Merge branch 'perf'

Signed-off-by: Avi Kivity <avi@redhat.com>

Documentation/kprobes.txt

@@ -165,8 +165,8 @@ the user entry_handler invocation is also skipped.
 
 1.4 How Does Jump Optimization Work?
 
-If you configured your kernel with CONFIG_OPTPROBES=y (currently
-this option is supported on x86/x86-64, non-preemptive kernel) and
+If your kernel is built with CONFIG_OPTPROBES=y (currently this flag
+is automatically set 'y' on x86/x86-64, non-preemptive kernel) and
 the "debug.kprobes_optimization" kernel parameter is set to 1 (see
 sysctl(8)), Kprobes tries to reduce probe-hit overhead by using a jump
 instruction instead of a breakpoint instruction at each probepoint.
@@ -271,8 +271,6 @@ tweak the kernel's execution path, you need to suppress optimization,
 using one of the following techniques:
 - Specify an empty function for the kprobe's post_handler or break_handler.
  or
-- Config CONFIG_OPTPROBES=n.
- or
 - Execute 'sysctl -w debug.kprobes_optimization=n'
 
 2. Architectures Supported
@@ -307,10 +305,6 @@ it useful to "Compile the kernel with debug info" (CONFIG_DEBUG_INFO),
 so you can use "objdump -d -l vmlinux" to see the source-to-object
 code mapping.
 
-If you want to reduce probing overhead, set "Kprobes jump optimization
-support" (CONFIG_OPTPROBES) to "y". You can find this option under the
-"Kprobes" line.
-
 4. API Reference
 
 The Kprobes API includes a "register" function and an "unregister"

Documentation/trace/kprobetrace.txt

@@ -40,7 +40,9 @@ Synopsis of kprobe_events
   $stack	: Fetch stack address.
   $retval	: Fetch return value.(*)
   +|-offs(FETCHARG) : Fetch memory at FETCHARG +|- offs address.(**)
-  NAME=FETCHARG: Set NAME as the argument name of FETCHARG.
+  NAME=FETCHARG : Set NAME as the argument name of FETCHARG.
+  FETCHARG:TYPE : Set TYPE as the type of FETCHARG. Currently, basic types
+		  (u8/u16/u32/u64/s8/s16/s32/s64) are supported.
 
   (*) only for return probe.
   (**) this is useful for fetching a field of data structures.

MAINTAINERS

@@ -4353,13 +4353,13 @@ M:	Paul Mackerras <paulus@samba.org>
 M:	Ingo Molnar <mingo@elte.hu>
 M:	Arnaldo Carvalho de Melo <acme@redhat.com>
 S:	Supported
-F:	kernel/perf_event.c
+F:	kernel/perf_event*.c
 F:	include/linux/perf_event.h
-F:	arch/*/kernel/perf_event.c
-F:	arch/*/kernel/*/perf_event.c
-F:	arch/*/kernel/*/*/perf_event.c
+F:	arch/*/kernel/perf_event*.c
+F:	arch/*/kernel/*/perf_event*.c
+F:	arch/*/kernel/*/*/perf_event*.c
 F:	arch/*/include/asm/perf_event.h
-F:	arch/*/lib/perf_event.c
+F:	arch/*/lib/perf_event*.c
 F:	arch/*/kernel/perf_callchain.c
 F:	tools/perf/
 

arch/Kconfig

@@ -42,15 +42,10 @@ config KPROBES
 	  If in doubt, say "N".
 
 config OPTPROBES
-	bool "Kprobes jump optimization support (EXPERIMENTAL)"
-	default y
-	depends on KPROBES
+	def_bool y
+	depends on KPROBES && HAVE_OPTPROBES
 	depends on !PREEMPT
-	depends on HAVE_OPTPROBES
 	select KALLSYMS_ALL
-	help
-	  This option will allow kprobes to optimize breakpoint to
-	  a jump for reducing its overhead.
 
 config HAVE_EFFICIENT_UNALIGNED_ACCESS
 	bool

arch/x86/Kconfig

@@ -58,6 +58,9 @@ config X86
 	select HAVE_ARCH_KMEMCHECK
 	select HAVE_USER_RETURN_NOTIFIER
 
+config INSTRUCTION_DECODER
+	def_bool (KPROBES || PERF_EVENTS)
+
 config OUTPUT_FORMAT
 	string
 	default "elf32-i386" if X86_32

arch/x86/Kconfig.cpu

@@ -502,23 +502,3 @@ config CPU_SUP_UMC_32
 	  CPU might render the kernel unbootable.
 
