
+		ioapic_res->start = ioapic_phys;
+		ioapic_res->end = ioapic_phys + (4 * 1024) - 1;
+		ioapic_res++;
 	}
 }
 
-- 
cgit v1.2.3


From 5051fd69773d2d044734b78516317a04d3774871 Mon Sep 17 00:00:00 2001
From: Cyrill Gorcunov <gorcunov@openvz.org>
Date: Mon, 24 Aug 2009 21:53:37 +0400
Subject: x86, e820: Guard against array overflowed in __e820_add_region()

Better to be paranoid against unpredicted nr_map modifications.

Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
LKML-Reference: <20090824175551.146070377@openvz.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/e820.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
index 7271fa33d791..2e5e0faa99b5 100644
--- a/arch/x86/kernel/e820.c
+++ b/arch/x86/kernel/e820.c
@@ -115,7 +115,7 @@ static void __init __e820_add_region(struct e820map *e820x, u64 start, u64 size,
 {
 	int x = e820x->nr_map;
 
-	if (x == ARRAY_SIZE(e820x->map)) {
+	if (x >= ARRAY_SIZE(e820x->map)) {
 		printk(KERN_ERR "Ooops! Too many entries in the memory map!\n");
 		return;
 	}
-- 
cgit v1.2.3


From d3a247bfb2c26f5b67367d58af7ad8c2efbbc6c1 Mon Sep 17 00:00:00 2001
From: Cyrill Gorcunov <gorcunov@openvz.org>
Date: Wed, 26 Aug 2009 21:13:24 +0400
Subject: x86, apic: Slim down stack usage in early_init_lapic_mapping()

As far as I see there is no external poking of mp_lapic_addr in
this procedure which could lead to unpredited changes and
require local storage unit for it. Lets use it plain forward.

Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
Cc: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <20090826171324.GC4548@lenovo>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic/apic.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c
index 3fc3a6c428da..159740decc41 100644
--- a/arch/x86/kernel/apic/apic.c
+++ b/arch/x86/kernel/apic/apic.c
@@ -1562,8 +1562,6 @@ no_apic:
 #ifdef CONFIG_X86_64
 void __init early_init_lapic_mapping(void)
 {
-	unsigned long phys_addr;
-
 	/*
 	 * If no local APIC can be found then go out
 	 * : it means there is no mpatable and MADT
@@ -1571,11 +1569,9 @@ void __init early_init_lapic_mapping(void)
 	if (!smp_found_config)
 		return;
 
-	phys_addr = mp_lapic_addr;
-
-	set_fixmap_nocache(FIX_APIC_BASE, phys_addr);
+	set_fixmap_nocache(FIX_APIC_BASE, mp_lapic_addr);
 	apic_printk(APIC_VERBOSE, "mapped APIC to %16lx (%16lx)\n",
-		    APIC_BASE, phys_addr);
+		    APIC_BASE, mp_lapic_addr);
 
 	/*
 	 * Fetch the APIC ID of the BSP in case we have a
-- 
cgit v1.2.3


From dd86dda24cc1dc70031a7d9250dc3c0c430a50e2 Mon Sep 17 00:00:00 2001
From: Jason Baron <jbaron@redhat.com>
Date: Mon, 24 Aug 2009 17:40:14 -0400
Subject: tracing: Define NR_syscalls for x86 (32)

Add a NR_syscalls #define for x86. This is used in the syscall events
tracing code. Todo: make it dynamic like x86_64.

NR_syscalls is the usual name used to determine the number of syscalls
supported by the current arch. We want to unify the use of this number
across archs that support the syscall tracing. This also prepare to move
some of the arch code to core code in the syscall tracing area.

Signed-off-by: Jason Baron <jbaron@redhat.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: Jiaying Zhang <jiayingz@google.com>
Cc: Martin Bligh <mbligh@google.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Josh Stone <jistone@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: H. Peter Anwin <hpa@zytor.com>
Cc: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
LKML-Reference: <0f33c0f96d198fccc3ddd9ff7f5334ff5cb42706.1251146513.git.jbaron@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
---
 arch/x86/include/asm/unistd_32.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/unistd_32.h b/arch/x86/include/asm/unistd_32.h
index 732a30706153..8deaada61bc8 100644
--- a/arch/x86/include/asm/unistd_32.h
+++ b/arch/x86/include/asm/unistd_32.h
@@ -345,6 +345,8 @@
 
 #ifdef __KERNEL__
 
+#define NR_syscalls 337
+
 #define __ARCH_WANT_IPC_PARSE_VERSION
 #define __ARCH_WANT_OLD_READDIR
 #define __ARCH_WANT_OLD_STAT
-- 
cgit v1.2.3


From a5a2f8e2acb991327952c45a13f5441fc09dffd6 Mon Sep 17 00:00:00 2001
From: Jason Baron <jbaron@redhat.com>
Date: Wed, 26 Aug 2009 12:09:10 -0400
Subject: tracing: Define NR_syscalls for x86_64

Express the available number of syscalls in a standard way by defining
NR_syscalls.

The common way to define it is to place its definition in asm/unistd.h
However, the number of syscalls is defined using __NR_syscall_max in
x86-64 after building a dynamic header file "asm-offsets.h"

The source file that generates this header, asm-offsets-64.c includes
unistd.h, then if we want to express NR_syscalls from __NR_syscall_max
in unistd.h only after generating the dynamic header file, we need a
watchguard.

If unistd.h is included from asm-offsets-64.c, then we are generating
asm-offset.h which defines __NR_syscall_max. At this time, we don't
want to (we can't) define NR_syscalls, then we do nothing.
Otherwise we define NR_syscalls because we know asm-offsets.h has
been generated.

Signed-off-by: Jason Baron <jbaron@redhat.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: Jiaying Zhang <jiayingz@google.com>
Cc: Martin Bligh <mbligh@google.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Josh Stone <jistone@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: H. Peter Anwin <hpa@zytor.com>
Cc: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
LKML-Reference: <20090826160910.GB2658@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
---
 arch/x86/include/asm/unistd_64.h | 6 ++++++
 arch/x86/kernel/asm-offsets_64.c | 1 +
 2 files changed, 7 insertions(+)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/unistd_64.h b/arch/x86/include/asm/unistd_64.h
index 900e1617e672..b9f3c60de5f7 100644
--- a/arch/x86/include/asm/unistd_64.h
+++ b/arch/x86/include/asm/unistd_64.h
@@ -688,6 +688,12 @@ __SYSCALL(__NR_perf_counter_open, sys_perf_counter_open)
 #endif	/* __NO_STUBS */
 
 #ifdef __KERNEL__
+
+#ifndef COMPILE_OFFSETS
+#include <asm/asm-offsets.h>
+#define NR_syscalls (__NR_syscall_max + 1)
+#endif
+
 /*
  * "Conditional" syscalls
  *
diff --git a/arch/x86/kernel/asm-offsets_64.c b/arch/x86/kernel/asm-offsets_64.c
index 898ecc47e129..4a6aeedcd965 100644
--- a/arch/x86/kernel/asm-offsets_64.c
+++ b/arch/x86/kernel/asm-offsets_64.c
@@ -3,6 +3,7 @@
  * This code generates raw asm output which is post-processed to extract
  * and format the required data.
  */
+#define COMPILE_OFFSETS
 
 #include <linux/crypto.h>
 #include <linux/sched.h> 
-- 
cgit v1.2.3


From 57421dbbdc932d65f0e6a41ebb027a2bfe3d0669 Mon Sep 17 00:00:00 2001
From: Jason Baron <jbaron@redhat.com>
Date: Mon, 24 Aug 2009 17:40:22 -0400
Subject: tracing: Convert event tracing code to use NR_syscalls

Convert the syscalls event tracing code to use NR_syscalls, instead of
FTRACE_SYSCALL_MAX. NR_syscalls is standard accross most arches, and
reduces code confusion/complexity.

Signed-off-by: Jason Baron <jbaron@redhat.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: Jiaying Zhang <jiayingz@google.com>
Cc: Martin Bligh <mbligh@google.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Josh Stone <jistone@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: H. Peter Anwin <hpa@zytor.com>
Cc: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
LKML-Reference: <9b4f1a84ecae57cc6599412772efa36f0d2b815b.1251146513.git.jbaron@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
---
 arch/x86/kernel/ftrace.c      |  8 ++++----
 kernel/trace/trace_syscalls.c | 24 ++++++++++++------------
 2 files changed, 16 insertions(+), 16 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c
index 3cff1214e176..9dbb527e1652 100644
--- a/arch/x86/kernel/ftrace.c
+++ b/arch/x86/kernel/ftrace.c
@@ -494,7 +494,7 @@ static struct syscall_metadata *find_syscall_meta(unsigned long *syscall)
 
 struct syscall_metadata *syscall_nr_to_meta(int nr)
 {
-	if (!syscalls_metadata || nr >= FTRACE_SYSCALL_MAX || nr < 0)
+	if (!syscalls_metadata || nr >= NR_syscalls || nr < 0)
 		return NULL;
 
 	return syscalls_metadata[nr];
@@ -507,7 +507,7 @@ int syscall_name_to_nr(char *name)
 	if (!syscalls_metadata)
 		return -1;
 
-	for (i = 0; i < FTRACE_SYSCALL_MAX; i++) {
+	for (i = 0; i < NR_syscalls; i++) {
 		if (syscalls_metadata[i]) {
 			if (!strcmp(syscalls_metadata[i]->name, name))
 				return i;
@@ -533,13 +533,13 @@ static int __init arch_init_ftrace_syscalls(void)
 	unsigned long **psys_syscall_table = &sys_call_table;
 
 	syscalls_metadata = kzalloc(sizeof(*syscalls_metadata) *
-					FTRACE_SYSCALL_MAX, GFP_KERNEL);
+					NR_syscalls, GFP_KERNEL);
 	if (!syscalls_metadata) {
 		WARN_ON(1);
 		return -ENOMEM;
 	}
 
-	for (i = 0; i < FTRACE_SYSCALL_MAX; i++) {
+	for (i = 0; i < NR_syscalls; i++) {
 		meta = find_syscall_meta(psys_syscall_table[i]);
 		syscalls_metadata[i] = meta;
 	}
diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c
index cb7f600cb02a..4f5fae6fad90 100644
--- a/kernel/trace/trace_syscalls.c
+++ b/kernel/trace/trace_syscalls.c
@@ -11,8 +11,8 @@
 static DEFINE_MUTEX(syscall_trace_lock);
 static int sys_refcount_enter;
 static int sys_refcount_exit;
-static DECLARE_BITMAP(enabled_enter_syscalls, FTRACE_SYSCALL_MAX);
-static DECLARE_BITMAP(enabled_exit_syscalls, FTRACE_SYSCALL_MAX);
+static DECLARE_BITMAP(enabled_enter_syscalls, NR_syscalls);
+static DECLARE_BITMAP(enabled_exit_syscalls, NR_syscalls);
 
 enum print_line_t
 print_syscall_enter(struct trace_iterator *iter, int flags)
@@ -289,7 +289,7 @@ int reg_event_syscall_enter(void *ptr)
 
 	name = (char *)ptr;
 	num = syscall_name_to_nr(name);
-	if (num < 0 || num >= FTRACE_SYSCALL_MAX)
+	if (num < 0 || num >= NR_syscalls)
 		return -ENOSYS;
 	mutex_lock(&syscall_trace_lock);
 	if (!sys_refcount_enter)
@@ -312,7 +312,7 @@ void unreg_event_syscall_enter(void *ptr)
 
 	name = (char *)ptr;
 	num = syscall_name_to_nr(name);
-	if (num < 0 || num >= FTRACE_SYSCALL_MAX)
+	if (num < 0 || num >= NR_syscalls)
 		return;
 	mutex_lock(&syscall_trace_lock);
 	sys_refcount_enter--;
@@ -330,7 +330,7 @@ int reg_event_syscall_exit(void *ptr)
 
 	name = (char *)ptr;
 	num = syscall_name_to_nr(name);
-	if (num < 0 || num >= FTRACE_SYSCALL_MAX)
+	if (num < 0 || num >= NR_syscalls)
 		return -ENOSYS;
 	mutex_lock(&syscall_trace_lock);
 	if (!sys_refcount_exit)
@@ -353,7 +353,7 @@ void unreg_event_syscall_exit(void *ptr)
 
 	name = (char *)ptr;
 	num = syscall_name_to_nr(name);
-	if (num < 0 || num >= FTRACE_SYSCALL_MAX)
+	if (num < 0 || num >= NR_syscalls)
 		return;
 	mutex_lock(&syscall_trace_lock);
 	sys_refcount_exit--;
@@ -373,8 +373,8 @@ struct trace_event event_syscall_exit = {
 
 #ifdef CONFIG_EVENT_PROFILE
 
-static DECLARE_BITMAP(enabled_prof_enter_syscalls, FTRACE_SYSCALL_MAX);
-static DECLARE_BITMAP(enabled_prof_exit_syscalls, FTRACE_SYSCALL_MAX);
+static DECLARE_BITMAP(enabled_prof_enter_syscalls, NR_syscalls);
+static DECLARE_BITMAP(enabled_prof_exit_syscalls, NR_syscalls);
 static int sys_prof_refcount_enter;
 static int sys_prof_refcount_exit;
 
@@ -420,7 +420,7 @@ int reg_prof_syscall_enter(char *name)
 	int num;
 
 	num = syscall_name_to_nr(name);
-	if (num < 0 || num >= FTRACE_SYSCALL_MAX)
+	if (num < 0 || num >= NR_syscalls)
 		return -ENOSYS;
 
 	mutex_lock(&syscall_trace_lock);
@@ -442,7 +442,7 @@ void unreg_prof_syscall_enter(char *name)
 	int num;
 
 	num = syscall_name_to_nr(name);
-	if (num < 0 || num >= FTRACE_SYSCALL_MAX)
+	if (num < 0 || num >= NR_syscalls)
 		return;
 
 	mutex_lock(&syscall_trace_lock);
@@ -481,7 +481,7 @@ int reg_prof_syscall_exit(char *name)
 	int num;
 
 	num = syscall_name_to_nr(name);
-	if (num < 0 || num >= FTRACE_SYSCALL_MAX)
+	if (num < 0 || num >= NR_syscalls)
 		return -ENOSYS;
 
 	mutex_lock(&syscall_trace_lock);
@@ -503,7 +503,7 @@ void unreg_prof_syscall_exit(char *name)
 	int num;
 
 	num = syscall_name_to_nr(name);
-	if (num < 0 || num >= FTRACE_SYSCALL_MAX)
+	if (num < 0 || num >= NR_syscalls)
 		return;
 
 	mutex_lock(&syscall_trace_lock);
-- 
cgit v1.2.3


From 117226d15850387b55fd01675917ee4fcb9699e8 Mon Sep 17 00:00:00 2001
From: Jason Baron <jbaron@redhat.com>
Date: Mon, 24 Aug 2009 17:40:26 -0400
Subject: tracing: Remove FTRACE_SYSCALL_MAX definitions

Remove the FTRACE_SYSCALL_MAX definitions now that we have converted the
syscall event tracing code to use NR_syscalls.

Signed-off-by: Jason Baron <jbaron@redhat.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: Jiaying Zhang <jiayingz@google.com>
Cc: Martin Bligh <mbligh@google.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Josh Stone <jistone@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: H. Peter Anwin <hpa@zytor.com>
Cc: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
LKML-Reference: <f2240cdc8f0b1ca7617390c8f5ec90ba2bd348cf.1251146513.git.jbaron@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
---
 arch/x86/include/asm/ftrace.h | 7 -------
 1 file changed, 7 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/ftrace.h b/arch/x86/include/asm/ftrace.h
index 71136545187a..db24c2278be0 100644
--- a/arch/x86/include/asm/ftrace.h
+++ b/arch/x86/include/asm/ftrace.h
@@ -28,13 +28,6 @@
 
 #endif
 
-/* FIXME: I don't want to stay hardcoded */
-#ifdef CONFIG_X86_64
-# define FTRACE_SYSCALL_MAX     299
-#else
-# define FTRACE_SYSCALL_MAX     337
-#endif
-
 #ifdef CONFIG_FUNCTION_TRACER
 #define MCOUNT_ADDR		((long)(mcount))
 #define MCOUNT_INSN_SIZE	5 /* sizeof mcount call */
-- 
cgit v1.2.3


From 2a4ab640d3c28c2952967e5f63ea495555bf2a5f Mon Sep 17 00:00:00 2001
From: Feng Tang <feng.tang@intel.com>
Date: Tue, 7 Jul 2009 23:01:15 -0400
Subject: ACPI, x86: expose some IO-APIC routines when CONFIG_ACPI=n

Some IO-APIC routines are ACPI specific now, but need to
be exposed when CONFIG_ACPI=n for the benefit of SFI.

Remove #ifdef ACPI around these routines:

io_apic_get_unique_id(int ioapic, int apic_id);
io_apic_get_version(int ioapic);
io_apic_get_redir_entries(int ioapic);

Move these routines from ACPI-specific boot.c to io_apic.c:

uniq_ioapic_id(u8 id)
mp_find_ioapic()
mp_find_ioapic_pin()
mp_register_ioapic()

Also, since uniq_ioapic_id() is now no longer static,
re-name it to io_apic_unique_id() for consistency
with the other public io_apic routines.

For simplicity, do not #ifdef the resulting code ACPI || SFI,
thought that could be done in the future if it is important
to optimize the !ACPI !SFI IO-APIC x86 kernel for size.

Signed-off-by: Feng Tang <feng.tang@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Cc: x86@kernel.org
---
 arch/x86/include/asm/io_apic.h |  13 +++++-
 arch/x86/kernel/acpi/boot.c    | 102 +---------------------------------------
 arch/x86/kernel/apic/io_apic.c | 103 ++++++++++++++++++++++++++++++++++++++---
 3 files changed, 109 insertions(+), 109 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/io_apic.h b/arch/x86/include/asm/io_apic.h
index 330ee807f89e..85232d32fcb8 100644
--- a/arch/x86/include/asm/io_apic.h
+++ b/arch/x86/include/asm/io_apic.h
@@ -150,11 +150,10 @@ extern int timer_through_8259;
 #define io_apic_assign_pci_irqs \
 	(mp_irq_entries && !skip_ioapic_setup && io_apic_irqs)
 
-#ifdef CONFIG_ACPI
+extern u8 io_apic_unique_id(u8 id);
 extern int io_apic_get_unique_id(int ioapic, int apic_id);
 extern int io_apic_get_version(int ioapic);
 extern int io_apic_get_redir_entries(int ioapic);
-#endif /* CONFIG_ACPI */
 
 struct io_apic_irq_attr;
 extern int io_apic_set_pci_routing(struct device *dev, int irq,
@@ -177,6 +176,16 @@ extern int setup_ioapic_entry(int apic, int irq,
 			      int polarity, int vector, int pin);
 extern void ioapic_write_entry(int apic, int pin,
 			       struct IO_APIC_route_entry e);
+
+struct mp_ioapic_gsi{
+	int gsi_base;
+	int gsi_end;
+};
+extern struct mp_ioapic_gsi  mp_gsi_routing[];
+int mp_find_ioapic(int gsi);
+int mp_find_ioapic_pin(int ioapic, int gsi);
+void __init mp_register_ioapic(int id, u32 address, u32 gsi_base);
+
 #else  /* !CONFIG_X86_IO_APIC */
 #define io_apic_assign_pci_irqs 0
 static const int timer_through_8259 = 0;
diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c
index 6b8ca3a0285d..7e62d1eb8e30 100644
--- a/arch/x86/kernel/acpi/boot.c
+++ b/arch/x86/kernel/acpi/boot.c
@@ -833,106 +833,6 @@ static int __init acpi_parse_madt_lapic_entries(void)
 extern int es7000_plat;
 #endif
 
-static struct {
-	int gsi_base;
-	int gsi_end;
-} mp_ioapic_routing[MAX_IO_APICS];
-
-int mp_find_ioapic(int gsi)
-{
-	int i = 0;
-
-	/* Find the IOAPIC that manages this GSI. */
-	for (i = 0; i < nr_ioapics; i++) {
-		if ((gsi >= mp_ioapic_routing[i].gsi_base)
-		    && (gsi <= mp_ioapic_routing[i].gsi_end))
-			return i;
-	}
-
-	printk(KERN_ERR "ERROR: Unable to locate IOAPIC for GSI %d\n", gsi);
-	return -1;
-}
-
-int mp_find_ioapic_pin(int ioapic, int gsi)
-{
-	if (WARN_ON(ioapic == -1))
-		return -1;
-	if (WARN_ON(gsi > mp_ioapic_routing[ioapic].gsi_end))
-		return -1;
-
-	return gsi - mp_ioapic_routing[ioapic].gsi_base;
-}
-
-static u8 __init uniq_ioapic_id(u8 id)
-{
-#ifdef CONFIG_X86_32
-	if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) &&
-	    !APIC_XAPIC(apic_version[boot_cpu_physical_apicid]))
-		return io_apic_get_unique_id(nr_ioapics, id);
-	else
-		return id;
-#else
-	int i;
-	DECLARE_BITMAP(used, 256);
-	bitmap_zero(used, 256);
-	for (i = 0; i < nr_ioapics; i++) {
-		struct mpc_ioapic *ia = &mp_ioapics[i];
-		__set_bit(ia->apicid, used);
-	}
-	if (!test_bit(id, used))
-		return id;
-	return find_first_zero_bit(used, 256);
-#endif
-}
-
-static int bad_ioapic(unsigned long address)
-{
-	if (nr_ioapics >= MAX_IO_APICS) {
-		printk(KERN_ERR "ERROR: Max # of I/O APICs (%d) exceeded "
-		       "(found %d)\n", MAX_IO_APICS, nr_ioapics);
-		panic("Recompile kernel with bigger MAX_IO_APICS!\n");
-	}
-	if (!address) {
-		printk(KERN_ERR "WARNING: Bogus (zero) I/O APIC address"
-		       " found in table, skipping!\n");
-		return 1;
-	}
-	return 0;
-}
-
-void __init mp_register_ioapic(int id, u32 address, u32 gsi_base)
-{
-	int idx = 0;
-
-	if (bad_ioapic(address))
-		return;
-
-	idx = nr_ioapics;
-
-	mp_ioapics[idx].type = MP_IOAPIC;
-	mp_ioapics[idx].flags = MPC_APIC_USABLE;
-	mp_ioapics[idx].apicaddr = address;
-
-	set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address);
-	mp_ioapics[idx].apicid = uniq_ioapic_id(id);
-	mp_ioapics[idx].apicver = io_apic_get_version(idx);
-
-	/*
-	 * Build basic GSI lookup table to facilitate gsi->io_apic lookups
-	 * and to prevent reprogramming of IOAPIC pins (PCI GSIs).
-	 */
-	mp_ioapic_routing[idx].gsi_base = gsi_base;
-	mp_ioapic_routing[idx].gsi_end = gsi_base +
-	    io_apic_get_redir_entries(idx);
-
-	printk(KERN_INFO "IOAPIC[%d]: apic_id %d, version %d, address 0x%x, "
-	       "GSI %d-%d\n", idx, mp_ioapics[idx].apicid,
-	       mp_ioapics[idx].apicver, mp_ioapics[idx].apicaddr,
-	       mp_ioapic_routing[idx].gsi_base, mp_ioapic_routing[idx].gsi_end);
-
-	nr_ioapics++;
-}
-
 int __init acpi_probe_gsi(void)
 {
 	int idx;
@@ -947,7 +847,7 @@ int __init acpi_probe_gsi(void)
 
 	max_gsi = 0;
 	for (idx = 0; idx < nr_ioapics; idx++) {
-		gsi = mp_ioapic_routing[idx].gsi_end;
+		gsi = mp_gsi_routing[idx].gsi_end;
 
 		if (gsi > max_gsi)
 			max_gsi = gsi;
diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index d2ed6c5ddc80..a8c0232b3893 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -85,6 +85,9 @@ int nr_ioapic_registers[MAX_IO_APICS];
 struct mpc_ioapic mp_ioapics[MAX_IO_APICS];
 int nr_ioapics;
 
+/* IO APIC gsi routing info */
+struct mp_ioapic_gsi  mp_gsi_routing[MAX_IO_APICS];
+
 /* MP IRQ source entries */
 struct mpc_intsrc mp_irqs[MAX_IRQ_SOURCES];
 
@@ -3941,11 +3944,28 @@ int io_apic_set_pci_routing(struct device *dev, int irq,
 	return __io_apic_set_pci_routing(dev, irq, irq_attr);
 }
 
-/* --------------------------------------------------------------------------
-                          ACPI-based IOAPIC Configuration
-   -------------------------------------------------------------------------- */
+u8 __init io_apic_unique_id(u8 id)
+{
+#ifdef CONFIG_X86_32
+	if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) &&
+	    !APIC_XAPIC(apic_version[boot_cpu_physical_apicid]))
+		return io_apic_get_unique_id(nr_ioapics, id);
+	else
+		return id;
+#else
+	int i;
+	DECLARE_BITMAP(used, 256);
 
-#ifdef CONFIG_ACPI
+	bitmap_zero(used, 256);
+	for (i = 0; i < nr_ioapics; i++) {
+		struct mpc_ioapic *ia = &mp_ioapics[i];
+		__set_bit(ia->apicid, used);
+	}
+	if (!test_bit(id, used))
+		return id;
+	return find_first_zero_bit(used, 256);
+#endif
+}
 
 #ifdef CONFIG_X86_32
 int __init io_apic_get_unique_id(int ioapic, int apic_id)
@@ -4054,8 +4074,6 @@ int acpi_get_override_irq(int bus_irq, int *trigger, int *polarity)
 	return 0;
 }
 
-#endif /* CONFIG_ACPI */
-
 /*
  * This function currently is only a helper for the i386 smp boot process where
  * we need to reprogram the ioredtbls to cater for the cpus which have come online
@@ -4201,3 +4219,76 @@ void __init ioapic_insert_resources(void)
 		r++;
 	}
 }
+
+int mp_find_ioapic(int gsi)
+{
+	int i = 0;
+
+	/* Find the IOAPIC that manages this GSI. */
+	for (i = 0; i < nr_ioapics; i++) {
+		if ((gsi >= mp_gsi_routing[i].gsi_base)
+		    && (gsi <= mp_gsi_routing[i].gsi_end))
+			return i;
+	}
+
+	printk(KERN_ERR "ERROR: Unable to locate IOAPIC for GSI %d\n", gsi);
+	return -1;
+}
+
+int mp_find_ioapic_pin(int ioapic, int gsi)
+{
+	if (WARN_ON(ioapic == -1))
+		return -1;
+	if (WARN_ON(gsi > mp_gsi_routing[ioapic].gsi_end))
+		return -1;
+
+	return gsi - mp_gsi_routing[ioapic].gsi_base;
+}
+
+static int bad_ioapic(unsigned long address)
+{
+	if (nr_ioapics >= MAX_IO_APICS) {
+		printk(KERN_WARNING "WARING: Max # of I/O APICs (%d) exceeded "
+		       "(found %d), skipping\n", MAX_IO_APICS, nr_ioapics);
+		return 1;
+	}
+	if (!address) {
+		printk(KERN_WARNING "WARNING: Bogus (zero) I/O APIC address"
+		       " found in table, skipping!\n");
+		return 1;
+	}
+	return 0;
+}
+
+void __init mp_register_ioapic(int id, u32 address, u32 gsi_base)
+{
+	int idx = 0;
+
+	if (bad_ioapic(address))
+		return;
+
+	idx = nr_ioapics;
+
+	mp_ioapics[idx].type = MP_IOAPIC;
+	mp_ioapics[idx].flags = MPC_APIC_USABLE;
+	mp_ioapics[idx].apicaddr = address;
+
+	set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address);
+	mp_ioapics[idx].apicid = io_apic_unique_id(id);
+	mp_ioapics[idx].apicver = io_apic_get_version(idx);
+
+	/*
+	 * Build basic GSI lookup table to facilitate gsi->io_apic lookups
+	 * and to prevent reprogramming of IOAPIC pins (PCI GSIs).
+	 */
+	mp_gsi_routing[idx].gsi_base = gsi_base;
+	mp_gsi_routing[idx].gsi_end = gsi_base +
+	    io_apic_get_redir_entries(idx);
+
+	printk(KERN_INFO "IOAPIC[%d]: apic_id %d, version %d, address 0x%x, "
+	       "GSI %d-%d\n", idx, mp_ioapics[idx].apicid,
+	       mp_ioapics[idx].apicver, mp_ioapics[idx].apicaddr,
+	       mp_gsi_routing[idx].gsi_base, mp_gsi_routing[idx].gsi_end);
+
+	nr_ioapics++;
+}
-- 
cgit v1.2.3


From 23386d63bbb3199cf247313ec088878d72debcfd Mon Sep 17 00:00:00 2001
From: Michal Schmidt <mschmidt@redhat.com>
Date: Sat, 29 Aug 2009 18:27:18 +0200
Subject: x86: Detect stack protector for i386 builds on x86_64

Stack protector support was not detected when building with
ARCH=i386 on x86_64 systems:

  arch/x86/Makefile:80: stack protector enabled but no compiler support

The "-m32" argument needs to be passed to the detection script.

Signed-off-by: Michal Schmidt <mschmidt@redhat.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: Arjan van de Ven <arjan@infradead.org>
LKML-Reference: <20090829182718.10f566b1@leela>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
--
---
 arch/x86/Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/Makefile b/arch/x86/Makefile
index 1b68659c41b4..5e7db44d709e 100644
--- a/arch/x86/Makefile
+++ b/arch/x86/Makefile
@@ -72,7 +72,7 @@ endif
 
 ifdef CONFIG_CC_STACKPROTECTOR
 	cc_has_sp := $(srctree)/scripts/gcc-x86_$(BITS)-has-stack-protector.sh
-        ifeq ($(shell $(CONFIG_SHELL) $(cc_has_sp) $(CC)),y)
+        ifeq ($(shell $(CONFIG_SHELL) $(cc_has_sp) $(CC) $(biarch)),y)
                 stackp-y := -fstack-protector
                 stackp-$(CONFIG_CC_STACKPROTECTOR_ALL) += -fstack-protector-all
                 KBUILD_CFLAGS += $(stackp-y)
-- 
cgit v1.2.3


From 132ec92f3f70fe365c1f4b8d46e66cf8a2a16880 Mon Sep 17 00:00:00 2001
From: Borislav Petkov <petkovbb@googlemail.com>
Date: Mon, 31 Aug 2009 09:50:09 +0200
Subject: x86, msr: Add rd/wrmsr interfaces with preset registers

native_{rdmsr,wrmsr}_safe_regs are two new interfaces which allow
presetting of a subset of eight x86 GPRs before executing the rd/wrmsr
instructions. This is needed at least on AMD K8 for accessing an erratum
workaround MSR.

Originally based on an idea by H. Peter Anvin.

Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
LKML-Reference: <1251705011-18636-1-git-send-email-petkovbb@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/include/asm/msr.h      | 13 ++++++
 arch/x86/include/asm/paravirt.h | 16 +++++++
 arch/x86/kernel/paravirt.c      |  2 +
 arch/x86/lib/Makefile           |  1 +
 arch/x86/lib/msr-reg.S          | 98 +++++++++++++++++++++++++++++++++++++++++
 5 files changed, 130 insertions(+)
 create mode 100644 arch/x86/lib/msr-reg.S

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/msr.h b/arch/x86/include/asm/msr.h
index 48ad9d29484a..184d4a113961 100644
--- a/arch/x86/include/asm/msr.h
+++ b/arch/x86/include/asm/msr.h
@@ -113,6 +113,9 @@ notrace static inline int native_write_msr_safe(unsigned int msr,
 
 extern unsigned long long native_read_tsc(void);
 
+extern int native_rdmsr_safe_regs(u32 *regs);
+extern int native_wrmsr_safe_regs(u32 *regs);
+
 static __always_inline unsigned long long __native_read_tsc(void)
 {
 	DECLARE_ARGS(val, low, high);
@@ -189,6 +192,16 @@ static inline int rdmsrl_amd_safe(unsigned msr, unsigned long long *p)
 	return err;
 }
 
+static inline int rdmsr_safe_regs(u32 *regs)
+{
+	return native_rdmsr_safe_regs(regs);
+}
+
+static inline int wrmsr_safe_regs(u32 *regs)
+{
+	return native_wrmsr_safe_regs(regs);
+}
+
 #define rdtscl(low)						\
 	((low) = (u32)__native_read_tsc())
 
diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h
index 4fb37c8a0832..1705944e0374 100644
--- a/arch/x86/include/asm/paravirt.h
+++ b/arch/x86/include/asm/paravirt.h
@@ -168,7 +168,9 @@ struct pv_cpu_ops {
 	   err = 0/-EFAULT.  wrmsr returns 0/-EFAULT. */
 	u64 (*read_msr_amd)(unsigned int msr, int *err);
 	u64 (*read_msr)(unsigned int msr, int *err);
+	int (*rdmsr_regs)(u32 *regs);
 	int (*write_msr)(unsigned int msr, unsigned low, unsigned high);
+	int (*wrmsr_regs)(u32 *regs);
 
 	u64 (*read_tsc)(void);
 	u64 (*read_pmc)(int counter);
@@ -820,6 +822,12 @@ static inline u64 paravirt_read_msr(unsigned msr, int *err)
 {
 	return PVOP_CALL2(u64, pv_cpu_ops.read_msr, msr, err);
 }
+
+static inline int paravirt_rdmsr_regs(u32 *regs)
+{
+	return PVOP_CALL1(int, pv_cpu_ops.rdmsr_regs, regs);
+}
+
 static inline u64 paravirt_read_msr_amd(unsigned msr, int *err)
 {
 	return PVOP_CALL2(u64, pv_cpu_ops.read_msr_amd, msr, err);
@@ -829,6 +837,11 @@ static inline int paravirt_write_msr(unsigned msr, unsigned low, unsigned high)
 	return PVOP_CALL3(int, pv_cpu_ops.write_msr, msr, low, high);
 }
 
+static inline int paravirt_wrmsr_regs(u32 *regs)
+{
+	return PVOP_CALL1(int, pv_cpu_ops.wrmsr_regs, regs);
+}
+
 /* These should all do BUG_ON(_err), but our headers are too tangled. */
 #define rdmsr(msr, val1, val2)			\
 do {						\
@@ -862,6 +875,9 @@ do {						\
 	_err;					\
 })
 
+#define rdmsr_safe_regs(regs)	paravirt_rdmsr_regs(regs)
+#define wrmsr_safe_regs(regs)	paravirt_wrmsr_regs(regs)
+
 static inline int rdmsrl_safe(unsigned msr, unsigned long long *p)
 {
 	int err;
diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c
index 70ec9b951d76..67594af43b38 100644
--- a/arch/x86/kernel/paravirt.c
+++ b/arch/x86/kernel/paravirt.c
@@ -362,8 +362,10 @@ struct pv_cpu_ops pv_cpu_ops = {
 #endif
 	.wbinvd = native_wbinvd,
 	.read_msr = native_read_msr_safe,
+	.rdmsr_regs = native_rdmsr_safe_regs,
 	.read_msr_amd = native_read_msr_amd_safe,
 	.write_msr = native_write_msr_safe,
+	.wrmsr_regs = native_wrmsr_safe_regs,
 	.read_tsc = native_read_tsc,
 	.read_pmc = native_read_pmc,
 	.read_tscp = native_read_tscp,
diff --git a/arch/x86/lib/Makefile b/arch/x86/lib/Makefile
index 07c31899c9c2..b59c0647d809 100644
--- a/arch/x86/lib/Makefile
+++ b/arch/x86/lib/Makefile
@@ -8,6 +8,7 @@ lib-y := delay.o
 lib-y += thunk_$(BITS).o
 lib-y += usercopy_$(BITS).o getuser.o putuser.o
 lib-y += memcpy_$(BITS).o
+lib-y += msr-reg.o
 
 ifeq ($(CONFIG_X86_32),y)
         obj-y += atomic64_32.o
diff --git a/arch/x86/lib/msr-reg.S b/arch/x86/lib/msr-reg.S
new file mode 100644
index 000000000000..51f1bb3f8c79
--- /dev/null
+++ b/arch/x86/lib/msr-reg.S
@@ -0,0 +1,98 @@
+#include <linux/linkage.h>
+#include <linux/errno.h>
+#include <asm/asm.h>
+#include <asm/msr.h>
+
+#ifdef CONFIG_X86_64
+/*
+ * int native_{rdmsr,wrmsr}_safe_regs(u32 gprs[8]);
+ *
+ * reg layout: u32 gprs[eax, ecx, edx, ebx, esp, ebp, esi, edi]
+ *
+ */
+.macro op_safe_regs op:req
+ENTRY(native_\op\()_safe_regs)
+	push    %rbx
+	push    %rbp
+	push    $0              /* Return value */
+	push    %rdi
+	movl    (%rdi), %eax
+	movl    4(%rdi), %ecx
+	movl    8(%rdi), %edx
+	movl    12(%rdi), %ebx
+	movl    20(%rdi), %ebp
+	movl    24(%rdi), %esi
+	movl    28(%rdi), %edi
+1:	\op
+2:	movl    %edi, %r10d
+	pop     %rdi
+	movl    %eax, (%rdi)
+	movl    %ecx, 4(%rdi)
+	movl    %edx, 8(%rdi)
+	movl    %ebx, 12(%rdi)
+	movl    %ebp, 20(%rdi)
+	movl    %esi, 24(%rdi)
+	movl    %r10d, 28(%rdi)
+	pop     %rax
+	pop     %rbp
+	pop     %rbx
+	ret
+3:
+	movq    $-EIO, 8(%rsp)
+	jmp     2b
+	.section __ex_table,"ax"
+	.balign 4
+	.quad   1b, 3b
+	.previous
+ENDPROC(native_\op\()_safe_regs)
+.endm
+
+#else /* X86_32 */
+
+.macro op_safe_regs op:req
+ENTRY(native_\op\()_safe_regs)
+	push    %ebx
+	push    %ebp
+	push    %esi
+	push    %edi
+	push    $0              /* Return value */
+	push    %eax
+	movl    4(%eax), %ecx
+	movl    8(%eax), %edx
+	movl    12(%eax), %ebx
+	movl    20(%eax), %ebp
+	movl    24(%eax), %esi
+	movl    28(%eax), %edi
+	movl    (%eax), %eax
+1:	\op
+2:	push    %eax
+	movl    4(%esp), %eax
+	pop     (%eax)
+	addl    $4, %esp
+	movl    %ecx, 4(%eax)
+	movl    %edx, 8(%eax)
+	movl    %ebx, 12(%eax)
+	movl    %ebp, 20(%eax)
+	movl    %esi, 24(%eax)
+	movl    %edi, 28(%eax)
+	pop     %eax
+	pop     %edi
+	pop     %esi
+	pop     %ebp
+	pop     %ebx
+	ret
+3:
+	movl    $-EIO, 4(%esp)
+	jmp     2b
+	.section __ex_table,"ax"
+	.balign 4
+	.long   1b, 3b
+	.previous
+ENDPROC(native_\op\()_safe_regs)
+.endm
+
+#endif
+
+op_safe_regs rdmsr
+op_safe_regs wrmsr
+
-- 
cgit v1.2.3


From 177fed1ee8d727c39601ce9fc2299b4cb25a718e Mon Sep 17 00:00:00 2001
From: Borislav Petkov <petkovbb@googlemail.com>
Date: Mon, 31 Aug 2009 09:50:10 +0200
Subject: x86, msr: Rewrite AMD rd/wrmsr variants

Switch them to native_{rd,wr}msr_safe_regs and remove
pv_cpu_ops.read_msr_amd.

Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
LKML-Reference: <1251705011-18636-2-git-send-email-petkovbb@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/include/asm/msr.h      | 38 +++++++++++++++++++++-----------------
 arch/x86/include/asm/paravirt.h | 26 ++++++++++++++++++++------
 arch/x86/kernel/paravirt.c      |  1 -
 3 files changed, 41 insertions(+), 24 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/msr.h b/arch/x86/include/asm/msr.h
index 184d4a113961..09c5ca70d49d 100644
--- a/arch/x86/include/asm/msr.h
+++ b/arch/x86/include/asm/msr.h
@@ -71,22 +71,6 @@ static inline unsigned long long native_read_msr_safe(unsigned int msr,
 	return EAX_EDX_VAL(val, low, high);
 }
 
-static inline unsigned long long native_read_msr_amd_safe(unsigned int msr,
-						      int *err)
-{
-	DECLARE_ARGS(val, low, high);
-
-	asm volatile("2: rdmsr ; xor %0,%0\n"
-		     "1:\n\t"
-		     ".section .fixup,\"ax\"\n\t"
-		     "3:  mov %3,%0 ; jmp 1b\n\t"
-		     ".previous\n\t"
-		     _ASM_EXTABLE(2b, 3b)
-		     : "=r" (*err), EAX_EDX_RET(val, low, high)
-		     : "c" (msr), "D" (0x9c5a203a), "i" (-EFAULT));
-	return EAX_EDX_VAL(val, low, high);
-}
-
 static inline void native_write_msr(unsigned int msr,
 				    unsigned low, unsigned high)
 {
@@ -184,14 +168,34 @@ static inline int rdmsrl_safe(unsigned msr, unsigned long long *p)
 	*p = native_read_msr_safe(msr, &err);
 	return err;
 }
+
 static inline int rdmsrl_amd_safe(unsigned msr, unsigned long long *p)
 {
+	u32 gprs[8] = { 0 };
 	int err;
 
-	*p = native_read_msr_amd_safe(msr, &err);
+	gprs[1] = msr;
+	gprs[7] = 0x9c5a203a;
+
+	err = native_rdmsr_safe_regs(gprs);
+
+	*p = gprs[0] | ((u64)gprs[2] << 32);
+
 	return err;
 }
 
+static inline int wrmsrl_amd_safe(unsigned msr, unsigned long long val)
+{
+	u32 gprs[8] = { 0 };
+
+	gprs[0] = (u32)val;
+	gprs[1] = msr;
+	gprs[2] = val >> 32;
+	gprs[7] = 0x9c5a203a;
+
+	return native_wrmsr_safe_regs(gprs);
+}
+
 static inline int rdmsr_safe_regs(u32 *regs)
 {
 	return native_rdmsr_safe_regs(regs);
diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h
index 1705944e0374..11574934a994 100644
--- a/arch/x86/include/asm/paravirt.h
+++ b/arch/x86/include/asm/paravirt.h
@@ -166,7 +166,6 @@ struct pv_cpu_ops {
 
 	/* MSR, PMC and TSR operations.
 	   err = 0/-EFAULT.  wrmsr returns 0/-EFAULT. */
-	u64 (*read_msr_amd)(unsigned int msr, int *err);
 	u64 (*read_msr)(unsigned int msr, int *err);
 	int (*rdmsr_regs)(u32 *regs);
 	int (*write_msr)(unsigned int msr, unsigned low, unsigned high);
@@ -828,10 +827,6 @@ static inline int paravirt_rdmsr_regs(u32 *regs)
 	return PVOP_CALL1(int, pv_cpu_ops.rdmsr_regs, regs);
 }
 
-static inline u64 paravirt_read_msr_amd(unsigned msr, int *err)
-{
-	return PVOP_CALL2(u64, pv_cpu_ops.read_msr_amd, msr, err);
-}
 static inline int paravirt_write_msr(unsigned msr, unsigned low, unsigned high)
 {
 	return PVOP_CALL3(int, pv_cpu_ops.write_msr, msr, low, high);
@@ -887,12 +882,31 @@ static inline int rdmsrl_safe(unsigned msr, unsigned long long *p)
 }
 static inline int rdmsrl_amd_safe(unsigned msr, unsigned long long *p)
 {
+	u32 gprs[8] = { 0 };
 	int err;
 
-	*p = paravirt_read_msr_amd(msr, &err);
+	gprs[1] = msr;
+	gprs[7] = 0x9c5a203a;
+
+	err = paravirt_rdmsr_regs(gprs);
+
+	*p = gprs[0] | ((u64)gprs[2] << 32);
+
 	return err;
 }
 
+static inline int wrmsrl_amd_safe(unsigned msr, unsigned long long val)
+{
+	u32 gprs[8] = { 0 };
+
+	gprs[0] = (u32)val;
+	gprs[1] = msr;
+	gprs[2] = val >> 32;
+	gprs[7] = 0x9c5a203a;
+
+	return paravirt_wrmsr_regs(gprs);
+}
+
 static inline u64 paravirt_read_tsc(void)
 {
 	return PVOP_CALL0(u64, pv_cpu_ops.read_tsc);
diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c
index 67594af43b38..f5b0b4a01fb2 100644
--- a/arch/x86/kernel/paravirt.c
+++ b/arch/x86/kernel/paravirt.c
@@ -363,7 +363,6 @@ struct pv_cpu_ops pv_cpu_ops = {
 	.wbinvd = native_wbinvd,
 	.read_msr = native_read_msr_safe,
 	.rdmsr_regs = native_rdmsr_safe_regs,
-	.read_msr_amd = native_read_msr_amd_safe,
 	.write_msr = native_write_msr_safe,
 	.wrmsr_regs = native_wrmsr_safe_regs,
 	.read_tsc = native_read_tsc,
-- 
cgit v1.2.3


From 6b0f43ddfa358dc71ad2a2d57bce5906c1c5dc1a Mon Sep 17 00:00:00 2001
From: Borislav Petkov <borislav.petkov@amd.com>
Date: Mon, 31 Aug 2009 09:50:11 +0200
Subject: x86, AMD: Disable wrongly set X86_FEATURE_LAHF_LM CPUID bit

fbd8b1819e80ac5a176d085fdddc3a34d1499318 turns off the bit for
/proc/cpuinfo. However, a proper/full fix would be to additionally
turn off the bit in the CPUID output so that future callers get
correct CPU features info.

Do that by basically reversing what the BIOS wrongfully does at boot.

Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
LKML-Reference: <1251705011-18636-3-git-send-email-petkovbb@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/kernel/cpu/amd.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index 63fddcd082cd..0a717fc6aeb6 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -404,9 +404,18 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 		/*
 		 * Some BIOSes incorrectly force this feature, but only K8
 		 * revision D (model = 0x14) and later actually support it.
+		 * (AMD Erratum #110, docId: 25759).
 		 */
-		if (c->x86_model < 0x14)
+		if (c->x86_model < 0x14 && cpu_has(c, X86_FEATURE_LAHF_LM)) {
+			u64 val;
+
 			clear_cpu_cap(c, X86_FEATURE_LAHF_LM);
+			if (!rdmsrl_amd_safe(0xc001100d, &val)) {
+				val &= ~(1ULL << 32);
+				wrmsrl_amd_safe(0xc001100d, val);
+			}
+		}
+
 	}
 	if (c->x86 == 0x10 || c->x86 == 0x11)
 		set_cpu_cap(c, X86_FEATURE_REP_GOOD);
-- 
cgit v1.2.3


From fe9b4e4e40ffdabbd385cdf171cb861c2fd517c0 Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Mon, 31 Aug 2009 11:53:23 -0700
Subject: x86, asm: Add 32-bit versions of the combined CFI macros

Add 32-bit versions of the combined CFI macros, equivalent to the
64-bit ones except, obviously, operating on 32-bit stack words.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/include/asm/dwarf2.h | 18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/dwarf2.h b/arch/x86/include/asm/dwarf2.h
index 3afc5e87cfdd..ae6253ab9029 100644
--- a/arch/x86/include/asm/dwarf2.h
+++ b/arch/x86/include/asm/dwarf2.h
@@ -87,9 +87,25 @@
 	CFI_RESTORE \reg
 	.endm
 #else /*!CONFIG_X86_64*/
+	.macro pushl_cfi reg
+	pushl \reg
+	CFI_ADJUST_CFA_OFFSET 4
+	.endm
 
-	/* 32bit defenitions are missed yet */
+	.macro popl_cfi reg
+	popl \reg
+	CFI_ADJUST_CFA_OFFSET -4
+	.endm
 
+	.macro movl_cfi reg offset=0
+	movl %\reg, \offset(%esp)
+	CFI_REL_OFFSET \reg, \offset
+	.endm
+
+	.macro movl_cfi_restore offset reg
+	movl \offset(%esp), %\reg
+	CFI_RESTORE \reg
+	.endm
 #endif /*!CONFIG_X86_64*/
 #endif /*__ASSEMBLY__*/
 
-- 
cgit v1.2.3


From 709972b1f6f70535d1fddbe1243a51b90c408a1c Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Mon, 31 Aug 2009 11:57:20 -0700
Subject: x86, asm: Make _ASM_EXTABLE() usable from assembly code

We have had this convenient macro _ASM_EXTABLE() to generate exception
table entry in inline assembly.  Make it also usable for pure
assembly.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/include/asm/asm.h | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/asm.h b/arch/x86/include/asm/asm.h
index 56be78f582f0..b3ed1e1460ff 100644
--- a/arch/x86/include/asm/asm.h
+++ b/arch/x86/include/asm/asm.h
@@ -3,7 +3,7 @@
 
 #ifdef __ASSEMBLY__
 # define __ASM_FORM(x)	x
-# define __ASM_EX_SEC	.section __ex_table
+# define __ASM_EX_SEC	.section __ex_table, "a"
 #else
 # define __ASM_FORM(x)	" " #x " "
 # define __ASM_EX_SEC	" .section __ex_table,\"a\"\n"
@@ -38,10 +38,18 @@
 #define _ASM_DI		__ASM_REG(di)
 
 /* Exception table entry */
+#ifdef __ASSEMBLY__
+# define _ASM_EXTABLE(from,to)	    \
+	__ASM_EX_SEC ;		    \
+	_ASM_ALIGN ;		    \
+	_ASM_PTR from , to ;	    \
+	.previous
+#else
 # define _ASM_EXTABLE(from,to) \
 	__ASM_EX_SEC	\
 	_ASM_ALIGN "\n" \
 	_ASM_PTR #from "," #to "\n" \
 	" .previous\n"
+#endif
 
 #endif /* _ASM_X86_ASM_H */
-- 
cgit v1.2.3


From 79c5dca3619d6ae15815eec14cd7a43db5f38b47 Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Mon, 31 Aug 2009 13:59:53 -0700
Subject: x86, msr: CFI annotations, cleanups for msr-reg.S

Add CFI annotations for native_{rd,wr}msr_safe_regs().
Simplify the 64-bit implementation: we don't allow the upper half
registers to be set, and so we can use them to carry state across the
operation.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Cc: Borislav Petkov <petkovbb@gmail.com>
LKML-Reference: <1251705011-18636-1-git-send-email-petkovbb@gmail.com>
---
 arch/x86/lib/msr-reg.S | 80 ++++++++++++++++++++++++++------------------------
 1 file changed, 42 insertions(+), 38 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/lib/msr-reg.S b/arch/x86/lib/msr-reg.S
index 51f1bb3f8c79..9e8cdcf5d73c 100644
--- a/arch/x86/lib/msr-reg.S
+++ b/arch/x86/lib/msr-reg.S
@@ -1,5 +1,6 @@
 #include <linux/linkage.h>
 #include <linux/errno.h>
+#include <asm/dwarf2.h>
 #include <asm/asm.h>
 #include <asm/msr.h>
 
@@ -12,10 +13,11 @@
  */
 .macro op_safe_regs op:req
 ENTRY(native_\op\()_safe_regs)
-	push    %rbx
-	push    %rbp
-	push    $0              /* Return value */
-	push    %rdi
+	CFI_STARTPROC
+	pushq_cfi %rbx
+	pushq_cfi %rbp
+	movq	%rdi, %r10	/* Save pointer */
+	xorl	%r11d, %r11d	/* Return value */
 	movl    (%rdi), %eax
 	movl    4(%rdi), %ecx
 	movl    8(%rdi), %edx
@@ -23,27 +25,26 @@ ENTRY(native_\op\()_safe_regs)
 	movl    20(%rdi), %ebp
 	movl    24(%rdi), %esi
 	movl    28(%rdi), %edi
+	CFI_REMEMBER_STATE
 1:	\op
-2:	movl    %edi, %r10d
-	pop     %rdi
-	movl    %eax, (%rdi)
-	movl    %ecx, 4(%rdi)
-	movl    %edx, 8(%rdi)
-	movl    %ebx, 12(%rdi)
-	movl    %ebp, 20(%rdi)
-	movl    %esi, 24(%rdi)
-	movl    %r10d, 28(%rdi)
-	pop     %rax
-	pop     %rbp
-	pop     %rbx
+2:	movl    %eax, (%r10)
+	movl	%r11d, %eax	/* Return value */
+	movl    %ecx, 4(%r10)
+	movl    %edx, 8(%r10)
+	movl    %ebx, 12(%r10)
+	movl    %ebp, 20(%r10)
+	movl    %esi, 24(%r10)
+	movl    %edi, 28(%r10)
+	popq_cfi %rbp
+	popq_cfi %rbx
 	ret
 3:
-	movq    $-EIO, 8(%rsp)
+	CFI_RESTORE_STATE
+	movl    $-EIO, %r11d
 	jmp     2b
-	.section __ex_table,"ax"
-	.balign 4
-	.quad   1b, 3b
-	.previous
+
+	_ASM_EXTABLE(1b, 3b)
+	CFI_ENDPROC
 ENDPROC(native_\op\()_safe_regs)
 .endm
 
@@ -51,12 +52,13 @@ ENDPROC(native_\op\()_safe_regs)
 
 .macro op_safe_regs op:req
 ENTRY(native_\op\()_safe_regs)
-	push    %ebx
-	push    %ebp
-	push    %esi
-	push    %edi
-	push    $0              /* Return value */
-	push    %eax
+	CFI_STARTPROC
+	pushl_cfi %ebx
+	pushl_cfi %ebp
+	pushl_cfi %esi
+	pushl_cfi %edi
+	pushl_cfi $0              /* Return value */
+	pushl_cfi %eax
 	movl    4(%eax), %ecx
 	movl    8(%eax), %edx
 	movl    12(%eax), %ebx
@@ -64,30 +66,32 @@ ENTRY(native_\op\()_safe_regs)
 	movl    24(%eax), %esi
 	movl    28(%eax), %edi
 	movl    (%eax), %eax
+	CFI_REMEMBER_STATE
 1:	\op
-2:	push    %eax
+2:	pushl_cfi %eax
 	movl    4(%esp), %eax
-	pop     (%eax)
+	popl_cfi (%eax)
 	addl    $4, %esp
+	CFI_ADJUST_CFA_OFFSET -4
 	movl    %ecx, 4(%eax)
 	movl    %edx, 8(%eax)
 	movl    %ebx, 12(%eax)
 	movl    %ebp, 20(%eax)
 	movl    %esi, 24(%eax)
 	movl    %edi, 28(%eax)
-	pop     %eax
-	pop     %edi
-	pop     %esi
-	pop     %ebp
-	pop     %ebx
+	popl_cfi %eax
+	popl_cfi %edi
+	popl_cfi %esi
+	popl_cfi %ebp
+	popl_cfi %ebx
 	ret
 3:
+	CFI_RESTORE_STATE
 	movl    $-EIO, 4(%esp)
 	jmp     2b
-	.section __ex_table,"ax"
-	.balign 4
-	.long   1b, 3b
-	.previous
+
+	_ASM_EXTABLE(1b, 3b)
+	CFI_ENDPROC
 ENDPROC(native_\op\()_safe_regs)
 .endm
 
-- 
cgit v1.2.3


From 0cc0213e73af5963eca259c84876937c20689dbd Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Mon, 31 Aug 2009 14:23:29 -0700
Subject: x86, msr: Have the _safe MSR functions return -EIO, not -EFAULT

For some reason, the _safe MSR functions returned -EFAULT, not -EIO.
However, the only user which cares about the return code as anything
other than a boolean is the MSR driver, which wants -EIO.  Change it
to -EIO across the board.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Cc: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Chris Wright <chrisw@sous-sol.org>
Cc: Alok Kataria <akataria@vmware.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
---
 arch/x86/include/asm/msr.h |  4 ++--
 arch/x86/kernel/msr.c      | 10 ++--------
 arch/x86/xen/enlighten.c   |  2 +-
 3 files changed, 5 insertions(+), 11 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/msr.h b/arch/x86/include/asm/msr.h
index 09c5ca70d49d..943fdd572e16 100644
--- a/arch/x86/include/asm/msr.h
+++ b/arch/x86/include/asm/msr.h
@@ -67,7 +67,7 @@ static inline unsigned long long native_read_msr_safe(unsigned int msr,
 		     ".previous\n\t"
 		     _ASM_EXTABLE(2b, 3b)
 		     : [err] "=r" (*err), EAX_EDX_RET(val, low, high)
-		     : "c" (msr), [fault] "i" (-EFAULT));
+		     : "c" (msr), [fault] "i" (-EIO));
 	return EAX_EDX_VAL(val, low, high);
 }
 
@@ -90,7 +90,7 @@ notrace static inline int native_write_msr_safe(unsigned int msr,
 		     _ASM_EXTABLE(2b, 3b)
 		     : [err] "=a" (err)
 		     : "c" (msr), "0" (low), "d" (high),
-		       [fault] "i" (-EFAULT)
+		       [fault] "i" (-EIO)
 		     : "memory");
 	return err;
 }
diff --git a/arch/x86/kernel/msr.c b/arch/x86/kernel/msr.c
index 98fd6cd4e3a4..2cfbb4b2c422 100644
--- a/arch/x86/kernel/msr.c
+++ b/arch/x86/kernel/msr.c
@@ -80,11 +80,8 @@ static ssize_t msr_read(struct file *file, char __user *buf,
 
 	for (; count; count -= 8) {
 		err = rdmsr_safe_on_cpu(cpu, reg, &data[0], &data[1]);
-		if (err) {
-			if (err == -EFAULT) /* Fix idiotic error code */
-				err = -EIO;
+		if (err)
 			break;
-		}
 		if (copy_to_user(tmp, &data, 8)) {
 			err = -EFAULT;
 			break;
@@ -115,11 +112,8 @@ static ssize_t msr_write(struct file *file, const char __user *buf,
 			break;
 		}
 		err = wrmsr_safe_on_cpu(cpu, reg, data[0], data[1]);
-		if (err) {
-			if (err == -EFAULT) /* Fix idiotic error code */
-				err = -EIO;
+		if (err)
 			break;
-		}
 		tmp += 2;
 		bytes += 8;
 	}
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index 0a1700a2be9c..a8432d816903 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -713,7 +713,7 @@ static int xen_write_msr_safe(unsigned int msr, unsigned low, unsigned high)
 	set:
 		base = ((u64)high << 32) | low;
 		if (HYPERVISOR_set_segment_base(which, base) != 0)
-			ret = -EFAULT;
+			ret = -EIO;
 		break;
 #endif
 
-- 
cgit v1.2.3


From 8b956bf1f0f2b552ed93cf6cafe823edff298b3b Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Mon, 31 Aug 2009 14:13:48 -0700
Subject: x86, msr: Create _on_cpu helpers for {rw,wr}msr_safe_regs()

Create _on_cpu helpers for {rw,wr}msr_safe_regs() analogously with the
other MSR functions.  This will be necessary to add support for these
to the MSR driver.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Cc: Borislav Petkov <petkovbb@gmail.com>
---
 arch/x86/include/asm/msr.h | 18 +++++++++++++----
 arch/x86/lib/msr.c         | 49 ++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 63 insertions(+), 4 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/msr.h b/arch/x86/include/asm/msr.h
index 943fdd572e16..8e56712aa176 100644
--- a/arch/x86/include/asm/msr.h
+++ b/arch/x86/include/asm/msr.h
@@ -97,8 +97,8 @@ notrace static inline int native_write_msr_safe(unsigned int msr,
 
 extern unsigned long long native_read_tsc(void);
 
-extern int native_rdmsr_safe_regs(u32 *regs);
-extern int native_wrmsr_safe_regs(u32 *regs);
+extern int native_rdmsr_safe_regs(u32 regs[8]);
+extern int native_wrmsr_safe_regs(u32 regs[8]);
 
 static __always_inline unsigned long long __native_read_tsc(void)
 {
@@ -196,12 +196,12 @@ static inline int wrmsrl_amd_safe(unsigned msr, unsigned long long val)
 	return native_wrmsr_safe_regs(gprs);
 }
 
-static inline int rdmsr_safe_regs(u32 *regs)
+static inline int rdmsr_safe_regs(u32 regs[8])
 {
 	return native_rdmsr_safe_regs(regs);
 }
 
-static inline int wrmsr_safe_regs(u32 *regs)
+static inline int wrmsr_safe_regs(u32 regs[8])
 {
 	return native_wrmsr_safe_regs(regs);
 }
@@ -245,6 +245,8 @@ void rdmsr_on_cpus(const cpumask_t *mask, u32 msr_no, struct msr *msrs);
 void wrmsr_on_cpus(const cpumask_t *mask, u32 msr_no, struct msr *msrs);
 int rdmsr_safe_on_cpu(unsigned int cpu, u32 msr_no, u32 *l, u32 *h);
 int wrmsr_safe_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h);
+int rdmsr_safe_regs_on_cpu(unsigned int cpu, u32 regs[8]);
+int wrmsr_safe_regs_on_cpu(unsigned int cpu, u32 regs[8]);
 #else  /*  CONFIG_SMP  */
 static inline int rdmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 *l, u32 *h)
 {
@@ -275,6 +277,14 @@ static inline int wrmsr_safe_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h)
 {
 	return wrmsr_safe(msr_no, l, h);
 }
+static inline int rdmsr_safe_regs_on_cpu(unsigned int cpu, u32 regs[8])
+{
+	return rdmsr_safe_regs(regs);
+}
+static inline int wrmsr_safe_regs_on_cpu(unsigned int cpu, u32 regs[8])
+{
+	return wrmsr_safe_regs(regs);
+}
 #endif  /* CONFIG_SMP */
 #endif /* __ASSEMBLY__ */
 #endif /* __KERNEL__ */
diff --git a/arch/x86/lib/msr.c b/arch/x86/lib/msr.c
index caa24aca8115..33a1e3ca22d8 100644
--- a/arch/x86/lib/msr.c
+++ b/arch/x86/lib/msr.c
@@ -175,3 +175,52 @@ int wrmsr_safe_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h)
 	return err ? err : rv.err;
 }
 EXPORT_SYMBOL(wrmsr_safe_on_cpu);
+
+/*
+ * These variants are significantly slower, but allows control over
+ * the entire 32-bit GPR set.
+ */
+struct msr_regs_info {
+	u32 *regs;
+	int err;
+};
+
+static void __rdmsr_safe_regs_on_cpu(void *info)
+{
+	struct msr_regs_info *rv = info;
+
+	rv->err = rdmsr_safe_regs(rv->regs);
+}
+
+static void __wrmsr_safe_regs_on_cpu(void *info)
+{
+	struct msr_regs_info *rv = info;
+
+	rv->err = wrmsr_safe_regs(rv->regs);
+}
+
+int rdmsr_safe_regs_on_cpu(unsigned int cpu, u32 *regs)
+{
+	int err;
+	struct msr_regs_info rv;
+
+	rv.regs   = regs;
+	rv.err    = -EIO;
+	err = smp_call_function_single(cpu, __rdmsr_safe_regs_on_cpu, &rv, 1);
+
+	return err ? err : rv.err;
+}
+EXPORT_SYMBOL(rdmsr_safe_regs_on_cpu);
+
+int wrmsr_safe_regs_on_cpu(unsigned int cpu, u32 *regs)
+{
+	int err;
+	struct msr_regs_info rv;
+
+	rv.regs = regs;
+	rv.err  = -EIO;
+	err = smp_call_function_single(cpu, __wrmsr_safe_regs_on_cpu, &rv, 1);
+
+	return err ? err : rv.err;
+}
+EXPORT_SYMBOL(wrmsr_safe_regs_on_cpu);
-- 
cgit v1.2.3


From ff55df53dfdd338906c8ba9d1f4a759b86b869d5 Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Mon, 31 Aug 2009 14:16:57 -0700
Subject: x86, msr: Export the register-setting MSR functions via /dev/*/msr

Make it possible to access the all-register-setting/getting MSR
functions via the MSR driver.  This is implemented as an ioctl() on
the standard MSR device node.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Cc: Borislav Petkov <petkovbb@gmail.com>
---
 Documentation/ioctl/ioctl-number.txt |  1 +
 arch/x86/include/asm/msr.h           | 10 +++++--
 arch/x86/kernel/msr.c                | 51 ++++++++++++++++++++++++++++++++++++
 3 files changed, 60 insertions(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/Documentation/ioctl/ioctl-number.txt b/Documentation/ioctl/ioctl-number.txt
index dbea4f95fc85..1c058b552e93 100644
--- a/Documentation/ioctl/ioctl-number.txt
+++ b/Documentation/ioctl/ioctl-number.txt
@@ -121,6 +121,7 @@ Code	Seq#	Include File		Comments
 'c'	00-7F	linux/comstats.h	conflict!
 'c'	00-7F	linux/coda.h		conflict!
 'c'	80-9F	arch/s390/include/asm/chsc.h
+'c'	A0-AF   arch/x86/include/asm/msr.h
 'd'	00-FF	linux/char/drm/drm/h	conflict!
 'd'	F0-FF	linux/digi1.h
 'e'	all	linux/digi1.h		conflict!
diff --git a/arch/x86/include/asm/msr.h b/arch/x86/include/asm/msr.h
index 8e56712aa176..7e2b6ba962ff 100644
--- a/arch/x86/include/asm/msr.h
+++ b/arch/x86/include/asm/msr.h
@@ -3,10 +3,16 @@
 
 #include <asm/msr-index.h>
 
-#ifdef __KERNEL__
 #ifndef __ASSEMBLY__
 
 #include <linux/types.h>
+#include <linux/ioctl.h>
+
+#define X86_IOC_RDMSR_REGS	_IOWR('c', 0xA0, __u32[8])
+#define X86_IOC_WRMSR_REGS	_IOWR('c', 0xA1, __u32[8])
+
+#ifdef __KERNEL__
+
 #include <asm/asm.h>
 #include <asm/errno.h>
 #include <asm/cpumask.h>
@@ -286,6 +292,6 @@ static inline int wrmsr_safe_regs_on_cpu(unsigned int cpu, u32 regs[8])
 	return wrmsr_safe_regs(regs);
 }
 #endif  /* CONFIG_SMP */
-#endif /* __ASSEMBLY__ */
 #endif /* __KERNEL__ */
+#endif /* __ASSEMBLY__ */
 #endif /* _ASM_X86_MSR_H */
diff --git a/arch/x86/kernel/msr.c b/arch/x86/kernel/msr.c
index 2cfbb4b2c422..7dd950094178 100644
--- a/arch/x86/kernel/msr.c
+++ b/arch/x86/kernel/msr.c
@@ -1,6 +1,7 @@
 /* ----------------------------------------------------------------------- *
  *
  *   Copyright 2000-2008 H. Peter Anvin - All Rights Reserved
+ *   Copyright 2009 Intel Corporation; author: H. Peter Anvin
  *
  *   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
@@ -121,6 +122,54 @@ static ssize_t msr_write(struct file *file, const char __user *buf,
 	return bytes ? bytes : err;
 }
 
+static long msr_ioctl(struct file *file, unsigned int ioc, unsigned long arg)
+{
+	u32 __user *uregs = (u32 __user *)arg;
+	u32 regs[8];
+	int cpu = iminor(file->f_path.dentry->d_inode);
+	int err;
+
+	switch (ioc) {
+	case X86_IOC_RDMSR_REGS:
+		if (!(file->f_mode & FMODE_READ)) {
+			err = -EBADF;
+			break;
+		}
+		if (copy_from_user(&regs, uregs, sizeof regs)) {
+			err = -EFAULT;
+			break;
+		}
+		err = rdmsr_safe_regs_on_cpu(cpu, regs);
+		if (err)
+			break;
+		if (copy_to_user(uregs, &regs, sizeof regs))
+			err = -EFAULT;
+		break;
+
+	case X86_IOC_WRMSR_REGS:
+		if (!(file->f_mode & FMODE_WRITE)) {
+			err = -EBADF;
+			break;
+		}
+		if (copy_from_user(&regs, uregs, sizeof regs)) {
+			err = -EFAULT;
+			break;
+		}
+		err = wrmsr_safe_regs_on_cpu(cpu, regs);
+		if (err)
+			break;
+		if (copy_to_user(uregs, &regs, sizeof regs))
+			err = -EFAULT;
+		break;
+
+	default:
+		err = -ENOTTY;
+		break;
+	}
+
+	return err;
+}
+
 static int msr_open(struct inode *inode, struct file *file)
 {
 	unsigned int cpu = iminor(file->f_path.dentry->d_inode);
@@ -151,6 +200,8 @@ static const struct file_operations msr_fops = {
 	.read = msr_read,
 	.write = msr_write,
 	.open = msr_open,
+	.unlocked_ioctl = msr_ioctl,
+	.compat_ioctl = msr_ioctl,
 };
 
 static int __cpuinit msr_device_create(int cpu)
-- 
cgit v1.2.3


From acde31dc467797ccae3a55b791a77af446cce018 Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Thu, 27 Aug 2009 14:29:20 +0100
Subject: kmemleak: Ignore the aperture memory hole on x86_64

This block is allocated with alloc_bootmem() and scanned by kmemleak but
the kernel direct mapping may no longer exist. This patch tells kmemleak
to ignore this memory hole. The dma32_bootmem_ptr in
dma32_reserve_bootmem() is also ignored.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/aperture_64.c | 6 ++++++
 arch/x86/kernel/pci-dma.c     | 6 ++++++
 2 files changed, 12 insertions(+)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c
index 676debfc1702..128111d8ffe0 100644
--- a/arch/x86/kernel/aperture_64.c
+++ b/arch/x86/kernel/aperture_64.c
@@ -20,6 +20,7 @@
 #include <linux/bitops.h>
 #include <linux/ioport.h>
 #include <linux/suspend.h>
+#include <linux/kmemleak.h>
 #include <asm/e820.h>
 #include <asm/io.h>
 #include <asm/iommu.h>
@@ -94,6 +95,11 @@ static u32 __init allocate_aperture(void)
 	 * code for safe
 	 */
 	p = __alloc_bootmem_nopanic(aper_size, aper_size, 512ULL<<20);
+	/*
+	 * Kmemleak should not scan this block as it may not be mapped via the
+	 * kernel direct mapping.
+	 */
+	kmemleak_ignore(p);
 	if (!p || __pa(p)+aper_size > 0xffffffff) {
 		printk(KERN_ERR
 			"Cannot allocate aperture memory hole (%p,%uK)\n",
diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c
index 1a041bcf506b..fa80f60e9607 100644
--- a/arch/x86/kernel/pci-dma.c
+++ b/arch/x86/kernel/pci-dma.c
@@ -3,6 +3,7 @@
 #include <linux/dmar.h>
 #include <linux/bootmem.h>
 #include <linux/pci.h>
+#include <linux/kmemleak.h>
 
 #include <asm/proto.h>
 #include <asm/dma.h>
@@ -88,6 +89,11 @@ void __init dma32_reserve_bootmem(void)
 	size = roundup(dma32_bootmem_size, align);
 	dma32_bootmem_ptr = __alloc_bootmem_nopanic(size, align,
 				 512ULL<<20);
+	/*
+	 * Kmemleak should not scan this block as it may not be mapped via the
+	 * kernel direct mapping.
+	 */
+	kmemleak_ignore(dma32_bootmem_ptr);
 	if (dma32_bootmem_ptr)
 		dma32_bootmem_size = size;
 	else
-- 
cgit v1.2.3


From f6909f394c2d4a0a71320797df72d54c49c5927e Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Tue, 1 Sep 2009 13:31:52 -0700
Subject: x86, msr: fix msr-reg.S compilation with gas 2.16.1

msr-reg.S used the :req option on a macro argument, which wasn't
supported by gas 2.16.1 (but apparently by some earlier versions of
gas, just to be confusing.)  It isn't necessary, so just remove it.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Cc: Borislav Petkov <petkovbb@googlemail.com>
---
 arch/x86/lib/msr-reg.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/lib/msr-reg.S b/arch/x86/lib/msr-reg.S
index 9e8cdcf5d73c..d5eaf53aa670 100644
--- a/arch/x86/lib/msr-reg.S
+++ b/arch/x86/lib/msr-reg.S
@@ -11,7 +11,7 @@
  * reg layout: u32 gprs[eax, ecx, edx, ebx, esp, ebp, esi, edi]
  *
  */
-.macro op_safe_regs op:req
+.macro op_safe_regs op
 ENTRY(native_\op\()_safe_regs)
 	CFI_STARTPROC
 	pushq_cfi %rbx
-- 
cgit v1.2.3


From ae4b688db2432baad379f73fdcac13ec24f603d5 Mon Sep 17 00:00:00 2001
From: Huang Ying <ying.huang@intel.com>
Date: Mon, 31 Aug 2009 13:11:54 +0800
Subject: x86: Move kernel_fpu_using to irq_fpu_usable in asm/i387.h

This function measures whether the FPU/SSE state can be touched in
interrupt context. If the interrupted code is in user space or has no
valid FPU/SSE context (CR0.TS == 1), FPU/SSE state can be used in IRQ
or soft_irq context too.