 	  If unsure, say N.
-
-config X86_DS
-	def_bool X86_PTRACE_BTS
-	depends on X86_DEBUGCTLMSR
-	select HAVE_HW_BRANCH_TRACER
-
-config X86_PTRACE_BTS
-	bool "Branch Trace Store"
-	default y
-	depends on X86_DEBUGCTLMSR
-	depends on BROKEN
-	---help---
-	  This adds a ptrace interface to the hardware's branch trace store.
-
-	  Debuggers may use it to collect an execution trace of the debugged
-	  application in order to answer the question 'how did I get here?'.
-	  Debuggers may trace user mode as well as kernel mode.
-
-	  Say Y unless there is no application development on this machine
-	  and you want to save a small amount of code size.

arch/x86/Kconfig.debug

@@ -174,15 +174,6 @@ config IOMMU_LEAK
 	  Add a simple leak tracer to the IOMMU code. This is useful when you
 	  are debugging a buggy device driver that leaks IOMMU mappings.
 
-config X86_DS_SELFTEST
-    bool "DS selftest"
-    default y
-    depends on DEBUG_KERNEL
-    depends on X86_DS
-	---help---
-	  Perform Debug Store selftests at boot time.
-	  If in doubt, say "N".
-
 config HAVE_MMIOTRACE_SUPPORT
 	def_bool y
 

arch/x86/include/asm/apic.h

@@ -373,6 +373,7 @@ extern atomic_t init_deasserted;
 extern int wakeup_secondary_cpu_via_nmi(int apicid, unsigned long start_eip);
 #endif
 
+#ifdef CONFIG_X86_LOCAL_APIC
 static inline u32 apic_read(u32 reg)
 {
 	return apic->read(reg);
@@ -403,10 +404,19 @@ static inline u32 safe_apic_wait_icr_idle(void)
 	return apic->safe_wait_icr_idle();
 }
 
+#else /* CONFIG_X86_LOCAL_APIC */
+
+static inline u32 apic_read(u32 reg) { return 0; }
+static inline void apic_write(u32 reg, u32 val) { }
+static inline u64 apic_icr_read(void) { return 0; }
+static inline void apic_icr_write(u32 low, u32 high) { }
+static inline void apic_wait_icr_idle(void) { }
+static inline u32 safe_apic_wait_icr_idle(void) { return 0; }
+
+#endif /* CONFIG_X86_LOCAL_APIC */
 
 static inline void ack_APIC_irq(void)
 {
-#ifdef CONFIG_X86_LOCAL_APIC
 	/*
 	 * ack_APIC_irq() actually gets compiled as a single instruction
 	 * ... yummie.
@@ -414,7 +424,6 @@ static inline void ack_APIC_irq(void)
 
 	/* Docs say use 0 for future compatibility */
 	apic_write(APIC_EOI, 0);
-#endif
 }
 
 static inline unsigned default_get_apic_id(unsigned long x)