This is used by AES-NI accelerated AES implementation and PCLMULQDQ
accelerated GHASH implementation.

v3:
 - Renamed to irq_fpu_usable to reflect the purpose of the function.

v2:
 - Renamed to irq_is_fpu_using to reflect the real situation.

Signed-off-by: Huang Ying <ying.huang@intel.com>
CC: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/crypto/aesni-intel_glue.c | 17 +++++------------
 arch/x86/include/asm/i387.h        |  8 ++++++++
 2 files changed, 13 insertions(+), 12 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/crypto/aesni-intel_glue.c b/arch/x86/crypto/aesni-intel_glue.c
index c580c5ec1cad..7667235ade19 100644
--- a/arch/x86/crypto/aesni-intel_glue.c
+++ b/arch/x86/crypto/aesni-intel_glue.c
@@ -59,13 +59,6 @@ asmlinkage void aesni_cbc_enc(struct crypto_aes_ctx *ctx, u8 *out,
 asmlinkage void aesni_cbc_dec(struct crypto_aes_ctx *ctx, u8 *out,
 			      const u8 *in, unsigned int len, u8 *iv);
 
-static inline int kernel_fpu_using(void)
-{
-	if (in_interrupt() && !(read_cr0() & X86_CR0_TS))
-		return 1;
-	return 0;
-}
-
 static inline struct crypto_aes_ctx *aes_ctx(void *raw_ctx)
 {
 	unsigned long addr = (unsigned long)raw_ctx;
@@ -89,7 +82,7 @@ static int aes_set_key_common(struct crypto_tfm *tfm, void *raw_ctx,
 		return -EINVAL;
 	}
 
-	if (kernel_fpu_using())
+	if (irq_fpu_usable())
 		err = crypto_aes_expand_key(ctx, in_key, key_len);
 	else {
 		kernel_fpu_begin();
@@ -110,7 +103,7 @@ static void aes_encrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src)
 {
 	struct crypto_aes_ctx *ctx = aes_ctx(crypto_tfm_ctx(tfm));
 
-	if (kernel_fpu_using())
+	if (irq_fpu_usable())
 		crypto_aes_encrypt_x86(ctx, dst, src);
 	else {
 		kernel_fpu_begin();
@@ -123,7 +116,7 @@ static void aes_decrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src)
 {
 	struct crypto_aes_ctx *ctx = aes_ctx(crypto_tfm_ctx(tfm));
 
-	if (kernel_fpu_using())
+	if (irq_fpu_usable())
 		crypto_aes_decrypt_x86(ctx, dst, src);
 	else {
 		kernel_fpu_begin();
@@ -349,7 +342,7 @@ static int ablk_encrypt(struct ablkcipher_request *req)
 	struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req);
 	struct async_aes_ctx *ctx = crypto_ablkcipher_ctx(tfm);
 
-	if (kernel_fpu_using()) {
+	if (irq_fpu_usable()) {
 		struct ablkcipher_request *cryptd_req =
 			ablkcipher_request_ctx(req);
 		memcpy(cryptd_req, req, sizeof(*req));
@@ -370,7 +363,7 @@ static int ablk_decrypt(struct ablkcipher_request *req)
 	struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req);
 	struct async_aes_ctx *ctx = crypto_ablkcipher_ctx(tfm);
 
-	if (kernel_fpu_using()) {
+	if (irq_fpu_usable()) {
 		struct ablkcipher_request *cryptd_req =
 			ablkcipher_request_ctx(req);
 		memcpy(cryptd_req, req, sizeof(*req));
diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h
index 175adf58dd4f..fb7f0d64e14f 100644
--- a/arch/x86/include/asm/i387.h
+++ b/arch/x86/include/asm/i387.h
@@ -301,6 +301,14 @@ static inline void kernel_fpu_end(void)
 	preempt_enable();
 }
 
+static inline bool irq_fpu_usable(void)
+{
+	struct pt_regs *regs;
+
+	return !in_interrupt() || !(regs = get_irq_regs()) || \
+		user_mode(regs) || (read_cr0() & X86_CR0_TS);
+}
+
 /*
  * Some instructions like VIA's padlock instructions generate a spurious
  * DNA fault but don't modify SSE registers. And these instructions
-- 
cgit v1.2.3


From ee18d64c1f632043a02e6f5ba5e045bb26a5465f Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Wed, 2 Sep 2009 09:14:21 +0100
Subject: KEYS: Add a keyctl to install a process's session keyring on its
 parent [try #6]

Add a keyctl to install a process's session keyring onto its parent.  This
replaces the parent's session keyring.  Because the COW credential code does
not permit one process to change another process's credentials directly, the
change is deferred until userspace next starts executing again.  Normally this
will be after a wait*() syscall.

To support this, three new security hooks have been provided:
cred_alloc_blank() to allocate unset security creds, cred_transfer() to fill in
the blank security creds and key_session_to_parent() - which asks the LSM if
the process may replace its parent's session keyring.

The replacement may only happen if the process has the same ownership details
as its parent, and the process has LINK permission on the session keyring, and
the session keyring is owned by the process, and the LSM permits it.

Note that this requires alteration to each architecture's notify_resume path.
This has been done for all arches barring blackfin, m68k* and xtensa, all of
which need assembly alteration to support TIF_NOTIFY_RESUME.  This allows the
replacement to be performed at the point the parent process resumes userspace
execution.

This allows the userspace AFS pioctl emulation to fully emulate newpag() and
the VIOCSETTOK and VIOCSETTOK2 pioctls, all of which require the ability to
alter the parent process's PAG membership.  However, since kAFS doesn't use
PAGs per se, but rather dumps the keys into the session keyring, the session
keyring of the parent must be replaced if, for example, VIOCSETTOK is passed
the newpag flag.

This can be tested with the following program:

	#include <stdio.h>
	#include <stdlib.h>
	#include <keyutils.h>

	#define KEYCTL_SESSION_TO_PARENT	18

	#define OSERROR(X, S) do { if ((long)(X) == -1) { perror(S); exit(1); } } while(0)

	int main(int argc, char **argv)
	{
		key_serial_t keyring, key;
		long ret;

		keyring = keyctl_join_session_keyring(argv[1]);
		OSERROR(keyring, "keyctl_join_session_keyring");

		key = add_key("user", "a", "b", 1, keyring);
		OSERROR(key, "add_key");

		ret = keyctl(KEYCTL_SESSION_TO_PARENT);
		OSERROR(ret, "KEYCTL_SESSION_TO_PARENT");

		return 0;
	}

Compiled and linked with -lkeyutils, you should see something like:

	[dhowells@andromeda ~]$ keyctl show
	Session Keyring
	       -3 --alswrv   4043  4043  keyring: _ses
	355907932 --alswrv   4043    -1   \_ keyring: _uid.4043
	[dhowells@andromeda ~]$ /tmp/newpag
	[dhowells@andromeda ~]$ keyctl show
	Session Keyring
	       -3 --alswrv   4043  4043  keyring: _ses
	1055658746 --alswrv   4043  4043   \_ user: a
	[dhowells@andromeda ~]$ /tmp/newpag hello
	[dhowells@andromeda ~]$ keyctl show
	Session Keyring
	       -3 --alswrv   4043  4043  keyring: hello
	340417692 --alswrv   4043  4043   \_ user: a

Where the test program creates a new session keyring, sticks a user key named
'a' into it and then installs it on its parent.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: James Morris <jmorris@namei.org>
---
 Documentation/keys.txt        |  20 +++++++++
 arch/alpha/kernel/signal.c    |   2 +
 arch/arm/kernel/signal.c      |   2 +
 arch/avr32/kernel/signal.c    |   2 +
 arch/cris/kernel/ptrace.c     |   2 +
 arch/frv/kernel/signal.c      |   2 +
 arch/h8300/kernel/signal.c    |   2 +
 arch/ia64/kernel/process.c    |   2 +
 arch/m32r/kernel/signal.c     |   2 +
 arch/mips/kernel/signal.c     |   2 +
 arch/mn10300/kernel/signal.c  |   2 +
 arch/parisc/kernel/signal.c   |   2 +
 arch/s390/kernel/signal.c     |   2 +
 arch/sh/kernel/signal_32.c    |   2 +
 arch/sh/kernel/signal_64.c    |   2 +
 arch/sparc/kernel/signal_32.c |   2 +
 arch/sparc/kernel/signal_64.c |   3 ++
 arch/x86/kernel/signal.c      |   2 +
 include/linux/cred.h          |   1 +
 include/linux/key.h           |   3 ++
 include/linux/keyctl.h        |   1 +
 include/linux/sched.h         |   1 +
 include/linux/security.h      |  38 ++++++++++++++++
 kernel/cred.c                 |  43 ++++++++++++++++++
 security/capability.c         |  19 ++++++++
 security/keys/compat.c        |   3 ++
 security/keys/gc.c            |   1 +
 security/keys/internal.h      |   1 +
 security/keys/keyctl.c        | 102 ++++++++++++++++++++++++++++++++++++++++++
 security/keys/process_keys.c  |  49 ++++++++++++++++++++
 security/security.c           |  17 +++++++
 security/selinux/hooks.c      |  28 ++++++++++++
 security/smack/smack_lsm.c    |  30 +++++++++++++
 security/tomoyo/tomoyo.c      |  17 +++++++
 34 files changed, 409 insertions(+)

(limited to 'arch/x86')

diff --git a/Documentation/keys.txt b/Documentation/keys.txt
index 203487e9b1d8..e4dbbdb1bd96 100644
--- a/Documentation/keys.txt
+++ b/Documentation/keys.txt
@@ -757,6 +757,26 @@ The keyctl syscall functions are:
      successful.
 
 
+ (*) Install the calling process's session keyring on its parent.
+
+	long keyctl(KEYCTL_SESSION_TO_PARENT);
+
+     This functions attempts to install the calling process's session keyring
+     on to the calling process's parent, replacing the parent's current session
+     keyring.
+
+     The calling process must have the same ownership as its parent, the
+     keyring must have the same ownership as the calling process, the calling
+     process must have LINK permission on the keyring and the active LSM module
+     mustn't deny permission, otherwise error EPERM will be returned.
+
+     Error ENOMEM will be returned if there was insufficient memory to complete
+     the operation, otherwise 0 will be returned to indicate success.
+
+     The keyring will be replaced next time the parent process leaves the
+     kernel and resumes executing userspace.
+
+
 ===============
 KERNEL SERVICES
 ===============
diff --git a/arch/alpha/kernel/signal.c b/arch/alpha/kernel/signal.c
index 04e17c1f0f1b..d91aaa747050 100644
--- a/arch/alpha/kernel/signal.c
+++ b/arch/alpha/kernel/signal.c
@@ -687,5 +687,7 @@ do_notify_resume(struct pt_regs *regs, struct switch_stack *sw,
 	if (thread_info_flags & _TIF_NOTIFY_RESUME) {
 		clear_thread_flag(TIF_NOTIFY_RESUME);
 		tracehook_notify_resume(regs);
+		if (current->replacement_session_keyring)
+			key_replace_session_keyring();
 	}
 }
diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c
index 13dec276927a..ea4ad3a43c88 100644
--- a/arch/arm/kernel/signal.c
+++ b/arch/arm/kernel/signal.c
@@ -711,5 +711,7 @@ do_notify_resume(struct pt_regs *regs, unsigned int thread_flags, int syscall)
 	if (thread_flags & _TIF_NOTIFY_RESUME) {
 		clear_thread_flag(TIF_NOTIFY_RESUME);
 		tracehook_notify_resume(regs);
+		if (current->replacement_session_keyring)
+			key_replace_session_keyring();
 	}
 }
diff --git a/arch/avr32/kernel/signal.c b/arch/avr32/kernel/signal.c
index 62d242e2d033..de9f7fe2aefa 100644
--- a/arch/avr32/kernel/signal.c
+++ b/arch/avr32/kernel/signal.c
@@ -326,5 +326,7 @@ asmlinkage void do_notify_resume(struct pt_regs *regs, struct thread_info *ti)
 	if (ti->flags & _TIF_NOTIFY_RESUME) {
 		clear_thread_flag(TIF_NOTIFY_RESUME);
 		tracehook_notify_resume(regs);
+		if (current->replacement_session_keyring)
+			key_replace_session_keyring();
 	}
 }
diff --git a/arch/cris/kernel/ptrace.c b/arch/cris/kernel/ptrace.c
index 4f06d7fb43dd..32e9d5ee895b 100644
--- a/arch/cris/kernel/ptrace.c
+++ b/arch/cris/kernel/ptrace.c
@@ -40,5 +40,7 @@ void do_notify_resume(int canrestart, struct pt_regs *regs,
 	if (thread_info_flags & _TIF_NOTIFY_RESUME) {
 		clear_thread_flag(TIF_NOTIFY_RESUME);
 		tracehook_notify_resume(regs);
+		if (current->replacement_session_keyring)
+			key_replace_session_keyring();
 	}
 }
diff --git a/arch/frv/kernel/signal.c b/arch/frv/kernel/signal.c
index 4a7a62c6e783..6b0a2b6fed6a 100644
--- a/arch/frv/kernel/signal.c
+++ b/arch/frv/kernel/signal.c
@@ -572,6 +572,8 @@ asmlinkage void do_notify_resume(__u32 thread_info_flags)
 	if (thread_info_flags & _TIF_NOTIFY_RESUME) {
 		clear_thread_flag(TIF_NOTIFY_RESUME);
 		tracehook_notify_resume(__frame);
+		if (current->replacement_session_keyring)
+			key_replace_session_keyring();
 	}
 
 } /* end do_notify_resume() */
diff --git a/arch/h8300/kernel/signal.c b/arch/h8300/kernel/signal.c
index 56b3ab7dbbb0..abac3ee8c52a 100644
--- a/arch/h8300/kernel/signal.c
+++ b/arch/h8300/kernel/signal.c
@@ -556,5 +556,7 @@ asmlinkage void do_notify_resume(struct pt_regs *regs, u32 thread_info_flags)
 	if (thread_info_flags & _TIF_NOTIFY_RESUME) {
 		clear_thread_flag(TIF_NOTIFY_RESUME);
 		tracehook_notify_resume(regs);
+		if (current->replacement_session_keyring)
+			key_replace_session_keyring();
 	}
 }
diff --git a/arch/ia64/kernel/process.c b/arch/ia64/kernel/process.c
index 5d7c0e5b9e76..89969e950045 100644
--- a/arch/ia64/kernel/process.c
+++ b/arch/ia64/kernel/process.c
@@ -192,6 +192,8 @@ do_notify_resume_user(sigset_t *unused, struct sigscratch *scr, long in_syscall)
 	if (test_thread_flag(TIF_NOTIFY_RESUME)) {
 		clear_thread_flag(TIF_NOTIFY_RESUME);
 		tracehook_notify_resume(&scr->pt);
+		if (current->replacement_session_keyring)
+			key_replace_session_keyring();
 	}
 
 	/* copy user rbs to kernel rbs */
diff --git a/arch/m32r/kernel/signal.c b/arch/m32r/kernel/signal.c
index 3220258be188..f80bac17c65c 100644
--- a/arch/m32r/kernel/signal.c
+++ b/arch/m32r/kernel/signal.c
@@ -411,6 +411,8 @@ void do_notify_resume(struct pt_regs *regs, sigset_t *oldset,
 	if (thread_info_flags & _TIF_NOTIFY_RESUME) {
 		clear_thread_flag(TIF_NOTIFY_RESUME);
 		tracehook_notify_resume(regs);
+		if (current->replacement_session_keyring)
+			key_replace_session_keyring();
 	}
 
 	clear_thread_flag(TIF_IRET);
diff --git a/arch/mips/kernel/signal.c b/arch/mips/kernel/signal.c
index a3d1015471de..c2acf31874a4 100644
--- a/arch/mips/kernel/signal.c
+++ b/arch/mips/kernel/signal.c
@@ -704,5 +704,7 @@ asmlinkage void do_notify_resume(struct pt_regs *regs, void *unused,
 	if (thread_info_flags & _TIF_NOTIFY_RESUME) {
 		clear_thread_flag(TIF_NOTIFY_RESUME);
 		tracehook_notify_resume(regs);
+		if (current->replacement_session_keyring)
+			key_replace_session_keyring();
 	}
 }
diff --git a/arch/mn10300/kernel/signal.c b/arch/mn10300/kernel/signal.c
index feb2f2e810db..a21f43bc68e2 100644
--- a/arch/mn10300/kernel/signal.c
+++ b/arch/mn10300/kernel/signal.c
@@ -568,5 +568,7 @@ asmlinkage void do_notify_resume(struct pt_regs *regs, u32 thread_info_flags)
 	if (thread_info_flags & _TIF_NOTIFY_RESUME) {
 		clear_thread_flag(TIF_NOTIFY_RESUME);
 		tracehook_notify_resume(__frame);
+		if (current->replacement_session_keyring)
+			key_replace_session_keyring();
 	}
 }
diff --git a/arch/parisc/kernel/signal.c b/arch/parisc/kernel/signal.c
index b3bfc4326703..5ca1c02b805a 100644
--- a/arch/parisc/kernel/signal.c
+++ b/arch/parisc/kernel/signal.c
@@ -649,5 +649,7 @@ void do_notify_resume(struct pt_regs *regs, long in_syscall)
 	if (test_thread_flag(TIF_NOTIFY_RESUME)) {
 		clear_thread_flag(TIF_NOTIFY_RESUME);
 		tracehook_notify_resume(regs);
+		if (current->replacement_session_keyring)
+			key_replace_session_keyring();
 	}
 }
diff --git a/arch/s390/kernel/signal.c b/arch/s390/kernel/signal.c
index 062bd64e65fa..6b4fef877f9d 100644
--- a/arch/s390/kernel/signal.c
+++ b/arch/s390/kernel/signal.c
@@ -536,4 +536,6 @@ void do_notify_resume(struct pt_regs *regs)
 {
 	clear_thread_flag(TIF_NOTIFY_RESUME);
 	tracehook_notify_resume(regs);
+	if (current->replacement_session_keyring)
+		key_replace_session_keyring();
 }
diff --git a/arch/sh/kernel/signal_32.c b/arch/sh/kernel/signal_32.c
index b5afbec1db59..04a21883f327 100644
--- a/arch/sh/kernel/signal_32.c
+++ b/arch/sh/kernel/signal_32.c
@@ -640,5 +640,7 @@ asmlinkage void do_notify_resume(struct pt_regs *regs, unsigned int save_r0,
 	if (thread_info_flags & _TIF_NOTIFY_RESUME) {
 		clear_thread_flag(TIF_NOTIFY_RESUME);
 		tracehook_notify_resume(regs);
+		if (current->replacement_session_keyring)
+			key_replace_session_keyring();
 	}
 }
diff --git a/arch/sh/kernel/signal_64.c b/arch/sh/kernel/signal_64.c
index 0663a0ee6021..9e5c9b1d7e98 100644
--- a/arch/sh/kernel/signal_64.c
+++ b/arch/sh/kernel/signal_64.c
@@ -772,5 +772,7 @@ asmlinkage void do_notify_resume(struct pt_regs *regs, unsigned long thread_info
 	if (thread_info_flags & _TIF_NOTIFY_RESUME) {
 		clear_thread_flag(TIF_NOTIFY_RESUME);
 		tracehook_notify_resume(regs);
+		if (current->replacement_session_keyring)
+			key_replace_session_keyring();
 	}
 }
diff --git a/arch/sparc/kernel/signal_32.c b/arch/sparc/kernel/signal_32.c
index 181d069a2d44..7ce1a1005b1d 100644
--- a/arch/sparc/kernel/signal_32.c
+++ b/arch/sparc/kernel/signal_32.c
@@ -590,6 +590,8 @@ void do_notify_resume(struct pt_regs *regs, unsigned long orig_i0,
 	if (thread_info_flags & _TIF_NOTIFY_RESUME) {
 		clear_thread_flag(TIF_NOTIFY_RESUME);
 		tracehook_notify_resume(regs);
+		if (current->replacement_session_keyring)
+			key_replace_session_keyring();
 	}
 }
 
diff --git a/arch/sparc/kernel/signal_64.c b/arch/sparc/kernel/signal_64.c
index ec82d76dc6f2..647afbda7ae1 100644
--- a/arch/sparc/kernel/signal_64.c
+++ b/arch/sparc/kernel/signal_64.c
@@ -613,5 +613,8 @@ void do_notify_resume(struct pt_regs *regs, unsigned long orig_i0, unsigned long
 	if (thread_info_flags & _TIF_NOTIFY_RESUME) {
 		clear_thread_flag(TIF_NOTIFY_RESUME);
 		tracehook_notify_resume(regs);
+		if (current->replacement_session_keyring)
+			key_replace_session_keyring();
 	}
 }
+
diff --git a/arch/x86/kernel/signal.c b/arch/x86/kernel/signal.c
index 4c578751e94e..81e58238c4ce 100644
--- a/arch/x86/kernel/signal.c
+++ b/arch/x86/kernel/signal.c
@@ -869,6 +869,8 @@ do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags)
 	if (thread_info_flags & _TIF_NOTIFY_RESUME) {
 		clear_thread_flag(TIF_NOTIFY_RESUME);
 		tracehook_notify_resume(regs);
+		if (current->replacement_session_keyring)
+			key_replace_session_keyring();
 	}
 
 #ifdef CONFIG_X86_32
diff --git a/include/linux/cred.h b/include/linux/cred.h
index 85439abdbc80..24520a539c6f 100644
--- a/include/linux/cred.h
+++ b/include/linux/cred.h
@@ -152,6 +152,7 @@ struct cred {
 extern void __put_cred(struct cred *);
 extern void exit_creds(struct task_struct *);
 extern int copy_creds(struct task_struct *, unsigned long);
+extern struct cred *cred_alloc_blank(void);
 extern struct cred *prepare_creds(void);
 extern struct cred *prepare_exec_creds(void);
 extern struct cred *prepare_usermodehelper_creds(void);
diff --git a/include/linux/key.h b/include/linux/key.h
index 33e0165de100..cd50dfa1d4c2 100644
--- a/include/linux/key.h
+++ b/include/linux/key.h
@@ -278,6 +278,8 @@ static inline key_serial_t key_serial(struct key *key)
 extern ctl_table key_sysctls[];
 #endif
 
+extern void key_replace_session_keyring(void);
+
 /*
  * the userspace interface
  */
@@ -300,6 +302,7 @@ extern void key_init(void);
 #define key_fsuid_changed(t)		do { } while(0)
 #define key_fsgid_changed(t)		do { } while(0)
 #define key_init()			do { } while(0)
+#define key_replace_session_keyring()	do { } while(0)
 
 #endif /* CONFIG_KEYS */
 #endif /* __KERNEL__ */
diff --git a/include/linux/keyctl.h b/include/linux/keyctl.h
index c0688eb72093..bd383f1944fb 100644
--- a/include/linux/keyctl.h
+++ b/include/linux/keyctl.h
@@ -52,5 +52,6 @@
 #define KEYCTL_SET_TIMEOUT		15	/* set key timeout */
 #define KEYCTL_ASSUME_AUTHORITY		16	/* assume request_key() authorisation */
 #define KEYCTL_GET_SECURITY		17	/* get key security label */
+#define KEYCTL_SESSION_TO_PARENT	18	/* apply session keyring to parent process */
 
 #endif /*  _LINUX_KEYCTL_H */
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 5c7ce13c1696..9304027673b0 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1292,6 +1292,7 @@ struct task_struct {
 	struct mutex cred_guard_mutex;	/* guard against foreign influences on
 					 * credential calculations
 					 * (notably. ptrace) */
+	struct cred *replacement_session_keyring; /* for KEYCTL_SESSION_TO_PARENT */
 
 	char comm[TASK_COMM_LEN]; /* executable name excluding path
 				     - access with [gs]et_task_comm (which lock
diff --git a/include/linux/security.h b/include/linux/security.h
index 40ba39ea68ce..97de3fe3dd0d 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -653,6 +653,11 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
  *	manual page for definitions of the @clone_flags.
  *	@clone_flags contains the flags indicating what should be shared.
  *	Return 0 if permission is granted.
+ * @cred_alloc_blank:
+ *	@cred points to the credentials.
+ *	@gfp indicates the atomicity of any memory allocations.
+ *	Only allocate sufficient memory and attach to @cred such that
+ *	cred_transfer() will not get ENOMEM.
  * @cred_free:
  *	@cred points to the credentials.
  *	Deallocate and clear the cred->security field in a set of credentials.
@@ -665,6 +670,10 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
  *	@new points to the new credentials.
  *	@old points to the original credentials.
  *	Install a new set of credentials.
+ * @cred_transfer:
+ *	@new points to the new credentials.
+ *	@old points to the original credentials.
+ *	Transfer data from original creds to new creds
  * @kernel_act_as:
  *	Set the credentials for a kernel service to act as (subjective context).
  *	@new points to the credentials to be modified.
@@ -1103,6 +1112,13 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
  *	Return the length of the string (including terminating NUL) or -ve if
  *      an error.
  *	May also return 0 (and a NULL buffer pointer) if there is no label.
+ * @key_session_to_parent:
+ *	Forcibly assign the session keyring from a process to its parent
+ *	process.
+ *	@cred: Pointer to process's credentials
+ *	@parent_cred: Pointer to parent process's credentials
+ *	@keyring: Proposed new session keyring
+ *	Return 0 if permission is granted, -ve error otherwise.
  *
  * Security hooks affecting all System V IPC operations.
  *
@@ -1498,10 +1514,12 @@ struct security_operations {
 	int (*dentry_open) (struct file *file, const struct cred *cred);
 
 	int (*task_create) (unsigned long clone_flags);
+	int (*cred_alloc_blank) (struct cred *cred, gfp_t gfp);
 	void (*cred_free) (struct cred *cred);
 	int (*cred_prepare)(struct cred *new, const struct cred *old,
 			    gfp_t gfp);
 	void (*cred_commit)(struct cred *new, const struct cred *old);
+	void (*cred_transfer)(struct cred *new, const struct cred *old);
 	int (*kernel_act_as)(struct cred *new, u32 secid);
 	int (*kernel_create_files_as)(struct cred *new, struct inode *inode);
 	int (*kernel_module_request)(void);
@@ -1639,6 +1657,9 @@ struct security_operations {
 			       const struct cred *cred,
 			       key_perm_t perm);
 	int (*key_getsecurity)(struct key *key, char **_buffer);
+	int (*key_session_to_parent)(const struct cred *cred,
+				     const struct cred *parent_cred,
+				     struct key *key);
 #endif	/* CONFIG_KEYS */
 
 #ifdef CONFIG_AUDIT
@@ -1755,9 +1776,11 @@ int security_file_send_sigiotask(struct task_struct *tsk,
 int security_file_receive(struct file *file);
 int security_dentry_open(struct file *file, const struct cred *cred);
 int security_task_create(unsigned long clone_flags);
+int security_cred_alloc_blank(struct cred *cred, gfp_t gfp);
 void security_cred_free(struct cred *cred);
 int security_prepare_creds(struct cred *new, const struct cred *old, gfp_t gfp);
 void security_commit_creds(struct cred *new, const struct cred *old);
+void security_transfer_creds(struct cred *new, const struct cred *old);
 int security_kernel_act_as(struct cred *new, u32 secid);
 int security_kernel_create_files_as(struct cred *new, struct inode *inode);
 int security_kernel_module_request(void);
@@ -2286,6 +2309,9 @@ static inline int security_task_create(unsigned long clone_flags)
 	return 0;
 }
 
+static inline void security_cred_alloc_blank(struct cred *cred, gfp_t gfp)
+{ }
+
 static inline void security_cred_free(struct cred *cred)
 { }
 
@@ -2301,6 +2327,11 @@ static inline void security_commit_creds(struct cred *new,
 {
 }
 
+static inline void security_transfer_creds(struct cred *new,
+					   const struct cred *old)
+{
+}
+
 static inline int security_kernel_act_as(struct cred *cred, u32 secid)
 {
 	return 0;
@@ -2923,6 +2954,9 @@ void security_key_free(struct key *key);
 int security_key_permission(key_ref_t key_ref,
 			    const struct cred *cred, key_perm_t perm);
 int security_key_getsecurity(struct key *key, char **_buffer);
+int security_key_session_to_parent(const struct cred *cred,
+				   const struct cred *parent_cred,
+				   struct key *key);
 
 #else
 
@@ -2950,6 +2984,10 @@ static inline int security_key_getsecurity(struct key *key, char **_buffer)
 	return 0;
 }
 
+static inline int security_key_session_to_parent(const struct cred *cred,
+						 const struct cred *parent_cred,
+						 struct key *key);
+
 #endif
 #endif /* CONFIG_KEYS */
 
diff --git a/kernel/cred.c b/kernel/cred.c
index 24dd2f5104b1..006fcab009d5 100644
--- a/kernel/cred.c
+++ b/kernel/cred.c
@@ -199,6 +199,49 @@ void exit_creds(struct task_struct *tsk)
 	validate_creds(cred);
 	alter_cred_subscribers(cred, -1);
 	put_cred(cred);
+
+	cred = (struct cred *) tsk->replacement_session_keyring;
+	if (cred) {
+		tsk->replacement_session_keyring = NULL;
+		validate_creds(cred);
+		put_cred(cred);
+	}
+}
+
+/*
+ * Allocate blank credentials, such that the credentials can be filled in at a
+ * later date without risk of ENOMEM.
+ */
+struct cred *cred_alloc_blank(void)
+{
+	struct cred *new;
+
+	new = kmem_cache_zalloc(cred_jar, GFP_KERNEL);
+	if (!new)
+		return NULL;
+
+#ifdef CONFIG_KEYS
+	new->tgcred = kzalloc(sizeof(*new->tgcred), GFP_KERNEL);
+	if (!new->tgcred) {
+		kfree(new);
+		return NULL;
+	}
+	atomic_set(&new->tgcred->usage, 1);
+#endif
+
+	atomic_set(&new->usage, 1);
+
+	if (security_cred_alloc_blank(new, GFP_KERNEL) < 0)
+		goto error;
+
+#ifdef CONFIG_DEBUG_CREDENTIALS
+	new->magic = CRED_MAGIC;
+#endif
+	return new;
+
+error:
+	abort_creds(new);
+	return NULL;
 }
 
 /**
diff --git a/security/capability.c b/security/capability.c
index 06400cf07757..93a2ffe65905 100644
--- a/security/capability.c
+++ b/security/capability.c
@@ -373,6 +373,11 @@ static int cap_task_create(unsigned long clone_flags)
 	return 0;
 }
 
+static int cap_cred_alloc_blank(struct cred *cred, gfp_t gfp)
+{
+	return 0;
+}
+
 static void cap_cred_free(struct cred *cred)
 {
 }
@@ -386,6 +391,10 @@ static void cap_cred_commit(struct cred *new, const struct cred *old)
 {
 }
 
+static void cap_cred_transfer(struct cred *new, const struct cred *old)
+{
+}
+
 static int cap_kernel_act_as(struct cred *new, u32 secid)
 {
 	return 0;
@@ -836,6 +845,13 @@ static int cap_key_getsecurity(struct key *key, char **_buffer)
 	return 0;
 }
 
+static int cap_key_session_to_parent(const struct cred *cred,
+				     const struct cred *parent_cred,
+				     struct key *key)
+{
+	return 0;
+}
+
 #endif /* CONFIG_KEYS */
 
 #ifdef CONFIG_AUDIT
@@ -961,9 +977,11 @@ void security_fixup_ops(struct security_operations *ops)
 	set_to_cap_if_null(ops, file_receive);
 	set_to_cap_if_null(ops, dentry_open);
 	set_to_cap_if_null(ops, task_create);
+	set_to_cap_if_null(ops, cred_alloc_blank);
 	set_to_cap_if_null(ops, cred_free);
 	set_to_cap_if_null(ops, cred_prepare);
 	set_to_cap_if_null(ops, cred_commit);
+	set_to_cap_if_null(ops, cred_transfer);
 	set_to_cap_if_null(ops, kernel_act_as);
 	set_to_cap_if_null(ops, kernel_create_files_as);
 	set_to_cap_if_null(ops, kernel_module_request);
@@ -1063,6 +1081,7 @@ void security_fixup_ops(struct security_operations *ops)
 	set_to_cap_if_null(ops, key_free);
 	set_to_cap_if_null(ops, key_permission);
 	set_to_cap_if_null(ops, key_getsecurity);
+	set_to_cap_if_null(ops, key_session_to_parent);
 #endif	/* CONFIG_KEYS */
 #ifdef CONFIG_AUDIT
 	set_to_cap_if_null(ops, audit_rule_init);
diff --git a/security/keys/compat.c b/security/keys/compat.c
index c766c68a63bc..792c0a611a6d 100644
--- a/security/keys/compat.c
+++ b/security/keys/compat.c
@@ -82,6 +82,9 @@ asmlinkage long compat_sys_keyctl(u32 option,
 	case KEYCTL_GET_SECURITY:
 		return keyctl_get_security(arg2, compat_ptr(arg3), arg4);
 
+	case KEYCTL_SESSION_TO_PARENT:
+		return keyctl_session_to_parent();
+
 	default:
 		return -EOPNOTSUPP;
 	}
diff --git a/security/keys/gc.c b/security/keys/gc.c
index 44adc325e15c..1e616aef55fd 100644
--- a/security/keys/gc.c
+++ b/security/keys/gc.c
@@ -65,6 +65,7 @@ static void key_gc_timer_func(unsigned long data)
  * - return true if we altered the keyring
  */
 static bool key_gc_keyring(struct key *keyring, time_t limit)
+	__releases(key_serial_lock)
 {
 	struct keyring_list *klist;
 	struct key *key;
diff --git a/security/keys/internal.h b/security/keys/internal.h
index fb830514c337..24ba0307b7ad 100644
--- a/security/keys/internal.h
+++ b/security/keys/internal.h
@@ -201,6 +201,7 @@ extern long keyctl_set_timeout(key_serial_t, unsigned);
 extern long keyctl_assume_authority(key_serial_t);
 extern long keyctl_get_security(key_serial_t keyid, char __user *buffer,
 				size_t buflen);
+extern long keyctl_session_to_parent(void);
 
 /*
  * debugging key validation
diff --git a/security/keys/keyctl.c b/security/keys/keyctl.c
index 736d7800f97f..74c968524592 100644
--- a/security/keys/keyctl.c
+++ b/security/keys/keyctl.c
@@ -1228,6 +1228,105 @@ long keyctl_get_security(key_serial_t keyid,
 	return ret;
 }
 
+/*
+ * attempt to install the calling process's session keyring on the process's
+ * parent process
+ * - the keyring must exist and must grant us LINK permission
+ * - implements keyctl(KEYCTL_SESSION_TO_PARENT)
+ */
+long keyctl_session_to_parent(void)
+{
+	struct task_struct *me, *parent;
+	const struct cred *mycred, *pcred;
+	struct cred *cred, *oldcred;
+	key_ref_t keyring_r;
+	int ret;
+
+	keyring_r = lookup_user_key(KEY_SPEC_SESSION_KEYRING, 0, KEY_LINK);
+	if (IS_ERR(keyring_r))
+		return PTR_ERR(keyring_r);
+
+	/* our parent is going to need a new cred struct, a new tgcred struct
+	 * and new security data, so we allocate them here to prevent ENOMEM in
+	 * our parent */
+	ret = -ENOMEM;
+	cred = cred_alloc_blank();
+	if (!cred)
+		goto error_keyring;
+
+	cred->tgcred->session_keyring = key_ref_to_ptr(keyring_r);
+	keyring_r = NULL;
+
+	me = current;
+	write_lock_irq(&tasklist_lock);
+
+	parent = me->real_parent;
+	ret = -EPERM;
+
+	/* the parent mustn't be init and mustn't be a kernel thread */
+	if (parent->pid <= 1 || !parent->mm)
+		goto not_permitted;
+
+	/* the parent must be single threaded */
+	if (atomic_read(&parent->signal->count) != 1)
+		goto not_permitted;
+
+	/* the parent and the child must have different session keyrings or
+	 * there's no point */
+	mycred = current_cred();
+	pcred = __task_cred(parent);
+	if (mycred == pcred ||
+	    mycred->tgcred->session_keyring == pcred->tgcred->session_keyring)
+		goto already_same;
+
+	/* the parent must have the same effective ownership and mustn't be
+	 * SUID/SGID */
+	if (pcred-> uid	!= mycred->euid	||
+	    pcred->euid	!= mycred->euid	||
+	    pcred->suid	!= mycred->euid	||
+	    pcred-> gid	!= mycred->egid	||
+	    pcred->egid	!= mycred->egid	||
+	    pcred->sgid	!= mycred->egid)
+		goto not_permitted;
+
+	/* the keyrings must have the same UID */
+	if (pcred ->tgcred->session_keyring->uid != mycred->euid ||
+	    mycred->tgcred->session_keyring->uid != mycred->euid)
+		goto not_permitted;
+
+	/* the LSM must permit the replacement of the parent's keyring with the
+	 * keyring from this process */
+	ret = security_key_session_to_parent(mycred, pcred,
+					     key_ref_to_ptr(keyring_r));
+	if (ret < 0)
+		goto not_permitted;
+
+	/* if there's an already pending keyring replacement, then we replace
+	 * that */
+	oldcred = parent->replacement_session_keyring;
+
+	/* the replacement session keyring is applied just prior to userspace
+	 * restarting */
+	parent->replacement_session_keyring = cred;
+	cred = NULL;
+	set_ti_thread_flag(task_thread_info(parent), TIF_NOTIFY_RESUME);
+
+	write_unlock_irq(&tasklist_lock);
+	if (oldcred)
+		put_cred(oldcred);
+	return 0;
+
+already_same:
+	ret = 0;
+not_permitted:
+	put_cred(cred);
+	return ret;
+
+error_keyring:
+	key_ref_put(keyring_r);
+	return ret;
+}
+
 /*****************************************************************************/
 /*
  * the key control system call
@@ -1313,6 +1412,9 @@ SYSCALL_DEFINE5(keyctl, int, option, unsigned long, arg2, unsigned long, arg3,
 					   (char __user *) arg3,
 					   (size_t) arg4);
 
+	case KEYCTL_SESSION_TO_PARENT:
+		return keyctl_session_to_parent();
+
 	default:
 		return -EOPNOTSUPP;
 	}
diff --git a/security/keys/process_keys.c b/security/keys/process_keys.c
index 4739cfbb41b7..5c23afb31ece 100644
--- a/security/keys/process_keys.c
+++ b/security/keys/process_keys.c
@@ -17,6 +17,7 @@
 #include <linux/fs.h>
 #include <linux/err.h>
 #include <linux/mutex.h>
+#include <linux/security.h>
 #include <linux/user_namespace.h>
 #include <asm/uaccess.h>
 #include "internal.h"
@@ -768,3 +769,51 @@ error:
 	abort_creds(new);
 	return ret;
 }
+
+/*
+ * Replace a process's session keyring when that process resumes userspace on
+ * behalf of one of its children
+ */
+void key_replace_session_keyring(void)
+{
+	const struct cred *old;
+	struct cred *new;
+
+	if (!current->replacement_session_keyring)
+		return;
+
+	write_lock_irq(&tasklist_lock);
+	new = current->replacement_session_keyring;
+	current->replacement_session_keyring = NULL;
+	write_unlock_irq(&tasklist_lock);
+
+	if (!new)
+		return;
+
+	old = current_cred();
+	new->  uid	= old->  uid;
+	new-> euid	= old-> euid;
+	new-> suid	= old-> suid;
+	new->fsuid	= old->fsuid;
+	new->  gid	= old->  gid;
+	new-> egid	= old-> egid;
+	new-> sgid	= old-> sgid;
+	new->fsgid	= old->fsgid;
+	new->user	= get_uid(old->user);
+	new->group_info	= get_group_info(old->group_info);
+
+	new->securebits	= old->securebits;
+	new->cap_inheritable	= old->cap_inheritable;
+	new->cap_permitted	= old->cap_permitted;
+	new->cap_effective	= old->cap_effective;
+	new->cap_bset		= old->cap_bset;
+
+	new->jit_keyring	= old->jit_keyring;
+	new->thread_keyring	= key_get(old->thread_keyring);
+	new->tgcred->tgid	= old->tgcred->tgid;
+	new->tgcred->process_keyring = key_get(old->tgcred->process_keyring);
+
+	security_transfer_creds(new, old);
+
+	commit_creds(new);
+}
diff --git a/security/security.c b/security/security.c
index f88eaf6b14cc..d8b727637f02 100644
--- a/security/security.c
+++ b/security/security.c
@@ -684,6 +684,11 @@ int security_task_create(unsigned long clone_flags)
 	return security_ops->task_create(clone_flags);
 }
 
+int security_cred_alloc_blank(struct cred *cred, gfp_t gfp)
+{
+	return security_ops->cred_alloc_blank(cred, gfp);
+}
+
 void security_cred_free(struct cred *cred)
 {
 	security_ops->cred_free(cred);
@@ -699,6 +704,11 @@ void security_commit_creds(struct cred *new, const struct cred *old)
 	security_ops->cred_commit(new, old);
 }
 
+void security_transfer_creds(struct cred *new, const struct cred *old)
+{
+	security_ops->cred_transfer(new, old);
+}
+
 int security_kernel_act_as(struct cred *new, u32 secid)
 {
 	return security_ops->kernel_act_as(new, secid);
@@ -1241,6 +1251,13 @@ int security_key_getsecurity(struct key *key, char **_buffer)
 	return security_ops->key_getsecurity(key, _buffer);
 }
 
+int security_key_session_to_parent(const struct cred *cred,
+				   const struct cred *parent_cred,
+				   struct key *key)
+{
+	return security_ops->key_session_to_parent(cred, parent_cred, key);
+}
+
 #endif	/* CONFIG_KEYS */
 
 #ifdef CONFIG_AUDIT
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index c3bb31ecc5aa..134a9c0d2004 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -3232,6 +3232,21 @@ static int selinux_task_create(unsigned long clone_flags)
 	return current_has_perm(current, PROCESS__FORK);
 }
 
+/*
+ * allocate the SELinux part of blank credentials
+ */
+static int selinux_cred_alloc_blank(struct cred *cred, gfp_t gfp)
+{
+	struct task_security_struct *tsec;
+
+	tsec = kzalloc(sizeof(struct task_security_struct), gfp);
+	if (!tsec)
+		return -ENOMEM;
+
+	cred->security = tsec;
+	return 0;
+}
+
 /*
  * detach and free the LSM part of a set of credentials
  */
@@ -3263,6 +3278,17 @@ static int selinux_cred_prepare(struct cred *new, const struct cred *old,
 	return 0;
 }
 
+/*
+ * transfer the SELinux data to a blank set of creds
+ */
+static void selinux_cred_transfer(struct cred *new, const struct cred *old)
+{
+	const struct task_security_struct *old_tsec = old->security;
+	struct task_security_struct *tsec = new->security;
+
+	*tsec = *old_tsec;
+}
+
 /*
  * set the security data for a kernel service
  * - all the creation contexts are set to unlabelled
@@ -5469,8 +5495,10 @@ static struct security_operations selinux_ops = {
 	.dentry_open =			selinux_dentry_open,
 
 	.task_create =			selinux_task_create,
+	.cred_alloc_blank =		selinux_cred_alloc_blank,
 	.cred_free =			selinux_cred_free,
 	.cred_prepare =			selinux_cred_prepare,
+	.cred_transfer =		selinux_cred_transfer,
 	.kernel_act_as =		selinux_kernel_act_as,
 	.kernel_create_files_as =	selinux_kernel_create_files_as,
 	.kernel_module_request =	selinux_kernel_module_request,
diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c
index c243a2b25832..969f5fee1906 100644
--- a/security/smack/smack_lsm.c
+++ b/security/smack/smack_lsm.c
@@ -1079,6 +1079,22 @@ static int smack_file_receive(struct file *file)
  * Task hooks
  */
 
+/**
+ * smack_cred_alloc_blank - "allocate" blank task-level security credentials
+ * @new: the new credentials
+ * @gfp: the atomicity of any memory allocations
+ *
+ * Prepare a blank set of credentials for modification.  This must allocate all
+ * the memory the LSM module might require such that cred_transfer() can
+ * complete without error.
+ */
+static int smack_cred_alloc_blank(struct cred *cred, gfp_t gfp)
+{
+	cred->security = NULL;
+	return 0;
+}
+
+
 /**
  * smack_cred_free - "free" task-level security credentials
  * @cred: the credentials in question
@@ -1116,6 +1132,18 @@ static void smack_cred_commit(struct cred *new, const struct cred *old)
 {
 }
 
+/**
+ * smack_cred_transfer - Transfer the old credentials to the new credentials
+ * @new: the new credentials
+ * @old: the original credentials
+ *
+ * Fill in a set of blank credentials from another set of credentials.
+ */
+static void smack_cred_transfer(struct cred *new, const struct cred *old)
+{
+	new->security = old->security;
+}
+
 /**
  * smack_kernel_act_as - Set the subjective context in a set of credentials
  * @new: points to the set of credentials to be modified.
@@ -3073,9 +3101,11 @@ struct security_operations smack_ops = {
 	.file_send_sigiotask = 		smack_file_send_sigiotask,
 	.file_receive = 		smack_file_receive,
 
+	.cred_alloc_blank =		smack_cred_alloc_blank,
 	.cred_free =			smack_cred_free,
 	.cred_prepare =			smack_cred_prepare,
 	.cred_commit =			smack_cred_commit,
+	.cred_transfer =		smack_cred_transfer,
 	.kernel_act_as =		smack_kernel_act_as,
 	.kernel_create_files_as =	smack_kernel_create_files_as,
 	.task_setpgid = 		smack_task_setpgid,
diff --git a/security/tomoyo/tomoyo.c b/security/tomoyo/tomoyo.c
index 35a13e7915e4..9548a0984cc4 100644
--- a/security/tomoyo/tomoyo.c
+++ b/security/tomoyo/tomoyo.c
@@ -14,6 +14,12 @@
 #include "tomoyo.h"
 #include "realpath.h"
 
+static int tomoyo_cred_alloc_blank(struct cred *new, gfp_t gfp)
+{
+	new->security = NULL;
+	return 0;
+}
+
 static int tomoyo_cred_prepare(struct cred *new, const struct cred *old,
 			       gfp_t gfp)
 {
@@ -25,6 +31,15 @@ static int tomoyo_cred_prepare(struct cred *new, const struct cred *old,
 	return 0;
 }
 
+static void tomoyo_cred_transfer(struct cred *new, const struct cred *old)
+{
+	/*
+	 * Since "struct tomoyo_domain_info *" is a sharable pointer,
+	 * we don't need to duplicate.
+	 */
+	new->security = old->security;
+}
+
 static int tomoyo_bprm_set_creds(struct linux_binprm *bprm)
 {
 	int rc;
@@ -262,7 +277,9 @@ static int tomoyo_dentry_open(struct file *f, const struct cred *cred)
  */
 static struct security_operations tomoyo_security_ops = {
 	.name                = "tomoyo",
+	.cred_alloc_blank    = tomoyo_cred_alloc_blank,
 	.cred_prepare        = tomoyo_cred_prepare,
+	.cred_transfer	     = tomoyo_cred_transfer,
 	.bprm_set_creds      = tomoyo_bprm_set_creds,
 	.bprm_check_security = tomoyo_bprm_check_security,
 #ifdef CONFIG_SYSCTL
-- 
cgit v1.2.3


From e3e59876e82a5e1a07f365d5474e7b6943524725 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Thu, 3 Sep 2009 14:02:10 +0200
Subject: x86/amd-iommu: Dump fault entry on DTE error

This patch adds code to dump the content of the device table
entry which caused an ILLEGAL_DEV_TABLE_ENTRY error from the
IOMMU hardware.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 6c99f5037801..364c6de26374 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -138,6 +138,15 @@ static int iommu_has_npcache(struct amd_iommu *iommu)
  *
  ****************************************************************************/
 
+static void dump_dte_entry(u16 devid)
+{
+	int i;
+
+	for (i = 0; i < 8; ++i)
+		pr_err("AMD-Vi: DTE[%d]: %08x\n", i,
+			amd_iommu_dev_table[devid].data[i]);
+}
+
 static void iommu_print_event(void *__evt)
 {
 	u32 *event = __evt;
@@ -155,6 +164,7 @@ static void iommu_print_event(void *__evt)
 		       "address=0x%016llx flags=0x%04x]\n",
 		       PCI_BUS(devid), PCI_SLOT(devid), PCI_FUNC(devid),
 		       address, flags);
+		dump_dte_entry(devid);
 		break;
 	case EVENT_TYPE_IO_FAULT:
 		printk("IO_PAGE_FAULT device=%02x:%02x.%x "
-- 
cgit v1.2.3


From 945b4ac44e5700acd3d974c176c8ace34b4d2e8e Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Thu, 3 Sep 2009 14:25:02 +0200
Subject: x86/amd-iommu: Dump illegal command on ILLEGAL_COMMAND_ERROR

This patch adds code to dump the command which caused an
ILLEGAL_COMMAND_ERROR raised by the IOMMU hardware.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 364c6de26374..e62b35f5df1b 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -147,6 +147,15 @@ static void dump_dte_entry(u16 devid)
 			amd_iommu_dev_table[devid].data[i]);
 }
 
+static void dump_command(unsigned long phys_addr)
+{
+	struct iommu_cmd *cmd = phys_to_virt(phys_addr);
+	int i;
+
+	for (i = 0; i < 4; ++i)
+		pr_err("AMD-Vi: CMD[%d]: %08x\n", i, cmd->data[i]);
+}
+
 static void iommu_print_event(void *__evt)
 {
 	u32 *event = __evt;
@@ -186,6 +195,7 @@ static void iommu_print_event(void *__evt)
 		break;
 	case EVENT_TYPE_ILL_CMD:
 		printk("ILLEGAL_COMMAND_ERROR address=0x%016llx]\n", address);
+		dump_command(address);
 		break;
 	case EVENT_TYPE_CMD_HARD_ERR:
 		printk("COMMAND_HARDWARE_ERROR address=0x%016llx "
-- 
cgit v1.2.3


From e394d72aa8b319211b8f947d151d9d50b0fde842 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Thu, 3 Sep 2009 15:28:33 +0200
Subject: x86/amd-iommu: Introduce function for iommu-local domain flush

This patch introduces a function to flush all domain tlbs
for on one given IOMMU. This is required later to reset the
command buffer on one IOMMU.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu.c | 49 ++++++++++++++++++++++++++++++---------------
 1 file changed, 33 insertions(+), 16 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index e62b35f5df1b..64cc582feb9b 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -465,38 +465,55 @@ static void iommu_flush_tlb_pde(struct amd_iommu *iommu, u16 domid)
 }
 
 /*
- * This function is used to flush the IO/TLB for a given protection domain
- * on every IOMMU in the system
+ * This function flushes one domain on one IOMMU
  */
-static void iommu_flush_domain(u16 domid)
+static void flush_domain_on_iommu(struct amd_iommu *iommu, u16 domid)
 {
-	unsigned long flags;
-	struct amd_iommu *iommu;
 	struct iommu_cmd cmd;
-
-	INC_STATS_COUNTER(domain_flush_all);
+	unsigned long flags;
 
 	__iommu_build_inv_iommu_pages(&cmd, CMD_INV_IOMMU_ALL_PAGES_ADDRESS,
 				      domid, 1, 1);
 
-	for_each_iommu(iommu) {
-		spin_lock_irqsave(&iommu->lock, flags);
-		__iommu_queue_command(iommu, &cmd);
-		__iommu_completion_wait(iommu);
-		__iommu_wait_for_completion(iommu);
-		spin_unlock_irqrestore(&iommu->lock, flags);
-	}
+	spin_lock_irqsave(&iommu->lock, flags);
+	__iommu_queue_command(iommu, &cmd);
+	__iommu_completion_wait(iommu);
+	__iommu_wait_for_completion(iommu);
+	spin_unlock_irqrestore(&iommu->lock, flags);
 }
 
-void amd_iommu_flush_all_domains(void)
+static void flush_all_domains_on_iommu(struct amd_iommu *iommu)
 {
 	int i;
 
 	for (i = 1; i < MAX_DOMAIN_ID; ++i) {
 		if (!test_bit(i, amd_iommu_pd_alloc_bitmap))
 			continue;
-		iommu_flush_domain(i);
+		flush_domain_on_iommu(iommu, i);
 	}
+
+}
+
+/*
+ * This function is used to flush the IO/TLB for a given protection domain
+ * on every IOMMU in the system
+ */
+static void iommu_flush_domain(u16 domid)
+{
+	struct amd_iommu *iommu;
+
+	INC_STATS_COUNTER(domain_flush_all);
+
+	for_each_iommu(iommu)
+		flush_domain_on_iommu(iommu, domid);
+}
+
+void amd_iommu_flush_all_domains(void)
+{
+	struct amd_iommu *iommu;
+
+	for_each_iommu(iommu)
+		flush_all_domains_on_iommu(iommu);
 }
 
 void amd_iommu_flush_all_devices(void)
-- 
cgit v1.2.3


From f2430bd104bec2706315e9e983a9d9f828ff9565 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 26 Aug 2009 12:10:19 +0200
Subject: x86/amd-iommu: Remove some merge helper code

This patch removes some left-overs which where put into the code to
simplify merging code which also depends on changes in other trees.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu.c | 6 ------
 1 file changed, 6 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 6c99f5037801..70fdef54e066 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -41,9 +41,7 @@ static DEFINE_RWLOCK(amd_iommu_devtable_lock);
 static LIST_HEAD(iommu_pd_list);
 static DEFINE_SPINLOCK(iommu_pd_list_lock);
 
-#ifdef CONFIG_IOMMU_API
 static struct iommu_ops amd_iommu_ops;
-#endif
 
 /*
  * general struct to manage commands send to an IOMMU
@@ -62,10 +60,6 @@ static void dma_ops_reserve_addresses(struct dma_ops_domain *dom,
 				      unsigned long start_page,
 				      unsigned int pages);
 
-#ifndef BUS_NOTIFY_UNBOUND_DRIVER
-#define BUS_NOTIFY_UNBOUND_DRIVER 0x0005
-#endif
-
 #ifdef CONFIG_AMD_IOMMU_STATS
 
 /*
-- 
cgit v1.2.3


From 4c6f40d4e0f0bba77a5f27eec4e1c6d1c457d324 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Tue, 1 Sep 2009 16:43:58 +0200
Subject: x86/amd-iommu: replace "AMD IOMMU" by "AMD-Vi"

This patch replaces the "AMD IOMMU" printk strings with the
official name for the hardware: "AMD-Vi".

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/include/asm/amd_iommu_types.h |  2 +-
 arch/x86/kernel/amd_iommu.c            |  2 +-
 arch/x86/kernel/amd_iommu_init.c       | 12 ++++++------
 3 files changed, 8 insertions(+), 8 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/amd_iommu_types.h b/arch/x86/include/asm/amd_iommu_types.h
index 0c878caaa0a2..106e1305ef86 100644
--- a/arch/x86/include/asm/amd_iommu_types.h
+++ b/arch/x86/include/asm/amd_iommu_types.h
@@ -198,7 +198,7 @@ extern bool amd_iommu_dump;
 #define DUMP_printk(format, arg...)					\
 	do {								\
 		if (amd_iommu_dump)						\
-			printk(KERN_INFO "AMD IOMMU: " format, ## arg);	\
+			printk(KERN_INFO "AMD-Vi: " format, ## arg);	\
 	} while(0);
 
 /*
diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 70fdef54e066..3e62d7836522 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -141,7 +141,7 @@ static void iommu_print_event(void *__evt)
 	int flags = (event[1] >> EVENT_FLAGS_SHIFT) & EVENT_FLAGS_MASK;
 	u64 address = (u64)(((u64)event[3]) << 32) | event[2];
 
-	printk(KERN_ERR "AMD IOMMU: Event logged [");
+	printk(KERN_ERR "AMD-Vi: Event logged [");
 
 	switch (type) {
 	case EVENT_TYPE_ILL_DEV:
diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c
index c1b17e97252e..169958ad6243 100644
--- a/arch/x86/kernel/amd_iommu_init.c
+++ b/arch/x86/kernel/amd_iommu_init.c
@@ -252,7 +252,7 @@ static void __init iommu_feature_disable(struct amd_iommu *iommu, u8 bit)
 /* Function to enable the hardware */
 static void iommu_enable(struct amd_iommu *iommu)
 {
-	printk(KERN_INFO "AMD IOMMU: Enabling IOMMU at %s cap 0x%hx\n",
+	printk(KERN_INFO "AMD-Vi: Enabling IOMMU at %s cap 0x%hx\n",
 	       dev_name(&iommu->dev->dev), iommu->cap_ptr);
 
 	iommu_feature_enable(iommu, CONTROL_IOMMU_EN);
@@ -902,7 +902,7 @@ static int __init iommu_setup_msi(struct amd_iommu *iommu)
 
 	r = request_irq(iommu->dev->irq, amd_iommu_int_handler,
 			IRQF_SAMPLE_RANDOM,
-			"AMD IOMMU",
+			"AMD-Vi",
 			NULL);
 
 	if (r) {
@@ -1150,7 +1150,7 @@ int __init amd_iommu_init(void)
 
 
 	if (no_iommu) {
-		printk(KERN_INFO "AMD IOMMU disabled by kernel command line\n");
+		printk(KERN_INFO "AMD-Vi disabled by kernel command line\n");
 		return 0;
 	}
 
@@ -1248,16 +1248,16 @@ int __init amd_iommu_init(void)
 
 	enable_iommus();
 
-	printk(KERN_INFO "AMD IOMMU: device isolation ");
+	printk(KERN_INFO "AMD-Vi: device isolation ");
 	if (amd_iommu_isolate)
 		printk("enabled\n");
 	else
 		printk("disabled\n");
 
 	if (amd_iommu_unmap_flush)
-		printk(KERN_INFO "AMD IOMMU: IO/TLB flush on unmap enabled\n");
+		printk(KERN_INFO "AMD-Vi: IO/TLB flush on unmap enabled\n");
 	else
-		printk(KERN_INFO "AMD IOMMU: Lazy IO/TLB flushing enabled\n");
+		printk(KERN_INFO "AMD-Vi: Lazy IO/TLB flushing enabled\n");
 
 out:
 	return ret;
-- 
cgit v1.2.3


From ae908c22aa2b9f7d4b41bd02d14e473f79c22dd3 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Tue, 1 Sep 2009 16:52:16 +0200
Subject: x86/amd-iommu: Remove redundant 'IOMMU' string

The 'IOMMU: ' prefix is not necessary because the
DUMP_printk macro already prints its own prefix.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu_init.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c
index 169958ad6243..264b3ef9dd6e 100644
--- a/arch/x86/kernel/amd_iommu_init.c
+++ b/arch/x86/kernel/amd_iommu_init.c
@@ -858,7 +858,7 @@ static int __init init_iommu_all(struct acpi_table_header *table)
 		switch (*p) {
 		case ACPI_IVHD_TYPE:
 
-			DUMP_printk("IOMMU: device: %02x:%02x.%01x cap: %04x "
+			DUMP_printk("device: %02x:%02x.%01x cap: %04x "
 				    "seg: %d flags: %01x info %04x\n",
 				    PCI_BUS(h->devid), PCI_SLOT(h->devid),
 				    PCI_FUNC(h->devid), h->cap_ptr,
-- 
cgit v1.2.3


From e0faf54ee82bf9c07f0307b4391caad4020bd659 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Thu, 3 Sep 2009 15:45:51 +0200
Subject: x86/amd-iommu: fix broken check in amd_iommu_flush_all_devices

The amd_iommu_pd_table is indexed by protection domain
number and not by device id. So this check is broken and
must be removed.

Cc: stable@kernel.org
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu.c | 2 --
 1 file changed, 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 3e62d7836522..009d722af009 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -479,8 +479,6 @@ void amd_iommu_flush_all_devices(void)
 	int i;
 
 	for (i = 0; i <= amd_iommu_last_bdf; ++i) {
-		if (amd_iommu_pd_table[i] == NULL)
-			continue;
 
 		iommu = amd_iommu_rlookup_table[i];
 		if (!iommu)
-- 
cgit v1.2.3


From d586d7852ccd0cecb502bf4809f827e60c486af0 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Thu, 3 Sep 2009 15:39:23 +0200
Subject: x86/amd-iommu: Add function to flush all DTEs on one IOMMU

This function flushes all DTE entries on one IOMMU for all
devices behind this IOMMU. This is required for command
buffer resetting later.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 64cc582feb9b..2dc093370d2d 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -516,6 +516,19 @@ void amd_iommu_flush_all_domains(void)
 		flush_all_domains_on_iommu(iommu);
 }
 
+static void flush_all_devices_for_iommu(struct amd_iommu *iommu)
+{
+	int i;
+
+	for (i = 0; i <= amd_iommu_last_bdf; ++i) {
+		if (iommu != amd_iommu_rlookup_table[i])
+			continue;
+
+		iommu_queue_inv_dev_entry(iommu, i);
+		iommu_completion_wait(iommu);
+	}
+}
+
 void amd_iommu_flush_all_devices(void)
 {
 	struct amd_iommu *iommu;
-- 
cgit v1.2.3


From 93f1cc67cf3196174412adca87321b25c1c986b0 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Thu, 3 Sep 2009 14:50:20 +0200
Subject: x86/amd-iommu: Add reset function for command buffers

This patch factors parts of the command buffer
initialization code into a seperate function which can be
used to reset the command buffer later.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/include/asm/amd_iommu_types.h |  3 +++
 arch/x86/kernel/amd_iommu_init.c       | 20 +++++++++++++++-----
 2 files changed, 18 insertions(+), 5 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/amd_iommu_types.h b/arch/x86/include/asm/amd_iommu_types.h
index 0c878caaa0a2..c54bc979dc17 100644
--- a/arch/x86/include/asm/amd_iommu_types.h
+++ b/arch/x86/include/asm/amd_iommu_types.h
@@ -457,4 +457,7 @@ static inline void amd_iommu_stats_init(void) { }
 
 #endif /* CONFIG_AMD_IOMMU_STATS */
 
+/* some function prototypes */
+extern void amd_iommu_reset_cmd_buffer(struct amd_iommu *iommu);
+
 #endif /* _ASM_X86_AMD_IOMMU_TYPES_H */
diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c
index c1b17e97252e..1752afef948a 100644
--- a/arch/x86/kernel/amd_iommu_init.c
+++ b/arch/x86/kernel/amd_iommu_init.c
@@ -434,6 +434,20 @@ static u8 * __init alloc_command_buffer(struct amd_iommu *iommu)
 	return cmd_buf;
 }
 
+/*
+ * This function resets the command buffer if the IOMMU stopped fetching
+ * commands from it.
+ */
+void amd_iommu_reset_cmd_buffer(struct amd_iommu *iommu)
+{
+	iommu_feature_disable(iommu, CONTROL_CMDBUF_EN);
+
+	writel(0x00, iommu->mmio_base + MMIO_CMD_HEAD_OFFSET);
+	writel(0x00, iommu->mmio_base + MMIO_CMD_TAIL_OFFSET);
+
+	iommu_feature_enable(iommu, CONTROL_CMDBUF_EN);
+}
+
 /*
  * This function writes the command buffer address to the hardware and
  * enables it.
@@ -450,11 +464,7 @@ static void iommu_enable_command_buffer(struct amd_iommu *iommu)
 	memcpy_toio(iommu->mmio_base + MMIO_CMD_BUF_OFFSET,
 		    &entry, sizeof(entry));
 
-	/* set head and tail to zero manually */
-	writel(0x00, iommu->mmio_base + MMIO_CMD_HEAD_OFFSET);
-	writel(0x00, iommu->mmio_base + MMIO_CMD_TAIL_OFFSET);
-
-	iommu_feature_enable(iommu, CONTROL_CMDBUF_EN);
+	amd_iommu_reset_cmd_buffer(iommu);
 }
 
 static void __init free_command_buffer(struct amd_iommu *iommu)
-- 
cgit v1.2.3


From a345b23b79f1900e7d87c3165182504419180de4 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Thu, 3 Sep 2009 15:01:43 +0200
Subject: x86/amd-iommu: Reset command buffer on ILLEGAL_COMMAND_ERROR

On an ILLEGAL_COMMAND_ERROR the IOMMU stops executing
further commands. This patch changes the code to handle this
case better by resetting the command buffer in the IOMMU.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu.c | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 2dc093370d2d..2333d615f5ee 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -61,6 +61,7 @@ static u64* alloc_pte(struct protection_domain *dom,
 static void dma_ops_reserve_addresses(struct dma_ops_domain *dom,
 				      unsigned long start_page,
 				      unsigned int pages);
+static void reset_iommu_command_buffer(struct amd_iommu *iommu);
 
 #ifndef BUS_NOTIFY_UNBOUND_DRIVER
 #define BUS_NOTIFY_UNBOUND_DRIVER 0x0005
@@ -156,7 +157,7 @@ static void dump_command(unsigned long phys_addr)
 		pr_err("AMD-Vi: CMD[%d]: %08x\n", i, cmd->data[i]);
 }
 
-static void iommu_print_event(void *__evt)
+static void iommu_print_event(struct amd_iommu *iommu, void *__evt)
 {
 	u32 *event = __evt;
 	int type  = (event[1] >> EVENT_TYPE_SHIFT)  & EVENT_TYPE_MASK;
@@ -195,6 +196,7 @@ static void iommu_print_event(void *__evt)
 		break;
 	case EVENT_TYPE_ILL_CMD:
 		printk("ILLEGAL_COMMAND_ERROR address=0x%016llx]\n", address);
+		reset_iommu_command_buffer(iommu);
 		dump_command(address);
 		break;
 	case EVENT_TYPE_CMD_HARD_ERR:
@@ -229,7 +231,7 @@ static void iommu_poll_events(struct amd_iommu *iommu)
 	tail = readl(iommu->mmio_base + MMIO_EVT_TAIL_OFFSET);
 
 	while (head != tail) {
-		iommu_print_event(iommu->evt_buf + head);
+		iommu_print_event(iommu, iommu->evt_buf + head);
 		head = (head + EVENT_ENTRY_SIZE) % iommu->evt_buf_size;
 	}
 
@@ -547,6 +549,15 @@ void amd_iommu_flush_all_devices(void)
 	}
 }
 
+static void reset_iommu_command_buffer(struct amd_iommu *iommu)
+{
+	pr_err("AMD-Vi: Resetting IOMMU command buffer\n");
+
+	amd_iommu_reset_cmd_buffer(iommu);
+	flush_all_devices_for_iommu(iommu);
+	flush_all_domains_on_iommu(iommu);
+}
+
 /****************************************************************************
  *
  * The functions below are used the create the page table mappings for
-- 
cgit v1.2.3


From b26e81b871bd18184968f0bb3f12945906eadfce Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Thu, 3 Sep 2009 15:08:09 +0200
Subject: x86/amd-iommu: Panic if IOMMU command buffer reset fails

To prevent the driver from doing recursive command buffer
resets, just panic when that recursion happens.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/include/asm/amd_iommu_types.h | 3 +++
 arch/x86/kernel/amd_iommu.c            | 7 +++++++
 2 files changed, 10 insertions(+)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/amd_iommu_types.h b/arch/x86/include/asm/amd_iommu_types.h
index c54bc979dc17..e676746a4a30 100644
--- a/arch/x86/include/asm/amd_iommu_types.h
+++ b/arch/x86/include/asm/amd_iommu_types.h
@@ -337,6 +337,9 @@ struct amd_iommu {
 	/* if one, we need to send a completion wait command */
 	bool need_sync;
 
+	/* becomes true if a command buffer reset is running */
+	bool reset_in_progress;
+
 	/* default dma_ops domain for that IOMMU */
 	struct dma_ops_domain *default_dom;
 };
diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 2333d615f5ee..b62a2f64dfc5 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -553,9 +553,16 @@ static void reset_iommu_command_buffer(struct amd_iommu *iommu)
 {
 	pr_err("AMD-Vi: Resetting IOMMU command buffer\n");
 
+	if (iommu->reset_in_progress)
+		panic("AMD-Vi: ILLEGAL_COMMAND_ERROR while resetting command buffer\n");
+
+	iommu->reset_in_progress = true;
+
 	amd_iommu_reset_cmd_buffer(iommu);
 	flush_all_devices_for_iommu(iommu);
 	flush_all_domains_on_iommu(iommu);
+
+	iommu->reset_in_progress = false;
 }
 
 /****************************************************************************
-- 
cgit v1.2.3


From 6a1eddd2f951656a6abbd42e2cddc2267c4a639d Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Thu, 3 Sep 2009 15:15:10 +0200
Subject: x86/amd-iommu: Reset command buffer if wait loop fails

Instead of a panic on an comletion wait loop failure, try to
recover from that event from resetting the command buffer.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index b62a2f64dfc5..cfca80bfe75f 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -318,8 +318,11 @@ static void __iommu_wait_for_completion(struct amd_iommu *iommu)
 	status &= ~MMIO_STATUS_COM_WAIT_INT_MASK;
 	writel(status, iommu->mmio_base + MMIO_STATUS_OFFSET);
 
-	if (unlikely(i == EXIT_LOOP_COUNT))
-		panic("AMD IOMMU: Completion wait loop failed\n");
+	if (unlikely(i == EXIT_LOOP_COUNT)) {
+		spin_unlock(&iommu->lock);
+		reset_iommu_command_buffer(iommu);
+		spin_lock(&iommu->lock);
+	}
 }
 
 /*
-- 
cgit v1.2.3


From 9355a08186e52b7c120adea91c984923b54efa10 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 2 Sep 2009 14:24:08 +0200
Subject: x86/amd-iommu: Make fetch_pte aware of dynamic mapping levels

This patch changes the fetch_pte function in the AMD IOMMU
driver to support dynamic mapping levels.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/include/asm/amd_iommu_types.h |  9 +++++++++
 arch/x86/kernel/amd_iommu.c            | 24 +++++++++++++-----------
 2 files changed, 22 insertions(+), 11 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/amd_iommu_types.h b/arch/x86/include/asm/amd_iommu_types.h
index 0c878caaa0a2..7fce4ef77bdd 100644
--- a/arch/x86/include/asm/amd_iommu_types.h
+++ b/arch/x86/include/asm/amd_iommu_types.h
@@ -146,12 +146,21 @@
 #define PAGE_MODE_1_LEVEL 0x01
 #define PAGE_MODE_2_LEVEL 0x02
 #define PAGE_MODE_3_LEVEL 0x03
+#define PAGE_MODE_4_LEVEL 0x04
+#define PAGE_MODE_5_LEVEL 0x05
+#define PAGE_MODE_6_LEVEL 0x06
 
 #define IOMMU_PDE_NL_0   0x000ULL
 #define IOMMU_PDE_NL_1   0x200ULL
 #define IOMMU_PDE_NL_2   0x400ULL
 #define IOMMU_PDE_NL_3   0x600ULL
 
+#define PM_LEVEL_SHIFT(x)	(12 + ((x) * 9))
+#define PM_LEVEL_SIZE(x)	(((x) < 6) ? \
+				  ((1ULL << PM_LEVEL_SHIFT((x))) - 1): \
+				   (0xffffffffffffffffULL))
+#define PM_LEVEL_INDEX(x, a)	(((a) >> PM_LEVEL_SHIFT((x))) & 0x1ffULL)
+
 #define IOMMU_PTE_L2_INDEX(address) (((address) >> 30) & 0x1ffULL)
 #define IOMMU_PTE_L1_INDEX(address) (((address) >> 21) & 0x1ffULL)
 #define IOMMU_PTE_L0_INDEX(address) (((address) >> 12) & 0x1ffULL)
diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 6c99f5037801..29bcd358b6ce 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -61,6 +61,8 @@ static u64* alloc_pte(struct protection_domain *dom,
 static void dma_ops_reserve_addresses(struct dma_ops_domain *dom,
 				      unsigned long start_page,
 				      unsigned int pages);
+static u64 *fetch_pte(struct protection_domain *domain,
+		      unsigned long address);
 
 #ifndef BUS_NOTIFY_UNBOUND_DRIVER
 #define BUS_NOTIFY_UNBOUND_DRIVER 0x0005
@@ -670,24 +672,24 @@ static int init_unity_mappings_for_device(struct dma_ops_domain *dma_dom,
  * This function checks if there is a PTE for a given dma address. If
  * there is one, it returns the pointer to it.
  */
-static u64* fetch_pte(struct protection_domain *domain,
+static u64 *fetch_pte(struct protection_domain *domain,
 		      unsigned long address)
 {
+	int level;
 	u64 *pte;
 
-	pte = &domain->pt_root[IOMMU_PTE_L2_INDEX(address)];
+	level =  domain->mode - 1;
+	pte   = &domain->pt_root[PM_LEVEL_INDEX(level, address)];
 
-	if (!IOMMU_PTE_PRESENT(*pte))
-		return NULL;
+	while (level > 0) {
+		if (!IOMMU_PTE_PRESENT(*pte))
+			return NULL;
 
-	pte = IOMMU_PTE_PAGE(*pte);
-	pte = &pte[IOMMU_PTE_L1_INDEX(address)];
+		level -= 1;
 