arch/x86/include/asm/ds.h

@@ -1,302 +0,0 @@
-/*
- * Debug Store (DS) support
- *
- * This provides a low-level interface to the hardware's Debug Store
- * feature that is used for branch trace store (BTS) and
- * precise-event based sampling (PEBS).
- *
- * It manages:
- * - DS and BTS hardware configuration
- * - buffer overflow handling (to be done)
- * - buffer access
- *
- * It does not do:
- * - security checking (is the caller allowed to trace the task)
- * - buffer allocation (memory accounting)
- *
- *
- * Copyright (C) 2007-2009 Intel Corporation.
- * Markus Metzger <markus.t.metzger@intel.com>, 2007-2009
- */
-
-#ifndef _ASM_X86_DS_H
-#define _ASM_X86_DS_H
-
-
-#include <linux/types.h>
-#include <linux/init.h>
-#include <linux/err.h>
-
-
-#ifdef CONFIG_X86_DS
-
-struct task_struct;
-struct ds_context;
-struct ds_tracer;
-struct bts_tracer;
-struct pebs_tracer;
-
-typedef void (*bts_ovfl_callback_t)(struct bts_tracer *);
-typedef void (*pebs_ovfl_callback_t)(struct pebs_tracer *);
-
-
-/*
- * A list of features plus corresponding macros to talk about them in
- * the ds_request function's flags parameter.
- *
- * We use the enum to index an array of corresponding control bits;
- * we use the macro to index a flags bit-vector.
- */
-enum ds_feature {
-	dsf_bts = 0,
-	dsf_bts_kernel,
-#define BTS_KERNEL (1 << dsf_bts_kernel)
-	/* trace kernel-mode branches */
-
-	dsf_bts_user,
-#define BTS_USER (1 << dsf_bts_user)
-	/* trace user-mode branches */
-
-	dsf_bts_overflow,
-	dsf_bts_max,
-	dsf_pebs = dsf_bts_max,
-
-	dsf_pebs_max,
-	dsf_ctl_max = dsf_pebs_max,
-	dsf_bts_timestamps = dsf_ctl_max,
-#define BTS_TIMESTAMPS (1 << dsf_bts_timestamps)
-	/* add timestamps into BTS trace */
-
-#define BTS_USER_FLAGS (BTS_KERNEL | BTS_USER | BTS_TIMESTAMPS)
-};
-
-
-/*
- * Request BTS or PEBS
- *
- * Due to alignement constraints, the actual buffer may be slightly
- * smaller than the requested or provided buffer.
- *
- * Returns a pointer to a tracer structure on success, or
- * ERR_PTR(errcode) on failure.
- *
- * The interrupt threshold is independent from the overflow callback
- * to allow users to use their own overflow interrupt handling mechanism.
- *
- * The function might sleep.
- *
- * task: the task to request recording for
- * cpu:  the cpu to request recording for
- * base: the base pointer for the (non-pageable) buffer;
- * size: the size of the provided buffer in bytes
- * ovfl: pointer to a function to be called on buffer overflow;
- *       NULL if cyclic buffer requested
- * th: the interrupt threshold in records from the end of the buffer;
- *     -1 if no interrupt threshold is requested.
- * flags: a bit-mask of the above flags
- */
-extern struct bts_tracer *ds_request_bts_task(struct task_struct *task,
-					      void *base, size_t size,
-					      bts_ovfl_callback_t ovfl,
-					      size_t th, unsigned int flags);
-extern struct bts_tracer *ds_request_bts_cpu(int cpu, void *base, size_t size,
-					     bts_ovfl_callback_t ovfl,
-					     size_t th, unsigned int flags);
-extern struct pebs_tracer *ds_request_pebs_task(struct task_struct *task,
-						void *base, size_t size,
-						pebs_ovfl_callback_t ovfl,
-						size_t th, unsigned int flags);
-extern struct pebs_tracer *ds_request_pebs_cpu(int cpu,
-					       void *base, size_t size,
-					       pebs_ovfl_callback_t ovfl,
-					       size_t th, unsigned int flags);
-
-/*
- * Release BTS or PEBS resources
- * Suspend and resume BTS or PEBS tracing
- *
- * Must be called with irq's enabled.
- *
- * tracer: the tracer handle returned from ds_request_~()
- */
-extern void ds_release_bts(struct bts_tracer *tracer);
-extern void ds_suspend_bts(struct bts_tracer *tracer);
-extern void ds_resume_bts(struct bts_tracer *tracer);
-extern void ds_release_pebs(struct pebs_tracer *tracer);
-extern void ds_suspend_pebs(struct pebs_tracer *tracer);
-extern void ds_resume_pebs(struct pebs_tracer *tracer);
-
-/*
- * Release BTS or PEBS resources
- * Suspend and resume BTS or PEBS tracing
- *
- * Cpu tracers must call this on the traced cpu.
- * Task tracers must call ds_release_~_noirq() for themselves.
- *
- * May be called with irq's disabled.
- *
- * Returns 0 if successful;
- * -EPERM if the cpu tracer does not trace the current cpu.
- * -EPERM if the task tracer does not trace itself.
- *
- * tracer: the tracer handle returned from ds_request_~()
- */
-extern int ds_release_bts_noirq(struct bts_tracer *tracer);
-extern int ds_suspend_bts_noirq(struct bts_tracer *tracer);
-extern int ds_resume_bts_noirq(struct bts_tracer *tracer);
-extern int ds_release_pebs_noirq(struct pebs_tracer *tracer);
-extern int ds_suspend_pebs_noirq(struct pebs_tracer *tracer);
-extern int ds_resume_pebs_noirq(struct pebs_tracer *tracer);
-
-
-/*
- * The raw DS buffer state as it is used for BTS and PEBS recording.
- *
- * This is the low-level, arch-dependent interface for working
- * directly on the raw trace data.
- */
-struct ds_trace {
-	/* the number of bts/pebs records */
-	size_t n;
-	/* the size of a bts/pebs record in bytes */
-	size_t size;
-	/* pointers into the raw buffer:
-	   - to the first entry */
-	void *begin;
-	/* - one beyond the last entry */
-	void *end;
-	/* - one beyond the newest entry */
-	void *top;
-	/* - the interrupt threshold */
-	void *ith;
-	/* flags given on ds_request() */
-	unsigned int flags;
-};
-
-/*
- * An arch-independent view on branch trace data.
- */
-enum bts_qualifier {
-	bts_invalid,
-#define BTS_INVALID bts_invalid
-
-	bts_branch,
-#define BTS_BRANCH bts_branch
-
-	bts_task_arrives,
-#define BTS_TASK_ARRIVES bts_task_arrives
-
-	bts_task_departs,
-#define BTS_TASK_DEPARTS bts_task_departs
-
-	bts_qual_bit_size = 4,
-	bts_qual_max = (1 << bts_qual_bit_size),
-};
-
-struct bts_struct {
-	__u64 qualifier;
-	union {
-		/* BTS_BRANCH */
-		struct {
-			__u64 from;
-			__u64 to;
-		} lbr;
-		/* BTS_TASK_ARRIVES or BTS_TASK_DEPARTS */
-		struct {
-			__u64 clock;
-			pid_t pid;
-		} event;
-	} variant;
-};
-
-
-/*
- * The BTS state.
- *
- * This gives access to the raw DS state and adds functions to provide
- * an arch-independent view of the BTS data.
- */
-struct bts_trace {
-	struct ds_trace ds;
-
-	int (*read)(struct bts_tracer *tracer, const void *at,
-		    struct bts_struct *out);
-	int (*write)(struct bts_tracer *tracer, const struct bts_struct *in);
-};
-
-
-/*
- * The PEBS state.
- *
- * This gives access to the raw DS state and the PEBS-specific counter
- * reset value.
- */
-struct pebs_trace {
-	struct ds_trace ds;
-
-	/* the number of valid counters in the below array */
-	unsigned int counters;
-
-#define MAX_PEBS_COUNTERS 4
-	/* the counter reset value */
-	unsigned long long counter_reset[MAX_PEBS_COUNTERS];
-};
-
-
-/*
- * Read the BTS or PEBS trace.
- *
- * Returns a view on the trace collected for the parameter tracer.
- *
- * The view remains valid as long as the traced task is not running or
- * the tracer is suspended.
- * Writes into the trace buffer are not reflected.
- *
- * tracer: the tracer handle returned from ds_request_~()
- */
-extern const struct bts_trace *ds_read_bts(struct bts_tracer *tracer);
-extern const struct pebs_trace *ds_read_pebs(struct pebs_tracer *tracer);
-
-
-/*
- * Reset the write pointer of the BTS/PEBS buffer.
- *
- * Returns 0 on success; -Eerrno on error
- *
- * tracer: the tracer handle returned from ds_request_~()
- */
-extern int ds_reset_bts(struct bts_tracer *tracer);
-extern int ds_reset_pebs(struct pebs_tracer *tracer);
-
-/*
- * Set the PEBS counter reset value.
- *
- * Returns 0 on success; -Eerrno on error
- *
- * tracer: the tracer handle returned from ds_request_pebs()
- * counter: the index of the counter
- * value: the new counter reset value
- */
-extern int ds_set_pebs_reset(struct pebs_tracer *tracer,
-			     unsigned int counter, u64 value);
-
-/*
- * Initialization
- */
-struct cpuinfo_x86;
-extern void __cpuinit ds_init_intel(struct cpuinfo_x86 *);
-
-/*
- * Context switch work
- */
-extern void ds_switch_to(struct task_struct *prev, struct task_struct *next);
-
-#else /* CONFIG_X86_DS */
-
-struct cpuinfo_x86;
-static inline void __cpuinit ds_init_intel(struct cpuinfo_x86 *ignored) {}
-static inline void ds_switch_to(struct task_struct *prev,
-				struct task_struct *next) {}
-
-#endif /* CONFIG_X86_DS */
-#endif /* _ASM_X86_DS_H */