-	if (!IOMMU_PTE_PRESENT(*pte))
-		return NULL;
-
-	pte = IOMMU_PTE_PAGE(*pte);
-	pte = &pte[IOMMU_PTE_L0_INDEX(address)];
+		pte = IOMMU_PTE_PAGE(*pte);
+		pte = &pte[PM_LEVEL_INDEX(level, address)];
+	}
 
 	return pte;
 }
-- 
cgit v1.2.3


From 38a76eeeafb251bf67d143a34b37a8105cff302e Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 2 Sep 2009 17:02:47 +0200
Subject: x86/amd-iommu: Use fetch_pte in iommu_unmap_page

Instead of reimplementing existing logic use fetch_pte to
walk the page table in iommu_unmap_page.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu.c | 19 +++----------------
 1 file changed, 3 insertions(+), 16 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 29bcd358b6ce..5e5279899994 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -546,23 +546,10 @@ static int iommu_map_page(struct protection_domain *dom,
 static void iommu_unmap_page(struct protection_domain *dom,
 			     unsigned long bus_addr)
 {
-	u64 *pte;
-
-	pte = &dom->pt_root[IOMMU_PTE_L2_INDEX(bus_addr)];
-
-	if (!IOMMU_PTE_PRESENT(*pte))
-		return;
-
-	pte = IOMMU_PTE_PAGE(*pte);
-	pte = &pte[IOMMU_PTE_L1_INDEX(bus_addr)];
-
-	if (!IOMMU_PTE_PRESENT(*pte))
-		return;
-
-	pte = IOMMU_PTE_PAGE(*pte);
-	pte = &pte[IOMMU_PTE_L1_INDEX(bus_addr)];
+	u64 *pte = fetch_pte(dom, bus_addr);
 
-	*pte = 0;
+	if (pte)
+		*pte = 0;
 }
 
 /*
-- 
cgit v1.2.3


From a6d41a4027b758a9473197a78fab45afb31003aa Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 2 Sep 2009 17:08:55 +0200
Subject: x86/amd-iommu: Use fetch_pte in amd_iommu_iova_to_phys

Don't reimplement the page table walker in this function.
Use the generic one.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu.c | 16 ++--------------
 1 file changed, 2 insertions(+), 14 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 5e5279899994..4a54366d422d 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -2140,21 +2140,9 @@ static phys_addr_t amd_iommu_iova_to_phys(struct iommu_domain *dom,
 	phys_addr_t paddr;
 	u64 *pte;
 
-	pte = &domain->pt_root[IOMMU_PTE_L2_INDEX(iova)];
+	pte = fetch_pte(domain, iova);
 
-	if (!IOMMU_PTE_PRESENT(*pte))
-		return 0;
-
-	pte = IOMMU_PTE_PAGE(*pte);
-	pte = &pte[IOMMU_PTE_L1_INDEX(iova)];
-
-	if (!IOMMU_PTE_PRESENT(*pte))
-		return 0;
-
-	pte = IOMMU_PTE_PAGE(*pte);
-	pte = &pte[IOMMU_PTE_L0_INDEX(iova)];
-
-	if (!IOMMU_PTE_PRESENT(*pte))
+	if (!pte || !IOMMU_PTE_PRESENT(*pte))
 		return 0;
 
 	paddr  = *pte & IOMMU_PAGE_MASK;
-- 
cgit v1.2.3


From 6a0dbcbe4e612fbc9d73cd4dde8ebef19295058a Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 2 Sep 2009 15:41:59 +0200
Subject: x86/amd-iommu: Add a gneric version of amd_iommu_flush_all_devices

This patch adds a generic variant of
amd_iommu_flush_all_devices function which flushes only the
DTEs for a given protection domain.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 4a54366d422d..5265dd17eb5c 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -481,13 +481,14 @@ void amd_iommu_flush_all_domains(void)
 	}
 }
 
-void amd_iommu_flush_all_devices(void)
+static void flush_devices_by_domain(struct protection_domain *domain)
 {
 	struct amd_iommu *iommu;
 	int i;
 
 	for (i = 0; i <= amd_iommu_last_bdf; ++i) {
-		if (amd_iommu_pd_table[i] == NULL)
+		if ((domain == NULL && amd_iommu_pd_table[i] == NULL) ||
+		    (amd_iommu_pd_table[i] != domain))
 			continue;
 
 		iommu = amd_iommu_rlookup_table[i];
@@ -499,6 +500,11 @@ void amd_iommu_flush_all_devices(void)
 	}
 }
 
+void amd_iommu_flush_all_devices(void)
+{
+	flush_devices_by_domain(NULL);
+}
+
 /****************************************************************************
  *
  * The functions below are used the create the page table mappings for
-- 
cgit v1.2.3


From 407d733e30a97daf5ea6f9eb5f9ebbd42a0a9ef2 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 2 Sep 2009 16:07:00 +0200
Subject: x86/amd-iommu: Introduce set_dte_entry function

This function factors out some logic of attach_device to a
seperate function. This new function will be used to update
device table entries when necessary.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu.c | 27 +++++++++++++++++----------
 1 file changed, 17 insertions(+), 10 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 5265dd17eb5c..0fab1f1d135e 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -1058,18 +1058,10 @@ static struct protection_domain *domain_for_device(u16 devid)
 	return dom;
 }
 
-/*
- * If a device is not yet associated with a domain, this function does
- * assigns it visible for the hardware
- */
-static void attach_device(struct amd_iommu *iommu,
-			  struct protection_domain *domain,
-			  u16 devid)
+static void set_dte_entry(u16 devid, struct protection_domain *domain)
 {
-	unsigned long flags;
 	u64 pte_root = virt_to_phys(domain->pt_root);
-
-	domain->dev_cnt += 1;
+	unsigned long flags;
 
 	pte_root |= (domain->mode & DEV_ENTRY_MODE_MASK)
 		    << DEV_ENTRY_MODE_SHIFT;
@@ -1082,6 +1074,21 @@ static void attach_device(struct amd_iommu *iommu,
 
 	amd_iommu_pd_table[devid] = domain;
 	write_unlock_irqrestore(&amd_iommu_devtable_lock, flags);
+}
+
+/*
+ * If a device is not yet associated with a domain, this function does
+ * assigns it visible for the hardware
+ */
+static void attach_device(struct amd_iommu *iommu,
+			  struct protection_domain *domain,
+			  u16 devid)
+{
+	/* set the DTE entry */
+	set_dte_entry(devid, domain);
+
+	/* increase reference counter */
+	domain->dev_cnt += 1;
 
        /*
         * We might boot into a crash-kernel here. The crashed kernel
-- 
cgit v1.2.3


From 04bfdd8406099fca2e6b8844748c4d6c5eba8c8d Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 2 Sep 2009 16:00:23 +0200
Subject: x86/amd-iommu: Flush domains if address space size was increased

Thist patch introduces the update_domain function which
propagates the larger address space of a protection domain
to the device table and flushes all relevant DTEs and the
domain TLB.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/include/asm/amd_iommu_types.h |  1 +
 arch/x86/kernel/amd_iommu.c            | 32 ++++++++++++++++++++++++++++++++
 2 files changed, 33 insertions(+)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/amd_iommu_types.h b/arch/x86/include/asm/amd_iommu_types.h
index 7fce4ef77bdd..97f3d09d3be5 100644
--- a/arch/x86/include/asm/amd_iommu_types.h
+++ b/arch/x86/include/asm/amd_iommu_types.h
@@ -235,6 +235,7 @@ struct protection_domain {
 	int mode;		/* paging mode (0-6 levels) */
 	u64 *pt_root;		/* page table root pointer */
 	unsigned long flags;	/* flags to find out type of domain */
+	bool updated;		/* complete domain flush required */
 	unsigned dev_cnt;	/* devices assigned to this domain */
 	void *priv;		/* private data */
 };
diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 0fab1f1d135e..5eab6a84b9cc 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -63,6 +63,7 @@ static void dma_ops_reserve_addresses(struct dma_ops_domain *dom,
 				      unsigned int pages);
 static u64 *fetch_pte(struct protection_domain *domain,
 		      unsigned long address);
+static void update_domain(struct protection_domain *domain);
 
 #ifndef BUS_NOTIFY_UNBOUND_DRIVER
 #define BUS_NOTIFY_UNBOUND_DRIVER 0x0005
@@ -546,6 +547,8 @@ static int iommu_map_page(struct protection_domain *dom,
 
 	*pte = __pte;
 
+	update_domain(dom);
+
 	return 0;
 }
 
@@ -762,9 +765,13 @@ static int alloc_new_range(struct amd_iommu *iommu,
 		dma_ops_reserve_addresses(dma_dom, i << PAGE_SHIFT, 1);
 	}
 
+	update_domain(&dma_dom->domain);
+
 	return 0;
 
 out_free:
+	update_domain(&dma_dom->domain);
+
 	free_page((unsigned long)dma_dom->aperture[index]->bitmap);
 
 	kfree(dma_dom->aperture[index]);
@@ -1294,6 +1301,29 @@ static int get_device_resources(struct device *dev,
 	return 1;
 }
 
+static void update_device_table(struct protection_domain *domain)
+{
+	int i;
+
+	for (i = 0; i <= amd_iommu_last_bdf; ++i) {
+		if (amd_iommu_pd_table[i] != domain)
+			continue;
+		set_dte_entry(i, domain);
+	}
+}
+
+static void update_domain(struct protection_domain *domain)
+{
+	if (!domain->updated)
+		return;
+
+	update_device_table(domain);
+	flush_devices_by_domain(domain);
+	iommu_flush_domain(domain->id);
+
+	domain->updated = false;
+}
+
 /*
  * If the pte_page is not yet allocated this function is called
  */
@@ -1351,6 +1381,8 @@ static u64* dma_ops_get_pte(struct dma_ops_domain *dom,
 	} else
 		pte += IOMMU_PTE_L0_INDEX(address);
 
+	update_domain(&dom->domain);
+
 	return pte;
 }
 
-- 
cgit v1.2.3


From 50020fb6324465e478d6c8cdbf3c695f0a60358d Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 2 Sep 2009 15:38:40 +0200
Subject: x86/amd-iommu: Introduce increase_address_space function

This function will be used to increase the address space
size of a protection domain.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/include/asm/amd_iommu_types.h |  4 ++++
 arch/x86/kernel/amd_iommu.c            | 27 +++++++++++++++++++++++++++
 2 files changed, 31 insertions(+)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/amd_iommu_types.h b/arch/x86/include/asm/amd_iommu_types.h
index 97f3d09d3be5..1b4b3d6c9f04 100644
--- a/arch/x86/include/asm/amd_iommu_types.h
+++ b/arch/x86/include/asm/amd_iommu_types.h
@@ -160,6 +160,10 @@
 				  ((1ULL << PM_LEVEL_SHIFT((x))) - 1): \
 				   (0xffffffffffffffffULL))
 #define PM_LEVEL_INDEX(x, a)	(((a) >> PM_LEVEL_SHIFT((x))) & 0x1ffULL)
+#define PM_LEVEL_ENC(x)		(((x) << 9) & 0xe00ULL)
+#define PM_LEVEL_PDE(x, a)	((a) | PM_LEVEL_ENC((x)) | \
+				 IOMMU_PTE_P | IOMMU_PTE_IR | IOMMU_PTE_IW)
+
 
 #define IOMMU_PTE_L2_INDEX(address) (((address) >> 30) & 0x1ffULL)
 #define IOMMU_PTE_L1_INDEX(address) (((address) >> 21) & 0x1ffULL)
diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 5eab6a84b9cc..fc97b51f0287 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -1324,6 +1324,33 @@ static void update_domain(struct protection_domain *domain)
 	domain->updated = false;
 }
 
+/*
+ * This function is used to add another level to an IO page table. Adding
+ * another level increases the size of the address space by 9 bits to a size up
+ * to 64 bits.
+ */
+static bool increase_address_space(struct protection_domain *domain,
+				   gfp_t gfp)
+{
+	u64 *pte;
+
+	if (domain->mode == PAGE_MODE_6_LEVEL)
+		/* address space already 64 bit large */
+		return false;
+
+	pte = (void *)get_zeroed_page(gfp);
+	if (!pte)
+		return false;
+
+	*pte             = PM_LEVEL_PDE(domain->mode,
+					virt_to_phys(domain->pt_root));
+	domain->pt_root  = pte;
+	domain->mode    += 1;
+	domain->updated  = true;
+
+	return true;
+}
+
 /*
  * If the pte_page is not yet allocated this function is called
  */
-- 
cgit v1.2.3


From 8bc3e127421bf3b735edbde05135892c12c5f615 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 2 Sep 2009 16:48:40 +0200
Subject: x86/amd-iommu: Change alloc_pte to support 64 bit address space

This patch changes the alloc_pte function to be able to map
pages into the whole 64 bit address space supported by AMD
IOMMU hardware from the old limit of 2**39 bytes.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu.c | 44 ++++++++++++++++++++------------------------
 1 file changed, 20 insertions(+), 24 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index fc97b51f0287..3be2b61fc31f 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -55,7 +55,7 @@ struct iommu_cmd {
 static int dma_ops_unity_map(struct dma_ops_domain *dma_dom,
 			     struct unity_map_entry *e);
 static struct dma_ops_domain *find_protection_domain(u16 devid);
-static u64* alloc_pte(struct protection_domain *dom,
+static u64 *alloc_pte(struct protection_domain *domain,
 		      unsigned long address, u64
 		      **pte_page, gfp_t gfp);
 static void dma_ops_reserve_addresses(struct dma_ops_domain *dom,
@@ -1351,39 +1351,35 @@ static bool increase_address_space(struct protection_domain *domain,
 	return true;
 }
 
-/*
- * If the pte_page is not yet allocated this function is called
- */
-static u64* alloc_pte(struct protection_domain *dom,
+static u64 *alloc_pte(struct protection_domain *domain,
 		      unsigned long address, u64 **pte_page, gfp_t gfp)
 {
 	u64 *pte, *page;
+	int level;
 
-	pte = &dom->pt_root[IOMMU_PTE_L2_INDEX(address)];
+	while (address > PM_LEVEL_SIZE(domain->mode))
+		increase_address_space(domain, gfp);
 
-	if (!IOMMU_PTE_PRESENT(*pte)) {
-		page = (u64 *)get_zeroed_page(gfp);
-		if (!page)
-			return NULL;
-		*pte = IOMMU_L2_PDE(virt_to_phys(page));
-	}
+	level =  domain->mode - 1;
+	pte   = &domain->pt_root[PM_LEVEL_INDEX(level, address)];
 
-	pte = IOMMU_PTE_PAGE(*pte);
-	pte = &pte[IOMMU_PTE_L1_INDEX(address)];
+	while (level > 0) {
+		if (!IOMMU_PTE_PRESENT(*pte)) {
+			page = (u64 *)get_zeroed_page(gfp);
+			if (!page)
+				return NULL;
+			*pte = PM_LEVEL_PDE(level, virt_to_phys(page));
+		}
 
-	if (!IOMMU_PTE_PRESENT(*pte)) {
-		page = (u64 *)get_zeroed_page(gfp);
-		if (!page)
-			return NULL;
-		*pte = IOMMU_L1_PDE(virt_to_phys(page));
-	}
+		level -= 1;
 
-	pte = IOMMU_PTE_PAGE(*pte);
+		pte = IOMMU_PTE_PAGE(*pte);
 
-	if (pte_page)
-		*pte_page = pte;
+		if (pte_page && level == 0)
+			*pte_page = pte;
 
-	pte = &pte[IOMMU_PTE_L0_INDEX(address)];
+		pte = &pte[PM_LEVEL_INDEX(level, address)];
+	}
 
 	return pte;
 }
-- 
cgit v1.2.3


From 8c8c143cdc95ebe50fd962917556e25e8912997b Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 2 Sep 2009 17:30:00 +0200
Subject: x86/amd-iommu: Remove last usages of IOMMU_PTE_L0_INDEX

This change allows to remove these old macros later.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 3be2b61fc31f..ebc1c8443926 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -1402,7 +1402,7 @@ static u64* dma_ops_get_pte(struct dma_ops_domain *dom,
 		pte = alloc_pte(&dom->domain, address, &pte_page, GFP_ATOMIC);
 		aperture->pte_pages[APERTURE_PAGE_INDEX(address)] = pte_page;
 	} else
-		pte += IOMMU_PTE_L0_INDEX(address);
+		pte += PM_LEVEL_INDEX(0, address);
 
 	update_domain(&dom->domain);
 
@@ -1466,7 +1466,7 @@ static void dma_ops_domain_unmap(struct amd_iommu *iommu,
 	if (!pte)
 		return;
 
-	pte += IOMMU_PTE_L0_INDEX(address);
+	pte += PM_LEVEL_INDEX(0, address);
 
 	WARN_ON(!*pte);
 
-- 
cgit v1.2.3


From bad1cac28a707c69301a4d0612da9ccbecd51953 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 2 Sep 2009 16:52:23 +0200
Subject: x86/amd-iommu: Remove bus_addr check in iommu_map_page

The driver now supports full 64 bit device address spaces.
So this check is not longer required.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index ebc1c8443926..6ffb3e637657 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -530,8 +530,7 @@ static int iommu_map_page(struct protection_domain *dom,
 	bus_addr  = PAGE_ALIGN(bus_addr);
 	phys_addr = PAGE_ALIGN(phys_addr);
 
-	/* only support 512GB address spaces for now */
-	if (bus_addr > IOMMU_MAP_SIZE_L3 || !(prot & IOMMU_PROT_MASK))
+	if (!(prot & IOMMU_PROT_MASK))
 		return -EINVAL;
 
 	pte = alloc_pte(dom, bus_addr, NULL, GFP_KERNEL);
-- 
cgit v1.2.3


From 8f7a017ce05ed4522809448e169daa44fe6edeb1 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 2 Sep 2009 16:55:24 +0200
Subject: x86/amd-iommu: Use 2-level page tables for dma_ops domains

The driver now supports a dynamic number of levels for IO
page tables. This allows to reduce the number of levels for
dma_ops domains by one because a dma_ops domain has usually
an address space size between 128MB and 4G.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 6ffb3e637657..addf6588c366 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -1010,7 +1010,7 @@ static struct dma_ops_domain *dma_ops_domain_alloc(struct amd_iommu *iommu)
 	dma_dom->domain.id = domain_id_alloc();
 	if (dma_dom->domain.id == 0)
 		goto free_dma_dom;
-	dma_dom->domain.mode = PAGE_MODE_3_LEVEL;
+	dma_dom->domain.mode = PAGE_MODE_2_LEVEL;
 	dma_dom->domain.pt_root = (void *)get_zeroed_page(GFP_KERNEL);
 	dma_dom->domain.flags = PD_DMA_OPS_MASK;
 	dma_dom->domain.priv = dma_dom;
-- 
cgit v1.2.3


From 674d798a80cb6ea1defa01899099f40d9124423c Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 2 Sep 2009 17:26:09 +0200
Subject: x86/amd-iommu: Remove old page table handling macros

These macros are not longer required. So remove them.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/include/asm/amd_iommu_types.h | 19 -------------------
 1 file changed, 19 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/amd_iommu_types.h b/arch/x86/include/asm/amd_iommu_types.h
index 1b4b3d6c9f04..d66430de5f7c 100644
--- a/arch/x86/include/asm/amd_iommu_types.h
+++ b/arch/x86/include/asm/amd_iommu_types.h
@@ -150,11 +150,6 @@
 #define PAGE_MODE_5_LEVEL 0x05
 #define PAGE_MODE_6_LEVEL 0x06
 
-#define IOMMU_PDE_NL_0   0x000ULL
-#define IOMMU_PDE_NL_1   0x200ULL
-#define IOMMU_PDE_NL_2   0x400ULL
-#define IOMMU_PDE_NL_3   0x600ULL
-
 #define PM_LEVEL_SHIFT(x)	(12 + ((x) * 9))
 #define PM_LEVEL_SIZE(x)	(((x) < 6) ? \
 				  ((1ULL << PM_LEVEL_SHIFT((x))) - 1): \
@@ -164,15 +159,6 @@
 #define PM_LEVEL_PDE(x, a)	((a) | PM_LEVEL_ENC((x)) | \
 				 IOMMU_PTE_P | IOMMU_PTE_IR | IOMMU_PTE_IW)
 
-
-#define IOMMU_PTE_L2_INDEX(address) (((address) >> 30) & 0x1ffULL)
-#define IOMMU_PTE_L1_INDEX(address) (((address) >> 21) & 0x1ffULL)
-#define IOMMU_PTE_L0_INDEX(address) (((address) >> 12) & 0x1ffULL)
-
-#define IOMMU_MAP_SIZE_L1 (1ULL << 21)
-#define IOMMU_MAP_SIZE_L2 (1ULL << 30)
-#define IOMMU_MAP_SIZE_L3 (1ULL << 39)
-
 #define IOMMU_PTE_P  (1ULL << 0)
 #define IOMMU_PTE_TV (1ULL << 1)
 #define IOMMU_PTE_U  (1ULL << 59)
@@ -180,11 +166,6 @@
 #define IOMMU_PTE_IR (1ULL << 61)
 #define IOMMU_PTE_IW (1ULL << 62)
 
-#define IOMMU_L1_PDE(address) \
-	((address) | IOMMU_PDE_NL_1 | IOMMU_PTE_P | IOMMU_PTE_IR | IOMMU_PTE_IW)
-#define IOMMU_L2_PDE(address) \
-	((address) | IOMMU_PDE_NL_2 | IOMMU_PTE_P | IOMMU_PTE_IR | IOMMU_PTE_IW)
-
 #define IOMMU_PAGE_MASK (((1ULL << 52) - 1) & ~0xfffULL)
 #define IOMMU_PTE_PRESENT(pte) ((pte) & IOMMU_PTE_P)
 #define IOMMU_PTE_PAGE(pte) (phys_to_virt((pte) & IOMMU_PAGE_MASK))
-- 
cgit v1.2.3


From a6b256b41357c33ccb2d105a4457e22bdc83e021 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Thu, 3 Sep 2009 12:21:31 +0200
Subject: x86/amd-iommu: Support higher level PTEs in iommu_page_unmap

This patch changes fetch_pte and iommu_page_unmap to support
different page sizes too.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/include/asm/amd_iommu_types.h |  1 +
 arch/x86/kernel/amd_iommu.c            | 21 +++++++++++++--------
 2 files changed, 14 insertions(+), 8 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/amd_iommu_types.h b/arch/x86/include/asm/amd_iommu_types.h
index d66430de5f7c..351ca39ece05 100644
--- a/arch/x86/include/asm/amd_iommu_types.h
+++ b/arch/x86/include/asm/amd_iommu_types.h
@@ -158,6 +158,7 @@
 #define PM_LEVEL_ENC(x)		(((x) << 9) & 0xe00ULL)
 #define PM_LEVEL_PDE(x, a)	((a) | PM_LEVEL_ENC((x)) | \
 				 IOMMU_PTE_P | IOMMU_PTE_IR | IOMMU_PTE_IW)
+#define PM_PTE_LEVEL(pte)	(((pte) >> 9) & 0x7ULL)
 
 #define IOMMU_PTE_P  (1ULL << 0)
 #define IOMMU_PTE_TV (1ULL << 1)
diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index addf6588c366..002cf9cab9e9 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -62,7 +62,7 @@ static void dma_ops_reserve_addresses(struct dma_ops_domain *dom,
 				      unsigned long start_page,
 				      unsigned int pages);
 static u64 *fetch_pte(struct protection_domain *domain,
-		      unsigned long address);
+		      unsigned long address, int map_size);
 static void update_domain(struct protection_domain *domain);
 
 #ifndef BUS_NOTIFY_UNBOUND_DRIVER
@@ -552,9 +552,9 @@ static int iommu_map_page(struct protection_domain *dom,
 }
 
 static void iommu_unmap_page(struct protection_domain *dom,
-			     unsigned long bus_addr)
+			     unsigned long bus_addr, int map_size)
 {
-	u64 *pte = fetch_pte(dom, bus_addr);
+	u64 *pte = fetch_pte(dom, bus_addr, map_size);
 
 	if (pte)
 		*pte = 0;
@@ -668,7 +668,7 @@ static int init_unity_mappings_for_device(struct dma_ops_domain *dma_dom,
  * there is one, it returns the pointer to it.
  */
 static u64 *fetch_pte(struct protection_domain *domain,
-		      unsigned long address)
+		      unsigned long address, int map_size)
 {
 	int level;
 	u64 *pte;
@@ -676,7 +676,7 @@ static u64 *fetch_pte(struct protection_domain *domain,
 	level =  domain->mode - 1;
 	pte   = &domain->pt_root[PM_LEVEL_INDEX(level, address)];
 
-	while (level > 0) {
+	while (level > map_size) {
 		if (!IOMMU_PTE_PRESENT(*pte))
 			return NULL;
 
@@ -684,6 +684,11 @@ static u64 *fetch_pte(struct protection_domain *domain,
 
 		pte = IOMMU_PTE_PAGE(*pte);
 		pte = &pte[PM_LEVEL_INDEX(level, address)];
+
+		if ((PM_PTE_LEVEL(*pte) == 0) && level != map_size) {
+			pte = NULL;
+			break;
+		}
 	}
 
 	return pte;
@@ -757,7 +762,7 @@ static int alloc_new_range(struct amd_iommu *iommu,
 	for (i = dma_dom->aperture[index]->offset;
 	     i < dma_dom->aperture_size;
 	     i += PAGE_SIZE) {
-		u64 *pte = fetch_pte(&dma_dom->domain, i);
+		u64 *pte = fetch_pte(&dma_dom->domain, i, PM_MAP_4k);
 		if (!pte || !IOMMU_PTE_PRESENT(*pte))
 			continue;
 
@@ -2192,7 +2197,7 @@ static void amd_iommu_unmap_range(struct iommu_domain *dom,
 	iova  &= PAGE_MASK;
 
 	for (i = 0; i < npages; ++i) {
-		iommu_unmap_page(domain, iova);
+		iommu_unmap_page(domain, iova, PM_MAP_4k);
 		iova  += PAGE_SIZE;
 	}
 
@@ -2207,7 +2212,7 @@ static phys_addr_t amd_iommu_iova_to_phys(struct iommu_domain *dom,
 	phys_addr_t paddr;
 	u64 *pte;
 
-	pte = fetch_pte(domain, iova);
+	pte = fetch_pte(domain, iova, PM_MAP_4k);
 
 	if (!pte || !IOMMU_PTE_PRESENT(*pte))
 		return 0;
-- 
cgit v1.2.3


From abdc5eb3d69279039ba6cb89719913d08013ab14 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Thu, 3 Sep 2009 11:33:51 +0200
Subject: x86/amd-iommu: Change iommu_map_page to support multiple page sizes

This patch adds a map_size parameter to the iommu_map_page
function which makes it generic enough to handle multiple
page sizes. This also requires a change to alloc_pte which
is also done in this patch.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/include/asm/amd_iommu_types.h |  6 ++++++
 arch/x86/kernel/amd_iommu.c            | 31 ++++++++++++++++++++-----------
 2 files changed, 26 insertions(+), 11 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/amd_iommu_types.h b/arch/x86/include/asm/amd_iommu_types.h
index 351ca39ece05..9e8bb9746dcf 100644
--- a/arch/x86/include/asm/amd_iommu_types.h
+++ b/arch/x86/include/asm/amd_iommu_types.h
@@ -160,6 +160,12 @@
 				 IOMMU_PTE_P | IOMMU_PTE_IR | IOMMU_PTE_IW)
 #define PM_PTE_LEVEL(pte)	(((pte) >> 9) & 0x7ULL)
 
+#define PM_MAP_4k		0
+#define PM_ADDR_MASK		0x000ffffffffff000ULL
+#define PM_MAP_MASK(lvl)	(PM_ADDR_MASK & \
+				(~((1ULL << (12 + ((lvl) * 9))) - 1)))
+#define PM_ALIGNED(lvl, addr)	((PM_MAP_MASK(lvl) & (addr)) == (addr))
+
 #define IOMMU_PTE_P  (1ULL << 0)
 #define IOMMU_PTE_TV (1ULL << 1)
 #define IOMMU_PTE_U  (1ULL << 59)
diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 002cf9cab9e9..45be9499c973 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -56,8 +56,8 @@ static int dma_ops_unity_map(struct dma_ops_domain *dma_dom,
 			     struct unity_map_entry *e);
 static struct dma_ops_domain *find_protection_domain(u16 devid);
 static u64 *alloc_pte(struct protection_domain *domain,
-		      unsigned long address, u64
-		      **pte_page, gfp_t gfp);
+		      unsigned long address, int end_lvl,
+		      u64 **pte_page, gfp_t gfp);
 static void dma_ops_reserve_addresses(struct dma_ops_domain *dom,
 				      unsigned long start_page,
 				      unsigned int pages);
@@ -523,17 +523,21 @@ void amd_iommu_flush_all_devices(void)
 static int iommu_map_page(struct protection_domain *dom,
 			  unsigned long bus_addr,
 			  unsigned long phys_addr,
-			  int prot)
+			  int prot,
+			  int map_size)
 {
 	u64 __pte, *pte;
 
 	bus_addr  = PAGE_ALIGN(bus_addr);
 	phys_addr = PAGE_ALIGN(phys_addr);
 
+	BUG_ON(!PM_ALIGNED(map_size, bus_addr));
+	BUG_ON(!PM_ALIGNED(map_size, phys_addr));
+
 	if (!(prot & IOMMU_PROT_MASK))
 		return -EINVAL;
 
-	pte = alloc_pte(dom, bus_addr, NULL, GFP_KERNEL);
+	pte = alloc_pte(dom, bus_addr, map_size, NULL, GFP_KERNEL);
 
 	if (IOMMU_PTE_PRESENT(*pte))
 		return -EBUSY;
@@ -612,7 +616,8 @@ static int dma_ops_unity_map(struct dma_ops_domain *dma_dom,
 
 	for (addr = e->address_start; addr < e->address_end;
 	     addr += PAGE_SIZE) {
-		ret = iommu_map_page(&dma_dom->domain, addr, addr, e->prot);
+		ret = iommu_map_page(&dma_dom->domain, addr, addr, e->prot,
+				     PM_MAP_4k);
 		if (ret)
 			return ret;
 		/*
@@ -729,7 +734,7 @@ static int alloc_new_range(struct amd_iommu *iommu,
 		u64 *pte, *pte_page;
 
 		for (i = 0; i < num_ptes; ++i) {
-			pte = alloc_pte(&dma_dom->domain, address,
+			pte = alloc_pte(&dma_dom->domain, address, PM_MAP_4k,
 					&pte_page, gfp);
 			if (!pte)
 				goto out_free;
@@ -1356,7 +1361,10 @@ static bool increase_address_space(struct protection_domain *domain,
 }
 
 static u64 *alloc_pte(struct protection_domain *domain,
-		      unsigned long address, u64 **pte_page, gfp_t gfp)
+		      unsigned long address,
+		      int end_lvl,
+		      u64 **pte_page,
+		      gfp_t gfp)
 {
 	u64 *pte, *page;
 	int level;
@@ -1367,7 +1375,7 @@ static u64 *alloc_pte(struct protection_domain *domain,
 	level =  domain->mode - 1;
 	pte   = &domain->pt_root[PM_LEVEL_INDEX(level, address)];
 
-	while (level > 0) {
+	while (level > end_lvl) {
 		if (!IOMMU_PTE_PRESENT(*pte)) {
 			page = (u64 *)get_zeroed_page(gfp);
 			if (!page)
@@ -1379,7 +1387,7 @@ static u64 *alloc_pte(struct protection_domain *domain,
 
 		pte = IOMMU_PTE_PAGE(*pte);
 
-		if (pte_page && level == 0)
+		if (pte_page && level == end_lvl)
 			*pte_page = pte;
 
 		pte = &pte[PM_LEVEL_INDEX(level, address)];
@@ -1403,7 +1411,8 @@ static u64* dma_ops_get_pte(struct dma_ops_domain *dom,
 
 	pte = aperture->pte_pages[APERTURE_PAGE_INDEX(address)];
 	if (!pte) {
-		pte = alloc_pte(&dom->domain, address, &pte_page, GFP_ATOMIC);
+		pte = alloc_pte(&dom->domain, address, PM_MAP_4k, &pte_page,
+				GFP_ATOMIC);
 		aperture->pte_pages[APERTURE_PAGE_INDEX(address)] = pte_page;
 	} else
 		pte += PM_LEVEL_INDEX(0, address);
@@ -2176,7 +2185,7 @@ static int amd_iommu_map_range(struct iommu_domain *dom,
 	paddr &= PAGE_MASK;
 
 	for (i = 0; i < npages; ++i) {
-		ret = iommu_map_page(domain, iova, paddr, prot);
+		ret = iommu_map_page(domain, iova, paddr, prot, PM_MAP_4k);
 		if (ret)
 			return ret;
 
-- 
cgit v1.2.3


From ac0101d396fee24994632f71b55b9f7f9ee16eff Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Tue, 1 Sep 2009 16:00:35 +0200
Subject: x86/dma: Mark iommu_pass_through as __read_mostly

This variable is read most of the time. This patch marks it
as such. It also documents the meaning the this variable
while at it.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/pci-dma.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c
index 1a041bcf506b..873aa079d166 100644
--- a/arch/x86/kernel/pci-dma.c
+++ b/arch/x86/kernel/pci-dma.c
@@ -32,7 +32,14 @@ int no_iommu __read_mostly;
 /* Set this to 1 if there is a HW IOMMU in the system */
 int iommu_detected __read_mostly = 0;
 
-int iommu_pass_through;
+/*
+ * This variable becomes 1 if iommu=pt is passed on the kernel command line.
+ * If this variable is 1, IOMMU implementations do no DMA ranslation for
+ * devices and allow every device to access to whole physical memory. This is
+ * useful if a user want to use an IOMMU only for KVM device assignment to
+ * guests and not for driver dma translation.
+ */
+int iommu_pass_through __read_mostly;
 
 dma_addr_t bad_dma_address __read_mostly = 0;
 EXPORT_SYMBOL(bad_dma_address);
-- 
cgit v1.2.3


From 2650815fb03fe2bf1e6701584087ba669dcf92cd Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 26 Aug 2009 16:52:40 +0200
Subject: x86/amd-iommu: Add core functions for pd allocation/freeing

This patch factors some code of protection domain allocation
into seperate functions. This way the logic can be used to
allocate the passthrough domain later. As a side effect this
patch fixes an unlikely domain id leakage bug.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu.c | 36 ++++++++++++++++++++++++++++++++----
 1 file changed, 32 insertions(+), 4 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 6c99f5037801..0934348abfad 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -1988,19 +1988,47 @@ static void cleanup_domain(struct protection_domain *domain)
 	write_unlock_irqrestore(&amd_iommu_devtable_lock, flags);
 }
 
-static int amd_iommu_domain_init(struct iommu_domain *dom)
+static void protection_domain_free(struct protection_domain *domain)
+{
+	if (!domain)
+		return;
+
+	if (domain->id)
+		domain_id_free(domain->id);
+
+	kfree(domain);
+}
+
+static struct protection_domain *protection_domain_alloc(void)
 {
 	struct protection_domain *domain;
 
 	domain = kzalloc(sizeof(*domain), GFP_KERNEL);
 	if (!domain)
-		return -ENOMEM;
+		return NULL;
 
 	spin_lock_init(&domain->lock);
-	domain->mode = PAGE_MODE_3_LEVEL;
 	domain->id = domain_id_alloc();
 	if (!domain->id)
+		goto out_err;
+
+	return domain;
+
+out_err:
+	kfree(domain);
+
+	return NULL;
+}
+
+static int amd_iommu_domain_init(struct iommu_domain *dom)
+{
+	struct protection_domain *domain;
+
+	domain = protection_domain_alloc();
+	if (!domain)
 		goto out_free;
+
+	domain->mode    = PAGE_MODE_3_LEVEL;
 	domain->pt_root = (void *)get_zeroed_page(GFP_KERNEL);
 	if (!domain->pt_root)
 		goto out_free;
@@ -2010,7 +2038,7 @@ static int amd_iommu_domain_init(struct iommu_domain *dom)
 	return 0;
 
 out_free:
-	kfree(domain);
+	protection_domain_free(domain);
 
 	return -ENOMEM;
 }
-- 
cgit v1.2.3


From 0feae533ddebe02cda6ccce5cac7349b446776a8 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 26 Aug 2009 15:26:30 +0200
Subject: x86/amd-iommu: Add passthrough mode initialization functions

When iommu=pt is passed on kernel command line the devices
should run untranslated. This requires the allocation of a
special domain for that purpose. This patch implements the
allocation and initialization path for iommu=pt.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/include/asm/amd_iommu.h       |  1 +
 arch/x86/include/asm/amd_iommu_types.h |  4 ++
 arch/x86/kernel/amd_iommu.c            | 72 ++++++++++++++++++++++++++++++----
 3 files changed, 69 insertions(+), 8 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/amd_iommu.h b/arch/x86/include/asm/amd_iommu.h
index bdf96f119f06..ac95995b7bad 100644
--- a/arch/x86/include/asm/amd_iommu.h
+++ b/arch/x86/include/asm/amd_iommu.h
@@ -25,6 +25,7 @@
 #ifdef CONFIG_AMD_IOMMU
 extern int amd_iommu_init(void);
 extern int amd_iommu_init_dma_ops(void);
+extern int amd_iommu_init_passthrough(void);
 extern void amd_iommu_detect(void);
 extern irqreturn_t amd_iommu_int_handler(int irq, void *data);
 extern void amd_iommu_flush_all_domains(void);
diff --git a/arch/x86/include/asm/amd_iommu_types.h b/arch/x86/include/asm/amd_iommu_types.h
index 0c878caaa0a2..49f7453bff76 100644
--- a/arch/x86/include/asm/amd_iommu_types.h
+++ b/arch/x86/include/asm/amd_iommu_types.h
@@ -143,6 +143,7 @@
 #define EVT_BUFFER_SIZE		8192 /* 512 entries */
 #define EVT_LEN_MASK		(0x9ULL << 56)
 
+#define PAGE_MODE_NONE    0x00
 #define PAGE_MODE_1_LEVEL 0x01
 #define PAGE_MODE_2_LEVEL 0x02
 #define PAGE_MODE_3_LEVEL 0x03
@@ -194,6 +195,9 @@
 #define PD_DMA_OPS_MASK		(1UL << 0) /* domain used for dma_ops */
 #define PD_DEFAULT_MASK		(1UL << 1) /* domain is a default dma_ops
 					      domain for an IOMMU */
+#define PD_PASSTHROUGH_MASK	(1UL << 2) /* domain has no page
+					      translation */
+
 extern bool amd_iommu_dump;
 #define DUMP_printk(format, arg...)					\
 	do {								\
diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 0934348abfad..7987f20499ad 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -41,6 +41,12 @@ static DEFINE_RWLOCK(amd_iommu_devtable_lock);
 static LIST_HEAD(iommu_pd_list);
 static DEFINE_SPINLOCK(iommu_pd_list_lock);
 
+/*
+ * Domain for untranslated devices - only allocated
+ * if iommu=pt passed on kernel cmd line.
+ */
+static struct protection_domain *pt_domain;
+
 #ifdef CONFIG_IOMMU_API
 static struct iommu_ops amd_iommu_ops;
 #endif
@@ -1067,9 +1073,9 @@ static struct protection_domain *domain_for_device(u16 devid)
  * If a device is not yet associated with a domain, this function does
  * assigns it visible for the hardware
  */
-static void attach_device(struct amd_iommu *iommu,
-			  struct protection_domain *domain,
-			  u16 devid)
+static void __attach_device(struct amd_iommu *iommu,
+			    struct protection_domain *domain,
+			    u16 devid)
 {
 	unsigned long flags;
 	u64 pte_root = virt_to_phys(domain->pt_root);
@@ -1087,12 +1093,19 @@ static void attach_device(struct amd_iommu *iommu,
 
 	amd_iommu_pd_table[devid] = domain;
 	write_unlock_irqrestore(&amd_iommu_devtable_lock, flags);
+}
 
-       /*
-        * We might boot into a crash-kernel here. The crashed kernel
-        * left the caches in the IOMMU dirty. So we have to flush
-        * here to evict all dirty stuff.
-        */
+static void attach_device(struct amd_iommu *iommu,
+			  struct protection_domain *domain,
+			  u16 devid)
+{
+	__attach_device(iommu, domain, devid);
+
+	/*
+	 * We might boot into a crash-kernel here. The crashed kernel
+	 * left the caches in the IOMMU dirty. So we have to flush
+	 * here to evict all dirty stuff.
+	 */
 	iommu_queue_inv_dev_entry(iommu, devid);
 	iommu_flush_tlb_pde(iommu, domain->id);
 }
@@ -2219,3 +2232,46 @@ static struct iommu_ops amd_iommu_ops = {
 	.domain_has_cap = amd_iommu_domain_has_cap,
 };
 
+/*****************************************************************************
+ *
+ * The next functions do a basic initialization of IOMMU for pass through
+ * mode
+ *
+ * In passthrough mode the IOMMU is initialized and enabled but not used for
+ * DMA-API translation.
+ *
+ *****************************************************************************/
+
+int __init amd_iommu_init_passthrough(void)
+{
+	struct pci_dev *dev = NULL;
+	u16 devid, devid2;
+
+	/* allocate passthroug domain */
+	pt_domain = protection_domain_alloc();
+	if (!pt_domain)
+		return -ENOMEM;
+
+	pt_domain->mode |= PAGE_MODE_NONE;
+
+	while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) {
+		struct amd_iommu *iommu;
+
+		devid = calc_devid(dev->bus->number, dev->devfn);
+		if (devid > amd_iommu_last_bdf)
+			continue;
+
+		devid2 = amd_iommu_alias_table[devid];
+
+		iommu = amd_iommu_rlookup_table[devid2];
+		if (!iommu)
+			continue;
+
+		__attach_device(iommu, pt_domain, devid);
+		__attach_device(iommu, pt_domain, devid2);
+	}
+
+	pr_info("AMD-Vi: Initialized for Passthrough Mode\n");
+
+	return 0;
+}
-- 
cgit v1.2.3


From aa879fff5d12318259816aa35023e941a1e4d3d9 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Mon, 31 Aug 2009 16:01:48 +0200
Subject: x86/amd-iommu: Fix device table write order

The V bit of the device table entry has to be set after the
rest of the entry is written to not confuse the hardware.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 7987f20499ad..2b1e77c714ff 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -1087,9 +1087,9 @@ static void __attach_device(struct amd_iommu *iommu,
 	pte_root |= IOMMU_PTE_IR | IOMMU_PTE_IW | IOMMU_PTE_P | IOMMU_PTE_TV;
 
 	write_lock_irqsave(&amd_iommu_devtable_lock, flags);
-	amd_iommu_dev_table[devid].data[0] = lower_32_bits(pte_root);
-	amd_iommu_dev_table[devid].data[1] = upper_32_bits(pte_root);
 	amd_iommu_dev_table[devid].data[2] = domain->id;
+	amd_iommu_dev_table[devid].data[1] = upper_32_bits(pte_root);
+	amd_iommu_dev_table[devid].data[0] = lower_32_bits(pte_root);
 
 	amd_iommu_pd_table[devid] = domain;
 	write_unlock_irqrestore(&amd_iommu_devtable_lock, flags);
-- 
cgit v1.2.3


From eba6ac60ba66c6bf6858938442204feaa67dea31 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Tue, 1 Sep 2009 12:07:08 +0200
Subject: x86/amd-iommu: Align locking between attach_device and detach_device

This patch makes the locking behavior between the functions
attach_device and __attach_device consistent with the
locking behavior between detach_device and __detach_device.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu.c | 19 ++++++++++++++-----
 1 file changed, 14 insertions(+), 5 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 2b1e77c714ff..9aa135d4453f 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -1077,29 +1077,38 @@ static void __attach_device(struct amd_iommu *iommu,
 			    struct protection_domain *domain,
 			    u16 devid)
 {
-	unsigned long flags;
-	u64 pte_root = virt_to_phys(domain->pt_root);
+	u64 pte_root;
 
-	domain->dev_cnt += 1;
+	/* lock domain */
+	spin_lock(&domain->lock);
+
+	pte_root = virt_to_phys(domain->pt_root);
 
 	pte_root |= (domain->mode & DEV_ENTRY_MODE_MASK)
 		    << DEV_ENTRY_MODE_SHIFT;
 	pte_root |= IOMMU_PTE_IR | IOMMU_PTE_IW | IOMMU_PTE_P | IOMMU_PTE_TV;
 
-	write_lock_irqsave(&amd_iommu_devtable_lock, flags);
 	amd_iommu_dev_table[devid].data[2] = domain->id;
 	amd_iommu_dev_table[devid].data[1] = upper_32_bits(pte_root);
 	amd_iommu_dev_table[devid].data[0] = lower_32_bits(pte_root);
 
 	amd_iommu_pd_table[devid] = domain;
-	write_unlock_irqrestore(&amd_iommu_devtable_lock, flags);
+
+	domain->dev_cnt += 1;
+
+	/* ready */
+	spin_unlock(&domain->lock);
 }
 
 static void attach_device(struct amd_iommu *iommu,
 			  struct protection_domain *domain,
 			  u16 devid)
 {
+	unsigned long flags;
+
+	write_lock_irqsave(&amd_iommu_devtable_lock, flags);
 	__attach_device(iommu, domain, devid);
+	write_unlock_irqrestore(&amd_iommu_devtable_lock, flags);
 
 	/*
 	 * We might boot into a crash-kernel here. The crashed kernel
-- 
cgit v1.2.3


From 21129f786f231f7a9dce5b504617b893f50a435f Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Tue, 1 Sep 2009 11:59:42 +0200
Subject: x86/amd-iommu: Make sure a device is assigned in passthrough mode

When the IOMMU driver runs in passthrough mode it has to
make sure that every device not assigned to an IOMMU-API
domain must be put into the passthrough domain instead of
keeping it unassigned.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu.c | 9 +++++++++
 1 file changed, 9 insertions(+)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index 9aa135d4453f..a8e74c34dd27 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -1141,6 +1141,15 @@ static void __detach_device(struct protection_domain *domain, u16 devid)
 
 	/* ready */
 	spin_unlock(&domain->lock);
+
+	/*
+	 * If we run in passthrough mode the device must be assigned to the
+	 * passthrough domain if it is detached from any other domain
+	 */
+	if (iommu_pass_through) {
+		struct amd_iommu *iommu = amd_iommu_rlookup_table[devid];
+		__attach_device(iommu, pt_domain, devid);
+	}
 }
 
 /*
-- 
cgit v1.2.3


From a1ca331c8aa75cd58fdf685e2e8745e1d3ec5c8f Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Tue, 1 Sep 2009 12:22:22 +0200
Subject: x86/amd-iommu: Don't detach device from pt domain on driver unbind

This patch makes sure a device is not detached from the
passthrough domain when the device driver is unloaded or
does otherwise release the device.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu.c | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index a8e74c34dd27..12a541deae5e 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -1195,6 +1195,8 @@ static int device_change_notifier(struct notifier_block *nb,
 	case BUS_NOTIFY_UNBOUND_DRIVER:
 		if (!domain)
 			goto out;
+		if (iommu_pass_through)
+			break;
 		detach_device(domain, devid);
 		break;
 	case BUS_NOTIFY_ADD_DEVICE:
-- 
cgit v1.2.3


From 4751a95134e05f1172131d2001c6991d671fa58c Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Tue, 1 Sep 2009 15:53:54 +0200
Subject: x86/amd-iommu: Initialize passthrough mode when requested

This patch enables the passthrough mode for AMD IOMMU by
running the initialization function when iommu=pt is passed
on the kernel command line.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kernel/amd_iommu_init.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c
index c1b17e97252e..f00f489ab150 100644
--- a/arch/x86/kernel/amd_iommu_init.c
+++ b/arch/x86/kernel/amd_iommu_init.c
@@ -1242,12 +1242,18 @@ int __init amd_iommu_init(void)
 	if (ret)
 		goto free;
 
-	ret = amd_iommu_init_dma_ops();
+	if (iommu_pass_through)
+		ret = amd_iommu_init_passthrough();
+	else
+		ret = amd_iommu_init_dma_ops();
 	if (ret)
 		goto free;
 
 	enable_iommus();
 
+	if (iommu_pass_through)
+		goto out;
+
 	printk(KERN_INFO "AMD IOMMU: device isolation ");
 	if (amd_iommu_isolate)
 		printk("enabled\n");
-- 
cgit v1.2.3


From 6ac162d6c01ac7626f46c68c0770556cf682ce34 Mon Sep 17 00:00:00 2001
From: Pavel Vasilyev <pavel@pavlinux.ru>
Date: Thu, 3 Sep 2009 16:20:55 +0200
Subject: x86/gart: Do not select AGP for GART_IOMMU

There is no dependency from the gart code to the agp code.
And since a lot of systems today do not have agp anymore
remove this dependency from the kernel configuration.

Signed-off-by: Pavel Vasilyev <pavel@pavlinux.ru>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/Kconfig | 1 -
 1 file changed, 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 13ffa5df37d7..1d9c18aa17eb 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -586,7 +586,6 @@ config GART_IOMMU
 	bool "GART IOMMU support" if EMBEDDED
 	default y
 	select SWIOTLB
-	select AGP
 	depends on X86_64 && PCI
 	---help---
 	  Support for full DMA access of devices with 32bit memory access only
-- 
cgit v1.2.3


From 8adf65cfae2d6c2ec5c06e4521f089c62f9eff05 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Thu, 3 Sep 2009 21:26:34 +0200
Subject: x86, msr: Fix msr-reg.S compilation with gas 2.16.1, on 32-bit too

The macro was defined in the 32-bit path as well - breaking the
build on 32-bit platforms:

  arch/x86/lib/msr-reg.S: Assembler messages:
  arch/x86/lib/msr-reg.S:53: Error: Bad macro parameter list
  arch/x86/lib/msr-reg.S:100: Error: invalid character '_' in mnemonic
  arch/x86/lib/msr-reg.S:101: Error: invalid character '_' in mnemonic

Cc: Borislav Petkov <petkovbb@googlemail.com>
Cc: H. Peter Anvin <hpa@zytor.com>
LKML-Reference: <tip-f6909f394c2d4a0a71320797df72d54c49c5927e@git.kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/lib/msr-reg.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/lib/msr-reg.S b/arch/x86/lib/msr-reg.S
index d5eaf53aa670..69fa10623f21 100644
--- a/arch/x86/lib/msr-reg.S
+++ b/arch/x86/lib/msr-reg.S
@@ -50,7 +50,7 @@ ENDPROC(native_\op\()_safe_regs)
 
 #else /* X86_32 */
 
-.macro op_safe_regs op:req
+.macro op_safe_regs op
 ENTRY(native_\op\()_safe_regs)
 	CFI_STARTPROC
 	pushl_cfi %ebx
-- 
cgit v1.2.3


From 1ea0d14e480c245683927eecc03a70faf06e80c8 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Thu, 3 Sep 2009 12:27:15 -0700
Subject: x86/i386: Make sure stack-protector segment base is cache aligned

The Intel Optimization Reference Guide says:

	In Intel Atom microarchitecture, the address generation unit
	assumes that the segment base will be 0 by default. Non-zero
	segment base will cause load and store operations to experience
	a delay.
		- If the segment base isn't aligned to a cache line
		  boundary, the max throughput of memory operations is
		  reduced to one [e]very 9 cycles.
	[...]
	Assembly/Compiler Coding Rule 15. (H impact, ML generality)
	For Intel Atom processors, use segments with base set to 0
	whenever possible; avoid non-zero segment base address that is
	not aligned to cache line boundary at all cost.

We can't avoid having a non-zero base for the stack-protector
segment, but we can make it cache-aligned.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: <stable@kernel.org>
LKML-Reference: <4AA01893.6000507@goop.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/include/asm/processor.h      | 12 +++++++++++-
 arch/x86/include/asm/stackprotector.h |  4 ++--
 arch/x86/include/asm/system.h         |  2 +-
 arch/x86/kernel/cpu/common.c          |  2 +-
 arch/x86/kernel/head_32.S             |  1 -
 5 files changed, 15 insertions(+), 6 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h
index c7768269b1cf..e597ecc8753c 100644
--- a/arch/x86/include/asm/processor.h
+++ b/arch/x86/include/asm/processor.h
@@ -403,7 +403,17 @@ extern unsigned long kernel_eflags;
 extern asmlinkage void ignore_sysret(void);
 #else	/* X86_64 */
 #ifdef CONFIG_CC_STACKPROTECTOR
-DECLARE_PER_CPU(unsigned long, stack_canary);
+/*
+ * Make sure stack canary segment base is cached-aligned:
+ *   "For Intel Atom processors, avoid non zero segment base address
+ *    that is not aligned to cache line boundary at all cost."
+ * (Optim Ref Manual Assembly/Compiler Coding Rule 15.)
+ */
+struct stack_canary {
+	char __pad[20];		/* canary at %gs:20 */
+	unsigned long canary;
+};
+DECLARE_PER_CPU(struct stack_canary, stack_canary) ____cacheline_aligned;
 #endif
 #endif	/* X86_64 */
 
diff --git a/arch/x86/include/asm/stackprotector.h b/arch/x86/include/asm/stackprotector.h
index 44efdff3975d..157517763565 100644
--- a/arch/x86/include/asm/stackprotector.h
+++ b/arch/x86/include/asm/stackprotector.h
@@ -78,14 +78,14 @@ static __always_inline void boot_init_stack_canary(void)
 #ifdef CONFIG_X86_64
 	percpu_write(irq_stack_union.stack_canary, canary);
 #else
-	percpu_write(stack_canary, canary);
+	percpu_write(stack_canary.canary, canary);
 #endif
 }
 
 static inline void setup_stack_canary_segment(int cpu)
 {
 #ifdef CONFIG_X86_32
-	unsigned long canary = (unsigned long)&per_cpu(stack_canary, cpu) - 20;
+	unsigned long canary = (unsigned long)&per_cpu(stack_canary, cpu);
 	struct desc_struct *gdt_table = get_cpu_gdt_table(cpu);
 	struct desc_struct desc;
 
diff --git a/arch/x86/include/asm/system.h b/arch/x86/include/asm/system.h
index 75c49c782e20..f08f97374892 100644
--- a/arch/x86/include/asm/system.h
+++ b/arch/x86/include/asm/system.h
@@ -31,7 +31,7 @@ void __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
 	"movl %P[task_canary](%[next]), %%ebx\n\t"			\
 	"movl %%ebx, "__percpu_arg([stack_canary])"\n\t"
 #define __switch_canary_oparam						\
-	, [stack_canary] "=m" (per_cpu_var(stack_canary))
+	, [stack_canary] "=m" (per_cpu_var(stack_canary.canary))
 #define __switch_canary_iparam						\
 	, [task_canary] "i" (offsetof(struct task_struct, stack_canary))
 #else	/* CC_STACKPROTECTOR */
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index ced07ba5e937..7d84bc4c1188 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -1043,7 +1043,7 @@ DEFINE_PER_CPU(struct orig_ist, orig_ist);
 #else	/* CONFIG_X86_64 */
 
 #ifdef CONFIG_CC_STACKPROTECTOR
-DEFINE_PER_CPU(unsigned long, stack_canary);
+DEFINE_PER_CPU(struct stack_canary, stack_canary) ____cacheline_aligned;
 #endif
 
 /* Make sure %fs and %gs are initialized properly in idle threads */
diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S
index cc827ac9e8d3..7ffec6b3b331 100644
--- a/arch/x86/kernel/head_32.S
+++ b/arch/x86/kernel/head_32.S
@@ -439,7 +439,6 @@ is386:	movl $2,%ecx		# set MP
 	jne 1f
 	movl $per_cpu__gdt_page,%eax
 	movl $per_cpu__stack_canary,%ecx
-	subl $20, %ecx
 	movw %cx, 8 * GDT_ENTRY_STACK_CANARY + 2(%eax)
 	shrl $16, %ecx
 	movb %cl, 8 * GDT_ENTRY_STACK_CANARY + 4(%eax)
-- 
cgit v1.2.3


From 4a376ec3a2599c02207cd4cbd5dbf73783548463 Mon Sep 17 00:00:00 2001
From: Andreas Herrmann <andreas.herrmann3@amd.com>
Date: Thu, 3 Sep 2009 09:40:21 +0200
Subject: x86: Fix CPU llc_shared_map information for AMD Magny-Cours

Construct entire NodeID and use it as cpu_llc_id. Thus internal node
siblings are stored in llc_shared_map.

Signed-off-by: Andreas Herrmann <andreas.herrmann3@amd.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/include/asm/cpufeature.h |  1 +
 arch/x86/kernel/cpu/amd.c         | 64 ++++++++++++++++++++++++++++++++++++++-
 2 files changed, 64 insertions(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h
index 4a28d22d4793..847fee6493a2 100644
--- a/arch/x86/include/asm/cpufeature.h
+++ b/arch/x86/include/asm/cpufeature.h
@@ -95,6 +95,7 @@
 #define X86_FEATURE_NONSTOP_TSC	(3*32+24) /* TSC does not stop in C states */
 #define X86_FEATURE_CLFLUSH_MONITOR (3*32+25) /* "" clflush reqd with monitor */
 #define X86_FEATURE_EXTD_APICID	(3*32+26) /* has extended APICID (8 bits) */
+#define X86_FEATURE_AMD_DCM     (3*32+27) /* multi-node processor */
 
 /* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */
 #define X86_FEATURE_XMM3	(4*32+ 0) /* "pni" SSE-3 */
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index 0a717fc6aeb6..a76d2c18f150 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -250,6 +250,64 @@ static int __cpuinit nearby_node(int apicid)
 }
 #endif
 
+/*
+ * Fixup core topology information for AMD multi-node processors.
+ * Assumption 1: Number of cores in each internal node is the same.
+ * Assumption 2: Mixed systems with both single-node and dual-node
+ *               processors are not supported.
+ */
+#ifdef CONFIG_X86_HT
+static void __cpuinit amd_fixup_dcm(struct cpuinfo_x86 *c)
+{
+#ifdef CONFIG_PCI
+	u32 t, cpn;
+	u8 n, n_id;
+	int cpu = smp_processor_id();
+
+	/* fixup topology information only once for a core */
+	if (cpu_has(c, X86_FEATURE_AMD_DCM))
+		return;
+
+	/* check for multi-node processor on boot cpu */
+	t = read_pci_config(0, 24, 3, 0xe8);
+	if (!(t & (1 << 29)))
+		return;
+
+	set_cpu_cap(c, X86_FEATURE_AMD_DCM);
+
+	/* cores per node: each internal node has half the number of cores */
+	cpn = c->x86_max_cores >> 1;
+
+	/* even-numbered NB_id of this dual-node processor */
+	n = c->phys_proc_id << 1;
+
+	/*
+	 * determine internal node id and assign cores fifty-fifty to
+	 * each node of the dual-node processor
+	 */
+	t = read_pci_config(0, 24 + n, 3, 0xe8);
+	n = (t>>30) & 0x3;
+	if (n == 0) {
+		if (c->cpu_core_id < cpn)
+			n_id = 0;
+		else
+			n_id = 1;
+	} else {
+		if (c->cpu_core_id < cpn)
+			n_id = 1;
+		else
+			n_id = 0;
+	}
+
+	/* compute entire NodeID, use llc_shared_map to store sibling info */
+	per_cpu(cpu_llc_id, cpu) = (c->phys_proc_id << 1) + n_id;
+
+	/* fixup core id to be in range from 0 to cpn */
+	c->cpu_core_id = c->cpu_core_id % cpn;
+#endif
+}
+#endif
+
 /*
  * On a AMD dual core setup the lower bits of the APIC id distingush the cores.
  * Assumes number of cores is a power of two.
@@ -267,6 +325,9 @@ static void __cpuinit amd_detect_cmp(struct cpuinfo_x86 *c)
 	c->phys_proc_id = c->initial_apicid >> bits;
 	/* use socket ID also for last level cache */
 	per_cpu(cpu_llc_id, cpu) = c->phys_proc_id;
+	/* fixup topology information on multi-node processors */
+	if ((c->x86 == 0x10) && (c->x86_model == 9))
+		amd_fixup_dcm(c);
 #endif
 }
 
@@ -277,7 +338,8 @@ static void __cpuinit srat_detect_node(struct cpuinfo_x86 *c)
 	int node;
 	unsigned apicid = cpu_has_apic ? hard_smp_processor_id() : c->apicid;
 
-	node = c->phys_proc_id;
+	node = per_cpu(cpu_llc_id, cpu);
+
 	if (apicid_to_node[apicid] != NUMA_NO_NODE)
 		node = apicid_to_node[apicid];
 	if (!node_online(node)) {
-- 
cgit v1.2.3


From a326e948c538e8ce998f30d92e146ecea8a30421 Mon Sep 17 00:00:00 2001
From: Andreas Herrmann <andreas.herrmann3@amd.com>
Date: Thu, 3 Sep 2009 09:41:19 +0200
Subject: x86, cacheinfo: Fixup L3 cache information for AMD multi-node
 processors

L3 cache size, associativity and shared_cpu information need to be
adapted to show information for an internal node instead of the
entire physical package.

Signed-off-by: Andreas Herrmann <andreas.herrmann3@amd.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/kernel/cpu/intel_cacheinfo.c | 32 ++++++++++++++++++++++----------
 1 file changed, 22 insertions(+), 10 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c
index 789efe217e1a..cb98a73e5167 100644
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -241,7 +241,7 @@ amd_cpuid4(int leaf, union _cpuid4_leaf_eax *eax,
 	case 0:
 		if (!l1->val)
 			return;
-		assoc = l1->assoc;
+		assoc = assocs[l1->assoc];
 		line_size = l1->line_size;
 		lines_per_tag = l1->lines_per_tag;
 		size_in_kb = l1->size_in_kb;
@@ -249,7 +249,7 @@ amd_cpuid4(int leaf, union _cpuid4_leaf_eax *eax,
 	case 2:
 		if (!l2.val)
 			return;
-		assoc = l2.assoc;
+		assoc = assocs[l2.assoc];
 		line_size = l2.line_size;
 		lines_per_tag = l2.lines_per_tag;
 		/* cpu_data has errata corrections for K7 applied */
@@ -258,10 +258,14 @@ amd_cpuid4(int leaf, union _cpuid4_leaf_eax *eax,
 	case 3:
 		if (!l3.val)
 			return;
-		assoc = l3.assoc;
+		assoc = assocs[l3.assoc];
 		line_size = l3.line_size;
 		lines_per_tag = l3.lines_per_tag;
 		size_in_kb = l3.size_encoded * 512;
+		if (boot_cpu_has(X86_FEATURE_AMD_DCM)) {
+			size_in_kb = size_in_kb >> 1;
+			assoc = assoc >> 1;
+		}
 		break;
 	default:
 		return;
@@ -270,18 +274,14 @@ amd_cpuid4(int leaf, union _cpuid4_leaf_eax *eax,
 	eax->split.is_self_initializing = 1;
 	eax->split.type = types[leaf];
 	eax->split.level = levels[leaf];
-	if (leaf == 3)
-		eax->split.num_threads_sharing =
-			current_cpu_data.x86_max_cores - 1;
-	else
-		eax->split.num_threads_sharing = 0;
+	eax->split.num_threads_sharing = 0;
 	eax->split.num_cores_on_die = current_cpu_data.x86_max_cores - 1;
 
 
-	if (assoc == 0xf)
+	if (assoc == 0xffff)
 		eax->split.is_fully_associative = 1;
 	ebx->split.coherency_line_size = line_size - 1;
-	ebx->split.ways_of_associativity = assocs[assoc] - 1;
+	ebx->split.ways_of_associativity = assoc - 1;
 	ebx->split.physical_line_partition = lines_per_tag - 1;
 	ecx->split.number_of_sets = (size_in_kb * 1024) / line_size /
 		(ebx->split.ways_of_associativity + 1) - 1;
@@ -522,6 +522,18 @@ static void __cpuinit cache_shared_cpu_map_setup(unsigned int cpu, int index)
 	int index_msb, i;
 	struct cpuinfo_x86 *c = &cpu_data(cpu);
 
+	if ((index == 3) && (c->x86_vendor == X86_VENDOR_AMD)) {
+		struct cpuinfo_x86 *d;
+		for_each_online_cpu(i) {
+			if (!per_cpu(cpuid4_info, i))
+				continue;
+			d = &cpu_data(i);
+			this_leaf = CPUID4_INFO_IDX(i, index);
+			cpumask_copy(to_cpumask(this_leaf->shared_cpu_map),
+				     d->llc_shared_map);
+		}
+		return;
+	}
 	this_leaf = CPUID4_INFO_IDX(cpu, index);
 	num_threads_sharing = 1 + this_leaf->eax.split.num_threads_sharing;
 
-- 
cgit v1.2.3


From cb9805ab5b97079f69a21b6b4e344a69d5c96157 Mon Sep 17 00:00:00 2001
From: Andreas Herrmann <andreas.herrmann3@amd.com>
Date: Thu, 3 Sep 2009 09:42:20 +0200
Subject: x86, mcheck: Use correct cpumask for shared bank4

This fixes threshold_bank4 support on multi-node processors.

The correct mask to use is llc_shared_map, representing an internal
node on Magny-Cours.

We need to create 2 sets of symlinks for sibling shared banks -- one
set for each internal node, symlinks of each set should target the
first core on same internal node.

Currently only one set is created where all symlinks are targeting
the first core of the entire socket.

Signed-off-by: Andreas Herrmann <andreas.herrmann3@amd.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/kernel/cpu/mcheck/mce_amd.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd.c b/arch/x86/kernel/cpu/mcheck/mce_amd.c
index ddae21620bda..1fecba404fd8 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_amd.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_amd.c
@@ -489,12 +489,14 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank)
 	int i, err = 0;
 	struct threshold_bank *b = NULL;
 	char name[32];
+	struct cpuinfo_x86 *c = &cpu_data(cpu);
+
 
 	sprintf(name, "threshold_bank%i", bank);
 
 #ifdef CONFIG_SMP
 	if (cpu_data(cpu).cpu_core_id && shared_bank[bank]) {	/* symlink */
-		i = cpumask_first(cpu_core_mask(cpu));
+		i = cpumask_first(c->llc_shared_map);
 
 		/* first core not up yet */
 		if (cpu_data(i).cpu_core_id)
@@ -514,7 +516,7 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank)
 		if (err)
 			goto out;
 
-		cpumask_copy(b->cpus, cpu_core_mask(cpu));
+		cpumask_copy(b->cpus, c->llc_shared_map);
 		per_cpu(threshold_banks, cpu)[bank] = b;
 
 		goto out;
@@ -539,7 +541,7 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank)
 #ifndef CONFIG_SMP
 	cpumask_setall(b->cpus);
 #else
-	cpumask_copy(b->cpus, cpu_core_mask(cpu));
+	cpumask_copy(b->cpus, c->llc_shared_map);
 #endif
 
 	per_cpu(threshold_banks, cpu)[bank] = b;
-- 
cgit v1.2.3


From 5a925b4282d7f805deafde62001a83dbaf8be275 Mon Sep 17 00:00:00 2001
From: Andreas Herrmann <andreas.herrmann3@amd.com>
Date: Thu, 3 Sep 2009 09:44:28 +0200
Subject: x86, sched: Workaround broken sched domain creation for AMD
 Magny-Cours

Current sched domain creation code can't handle multi-node processors.
When switching to power_savings scheduling errors show up and
system might hang later on (due to broken sched domain hierarchy):

  # echo 0  >> /sys/devices/system/cpu/sched_mc_power_savings
  CPU0 attaching sched-domain:
   domain 0: span 0-5 level MC
    groups: 0 1 2 3 4 5
    domain 1: span 0-23 level NODE
     groups: 0-5 6-11 18-23 12-17
  ...
  # echo 1  >> /sys/devices/system/cpu/sched_mc_power_savings
  CPU0 attaching sched-domain:
   domain 0: span 0-11 level MC
    groups: 0 1 2 3 4 5 6 7 8 9 10 11
  ERROR: parent span is not a superset of domain->span
    domain 1: span 0-5 level CPU
  ERROR: domain->groups does not contain CPU0
     groups: 6-11 (__cpu_power = 12288)
  ERROR: groups don't span domain->span
     domain 2: span 0-23 level NODE
      groups:
  ERROR: domain->cpu_power not set

  ERROR: groups don't span domain->span
  ...

Fixing all aspects of power-savings scheduling for Magny-Cours needs
some larger changes in the sched domain creation code.

As a short-term and temporary workaround avoid the problems by
extending "the worst possible hack" ;-(
and always use llc_shared_map on AMD Magny-Cours when MC domain span
is calculated.

With this I get:

  # echo 1  >> /sys/devices/system/cpu/sched_mc_power_savings
  CPU0 attaching sched-domain:
   domain 0: span 0-5 level MC
    groups: 0 1 2 3 4 5
    domain 1: span 0-5 level CPU
     groups: 0-5 (__cpu_power = 6144)
     domain 2: span 0-23 level NODE
      groups: 0-5 (__cpu_power = 6144) 6-11 (__cpu_power = 6144) 18-23 (__cpu_power = 6144) 12-17 (__cpu_power = 6144)
  ...

I.e. no errors during sched domain creation, no system hangs, and also
mc_power_savings scheduling works to a certain extend.

Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Andreas Herrmann <andreas.herrmann3@amd.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/kernel/smpboot.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 2fecda69ee64..c36cc1452cdc 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -434,7 +434,8 @@ const struct cpumask *cpu_coregroup_mask(int cpu)
 	 * For perf, we return last level cache shared map.
 	 * And for power savings, we return cpu_core_map
 	 */
-	if (sched_mc_power_savings || sched_smt_power_savings)
+	if ((sched_mc_power_savings || sched_smt_power_savings) &&
+	    !(cpu_has(c, X86_FEATURE_AMD_DCM)))
 		return cpu_core_mask(cpu);
 	else
 		return c->llc_shared_map;
-- 
cgit v1.2.3


From 53f824520b6d84ca5b4a8fd71addc91dbf64357e Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Thu, 3 Sep 2009 14:31:44 -0700
Subject: x86/i386: Put aligned stack-canary in percpu shared_aligned section

Pack aligned things together into a special section to minimize
padding holes.

Suggested-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Tejun Heo <tj@kernel.org>
LKML-Reference: <4AA035C0.9070202@goop.org>
[ queued up in tip:x86/asm because it depends on this commit:
  x86/i386: Make sure stack-protector segment base is cache aligned ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/include/asm/processor.h | 2 +-
 arch/x86/kernel/cpu/common.c     | 2 +-
 include/asm-generic/percpu.h     | 3 +++
 include/linux/percpu-defs.h      | 8 ++++++++
 4 files changed, 13 insertions(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h
index e597ecc8753c..ac7e79654f3a 100644
--- a/arch/x86/include/asm/processor.h
+++ b/arch/x86/include/asm/processor.h
@@ -413,7 +413,7 @@ struct stack_canary {
 	char __pad[20];		/* canary at %gs:20 */
 	unsigned long canary;
 };
-DECLARE_PER_CPU(struct stack_canary, stack_canary) ____cacheline_aligned;
+DECLARE_PER_CPU_ALIGNED(struct stack_canary, stack_canary);
 #endif
 #endif	/* X86_64 */
 
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 7d84bc4c1188..f23e236391a3 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -1043,7 +1043,7 @@ DEFINE_PER_CPU(struct orig_ist, orig_ist);
 #else	/* CONFIG_X86_64 */
 
 #ifdef CONFIG_CC_STACKPROTECTOR
-DEFINE_PER_CPU(struct stack_canary, stack_canary) ____cacheline_aligned;
+DEFINE_PER_CPU_ALIGNED(struct stack_canary, stack_canary);
 #endif
 
 /* Make sure %fs and %gs are initialized properly in idle threads */
diff --git a/include/asm-generic/percpu.h b/include/asm-generic/percpu.h
index aa00800adacc..90079c373f1c 100644
--- a/include/asm-generic/percpu.h
+++ b/include/asm-generic/percpu.h
@@ -81,14 +81,17 @@ extern void setup_per_cpu_areas(void);
 
 #ifdef MODULE
 #define PER_CPU_SHARED_ALIGNED_SECTION ""
+#define PER_CPU_ALIGNED_SECTION ""
 #else
 #define PER_CPU_SHARED_ALIGNED_SECTION ".shared_aligned"
+#define PER_CPU_ALIGNED_SECTION ".shared_aligned"
 #endif
 #define PER_CPU_FIRST_SECTION ".first"
 
 #else
 
 #define PER_CPU_SHARED_ALIGNED_SECTION ""
+#define PER_CPU_ALIGNED_SECTION ".shared_aligned"
 #define PER_CPU_FIRST_SECTION ""
 
 #endif
diff --git a/include/linux/percpu-defs.h b/include/linux/percpu-defs.h
index 68438e18fff4..3058cf9dd3d4 100644
--- a/include/linux/percpu-defs.h
+++ b/include/linux/percpu-defs.h
@@ -66,6 +66,14 @@
 	DEFINE_PER_CPU_SECTION(type, name, PER_CPU_SHARED_ALIGNED_SECTION) \
 	____cacheline_aligned_in_smp
 
+#define DECLARE_PER_CPU_ALIGNED(type, name)				\
+	DECLARE_PER_CPU_SECTION(type, name, PER_CPU_ALIGNED_SECTION)	\
+	____cacheline_aligned
+
+#define DEFINE_PER_CPU_ALIGNED(type, name)				\
+	DEFINE_PER_CPU_SECTION(type, name, PER_CPU_ALIGNED_SECTION)	\
+	____cacheline_aligned
+
 /*
  * Declaration/definition used for per-CPU variables that must be page aligned.
  */
-- 
cgit v1.2.3


From 747b50aaf728987732e6ff3ba10aba4acc4e0277 Mon Sep 17 00:00:00 2001
From: "markus.t.metzger@intel.com" <markus.t.metzger@intel.com>
Date: Wed, 2 Sep 2009 16:04:46 +0200
Subject: x86, perf_counter, bts: Fail if BTS is not available

Reserve PERF_COUNT_HW_BRANCH_INSTRUCTIONS with sample_period ==
1 for BTS tracing and fail, if BTS is not available.