arch/x86/include/asm/insn.h

@@ -68,6 +68,8 @@ struct insn {
 	const insn_byte_t *next_byte;
 };
 
+#define MAX_INSN_SIZE	16
+
 #define X86_MODRM_MOD(modrm) (((modrm) & 0xc0) >> 6)
 #define X86_MODRM_REG(modrm) (((modrm) & 0x38) >> 3)
 #define X86_MODRM_RM(modrm) ((modrm) & 0x07)

arch/x86/include/asm/kprobes.h

@@ -24,6 +24,7 @@
 #include <linux/types.h>
 #include <linux/ptrace.h>
 #include <linux/percpu.h>
+#include <asm/insn.h>
 
 #define  __ARCH_WANT_KPROBES_INSN_SLOT
 
@@ -36,7 +37,6 @@ typedef u8 kprobe_opcode_t;
 #define RELATIVEJUMP_SIZE 5
 #define RELATIVECALL_OPCODE 0xe8
 #define RELATIVE_ADDR_SIZE 4
-#define MAX_INSN_SIZE 16
 #define MAX_STACK_SIZE 64
 #define MIN_STACK_SIZE(ADDR)					       \
 	(((MAX_STACK_SIZE) < (((unsigned long)current_thread_info()) + \

arch/x86/include/asm/msr-index.h

@@ -71,11 +71,14 @@
 #define MSR_IA32_LASTINTTOIP		0x000001de
 
 /* DEBUGCTLMSR bits (others vary by model): */
-#define _DEBUGCTLMSR_LBR	0 /* last branch recording */
-#define _DEBUGCTLMSR_BTF	1 /* single-step on branches */
-
-#define DEBUGCTLMSR_LBR		(1UL << _DEBUGCTLMSR_LBR)
-#define DEBUGCTLMSR_BTF		(1UL << _DEBUGCTLMSR_BTF)
+#define DEBUGCTLMSR_LBR			(1UL <<  0) /* last branch recording */
+#define DEBUGCTLMSR_BTF			(1UL <<  1) /* single-step on branches */
+#define DEBUGCTLMSR_TR			(1UL <<  6)
+#define DEBUGCTLMSR_BTS			(1UL <<  7)
+#define DEBUGCTLMSR_BTINT		(1UL <<  8)
+#define DEBUGCTLMSR_BTS_OFF_OS		(1UL <<  9)
+#define DEBUGCTLMSR_BTS_OFF_USR		(1UL << 10)
+#define DEBUGCTLMSR_FREEZE_LBRS_ON_PMI	(1UL << 11)
 
 #define MSR_IA32_MC0_CTL		0x00000400
 #define MSR_IA32_MC0_STATUS		0x00000401
@@ -359,6 +362,8 @@
 #define MSR_P4_U2L_ESCR0		0x000003b0
 #define MSR_P4_U2L_ESCR1		0x000003b1
 
+#define MSR_P4_PEBS_MATRIX_VERT		0x000003f2
+
 /* Intel Core-based CPU performance counters */
 #define MSR_CORE_PERF_FIXED_CTR0	0x00000309
 #define MSR_CORE_PERF_FIXED_CTR1	0x0000030a

arch/x86/include/asm/perf_event.h

@@ -5,7 +5,7 @@
  * Performance event hw details:
  */
 
-#define X86_PMC_MAX_GENERIC					8
+#define X86_PMC_MAX_GENERIC				       32
 #define X86_PMC_MAX_FIXED					3
 
 #define X86_PMC_IDX_GENERIC				        0
@@ -18,39 +18,31 @@
 #define MSR_ARCH_PERFMON_EVENTSEL0			     0x186
 #define MSR_ARCH_PERFMON_EVENTSEL1			     0x187
 
-#define ARCH_PERFMON_EVENTSEL_ENABLE			  (1 << 22)
-#define ARCH_PERFMON_EVENTSEL_ANY			  (1 << 21)
-#define ARCH_PERFMON_EVENTSEL_INT			  (1 << 20)
-#define ARCH_PERFMON_EVENTSEL_OS			  (1 << 17)
-#define ARCH_PERFMON_EVENTSEL_USR			  (1 << 16)
-
-/*
- * Includes eventsel and unit mask as well:
- */
-
-
-#define INTEL_ARCH_EVTSEL_MASK		0x000000FFULL
-#define INTEL_ARCH_UNIT_MASK		0x0000FF00ULL
-#define INTEL_ARCH_EDGE_MASK		0x00040000ULL
-#define INTEL_ARCH_INV_MASK		0x00800000ULL
-#define INTEL_ARCH_CNT_MASK		0xFF000000ULL
-#define INTEL_ARCH_EVENT_MASK	(INTEL_ARCH_UNIT_MASK|INTEL_ARCH_EVTSEL_MASK)
-
-/*
- * filter mask to validate fixed counter events.
- * the following filters disqualify for fixed counters:
- *  - inv
- *  - edge
- *  - cnt-mask
- *  The other filters are supported by fixed counters.
- *  The any-thread option is supported starting with v3.
- */
-#define INTEL_ARCH_FIXED_MASK \
-	(INTEL_ARCH_CNT_MASK| \
-	 INTEL_ARCH_INV_MASK| \
-	 INTEL_ARCH_EDGE_MASK|\
-	 INTEL_ARCH_UNIT_MASK|\
-	 INTEL_ARCH_EVENT_MASK)
+#define ARCH_PERFMON_EVENTSEL_EVENT			0x000000FFULL
+#define ARCH_PERFMON_EVENTSEL_UMASK			0x0000FF00ULL
+#define ARCH_PERFMON_EVENTSEL_USR			(1ULL << 16)
+#define ARCH_PERFMON_EVENTSEL_OS			(1ULL << 17)
+#define ARCH_PERFMON_EVENTSEL_EDGE			(1ULL << 18)
+#define ARCH_PERFMON_EVENTSEL_INT			(1ULL << 20)
+#define ARCH_PERFMON_EVENTSEL_ANY			(1ULL << 21)
+#define ARCH_PERFMON_EVENTSEL_ENABLE			(1ULL << 22)
+#define ARCH_PERFMON_EVENTSEL_INV			(1ULL << 23)
+#define ARCH_PERFMON_EVENTSEL_CMASK			0xFF000000ULL
+
+#define AMD64_EVENTSEL_EVENT	\
+	(ARCH_PERFMON_EVENTSEL_EVENT | (0x0FULL << 32))
+#define INTEL_ARCH_EVENT_MASK	\
+	(ARCH_PERFMON_EVENTSEL_UMASK | ARCH_PERFMON_EVENTSEL_EVENT)
+
+#define X86_RAW_EVENT_MASK		\
+	(ARCH_PERFMON_EVENTSEL_EVENT |	\
+	 ARCH_PERFMON_EVENTSEL_UMASK |	\
+	 ARCH_PERFMON_EVENTSEL_EDGE  |	\
+	 ARCH_PERFMON_EVENTSEL_INV   |	\
+	 ARCH_PERFMON_EVENTSEL_CMASK)
+#define AMD64_RAW_EVENT_MASK		\
+	(X86_RAW_EVENT_MASK          |  \
+	 AMD64_EVENTSEL_EVENT)
 