Signed-off-by: Markus Metzger <markus.t.metzger@intel.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20090902140612.943801000@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/perf_counter.c | 25 +++++++++++--------------
 1 file changed, 11 insertions(+), 14 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/perf_counter.c b/arch/x86/kernel/cpu/perf_counter.c
index 396e35db7058..2f41874ffb86 100644
--- a/arch/x86/kernel/cpu/perf_counter.c
+++ b/arch/x86/kernel/cpu/perf_counter.c
@@ -769,7 +769,7 @@ static int reserve_bts_hardware(void)
 	int cpu, err = 0;
 
 	if (!bts_available())
-		return -EOPNOTSUPP;
+		return 0;
 
 	get_online_cpus();
 
@@ -914,7 +914,7 @@ static int __hw_perf_counter_init(struct perf_counter *counter)
 			if (!reserve_pmc_hardware())
 				err = -EBUSY;
 			else
-				reserve_bts_hardware();
+				err = reserve_bts_hardware();
 		}
 		if (!err)
 			atomic_inc(&active_counters);
@@ -979,6 +979,13 @@ static int __hw_perf_counter_init(struct perf_counter *counter)
 	if (config == -1LL)
 		return -EINVAL;
 
+	/*
+	 * Branch tracing:
+	 */
+	if ((attr->config == PERF_COUNT_HW_BRANCH_INSTRUCTIONS) &&
+	    (hwc->sample_period == 1) && !bts_available())
+		return -EOPNOTSUPP;
+
 	hwc->config |= config;
 
 	return 0;
@@ -1355,19 +1362,9 @@ static int x86_pmu_enable(struct perf_counter *counter)
 
 	idx = fixed_mode_idx(counter, hwc);
 	if (idx == X86_PMC_IDX_FIXED_BTS) {
-		/*
-		 * Try to use BTS for branch tracing. If that is not
-		 * available, try to get a generic counter.
-		 */
-		if (unlikely(!cpuc->ds))
-			goto try_generic;
-
-		/*
-		 * Try to get the fixed counter, if that is already taken
-		 * then try to get a generic counter:
-		 */
+		/* BTS is already occupied. */
 		if (test_and_set_bit(idx, cpuc->used_mask))
-			goto try_generic;
+			return -EAGAIN;
 
 		hwc->config_base	= 0;
 		hwc->counter_base	= 0;
-- 
cgit v1.2.3


From 596da17f94c103348ebe04129c00d536ea0e80e2 Mon Sep 17 00:00:00 2001
From: "markus.t.metzger@intel.com" <markus.t.metzger@intel.com>
Date: Wed, 2 Sep 2009 16:04:47 +0200
Subject: x86, perf_counter, bts: Correct pointer-to-u64 casts

On 32bit, pointers in the DS AREA configuration are cast to
u64. The current (long) cast to avoid compiler warnings results
in a signed 64bit address.

Signed-off-by: Markus Metzger <markus.t.metzger@intel.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20090902140615.305889000@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/perf_counter.c | 24 ++++++++++++------------
 1 file changed, 12 insertions(+), 12 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/perf_counter.c b/arch/x86/kernel/cpu/perf_counter.c
index 2f41874ffb86..3776b0b630c8 100644
--- a/arch/x86/kernel/cpu/perf_counter.c
+++ b/arch/x86/kernel/cpu/perf_counter.c
@@ -726,7 +726,8 @@ static inline void init_debug_store_on_cpu(int cpu)
 		return;
 
 	wrmsr_on_cpu(cpu, MSR_IA32_DS_AREA,
-		     (u32)((u64)(long)ds), (u32)((u64)(long)ds >> 32));
+		     (u32)((u64)(unsigned long)ds),
+		     (u32)((u64)(unsigned long)ds >> 32));
 }
 
 static inline void fini_debug_store_on_cpu(int cpu)
@@ -757,7 +758,7 @@ static void release_bts_hardware(void)
 
 		per_cpu(cpu_hw_counters, cpu).ds = NULL;
 
-		kfree((void *)(long)ds->bts_buffer_base);
+		kfree((void *)(unsigned long)ds->bts_buffer_base);
 		kfree(ds);
 	}
 
@@ -788,7 +789,7 @@ static int reserve_bts_hardware(void)
 			break;
 		}
 
-		ds->bts_buffer_base = (u64)(long)buffer;
+		ds->bts_buffer_base = (u64)(unsigned long)buffer;
 		ds->bts_index = ds->bts_buffer_base;
 		ds->bts_absolute_maximum =
 			ds->bts_buffer_base + BTS_BUFFER_SIZE;
@@ -1491,7 +1492,7 @@ static void intel_pmu_drain_bts_buffer(struct cpu_hw_counters *cpuc,
 	};
 	struct perf_counter *counter = cpuc->counters[X86_PMC_IDX_FIXED_BTS];
 	unsigned long orig_ip = data->regs->ip;
-	u64 at;
+	struct bts_record *at, *top;
 
 	if (!counter)
 		return;
@@ -1499,19 +1500,18 @@ static void intel_pmu_drain_bts_buffer(struct cpu_hw_counters *cpuc,
 	if (!ds)
 		return;
 
-	for (at = ds->bts_buffer_base;
-	     at < ds->bts_index;
-	     at += sizeof(struct bts_record)) {
-		struct bts_record *rec = (struct bts_record *)(long)at;
+	at  = (struct bts_record *)(unsigned long)ds->bts_buffer_base;
+	top = (struct bts_record *)(unsigned long)ds->bts_index;
 
-		data->regs->ip	= rec->from;
-		data->addr	= rec->to;
+	ds->bts_index = ds->bts_buffer_base;
+
+	for (; at < top; at++) {
+		data->regs->ip	= at->from;
+		data->addr	= at->to;
 
 		perf_counter_output(counter, 1, data);
 	}
 
-	ds->bts_index = ds->bts_buffer_base;
-
 	data->regs->ip	= orig_ip;
 	data->addr	= 0;
 
-- 
cgit v1.2.3


From 1653192f510bd8114b7b133d7289e6e5c3e95046 Mon Sep 17 00:00:00 2001
From: "markus.t.metzger@intel.com" <markus.t.metzger@intel.com>
Date: Wed, 2 Sep 2009 16:04:48 +0200
Subject: x86, perf_counter, bts: Do not allow kernel BTS tracing for now

Kernel BTS tracing generates too much data too fast for us to
handle, causing the kernel to hang.

Fail for BTS requests for kernel code.

Signed-off-by: Markus Metzger <markus.t.metzger@intel.com>
Acked-by: Peter Zijlstra <a.p.zjilstra@chello.nl>
LKML-Reference: <20090902140616.901253000@intel.com>
[ This is really a workaround - but we want BTS tracing in .32
  so make sure we dont regress. The lockup should be fixed
  ASAP. ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/perf_counter.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/perf_counter.c b/arch/x86/kernel/cpu/perf_counter.c
index 3776b0b630c8..f9cd0849bd42 100644
--- a/arch/x86/kernel/cpu/perf_counter.c
+++ b/arch/x86/kernel/cpu/perf_counter.c
@@ -984,8 +984,15 @@ static int __hw_perf_counter_init(struct perf_counter *counter)
 	 * Branch tracing:
 	 */
 	if ((attr->config == PERF_COUNT_HW_BRANCH_INSTRUCTIONS) &&
-	    (hwc->sample_period == 1) && !bts_available())
-		return -EOPNOTSUPP;
+	    (hwc->sample_period == 1)) {
+		/* BTS is not supported by this architecture. */
+		if (!bts_available())
+			return -EOPNOTSUPP;
+
+		/* BTS is currently only allowed for user-mode. */
+		if (hwc->config & ARCH_PERFMON_EVENTSEL_OS)
+			return -EOPNOTSUPP;
+	}
 
 	hwc->config |= config;
 
-- 
cgit v1.2.3


From 0d96b9ff748b5f57d6f1d6d21209f5745245aadc Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yinghai@kernel.org>
Date: Sat, 29 Aug 2009 13:17:14 -0700
Subject: x86: Use hard_smp_processor_id() to get apic id for AMD K8 cpus

Otherwise, system with apci id lifting will have wrong apicid in
/proc/cpuinfo.

and use that in srat_detect_node().

Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Cc: Andreas Herrmann <andreas.herrmann3@amd.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Cyrill Gorcunov <gorcunov@openvz.org>
LKML-Reference: <4A998CCA.1040407@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/amd.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index a76d2c18f150..e1600c775e1d 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -336,7 +336,7 @@ static void __cpuinit srat_detect_node(struct cpuinfo_x86 *c)
 #if defined(CONFIG_NUMA) && defined(CONFIG_X86_64)
 	int cpu = smp_processor_id();
 	int node;
-	unsigned apicid = cpu_has_apic ? hard_smp_processor_id() : c->apicid;
+	unsigned apicid = c->apicid;
 
 	node = per_cpu(cpu_llc_id, cpu);
 
@@ -481,6 +481,9 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 	}
 	if (c->x86 == 0x10 || c->x86 == 0x11)
 		set_cpu_cap(c, X86_FEATURE_REP_GOOD);
+
+	/* get apicid instead of initial apic id from cpuid */
+	c->apicid = hard_smp_processor_id();
 #else
 
 	/*
-- 
cgit v1.2.3


From 47734f89be0614b5acbd6a532390f9c72f019648 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 4 Sep 2009 11:21:24 +0200
Subject: sched: Clean up topology.h

Re-organize the flag settings so that it's visible at a glance
which sched-domains flags are set and which not.

With the new balancer code we'll need to re-tune these details
anyway, so make it cleaner to make fewer mistakes down the
road ;-)

Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Andreas Herrmann <andreas.herrmann3@amd.com>
Cc: Andreas Herrmann <andreas.herrmann3@amd.com>
Cc: Gautham R Shenoy <ego@in.ibm.com>
Cc: Balbir Singh <balbir@in.ibm.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/include/asm/topology.h |  47 ++++++-----
 include/linux/topology.h        | 169 ++++++++++++++++++++++++----------------
 2 files changed, 129 insertions(+), 87 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/topology.h b/arch/x86/include/asm/topology.h
index 066ef590d7e0..be29eb81fb06 100644
--- a/arch/x86/include/asm/topology.h
+++ b/arch/x86/include/asm/topology.h
@@ -129,25 +129,34 @@ extern unsigned long node_remap_size[];
 #endif
 
 /* sched_domains SD_NODE_INIT for NUMA machines */
-#define SD_NODE_INIT (struct sched_domain) {		\
-	.min_interval		= 8,			\
-	.max_interval		= 32,			\
-	.busy_factor		= 32,			\
-	.imbalance_pct		= 125,			\
-	.cache_nice_tries	= SD_CACHE_NICE_TRIES,	\
-	.busy_idx		= 3,			\
-	.idle_idx		= SD_IDLE_IDX,		\
-	.newidle_idx		= SD_NEWIDLE_IDX,	\
-	.wake_idx		= 1,			\
-	.forkexec_idx		= SD_FORKEXEC_IDX,	\
-	.flags			= SD_LOAD_BALANCE	\
-				| SD_BALANCE_EXEC	\
-				| SD_BALANCE_FORK	\
-				| SD_WAKE_AFFINE	\
-				| SD_WAKE_BALANCE	\
-				| SD_SERIALIZE,		\
-	.last_balance		= jiffies,		\
-	.balance_interval	= 1,			\
+#define SD_NODE_INIT (struct sched_domain) {				\
+	.min_interval		= 8,					\
+	.max_interval		= 32,					\
+	.busy_factor		= 32,					\
+	.imbalance_pct		= 125,					\
+	.cache_nice_tries	= SD_CACHE_NICE_TRIES,			\
+	.busy_idx		= 3,					\
+	.idle_idx		= SD_IDLE_IDX,				\
+	.newidle_idx		= SD_NEWIDLE_IDX,			\
+	.wake_idx		= 1,					\
+	.forkexec_idx		= SD_FORKEXEC_IDX,			\
+									\
+	.flags			= 1*SD_LOAD_BALANCE			\
+				| 0*SD_BALANCE_NEWIDLE			\
+				| 1*SD_BALANCE_EXEC			\
+				| 1*SD_BALANCE_FORK			\
+				| 0*SD_WAKE_IDLE			\
+				| 1*SD_WAKE_AFFINE			\
+				| 1*SD_WAKE_BALANCE			\
+				| 0*SD_SHARE_CPUPOWER			\
+				| 0*SD_POWERSAVINGS_BALANCE		\
+				| 0*SD_SHARE_PKG_RESOURCES		\
+				| 1*SD_SERIALIZE			\
+				| 0*SD_WAKE_IDLE_FAR			\
+				| 0*SD_PREFER_SIBLING			\
+				,					\
+	.last_balance		= jiffies,				\
+	.balance_interval	= 1,					\
 }
 
 #ifdef CONFIG_X86_64_ACPI_NUMA
diff --git a/include/linux/topology.h b/include/linux/topology.h
index 6203ae5067ce..fe2c0329f82f 100644
--- a/include/linux/topology.h
+++ b/include/linux/topology.h
@@ -85,21 +85,29 @@ int arch_update_cpu_topology(void);
 #define ARCH_HAS_SCHED_WAKE_IDLE
 /* Common values for SMT siblings */
 #ifndef SD_SIBLING_INIT
-#define SD_SIBLING_INIT (struct sched_domain) {		\
-	.min_interval		= 1,			\
-	.max_interval		= 2,			\
-	.busy_factor		= 64,			\
-	.imbalance_pct		= 110,			\
-	.flags			= SD_LOAD_BALANCE	\
-				| SD_BALANCE_NEWIDLE	\
-				| SD_BALANCE_FORK	\
-				| SD_BALANCE_EXEC	\
-				| SD_WAKE_AFFINE	\
-				| SD_WAKE_BALANCE	\
-				| SD_SHARE_CPUPOWER,	\
-	.last_balance		= jiffies,		\
-	.balance_interval	= 1,			\
-	.smt_gain		= 1178,	/* 15% */	\
+#define SD_SIBLING_INIT (struct sched_domain) {				\
+	.min_interval		= 1,					\
+	.max_interval		= 2,					\
+	.busy_factor		= 64,					\
+	.imbalance_pct		= 110,					\
+									\
+	.flags			= 1*SD_LOAD_BALANCE			\
+				| 1*SD_BALANCE_NEWIDLE			\
+				| 1*SD_BALANCE_EXEC			\
+				| 1*SD_BALANCE_FORK			\
+				| 0*SD_WAKE_IDLE			\
+				| 1*SD_WAKE_AFFINE			\
+				| 1*SD_WAKE_BALANCE			\
+				| 1*SD_SHARE_CPUPOWER			\
+				| 0*SD_POWERSAVINGS_BALANCE		\
+				| 0*SD_SHARE_PKG_RESOURCES		\
+				| 0*SD_SERIALIZE			\
+				| 0*SD_WAKE_IDLE_FAR			\
+				| 0*SD_PREFER_SIBLING			\
+				,					\
+	.last_balance		= jiffies,				\
+	.balance_interval	= 1,					\
+	.smt_gain		= 1178,	/* 15% */			\
 }
 #endif
 #endif /* CONFIG_SCHED_SMT */
@@ -107,69 +115,94 @@ int arch_update_cpu_topology(void);
 #ifdef CONFIG_SCHED_MC
 /* Common values for MC siblings. for now mostly derived from SD_CPU_INIT */
 #ifndef SD_MC_INIT
-#define SD_MC_INIT (struct sched_domain) {		\
-	.min_interval		= 1,			\
-	.max_interval		= 4,			\
-	.busy_factor		= 64,			\
-	.imbalance_pct		= 125,			\
-	.cache_nice_tries	= 1,			\
-	.busy_idx		= 2,			\
-	.wake_idx		= 1,			\
-	.forkexec_idx		= 1,			\
-	.flags			= SD_LOAD_BALANCE	\
-				| SD_BALANCE_FORK	\
-				| SD_BALANCE_EXEC	\
-				| SD_WAKE_AFFINE	\
-				| SD_WAKE_BALANCE	\
-				| SD_SHARE_PKG_RESOURCES\
-				| sd_balance_for_mc_power()\
-				| sd_power_saving_flags(),\
-	.last_balance		= jiffies,		\
-	.balance_interval	= 1,			\
+#define SD_MC_INIT (struct sched_domain) {				\
+	.min_interval		= 1,					\
+	.max_interval		= 4,					\
+	.busy_factor		= 64,					\
+	.imbalance_pct		= 125,					\
+	.cache_nice_tries	= 1,					\
+	.busy_idx		= 2,					\
+	.wake_idx		= 1,					\
+	.forkexec_idx		= 1,					\
+									\
+	.flags			= 1*SD_LOAD_BALANCE			\
+				| 0*SD_BALANCE_NEWIDLE			\
+				| 1*SD_BALANCE_EXEC			\
+				| 1*SD_BALANCE_FORK			\
+				| 0*SD_WAKE_IDLE			\
+				| 1*SD_WAKE_AFFINE			\
+				| 1*SD_WAKE_BALANCE			\
+				| 0*SD_SHARE_CPUPOWER			\
+				| 1*SD_SHARE_PKG_RESOURCES		\
+				| 0*SD_SERIALIZE			\
+				| 0*SD_WAKE_IDLE_FAR			\
+				| sd_balance_for_mc_power()		\
+				| sd_power_saving_flags()		\
+				,					\
+	.last_balance		= jiffies,				\
+	.balance_interval	= 1,					\
 }
 #endif
 #endif /* CONFIG_SCHED_MC */
 
 /* Common values for CPUs */
 #ifndef SD_CPU_INIT
-#define SD_CPU_INIT (struct sched_domain) {		\
-	.min_interval		= 1,			\
-	.max_interval		= 4,			\
-	.busy_factor		= 64,			\
-	.imbalance_pct		= 125,			\
-	.cache_nice_tries	= 1,			\
-	.busy_idx		= 2,			\
-	.idle_idx		= 1,			\
-	.newidle_idx		= 2,			\
-	.wake_idx		= 1,			\
-	.forkexec_idx		= 1,			\
-	.flags			= SD_LOAD_BALANCE	\
-				| SD_BALANCE_EXEC	\
-				| SD_BALANCE_FORK	\
-				| SD_WAKE_AFFINE	\
-				| SD_WAKE_BALANCE	\
-				| sd_balance_for_package_power()\
-				| sd_power_saving_flags(),\
-	.last_balance		= jiffies,		\
-	.balance_interval	= 1,			\
+#define SD_CPU_INIT (struct sched_domain) {				\
+	.min_interval		= 1,					\
+	.max_interval		= 4,					\
+	.busy_factor		= 64,					\
+	.imbalance_pct		= 125,					\
+	.cache_nice_tries	= 1,					\
+	.busy_idx		= 2,					\
+	.idle_idx		= 1,					\
+	.newidle_idx		= 2,					\
+	.wake_idx		= 1,					\
+	.forkexec_idx		= 1,					\
+									\
+	.flags			= 1*SD_LOAD_BALANCE			\
+				| 0*SD_BALANCE_NEWIDLE			\
+				| 1*SD_BALANCE_EXEC			\
+				| 1*SD_BALANCE_FORK			\
+				| 0*SD_WAKE_IDLE			\
+				| 0*SD_WAKE_AFFINE			\
+				| 1*SD_WAKE_BALANCE			\
+				| 0*SD_SHARE_CPUPOWER			\
+				| 0*SD_SHARE_PKG_RESOURCES		\
+				| 0*SD_SERIALIZE			\
+				| 0*SD_WAKE_IDLE_FAR			\
+				| sd_balance_for_package_power()	\
+				| sd_power_saving_flags()		\
+				,					\
+	.last_balance		= jiffies,				\
+	.balance_interval	= 1,					\
 }
 #endif
 
 /* sched_domains SD_ALLNODES_INIT for NUMA machines */
-#define SD_ALLNODES_INIT (struct sched_domain) {	\
-	.min_interval		= 64,			\
-	.max_interval		= 64*num_online_cpus(),	\
-	.busy_factor		= 128,			\
-	.imbalance_pct		= 133,			\
-	.cache_nice_tries	= 1,			\
-	.busy_idx		= 3,			\
-	.idle_idx		= 3,			\
-	.flags			= SD_LOAD_BALANCE	\
-				| SD_BALANCE_NEWIDLE	\
-				| SD_WAKE_AFFINE	\
-				| SD_SERIALIZE,		\
-	.last_balance		= jiffies,		\
-	.balance_interval	= 64,			\
+#define SD_ALLNODES_INIT (struct sched_domain) {			\
+	.min_interval		= 64,					\
+	.max_interval		= 64*num_online_cpus(),			\
+	.busy_factor		= 128,					\
+	.imbalance_pct		= 133,					\
+	.cache_nice_tries	= 1,					\
+	.busy_idx		= 3,					\
+	.idle_idx		= 3,					\
+	.flags			= 1*SD_LOAD_BALANCE			\
+				| 1*SD_BALANCE_NEWIDLE			\
+				| 0*SD_BALANCE_EXEC			\
+				| 0*SD_BALANCE_FORK			\
+				| 0*SD_WAKE_IDLE			\
+				| 1*SD_WAKE_AFFINE			\
+				| 0*SD_WAKE_BALANCE			\
+				| 0*SD_SHARE_CPUPOWER			\
+				| 0*SD_POWERSAVINGS_BALANCE		\
+				| 0*SD_SHARE_PKG_RESOURCES		\
+				| 1*SD_SERIALIZE			\
+				| 0*SD_WAKE_IDLE_FAR			\
+				| 0*SD_PREFER_SIBLING			\
+				,					\
+	.last_balance		= jiffies,				\
+	.balance_interval	= 64,					\
 }
 
 #ifdef CONFIG_NUMA
-- 
cgit v1.2.3


From 840a0653100dbde599ae8ddf83fa214dfa5fd1aa Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 4 Sep 2009 11:32:54 +0200
Subject: sched: Turn on SD_BALANCE_NEWIDLE

Start the re-tuning of the balancer by turning on newidle.

It improves hackbench performance and parallelism on a 4x4 box.
The "perf stat --repeat 10" measurements give us:

  domain0             domain1
  .......................................
 -SD_BALANCE_NEWIDLE -SD_BALANCE_NEWIDLE:
   2041.273208  task-clock-msecs         #      9.354 CPUs    ( +-   0.363% )

 +SD_BALANCE_NEWIDLE -SD_BALANCE_NEWIDLE:
   2086.326925  task-clock-msecs         #     11.934 CPUs    ( +-   0.301% )

 +SD_BALANCE_NEWIDLE +SD_BALANCE_NEWIDLE:
   2115.289791  task-clock-msecs         #     12.158 CPUs    ( +-   0.263% )

Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Andreas Herrmann <andreas.herrmann3@amd.com>
Cc: Andreas Herrmann <andreas.herrmann3@amd.com>
Cc: Gautham R Shenoy <ego@in.ibm.com>
Cc: Balbir Singh <balbir@in.ibm.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/include/asm/topology.h | 2 +-
 include/linux/topology.h        | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/topology.h b/arch/x86/include/asm/topology.h
index be29eb81fb06..ef7bc7fc2528 100644
--- a/arch/x86/include/asm/topology.h
+++ b/arch/x86/include/asm/topology.h
@@ -142,7 +142,7 @@ extern unsigned long node_remap_size[];
 	.forkexec_idx		= SD_FORKEXEC_IDX,			\
 									\
 	.flags			= 1*SD_LOAD_BALANCE			\
-				| 0*SD_BALANCE_NEWIDLE			\
+				| 1*SD_BALANCE_NEWIDLE			\
 				| 1*SD_BALANCE_EXEC			\
 				| 1*SD_BALANCE_FORK			\
 				| 0*SD_WAKE_IDLE			\
diff --git a/include/linux/topology.h b/include/linux/topology.h
index fe2c0329f82f..66774fddec9b 100644
--- a/include/linux/topology.h
+++ b/include/linux/topology.h
@@ -126,7 +126,7 @@ int arch_update_cpu_topology(void);
 	.forkexec_idx		= 1,					\
 									\
 	.flags			= 1*SD_LOAD_BALANCE			\
-				| 0*SD_BALANCE_NEWIDLE			\
+				| 1*SD_BALANCE_NEWIDLE			\
 				| 1*SD_BALANCE_EXEC			\
 				| 1*SD_BALANCE_FORK			\
 				| 0*SD_WAKE_IDLE			\
@@ -160,7 +160,7 @@ int arch_update_cpu_topology(void);
 	.forkexec_idx		= 1,					\
 									\
 	.flags			= 1*SD_LOAD_BALANCE			\
-				| 0*SD_BALANCE_NEWIDLE			\
+				| 1*SD_BALANCE_NEWIDLE			\
 				| 1*SD_BALANCE_EXEC			\
 				| 1*SD_BALANCE_FORK			\
 				| 0*SD_WAKE_IDLE			\
-- 
cgit v1.2.3


From 8e019366ba749a536131cde1947af6dcaccf8e8f Mon Sep 17 00:00:00 2001
From: Pekka Enberg <penberg@cs.helsinki.fi>
Date: Thu, 27 Aug 2009 14:50:00 +0100
Subject: kmemleak: Don't scan uninitialized memory when kmemcheck is enabled

Ingo Molnar reported the following kmemcheck warning when running both
kmemleak and kmemcheck enabled:

  PM: Adding info for No Bus:vcsa7
  WARNING: kmemcheck: Caught 32-bit read from uninitialized memory
  (f6f6e1a4)
  d873f9f600000000c42ae4c1005c87f70000000070665f666978656400000000
   i i i i u u u u i i i i i i i i i i i i i i i i i i i i i u u u
           ^

  Pid: 3091, comm: kmemleak Not tainted (2.6.31-rc7-tip #1303) P4DC6
  EIP: 0060:[<c110301f>] EFLAGS: 00010006 CPU: 0
  EIP is at scan_block+0x3f/0xe0
  EAX: f40bd700 EBX: f40bd780 ECX: f16b46c0 EDX: 00000001
  ESI: f6f6e1a4 EDI: 00000000 EBP: f10f3f4c ESP: c2605fcc
   DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068
  CR0: 8005003b CR2: e89a4844 CR3: 30ff1000 CR4: 000006f0
  DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000
  DR6: ffff4ff0 DR7: 00000400
   [<c110313c>] scan_object+0x7c/0xf0
   [<c1103389>] kmemleak_scan+0x1d9/0x400
   [<c1103a3c>] kmemleak_scan_thread+0x4c/0xb0
   [<c10819d4>] kthread+0x74/0x80
   [<c10257db>] kernel_thread_helper+0x7/0x3c
   [<ffffffff>] 0xffffffff
  kmemleak: 515 new suspected memory leaks (see
  /sys/kernel/debug/kmemleak)
  kmemleak: 42 new suspected memory leaks (see /sys/kernel/debug/kmemleak)

The problem here is that kmemleak will scan partially initialized
objects that makes kmemcheck complain. Fix that up by skipping
uninitialized memory regions when kmemcheck is enabled.

Reported-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
---
 arch/x86/mm/kmemcheck/kmemcheck.c | 14 ++++++++++++++
 include/linux/kmemcheck.h         |  7 +++++++
 mm/kmemleak.c                     | 12 ++++++++++--
 3 files changed, 31 insertions(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/mm/kmemcheck/kmemcheck.c b/arch/x86/mm/kmemcheck/kmemcheck.c
index 2c55ed098654..528bf954eb74 100644
--- a/arch/x86/mm/kmemcheck/kmemcheck.c
+++ b/arch/x86/mm/kmemcheck/kmemcheck.c
@@ -331,6 +331,20 @@ static void kmemcheck_read_strict(struct pt_regs *regs,
 	kmemcheck_shadow_set(shadow, size);
 }
 
+bool kmemcheck_is_obj_initialized(unsigned long addr, size_t size)
+{
+	enum kmemcheck_shadow status;
+	void *shadow;
+
+	shadow = kmemcheck_shadow_lookup(addr);
+	if (!shadow)
+		return true;
+
+	status = kmemcheck_shadow_test(shadow, size);
+
+	return status == KMEMCHECK_SHADOW_INITIALIZED;
+}
+
 /* Access may cross page boundary */
 static void kmemcheck_read(struct pt_regs *regs,
 	unsigned long addr, unsigned int size)
diff --git a/include/linux/kmemcheck.h b/include/linux/kmemcheck.h
index 47b39b7c7e84..dc2fd545db00 100644
--- a/include/linux/kmemcheck.h
+++ b/include/linux/kmemcheck.h
@@ -34,6 +34,8 @@ void kmemcheck_mark_initialized_pages(struct page *p, unsigned int n);
 int kmemcheck_show_addr(unsigned long address);
 int kmemcheck_hide_addr(unsigned long address);
 
+bool kmemcheck_is_obj_initialized(unsigned long addr, size_t size);
+
 #else
 #define kmemcheck_enabled 0
 
@@ -99,6 +101,11 @@ static inline void kmemcheck_mark_initialized_pages(struct page *p,
 {
 }
 
+static inline bool kmemcheck_is_obj_initialized(unsigned long addr, size_t size)
+{
+	return true;
+}
+
 #endif /* CONFIG_KMEMCHECK */
 
 /*
diff --git a/mm/kmemleak.c b/mm/kmemleak.c
index 1d7645b0a97c..c494fee7a2b5 100644
--- a/mm/kmemleak.c
+++ b/mm/kmemleak.c
@@ -97,6 +97,7 @@
 #include <asm/processor.h>
 #include <asm/atomic.h>
 
+#include <linux/kmemcheck.h>
 #include <linux/kmemleak.h>
 
 /*
@@ -967,15 +968,22 @@ static void scan_block(void *_start, void *_end,
 	unsigned long *end = _end - (BYTES_PER_POINTER - 1);
 
 	for (ptr = start; ptr < end; ptr++) {
-		unsigned long flags;
-		unsigned long pointer = *ptr;
 		struct kmemleak_object *object;
+		unsigned long flags;
+		unsigned long pointer;
 
 		if (allow_resched)
 			cond_resched();
 		if (scan_should_stop())
 			break;
 
+		/* don't scan uninitialized memory */
+		if (!kmemcheck_is_obj_initialized((unsigned long)ptr,
+						  BYTES_PER_POINTER))
+			continue;
+
+		pointer = *ptr;
+
 		object = find_and_get_object(pointer, 1);
 		if (!object)
 			continue;
-- 
cgit v1.2.3


From b19ae3999891cad21a3995c34d313dda5df014e2 Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Fri, 4 Sep 2009 10:00:09 -0700
Subject: x86, msr: change msr-reg.o to obj-y, and export its symbols

Change msr-reg.o to obj-y (it will be included in virtually every
kernel since it is used by the initialization code for AMD processors)
and add a separate C file to export its symbols to modules, so that
msr.ko can use them; on uniprocessors we bypass the helper functions
in msr.o and use the accessor functions directly via inlines.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
LKML-Reference: <20090904140834.GA15789@elte.hu>
Cc: Borislav Petkov <petkovbb@googlemail.com>
---
 arch/x86/lib/Makefile         | 3 ++-
 arch/x86/lib/msr-reg-export.c | 5 +++++
 2 files changed, 7 insertions(+), 1 deletion(-)
 create mode 100644 arch/x86/lib/msr-reg-export.c

(limited to 'arch/x86')

diff --git a/arch/x86/lib/Makefile b/arch/x86/lib/Makefile
index b59c0647d809..9e609206fac9 100644
--- a/arch/x86/lib/Makefile
+++ b/arch/x86/lib/Makefile
@@ -8,7 +8,8 @@ lib-y := delay.o
 lib-y += thunk_$(BITS).o
 lib-y += usercopy_$(BITS).o getuser.o putuser.o
 lib-y += memcpy_$(BITS).o
-lib-y += msr-reg.o
+
+obj-y += msr-reg.o msr-reg-export.o
 
 ifeq ($(CONFIG_X86_32),y)
         obj-y += atomic64_32.o
diff --git a/arch/x86/lib/msr-reg-export.c b/arch/x86/lib/msr-reg-export.c
new file mode 100644
index 000000000000..a311cc59b65d
--- /dev/null
+++ b/arch/x86/lib/msr-reg-export.c
@@ -0,0 +1,5 @@
+#include <linux/module.h>
+#include <asm/msr.h>
+
+EXPORT_SYMBOL(native_rdmsr_safe_regs);
+EXPORT_SYMBOL(native_wrmsr_safe_regs);
-- 
cgit v1.2.3


From 23b6c52cf558f2a1dc6c7010c601b3624eb4591f Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Sat, 5 Sep 2009 01:07:45 +0200
Subject: x86: Decrease the level of some NUMA messages to KERN_DEBUG

Some NUMA messages in srat_32.c are confusing to users,
because they seem to indicate errors, while in fact they
reflect normal behaviour.

Decrease the level of these messages to KERN_DEBUG so that
they don't show up unnecessarily.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
LKML-Reference: <200909050107.45175.rjw@sisk.pl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/srat_32.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/mm/srat_32.c b/arch/x86/mm/srat_32.c
index 29a0e37114f8..6f8aa33031c7 100644
--- a/arch/x86/mm/srat_32.c
+++ b/arch/x86/mm/srat_32.c
@@ -215,7 +215,7 @@ int __init get_memcfg_from_srat(void)
 		goto out_fail;
 
 	if (num_memory_chunks == 0) {
-		printk(KERN_WARNING
+		printk(KERN_DEBUG
 			 "could not find any ACPI SRAT memory areas.\n");
 		goto out_fail;
 	}
@@ -277,7 +277,7 @@ int __init get_memcfg_from_srat(void)
 	}
 	return 1;
 out_fail:
-	printk(KERN_ERR "failed to get NUMA memory information from SRAT"
+	printk(KERN_DEBUG "failed to get NUMA memory information from SRAT"
 			" table\n");
 	return 0;
 }
-- 
cgit v1.2.3


From d535e4319a2e598ba25dc966ada4e52ea774e33f Mon Sep 17 00:00:00 2001
From: Tobias Klauser <tklauser@distanz.ch>
Date: Fri, 4 Sep 2009 15:53:09 +0200
Subject: x86: Make memtype_seq_ops const

Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/pat.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch/x86')

diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c
index e6718bb28065..e2900a36de0e 100644
--- a/arch/x86/mm/pat.c
+++ b/arch/x86/mm/pat.c
@@ -826,7 +826,7 @@ static int memtype_seq_show(struct seq_file *seq, void *v)
 	return 0;
 }
 
-static struct seq_operations memtype_seq_ops = {
+static const struct seq_operations memtype_seq_ops = {
 	.start = memtype_seq_start,
 	.next  = memtype_seq_next,
 	.stop  = memtype_seq_stop,
-- 
cgit v1.2.3


From a8fae3ec5f118dc92517dcbed3ecf69ddb641d0f Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 7 Sep 2009 18:32:32 +0200
Subject: sched: enable SD_WAKE_IDLE

Now that SD_WAKE_IDLE doesn't make pipe-test suck anymore,
enable it by default for MC, CPU and NUMA domains.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/include/asm/topology.h | 2 +-
 include/linux/topology.h        | 6 +++---
 2 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/topology.h b/arch/x86/include/asm/topology.h
index ef7bc7fc2528..26d06e052a18 100644
--- a/arch/x86/include/asm/topology.h
+++ b/arch/x86/include/asm/topology.h
@@ -152,7 +152,7 @@ extern unsigned long node_remap_size[];
 				| 0*SD_POWERSAVINGS_BALANCE		\
 				| 0*SD_SHARE_PKG_RESOURCES		\
 				| 1*SD_SERIALIZE			\
-				| 0*SD_WAKE_IDLE_FAR			\
+				| 1*SD_WAKE_IDLE_FAR			\
 				| 0*SD_PREFER_SIBLING			\
 				,					\
 	.last_balance		= jiffies,				\
diff --git a/include/linux/topology.h b/include/linux/topology.h
index 66774fddec9b..85e8cf7d393c 100644
--- a/include/linux/topology.h
+++ b/include/linux/topology.h
@@ -129,7 +129,7 @@ int arch_update_cpu_topology(void);
 				| 1*SD_BALANCE_NEWIDLE			\
 				| 1*SD_BALANCE_EXEC			\
 				| 1*SD_BALANCE_FORK			\
-				| 0*SD_WAKE_IDLE			\
+				| 1*SD_WAKE_IDLE			\
 				| 1*SD_WAKE_AFFINE			\
 				| 1*SD_WAKE_BALANCE			\
 				| 0*SD_SHARE_CPUPOWER			\
@@ -163,7 +163,7 @@ int arch_update_cpu_topology(void);
 				| 1*SD_BALANCE_NEWIDLE			\
 				| 1*SD_BALANCE_EXEC			\
 				| 1*SD_BALANCE_FORK			\
-				| 0*SD_WAKE_IDLE			\
+				| 1*SD_WAKE_IDLE			\
 				| 0*SD_WAKE_AFFINE			\
 				| 1*SD_WAKE_BALANCE			\
 				| 0*SD_SHARE_CPUPOWER			\
@@ -198,7 +198,7 @@ int arch_update_cpu_topology(void);
 				| 0*SD_POWERSAVINGS_BALANCE		\
 				| 0*SD_SHARE_PKG_RESOURCES		\
 				| 1*SD_SERIALIZE			\
-				| 0*SD_WAKE_IDLE_FAR			\
+				| 1*SD_WAKE_IDLE_FAR			\
 				| 0*SD_PREFER_SIBLING			\
 				,					\
 	.last_balance		= jiffies,				\
-- 
cgit v1.2.3


From 3e5cd1f2576c720f1d0705fdd7ba64f27e8836b7 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Sun, 16 Aug 2009 21:02:36 +0900
Subject: dmi: extend dmi_get_year() to dmi_get_date()

There are cases where full date information is required instead of
just the year.  Add month and day parsing to dmi_get_year() and rename
it to dmi_get_date().