 #define ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL		      0x3c
 #define ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK		(0x00 << 8)
@@ -67,7 +59,7 @@
 union cpuid10_eax {
 	struct {
 		unsigned int version_id:8;
-		unsigned int num_events:8;
+		unsigned int num_counters:8;
 		unsigned int bit_width:8;
 		unsigned int mask_length:8;
 	} split;
@@ -76,7 +68,7 @@ union cpuid10_eax {
 
 union cpuid10_edx {
 	struct {
-		unsigned int num_events_fixed:4;
+		unsigned int num_counters_fixed:4;
 		unsigned int reserved:28;
 	} split;
 	unsigned int full;
@@ -136,6 +128,18 @@ extern void perf_events_lapic_init(void);
 
 #define PERF_EVENT_INDEX_OFFSET			0
 
+/*
+ * Abuse bit 3 of the cpu eflags register to indicate proper PEBS IP fixups.
+ * This flag is otherwise unused and ABI specified to be 0, so nobody should
+ * care what we do with it.
+ */
+#define PERF_EFLAGS_EXACT	(1UL << 3)
+
+struct pt_regs;
+extern unsigned long perf_instruction_pointer(struct pt_regs *regs);
+extern unsigned long perf_misc_flags(struct pt_regs *regs);
+#define perf_misc_flags(regs)	perf_misc_flags(regs)
+
 #else
 static inline void init_hw_perf_events(void)		{ }
 static inline void perf_events_lapic_init(void)	{ }

arch/x86/include/asm/perf_event_p4.h

@@ -0,0 +1,794 @@
+/*
+ * Netburst Perfomance Events (P4, old Xeon)
+ */
+
+#ifndef PERF_EVENT_P4_H
+#define PERF_EVENT_P4_H
+
+#include <linux/cpu.h>
+#include <linux/bitops.h>
+
+/*
+ * NetBurst has perfomance MSRs shared between
+ * threads if HT is turned on, ie for both logical
+ * processors (mem: in turn in Atom with HT support
+ * perf-MSRs are not shared and every thread has its
+ * own perf-MSRs set)
+ */
+#define ARCH_P4_TOTAL_ESCR	(46)
+#define ARCH_P4_RESERVED_ESCR	(2) /* IQ_ESCR(0,1) not always present */
+#define ARCH_P4_MAX_ESCR	(ARCH_P4_TOTAL_ESCR - ARCH_P4_RESERVED_ESCR)
+#define ARCH_P4_MAX_CCCR	(18)
+#define ARCH_P4_MAX_COUNTER	(ARCH_P4_MAX_CCCR / 2)
+
+#define P4_ESCR_EVENT_MASK	0x7e000000U
+#define P4_ESCR_EVENT_SHIFT	25
+#define P4_ESCR_EVENTMASK_MASK	0x01fffe00U
+#define P4_ESCR_EVENTMASK_SHIFT	9
+#define P4_ESCR_TAG_MASK	0x000001e0U
+#define P4_ESCR_TAG_SHIFT	5
+#define P4_ESCR_TAG_ENABLE	0x00000010U
+#define P4_ESCR_T0_OS		0x00000008U
+#define P4_ESCR_T0_USR		0x00000004U
+#define P4_ESCR_T1_OS		0x00000002U
+#define P4_ESCR_T1_USR		0x00000001U
+
+#define P4_ESCR_EVENT(v)	((v) << P4_ESCR_EVENT_SHIFT)
+#define P4_ESCR_EMASK(v)	((v) << P4_ESCR_EVENTMASK_SHIFT)
+#define P4_ESCR_TAG(v)		((v) << P4_ESCR_TAG_SHIFT)
+
+/* Non HT mask */
+#define P4_ESCR_MASK			\
+	(P4_ESCR_EVENT_MASK	|	\
+	P4_ESCR_EVENTMASK_MASK	|	\
+	P4_ESCR_TAG_MASK	|	\
+	P4_ESCR_TAG_ENABLE	|	\
+	P4_ESCR_T0_OS		|	\
+	P4_ESCR_T0_USR)
+
+/* HT mask */
+#define P4_ESCR_MASK_HT			\
+	(P4_ESCR_MASK |	P4_ESCR_T1_OS | P4_ESCR_T1_USR)
+
+#define P4_CCCR_OVF			0x80000000U
+#define P4_CCCR_CASCADE			0x40000000U
+#define P4_CCCR_OVF_PMI_T0		0x04000000U
+#define P4_CCCR_OVF_PMI_T1		0x08000000U
+#define P4_CCCR_FORCE_OVF		0x02000000U
+#define P4_CCCR_EDGE			0x01000000U
+#define P4_CCCR_THRESHOLD_MASK		0x00f00000U
+#define P4_CCCR_THRESHOLD_SHIFT		20
+#define P4_CCCR_COMPLEMENT		0x00080000U
+#define P4_CCCR_COMPARE			0x00040000U
+#define P4_CCCR_ESCR_SELECT_MASK	0x0000e000U
+#define P4_CCCR_ESCR_SELECT_SHIFT	13
+#define P4_CCCR_ENABLE			0x00001000U
+#define P4_CCCR_THREAD_SINGLE		0x00010000U
+#define P4_CCCR_THREAD_BOTH		0x00020000U
+#define P4_CCCR_THREAD_ANY		0x00030000U
+#define P4_CCCR_RESERVED		0x00000fffU
+
+#define P4_CCCR_THRESHOLD(v)		((v) << P4_CCCR_THRESHOLD_SHIFT)
+#define P4_CCCR_ESEL(v)			((v) << P4_CCCR_ESCR_SELECT_SHIFT)
+
+/* Custom bits in reerved CCCR area */
+#define P4_CCCR_CACHE_OPS_MASK		0x0000003fU
+
+
+/* Non HT mask */
+#define P4_CCCR_MASK				\
+	(P4_CCCR_OVF			|	\
+	P4_CCCR_CASCADE			|	\
+	P4_CCCR_OVF_PMI_T0		|	\
+	P4_CCCR_FORCE_OVF		|	\
+	P4_CCCR_EDGE			|	\
+	P4_CCCR_THRESHOLD_MASK		|	\
+	P4_CCCR_COMPLEMENT		|	\
+	P4_CCCR_COMPARE			|	\
+	P4_CCCR_ESCR_SELECT_MASK	|	\
+	P4_CCCR_ENABLE)
+
+/* HT mask */
+#define P4_CCCR_MASK_HT	(P4_CCCR_MASK | P4_CCCR_THREAD_ANY)
+
+#define P4_GEN_ESCR_EMASK(class, name, bit)	\
+	class##__##name = ((1 << bit) << P4_ESCR_EVENTMASK_SHIFT)
+#define P4_ESCR_EMASK_BIT(class, name)		class##__##name
+
+/*
+ * config field is 64bit width and consists of
+ * HT << 63 | ESCR << 32 | CCCR
+ * where HT is HyperThreading bit (since ESCR
+ * has it reserved we may use it for own purpose)
+ *
+ * note that this is NOT the addresses of respective
+ * ESCR and CCCR but rather an only packed value should
+ * be unpacked and written to a proper addresses
+ *
+ * the base idea is to pack as much info as
+ * possible
+ */
+#define p4_config_pack_escr(v)		(((u64)(v)) << 32)
+#define p4_config_pack_cccr(v)		(((u64)(v)) & 0xffffffffULL)
+#define p4_config_unpack_escr(v)	(((u64)(v)) >> 32)
+#define p4_config_unpack_cccr(v)	(((u64)(v)) & 0xffffffffULL)
+
+#define p4_config_unpack_emask(v)			\
+	({						\
+		u32 t = p4_config_unpack_escr((v));	\
+		t = t &  P4_ESCR_EVENTMASK_MASK;	\
+		t = t >> P4_ESCR_EVENTMASK_SHIFT;	\
+		t;					\
+	})
+
+#define p4_config_unpack_event(v)			\
+	({						\
+		u32 t = p4_config_unpack_escr((v));	\
+		t = t &  P4_ESCR_EVENT_MASK;		\