As the original function only required '/' followed by any number of
parseable characters at the end of the string, keep that behavior to
avoid upsetting existing users.

The new function takes dates of format [mm[/dd]]/yy[yy].  Year, month
and date are checked to be in the ranges of [1-9999], [1-12] and
[1-31] respectively and any invalid or out-of-range component is
returned as zero.

The dummy implementation is updated accordingly but the return value
is updated to indicate field not found which is consistent with how
other dummy functions behave.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 arch/x86/pci/direct.c       |  5 +--
 drivers/acpi/blacklist.c    |  5 +--
 drivers/ata/ahci.c          |  2 +-
 drivers/firmware/dmi_scan.c | 76 ++++++++++++++++++++++++++++++++++-----------
 include/linux/dmi.h         | 13 ++++++--
 5 files changed, 76 insertions(+), 25 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/pci/direct.c b/arch/x86/pci/direct.c
index bd13c3e4c6db..347d882b3bb3 100644
--- a/arch/x86/pci/direct.c
+++ b/arch/x86/pci/direct.c
@@ -192,13 +192,14 @@ struct pci_raw_ops pci_direct_conf2 = {
 static int __init pci_sanity_check(struct pci_raw_ops *o)
 {
 	u32 x = 0;
-	int devfn;
+	int year, devfn;
 
 	if (pci_probe & PCI_NO_CHECKS)
 		return 1;
 	/* Assume Type 1 works for newer systems.
 	   This handles machines that don't have anything on PCI Bus 0. */
-	if (dmi_get_year(DMI_BIOS_DATE) >= 2001)
+	dmi_get_date(DMI_BIOS_DATE, &year, NULL, NULL);
+	if (year >= 2001)
 		return 1;
 
 	for (devfn = 0; devfn < 0x100; devfn++) {
diff --git a/drivers/acpi/blacklist.c b/drivers/acpi/blacklist.c
index f6baa77deefb..0c4ca4d318b3 100644
--- a/drivers/acpi/blacklist.c
+++ b/drivers/acpi/blacklist.c
@@ -78,9 +78,10 @@ static struct acpi_blacklist_item acpi_blacklist[] __initdata = {
 
 static int __init blacklist_by_year(void)
 {
-	int year = dmi_get_year(DMI_BIOS_DATE);
+	int year;
+
 	/* Doesn't exist? Likely an old system */
-	if (year == -1) {
+	if (!dmi_get_date(DMI_BIOS_DATE, &year, NULL, NULL)) {
 		printk(KERN_ERR PREFIX "no DMI BIOS year, "
 			"acpi=force is required to enable ACPI\n" );
 		return 1;
diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index be4c39f8ab81..147b9be3b4d2 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -2679,7 +2679,7 @@ static bool ahci_asus_m2a_vm_32bit_only(struct pci_dev *pdev)
 	 * different versions.
 	 */
 	date = dmi_get_system_info(DMI_BIOS_DATE);
-	year = dmi_get_year(DMI_BIOS_DATE);
+	dmi_get_date(DMI_BIOS_DATE, &year, NULL, NULL);
 	if (date && strlen(date) >= 10 && date[2] == '/' && date[5] == '/' &&
 	    (year > 2007 ||
 	     (year == 2007 && strncmp(date, cutoff_mmdd, 5) >= 0)))
diff --git a/drivers/firmware/dmi_scan.c b/drivers/firmware/dmi_scan.c
index 531e621677ce..938100f14b16 100644
--- a/drivers/firmware/dmi_scan.c
+++ b/drivers/firmware/dmi_scan.c
@@ -568,36 +568,76 @@ const struct dmi_device * dmi_find_device(int type, const char *name,
 EXPORT_SYMBOL(dmi_find_device);
 
 /**
- *	dmi_get_year - Return year of a DMI date
- *	@field:	data index (like dmi_get_system_info)
+ *	dmi_get_date - parse a DMI date
+ *	@field:	data index (see enum dmi_field)
+ *	@yearp: optional out parameter for the year
+ *	@monthp: optional out parameter for the month
+ *	@dayp: optional out parameter for the day
  *
- *	Returns -1 when the field doesn't exist. 0 when it is broken.
+ *	The date field is assumed to be in the form resembling
+ *	[mm[/dd]]/yy[yy] and the result is stored in the out
+ *	parameters any or all of which can be omitted.
+ *
+ *	If the field doesn't exist, all out parameters are set to zero
+ *	and false is returned.  Otherwise, true is returned with any
+ *	invalid part of date set to zero.
+ *
+ *	On return, year, month and day are guaranteed to be in the
+ *	range of [0,9999], [0,12] and [0,31] respectively.
  */
-int dmi_get_year(int field)
+bool dmi_get_date(int field, int *yearp, int *monthp, int *dayp)
 {
-	int year;
-	const char *s = dmi_get_system_info(field);
+	int year = 0, month = 0, day = 0;
+	bool exists;
+	const char *s, *y;
 	char *e;
 
-	if (!s)
-		return -1;
-	if (*s == '\0')
-		return 0;
-	s = strrchr(s, '/');
-	if (!s)
-		return 0;
+	s = dmi_get_system_info(field);
+	exists = s;
+	if (!exists)
+		goto out;
 
-	s += 1;
-	year = simple_strtoul(s, &e, 10);
-	if (s != e && year < 100) {	/* 2-digit year */
+	/*
+	 * Determine year first.  We assume the date string resembles
+	 * mm/dd/yy[yy] but the original code extracted only the year
+	 * from the end.  Keep the behavior in the spirit of no
+	 * surprises.
+	 */
+	y = strrchr(s, '/');
+	if (!y)
+		goto out;
+
+	y++;
+	year = simple_strtoul(y, &e, 10);
+	if (y != e && year < 100) {	/* 2-digit year */
 		year += 1900;
 		if (year < 1996)	/* no dates < spec 1.0 */
 			year += 100;
 	}
+	if (year > 9999)		/* year should fit in %04d */
+		year = 0;
+
+	/* parse the mm and dd */
+	month = simple_strtoul(s, &e, 10);
+	if (s == e || *e != '/' || !month || month > 12) {
+		month = 0;
+		goto out;
+	}
 
-	return year;
+	s = e + 1;
+	day = simple_strtoul(s, &e, 10);
+	if (s == y || s == e || *e != '/' || day > 31)
+		day = 0;
+out:
+	if (yearp)
+		*yearp = year;
+	if (monthp)
+		*monthp = month;
+	if (dayp)
+		*dayp = day;
+	return exists;
 }
-EXPORT_SYMBOL(dmi_get_year);
+EXPORT_SYMBOL(dmi_get_date);
 
 /**
  *	dmi_walk - Walk the DMI table and get called back for every record
diff --git a/include/linux/dmi.h b/include/linux/dmi.h
index bb5489c82c99..a8a3e1ac281d 100644
--- a/include/linux/dmi.h
+++ b/include/linux/dmi.h
@@ -43,7 +43,7 @@ extern const char * dmi_get_system_info(int field);
 extern const struct dmi_device * dmi_find_device(int type, const char *name,
 	const struct dmi_device *from);
 extern void dmi_scan_machine(void);
-extern int dmi_get_year(int field);
+extern bool dmi_get_date(int field, int *yearp, int *monthp, int *dayp);
 extern int dmi_name_in_vendors(const char *str);
 extern int dmi_name_in_serial(const char *str);
 extern int dmi_available;
@@ -58,7 +58,16 @@ static inline const char * dmi_get_system_info(int field) { return NULL; }
 static inline const struct dmi_device * dmi_find_device(int type, const char *name,
 	const struct dmi_device *from) { return NULL; }
 static inline void dmi_scan_machine(void) { return; }
-static inline int dmi_get_year(int year) { return 0; }
+static inline bool dmi_get_date(int field, int *yearp, int *monthp, int *dayp)
+{
+	if (yearp)
+		*yearp = 0;
+	if (monthp)
+		*monthp = 0;
+	if (dayp)
+		*dayp = 0;
+	return false;
+}
 static inline int dmi_name_in_vendors(const char *s) { return 0; }
 static inline int dmi_name_in_serial(const char *s) { return 0; }
 #define dmi_available 0
-- 
cgit v1.2.3


From 577eebeae34d340685d8985dfdb7dfe337c511e8 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Date: Thu, 27 Aug 2009 12:46:35 -0700
Subject: xen: make -fstack-protector work under Xen

-fstack-protector uses a special per-cpu "stack canary" value.
gcc generates special code in each function to test the canary to make
sure that the function's stack hasn't been overrun.

On x86-64, this is simply an offset of %gs, which is the usual per-cpu
base segment register, so setting it up simply requires loading %gs's
base as normal.

On i386, the stack protector segment is %gs (rather than the usual kernel
percpu %fs segment register).  This requires setting up the full kernel
GDT and then loading %gs accordingly.  We also need to make sure %gs is
initialized when bringing up secondary cpus too.

To keep things consistent, we do the full GDT/segment register setup on
both architectures.

Because we need to avoid -fstack-protected code before setting up the GDT
and because there's no way to disable it on a per-function basis, several
files need to have stack-protector inhibited.

[ Impact: allow Xen booting with stack-protector enabled ]

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
---
 arch/x86/mm/Makefile     |   4 ++
 arch/x86/xen/Makefile    |   2 +
 arch/x86/xen/enlighten.c | 131 ++++++++++++++++++++++++++++++++++++++++-------
 arch/x86/xen/smp.c       |   1 +
 drivers/xen/Makefile     |   3 ++
 5 files changed, 122 insertions(+), 19 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile
index eefdeee8a871..72bb3a26c738 100644
--- a/arch/x86/mm/Makefile
+++ b/arch/x86/mm/Makefile
@@ -1,6 +1,10 @@
 obj-y	:=  init.o init_$(BITS).o fault.o ioremap.o extable.o pageattr.o mmap.o \
 	    pat.o pgtable.o gup.o
 
+# Make sure __phys_addr has no stackprotector
+nostackp := $(call cc-option, -fno-stack-protector)
+CFLAGS_ioremap.o		:= $(nostackp)
+
 obj-$(CONFIG_SMP)		+= tlb.o
 
 obj-$(CONFIG_X86_32)		+= pgtable_32.o iomap_32.o
diff --git a/arch/x86/xen/Makefile b/arch/x86/xen/Makefile
index 7410640db173..3bb4fc21f4f2 100644
--- a/arch/x86/xen/Makefile
+++ b/arch/x86/xen/Makefile
@@ -8,6 +8,7 @@ endif
 # Make sure early boot has no stackprotector
 nostackp := $(call cc-option, -fno-stack-protector)
 CFLAGS_enlighten.o		:= $(nostackp)
+CFLAGS_mmu.o			:= $(nostackp)
 
 obj-y		:= enlighten.o setup.o multicalls.o mmu.o irq.o \
 			time.o xen-asm.o xen-asm_$(BITS).o \
@@ -16,3 +17,4 @@ obj-y		:= enlighten.o setup.o multicalls.o mmu.o irq.o \
 obj-$(CONFIG_SMP)		+= smp.o
 obj-$(CONFIG_PARAVIRT_SPINLOCKS)+= spinlock.o
 obj-$(CONFIG_XEN_DEBUG_FS)	+= debugfs.o
+
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index eb33aaa8415d..76143136dead 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -51,6 +51,7 @@
 #include <asm/pgtable.h>
 #include <asm/tlbflush.h>
 #include <asm/reboot.h>
+#include <asm/stackprotector.h>
 
 #include "xen-ops.h"
 #include "mmu.h"
@@ -330,18 +331,28 @@ static void xen_load_gdt(const struct desc_ptr *dtr)
 	unsigned long frames[pages];
 	int f;
 
-	/* A GDT can be up to 64k in size, which corresponds to 8192
-	   8-byte entries, or 16 4k pages.. */
+	/*
+	 * A GDT can be up to 64k in size, which corresponds to 8192
+	 * 8-byte entries, or 16 4k pages..
+	 */
 
 	BUG_ON(size > 65536);
 	BUG_ON(va & ~PAGE_MASK);
 
 	for (f = 0; va < dtr->address + size; va += PAGE_SIZE, f++) {
 		int level;
-		pte_t *ptep = lookup_address(va, &level);
+		pte_t *ptep;
 		unsigned long pfn, mfn;
 		void *virt;
 
+		/*
+		 * The GDT is per-cpu and is in the percpu data area.
+		 * That can be virtually mapped, so we need to do a
+		 * page-walk to get the underlying MFN for the
+		 * hypercall.  The page can also be in the kernel's
+		 * linear range, so we need to RO that mapping too.
+		 */
+		ptep = lookup_address(va, &level);
 		BUG_ON(ptep == NULL);
 
 		pfn = pte_pfn(*ptep);
@@ -358,6 +369,44 @@ static void xen_load_gdt(const struct desc_ptr *dtr)
 		BUG();
 }
 
+/*
+ * load_gdt for early boot, when the gdt is only mapped once
+ */
+static __init void xen_load_gdt_boot(const struct desc_ptr *dtr)
+{
+	unsigned long va = dtr->address;
+	unsigned int size = dtr->size + 1;
+	unsigned pages = (size + PAGE_SIZE - 1) / PAGE_SIZE;
+	unsigned long frames[pages];
+	int f;
+
+	/*
+	 * A GDT can be up to 64k in size, which corresponds to 8192
+	 * 8-byte entries, or 16 4k pages..
+	 */
+
+	BUG_ON(size > 65536);
+	BUG_ON(va & ~PAGE_MASK);
+
+	for (f = 0; va < dtr->address + size; va += PAGE_SIZE, f++) {
+		pte_t pte;
+		unsigned long pfn, mfn;
+
+		pfn = virt_to_pfn(va);
+		mfn = pfn_to_mfn(pfn);
+
+		pte = pfn_pte(pfn, PAGE_KERNEL_RO);
+
+		if (HYPERVISOR_update_va_mapping((unsigned long)va, pte, 0))
+			BUG();
+
+		frames[f] = mfn;
+	}
+
+	if (HYPERVISOR_set_gdt(frames, size / sizeof(struct desc_struct)))
+		BUG();
+}
+
 static void load_TLS_descriptor(struct thread_struct *t,
 				unsigned int cpu, unsigned int i)
 {
@@ -581,6 +630,29 @@ static void xen_write_gdt_entry(struct desc_struct *dt, int entry,
 	preempt_enable();
 }
 
+/*
+ * Version of write_gdt_entry for use at early boot-time needed to
+ * update an entry as simply as possible.
+ */
+static __init void xen_write_gdt_entry_boot(struct desc_struct *dt, int entry,
+					    const void *desc, int type)
+{
+	switch (type) {
+	case DESC_LDT:
+	case DESC_TSS:
+		/* ignore */
+		break;
+
+	default: {
+		xmaddr_t maddr = virt_to_machine(&dt[entry]);
+
+		if (HYPERVISOR_update_descriptor(maddr.maddr, *(u64 *)desc))
+			dt[entry] = *(struct desc_struct *)desc;
+	}
+
+	}
+}
+
 static void xen_load_sp0(struct tss_struct *tss,
 			 struct thread_struct *thread)
 {
@@ -965,6 +1037,23 @@ static const struct machine_ops __initdata xen_machine_ops = {
 	.emergency_restart = xen_emergency_restart,
 };
 
+/*
+ * Set up the GDT and segment registers for -fstack-protector.  Until
+ * we do this, we have to be careful not to call any stack-protected
+ * function, which is most of the kernel.
+ */
+static void __init xen_setup_stackprotector(void)
+{
+	pv_cpu_ops.write_gdt_entry = xen_write_gdt_entry_boot;
+	pv_cpu_ops.load_gdt = xen_load_gdt_boot;
+
+	setup_stack_canary_segment(0);
+	switch_to_new_gdt(0);
+
+	pv_cpu_ops.write_gdt_entry = xen_write_gdt_entry;
+	pv_cpu_ops.load_gdt = xen_load_gdt;
+}
+
 /* First C function to be called on Xen boot */
 asmlinkage void __init xen_start_kernel(void)
 {
@@ -983,13 +1072,28 @@ asmlinkage void __init xen_start_kernel(void)
 	pv_apic_ops = xen_apic_ops;
 	pv_mmu_ops = xen_mmu_ops;
 
-#ifdef CONFIG_X86_64
 	/*
-	 * Setup percpu state.  We only need to do this for 64-bit
-	 * because 32-bit already has %fs set properly.
+	 * Set up some pagetable state before starting to set any ptes.
 	 */
-	load_percpu_segment(0);
-#endif
+
+	/* Prevent unwanted bits from being set in PTEs. */
+	__supported_pte_mask &= ~_PAGE_GLOBAL;
+	if (!xen_initial_domain())
+		__supported_pte_mask &= ~(_PAGE_PWT | _PAGE_PCD);
+
+	__supported_pte_mask |= _PAGE_IOMAP;
+
+	xen_setup_features();
+
+	/* Get mfn list */
+	if (!xen_feature(XENFEAT_auto_translated_physmap))
+		xen_build_dynamic_phys_to_machine();
+
+	/*
+	 * Set up kernel GDT and segment registers, mainly so that
+	 * -fstack-protector code can be executed.
+	 */
+	xen_setup_stackprotector();
 
 	xen_init_irq_ops();
 	xen_init_cpuid_mask();
@@ -1001,8 +1105,6 @@ asmlinkage void __init xen_start_kernel(void)
 	set_xen_basic_apic_ops();
 #endif
 
-	xen_setup_features();
-
 	if (xen_feature(XENFEAT_mmu_pt_update_preserve_ad)) {
 		pv_mmu_ops.ptep_modify_prot_start = xen_ptep_modify_prot_start;
 		pv_mmu_ops.ptep_modify_prot_commit = xen_ptep_modify_prot_commit;
@@ -1019,17 +1121,8 @@ asmlinkage void __init xen_start_kernel(void)
 
 	xen_smp_init();
 
-	/* Get mfn list */
-	if (!xen_feature(XENFEAT_auto_translated_physmap))
-		xen_build_dynamic_phys_to_machine();
-
 	pgd = (pgd_t *)xen_start_info->pt_base;
 
-	/* Prevent unwanted bits from being set in PTEs. */
-	__supported_pte_mask &= ~_PAGE_GLOBAL;
-	if (!xen_initial_domain())
-		__supported_pte_mask &= ~(_PAGE_PWT | _PAGE_PCD);
-
 #ifdef CONFIG_X86_64
 	/* Work out if we support NX */
 	check_efer();
diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c
index 429834ec1687..fe03eeed7b48 100644
--- a/arch/x86/xen/smp.c
+++ b/arch/x86/xen/smp.c
@@ -236,6 +236,7 @@ cpu_initialize_context(unsigned int cpu, struct task_struct *idle)
 	ctxt->user_regs.ss = __KERNEL_DS;
 #ifdef CONFIG_X86_32
 	ctxt->user_regs.fs = __KERNEL_PERCPU;
+	ctxt->user_regs.gs = __KERNEL_STACK_CANARY;
 #else
 	ctxt->gs_base_kernel = per_cpu_offset(cpu);
 #endif
diff --git a/drivers/xen/Makefile b/drivers/xen/Makefile
index ec2a39b1e26f..7c284342f30f 100644
--- a/drivers/xen/Makefile
+++ b/drivers/xen/Makefile
@@ -1,6 +1,9 @@
 obj-y	+= grant-table.o features.o events.o manage.o
 obj-y	+= xenbus/
 
+nostackp := $(call cc-option, -fno-stack-protector)
+CFLAGS_features.o			:= $(nostackp)
+
 obj-$(CONFIG_HOTPLUG_CPU)	+= cpu_hotplug.o
 obj-$(CONFIG_XEN_XENCOMM)	+= xencomm.o
 obj-$(CONFIG_XEN_BALLOON)	+= balloon.o
-- 
cgit v1.2.3


From 4d576b57b50a92801e6493e76e5243d6cff193d2 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Date: Wed, 9 Sep 2009 12:33:51 -0700
Subject: xen: only enable interrupts while actually blocking for spinlock

Where possible we enable interrupts while waiting for a spinlock to
become free, in order to reduce big latency spikes in interrupt handling.

However, at present if we manage to pick up the spinlock just before
blocking, we'll end up holding the lock with interrupts enabled for a
while.  This will cause a deadlock if we recieve an interrupt in that
window, and the interrupt handler tries to take the lock too.

Solve this by shrinking the interrupt-enabled region to just around the
blocking call.

[ Impact: avoid race/deadlock when using Xen PV spinlocks ]

Reported-by: "Yang, Xiaowei" <xiaowei.yang@intel.com>
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
---
 arch/x86/xen/spinlock.c | 19 +++++++++++--------
 1 file changed, 11 insertions(+), 8 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/xen/spinlock.c b/arch/x86/xen/spinlock.c
index 5601506f2dd9..2f91e5651926 100644
--- a/arch/x86/xen/spinlock.c
+++ b/arch/x86/xen/spinlock.c
@@ -187,7 +187,6 @@ static noinline int xen_spin_lock_slow(struct raw_spinlock *lock, bool irq_enabl
 	struct xen_spinlock *prev;
 	int irq = __get_cpu_var(lock_kicker_irq);
 	int ret;
-	unsigned long flags;
 	u64 start;
 
 	/* If kicker interrupts not initialized yet, just spin */
@@ -199,16 +198,12 @@ static noinline int xen_spin_lock_slow(struct raw_spinlock *lock, bool irq_enabl
 	/* announce we're spinning */
 	prev = spinning_lock(xl);
 
-	flags = __raw_local_save_flags();
-	if (irq_enable) {
-		ADD_STATS(taken_slow_irqenable, 1);
-		raw_local_irq_enable();
-	}
-
 	ADD_STATS(taken_slow, 1);
 	ADD_STATS(taken_slow_nested, prev != NULL);
 
 	do {
+		unsigned long flags;
+
 		/* clear pending */
 		xen_clear_irq_pending(irq);
 
@@ -228,6 +223,12 @@ static noinline int xen_spin_lock_slow(struct raw_spinlock *lock, bool irq_enabl
 			goto out;
 		}
 
+		flags = __raw_local_save_flags();
+		if (irq_enable) {
+			ADD_STATS(taken_slow_irqenable, 1);
+			raw_local_irq_enable();
+		}
+
 		/*
 		 * Block until irq becomes pending.  If we're
 		 * interrupted at this point (after the trylock but
@@ -238,13 +239,15 @@ static noinline int xen_spin_lock_slow(struct raw_spinlock *lock, bool irq_enabl
 		 * pending.
 		 */
 		xen_poll_irq(irq);
+
+		raw_local_irq_restore(flags);
+
 		ADD_STATS(taken_slow_spurious, !xen_test_irq_pending(irq));
 	} while (!xen_test_irq_pending(irq)); /* check for spurious wakeups */
 
 	kstat_incr_irqs_this_cpu(irq, irq_to_desc(irq));
 
 out:
-	raw_local_irq_restore(flags);
 	unspinning_lock(xl, prev);
 	spin_time_accum_blocked(start);
 
-- 
cgit v1.2.3


From 2496afbf1e50c70f80992656bcb730c8583ddac3 Mon Sep 17 00:00:00 2001
From: Yang Xiaowei <xiaowei.yang@intel.com>
Date: Wed, 9 Sep 2009 12:44:52 -0700
Subject: xen: use stronger barrier after unlocking lock

We need to have a stronger barrier between releasing the lock and
checking for any waiting spinners.  A compiler barrier is not sufficient
because the CPU's ordering rules do not prevent the read xl->spinners
from happening before the unlock assignment, as they are different
memory locations.

We need to have an explicit barrier to enforce the write-read ordering
to different memory locations.

Because of it, I can't bring up > 4 HVM guests on one SMP machine.

[ Code and commit comments expanded -J ]

[ Impact: avoid deadlock when using Xen PV spinlocks ]

Signed-off-by: Yang Xiaowei <xiaowei.yang@intel.com>
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
---
 arch/x86/xen/spinlock.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/xen/spinlock.c b/arch/x86/xen/spinlock.c
index 2f91e5651926..36a5141108df 100644
--- a/arch/x86/xen/spinlock.c
+++ b/arch/x86/xen/spinlock.c
@@ -326,8 +326,13 @@ static void xen_spin_unlock(struct raw_spinlock *lock)
 	smp_wmb();		/* make sure no writes get moved after unlock */
 	xl->lock = 0;		/* release lock */
 
-	/* make sure unlock happens before kick */
-	barrier();
+	/*
+	 * Make sure unlock happens before checking for waiting
+	 * spinners.  We need a strong barrier to enforce the
+	 * write-read ordering to different memory locations, as the
+	 * CPU makes no implied guarantees about their ordering.
+	 */
+	mb();
 
 	if (unlikely(xl->spinners))
 		xen_spin_unlock_slow(xl);
-- 
cgit v1.2.3


From 78c86e5e5691fc84d5fbea0cd4ac7147e87b7490 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Date: Thu, 10 Sep 2009 10:09:38 -0700
Subject: x86: split __phys_addr out into separate file

Split __phys_addr out into its own file so we can disable
-fstack-protector in a fine-grained fashion.  Also it doesn't
have terribly much to do with the rest of ioremap.c.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
---
 arch/x86/mm/Makefile   |  4 +--
 arch/x86/mm/ioremap.c  | 72 +-------------------------------------------------
 arch/x86/mm/physaddr.c | 70 ++++++++++++++++++++++++++++++++++++++++++++++++
 arch/x86/mm/physaddr.h | 10 +++++++
 4 files changed, 83 insertions(+), 73 deletions(-)
 create mode 100644 arch/x86/mm/physaddr.c
 create mode 100644 arch/x86/mm/physaddr.h

(limited to 'arch/x86')

diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile
index 72bb3a26c738..9b5a9f59a478 100644
--- a/arch/x86/mm/Makefile
+++ b/arch/x86/mm/Makefile
@@ -1,9 +1,9 @@
 obj-y	:=  init.o init_$(BITS).o fault.o ioremap.o extable.o pageattr.o mmap.o \
-	    pat.o pgtable.o gup.o
+	    pat.o pgtable.o physaddr.o gup.o
 
 # Make sure __phys_addr has no stackprotector
 nostackp := $(call cc-option, -fno-stack-protector)
-CFLAGS_ioremap.o		:= $(nostackp)
+CFLAGS_physaddr.o		:= $(nostackp)
 
 obj-$(CONFIG_SMP)		+= tlb.o
 
diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c
index 8a450930834f..04e1ad60c63a 100644
--- a/arch/x86/mm/ioremap.c
+++ b/arch/x86/mm/ioremap.c
@@ -22,77 +22,7 @@
 #include <asm/pgalloc.h>
 #include <asm/pat.h>
 
-static inline int phys_addr_valid(resource_size_t addr)
-{
-#ifdef CONFIG_PHYS_ADDR_T_64BIT
-	return !(addr >> boot_cpu_data.x86_phys_bits);
-#else
-	return 1;
-#endif
-}
-
-#ifdef CONFIG_X86_64
-
-unsigned long __phys_addr(unsigned long x)
-{
-	if (x >= __START_KERNEL_map) {
-		x -= __START_KERNEL_map;
-		VIRTUAL_BUG_ON(x >= KERNEL_IMAGE_SIZE);
-		x += phys_base;
-	} else {
-		VIRTUAL_BUG_ON(x < PAGE_OFFSET);
-		x -= PAGE_OFFSET;
-		VIRTUAL_BUG_ON(!phys_addr_valid(x));
-	}
-	return x;
-}
-EXPORT_SYMBOL(__phys_addr);
-
-bool __virt_addr_valid(unsigned long x)
-{
-	if (x >= __START_KERNEL_map) {
-		x -= __START_KERNEL_map;
-		if (x >= KERNEL_IMAGE_SIZE)
-			return false;
-		x += phys_base;
-	} else {
-		if (x < PAGE_OFFSET)
-			return false;
-		x -= PAGE_OFFSET;
-		if (!phys_addr_valid(x))
-			return false;
-	}
-
-	return pfn_valid(x >> PAGE_SHIFT);
-}
-EXPORT_SYMBOL(__virt_addr_valid);
-
-#else
-
-#ifdef CONFIG_DEBUG_VIRTUAL
-unsigned long __phys_addr(unsigned long x)
-{
-	/* VMALLOC_* aren't constants  */
-	VIRTUAL_BUG_ON(x < PAGE_OFFSET);
-	VIRTUAL_BUG_ON(__vmalloc_start_set && is_vmalloc_addr((void *) x));
-	return x - PAGE_OFFSET;
-}
-EXPORT_SYMBOL(__phys_addr);
-#endif
-
-bool __virt_addr_valid(unsigned long x)
-{
-	if (x < PAGE_OFFSET)
-		return false;
-	if (__vmalloc_start_set && is_vmalloc_addr((void *) x))
-		return false;
-	if (x >= FIXADDR_START)
-		return false;
-	return pfn_valid((x - PAGE_OFFSET) >> PAGE_SHIFT);
-}
-EXPORT_SYMBOL(__virt_addr_valid);
-
-#endif
+#include "physaddr.h"
 
 int page_is_ram(unsigned long pagenr)
 {
diff --git a/arch/x86/mm/physaddr.c b/arch/x86/mm/physaddr.c
new file mode 100644
index 000000000000..d2e2735327b4
--- /dev/null
+++ b/arch/x86/mm/physaddr.c
@@ -0,0 +1,70 @@
+#include <linux/mmdebug.h>
+#include <linux/module.h>
+#include <linux/mm.h>
+
+#include <asm/page.h>
+
+#include "physaddr.h"
+
+#ifdef CONFIG_X86_64
+
+unsigned long __phys_addr(unsigned long x)
+{
+	if (x >= __START_KERNEL_map) {
+		x -= __START_KERNEL_map;
+		VIRTUAL_BUG_ON(x >= KERNEL_IMAGE_SIZE);
+		x += phys_base;
+	} else {
+		VIRTUAL_BUG_ON(x < PAGE_OFFSET);
+		x -= PAGE_OFFSET;
+		VIRTUAL_BUG_ON(!phys_addr_valid(x));
+	}
+	return x;
+}
+EXPORT_SYMBOL(__phys_addr);
+
+bool __virt_addr_valid(unsigned long x)
+{
+	if (x >= __START_KERNEL_map) {
+		x -= __START_KERNEL_map;
+		if (x >= KERNEL_IMAGE_SIZE)
+			return false;
+		x += phys_base;
+	} else {
+		if (x < PAGE_OFFSET)
+			return false;
+		x -= PAGE_OFFSET;
+		if (!phys_addr_valid(x))
+			return false;
+	}
+
+	return pfn_valid(x >> PAGE_SHIFT);
+}
+EXPORT_SYMBOL(__virt_addr_valid);
+
+#else
+
+#ifdef CONFIG_DEBUG_VIRTUAL
+unsigned long __phys_addr(unsigned long x)
+{
+	/* VMALLOC_* aren't constants  */
+	VIRTUAL_BUG_ON(x < PAGE_OFFSET);
+	VIRTUAL_BUG_ON(__vmalloc_start_set && is_vmalloc_addr((void *) x));
+	return x - PAGE_OFFSET;
+}
+EXPORT_SYMBOL(__phys_addr);
+#endif
+
+bool __virt_addr_valid(unsigned long x)
+{
+	if (x < PAGE_OFFSET)
+		return false;
+	if (__vmalloc_start_set && is_vmalloc_addr((void *) x))
+		return false;
+	if (x >= FIXADDR_START)
+		return false;
+	return pfn_valid((x - PAGE_OFFSET) >> PAGE_SHIFT);
+}
+EXPORT_SYMBOL(__virt_addr_valid);
+
+#endif	/* CONFIG_X86_64 */
diff --git a/arch/x86/mm/physaddr.h b/arch/x86/mm/physaddr.h
new file mode 100644
index 000000000000..a3cd5a0c97b3
--- /dev/null
+++ b/arch/x86/mm/physaddr.h
@@ -0,0 +1,10 @@
+#include <asm/processor.h>
+
+static inline int phys_addr_valid(resource_size_t addr)
+{
+#ifdef CONFIG_PHYS_ADDR_T_64BIT
+	return !(addr >> boot_cpu_data.x86_phys_bits);
+#else
+	return 1;
+#endif
+}
-- 
cgit v1.2.3


From 5367b6887e7d8c870a5da7d9b8c6e9c207684e43 Mon Sep 17 00:00:00 2001
From: Ben Hutchings <ben@decadent.org.uk>
Date: Thu, 10 Sep 2009 02:53:50 +0100
Subject: x86: Fix code patching for paravirt-alternatives on 486

As reported in <http://bugs.debian.org/511703> and
<http://bugs.debian.org/515982>, kernels with paravirt-alternatives
enabled crash in text_poke_early() on at least some 486-class
processors.

The problem is that text_poke_early() itself uses inline functions
affected by paravirt-alternatives and so will modify instructions that
have already been prefetched.  Pentium and later processors will
invalidate the prefetched instructions in this case, but 486-class
processors do not.

Change sync_core() to limit prefetching on 486-class (and 386-class)
processors, and move the call to sync_core() above the call to the
modifiable local_irq_restore().

Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
LKML-Reference: <1252547631.3423.134.camel@localhost>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/include/asm/processor.h | 16 +++++++++++++---
 arch/x86/kernel/alternative.c    |  2 +-
 2 files changed, 14 insertions(+), 4 deletions(-)

(limited to 'arch/x86')

diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h
index c7768269b1cf..2db56c57a281 100644
--- a/arch/x86/include/asm/processor.h
+++ b/arch/x86/include/asm/processor.h
@@ -703,13 +703,23 @@ static inline void cpu_relax(void)
 	rep_nop();
 }
 
-/* Stop speculative execution: */
+/* Stop speculative execution and prefetching of modified code. */
 static inline void sync_core(void)
 {
 	int tmp;
 
-	asm volatile("cpuid" : "=a" (tmp) : "0" (1)
-		     : "ebx", "ecx", "edx", "memory");
+#if defined(CONFIG_M386) || defined(CONFIG_M486)
+	if (boot_cpu_data.x86 < 5)
+		/* There is no speculative execution.
+		 * jmp is a barrier to prefetching. */
+		asm volatile("jmp 1f\n1:\n" ::: "memory");
+	else
+#endif
+		/* cpuid is a barrier to speculative execution.
+		 * Prefetched instructions are automatically
+		 * invalidated when modified. */
+		asm volatile("cpuid" : "=a" (tmp) : "0" (1)
+			     : "ebx", "ecx", "edx", "memory");
 }
 
 static inline void __monitor(const void *eax, unsigned long ecx,
diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c
index f57658702571..b8ebd0b689b1 100644
--- a/arch/x86/kernel/alternative.c
+++ b/arch/x86/kernel/alternative.c
@@ -490,8 +490,8 @@ void *text_poke_early(void *addr, const void *opcode, size_t len)
 	unsigned long flags;
 	local_irq_save(flags);
 	memcpy(addr, opcode, len);
-	local_irq_restore(flags);
 	sync_core();
+	local_irq_restore(flags);
 	/* Could also do a CLFLUSH here to speed up CPU recovery; but
 	   that causes hangs on some VIA CPUs. */
 	return addr;
-- 
cgit v1.2.3