+		t = t >> P4_ESCR_EVENT_SHIFT;		\
+		t;					\
+	})
+
+#define p4_config_unpack_cache_event(v)	(((u64)(v)) & P4_CCCR_CACHE_OPS_MASK)
+
+#define P4_CONFIG_HT_SHIFT		63
+#define P4_CONFIG_HT			(1ULL << P4_CONFIG_HT_SHIFT)
+
+static inline bool p4_is_event_cascaded(u64 config)
+{
+	u32 cccr = p4_config_unpack_cccr(config);
+	return !!(cccr & P4_CCCR_CASCADE);
+}
+
+static inline int p4_ht_config_thread(u64 config)
+{
+	return !!(config & P4_CONFIG_HT);
+}
+
+static inline u64 p4_set_ht_bit(u64 config)
+{
+	return config | P4_CONFIG_HT;
+}
+
+static inline u64 p4_clear_ht_bit(u64 config)
+{
+	return config & ~P4_CONFIG_HT;
+}
+
+static inline int p4_ht_active(void)
+{
+#ifdef CONFIG_SMP
+	return smp_num_siblings > 1;
+#endif
+	return 0;
+}
+
+static inline int p4_ht_thread(int cpu)
+{
+#ifdef CONFIG_SMP
+	if (smp_num_siblings == 2)
+		return cpu != cpumask_first(__get_cpu_var(cpu_sibling_map));
+#endif
+	return 0;
+}
+
+static inline int p4_should_swap_ts(u64 config, int cpu)
+{
+	return p4_ht_config_thread(config) ^ p4_ht_thread(cpu);
+}
+
+static inline u32 p4_default_cccr_conf(int cpu)
+{
+	/*
+	 * Note that P4_CCCR_THREAD_ANY is "required" on
+	 * non-HT machines (on HT machines we count TS events
+	 * regardless the state of second logical processor
+	 */
+	u32 cccr = P4_CCCR_THREAD_ANY;
+
+	if (!p4_ht_thread(cpu))
+		cccr |= P4_CCCR_OVF_PMI_T0;
+	else
+		cccr |= P4_CCCR_OVF_PMI_T1;
+
+	return cccr;
+}
+
+static inline u32 p4_default_escr_conf(int cpu, int exclude_os, int exclude_usr)
+{
+	u32 escr = 0;
+
+	if (!p4_ht_thread(cpu)) {
+		if (!exclude_os)
+			escr |= P4_ESCR_T0_OS;
+		if (!exclude_usr)
+			escr |= P4_ESCR_T0_USR;
+	} else {
+		if (!exclude_os)
+			escr |= P4_ESCR_T1_OS;
+		if (!exclude_usr)
+			escr |= P4_ESCR_T1_USR;
+	}
+
+	return escr;
+}
+
+enum P4_EVENTS {
+	P4_EVENT_TC_DELIVER_MODE,
+	P4_EVENT_BPU_FETCH_REQUEST,
+	P4_EVENT_ITLB_REFERENCE,
+	P4_EVENT_MEMORY_CANCEL,
+	P4_EVENT_MEMORY_COMPLETE,
+	P4_EVENT_LOAD_PORT_REPLAY,
+	P4_EVENT_STORE_PORT_REPLAY,
+	P4_EVENT_MOB_LOAD_REPLAY,
+	P4_EVENT_PAGE_WALK_TYPE,
+	P4_EVENT_BSQ_CACHE_REFERENCE,
+	P4_EVENT_IOQ_ALLOCATION,
+	P4_EVENT_IOQ_ACTIVE_ENTRIES,
+	P4_EVENT_FSB_DATA_ACTIVITY,
+	P4_EVENT_BSQ_ALLOCATION,
+	P4_EVENT_BSQ_ACTIVE_ENTRIES,
+	P4_EVENT_SSE_INPUT_ASSIST,
+	P4_EVENT_PACKED_SP_UOP,
+	P4_EVENT_PACKED_DP_UOP,
+	P4_EVENT_SCALAR_SP_UOP,
+	P4_EVENT_SCALAR_DP_UOP,
+	P4_EVENT_64BIT_MMX_UOP,
+	P4_EVENT_128BIT_MMX_UOP,
+	P4_EVENT_X87_FP_UOP,
+	P4_EVENT_TC_MISC,
+	P4_EVENT_GLOBAL_POWER_EVENTS,
+	P4_EVENT_TC_MS_XFER,
+	P4_EVENT_UOP_QUEUE_WRITES,
+	P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE,
+	P4_EVENT_RETIRED_BRANCH_TYPE,
+	P4_EVENT_RESOURCE_STALL,
+	P4_EVENT_WC_BUFFER,
+	P4_EVENT_B2B_CYCLES,
+	P4_EVENT_BNR,
+	P4_EVENT_SNOOP,
+	P4_EVENT_RESPONSE,
+	P4_EVENT_FRONT_END_EVENT,
+	P4_EVENT_EXECUTION_EVENT,
+	P4_EVENT_REPLAY_EVENT,
+	P4_EVENT_INSTR_RETIRED,
+	P4_EVENT_UOPS_RETIRED,
+	P4_EVENT_UOP_TYPE,
+	P4_EVENT_BRANCH_RETIRED,
+	P4_EVENT_MISPRED_BRANCH_RETIRED,
+	P4_EVENT_X87_ASSIST,
+	P4_EVENT_MACHINE_CLEAR,
+	P4_EVENT_INSTR_COMPLETED,
+};
+
+#define P4_OPCODE(event)		event##_OPCODE
+#define P4_OPCODE_ESEL(opcode)		((opcode & 0x00ff) >> 0)
+#define P4_OPCODE_EVNT(opcode)		((opcode & 0xff00) >> 8)
+#define P4_OPCODE_PACK(event, sel)	(((event) << 8) | sel)
+
+/*
+ * Comments below the event represent ESCR restriction
+ * for this event and counter index per ESCR
+ *
+ * MSR_P4_IQ_ESCR0 and MSR_P4_IQ_ESCR1 are available only on early
+ * processor builds (family 0FH, models 01H-02H). These MSRs
+ * are not available on later versions, so that we don't use
+ * them completely
+ *
+ * Also note that CCCR1 do not have P4_CCCR_ENABLE bit properly
+ * working so that we should not use this CCCR and respective
+ * counter as result
+ */
+enum P4_EVENT_OPCODES {
+	P4_OPCODE(P4_EVENT_TC_DELIVER_MODE)		= P4_OPCODE_PACK(0x01, 0x01),
+	/*
+	 * MSR_P4_TC_ESCR0:	4, 5
+	 * MSR_P4_TC_ESCR1:	6, 7
+	 */
+
+	P4_OPCODE(P4_EVENT_BPU_FETCH_REQUEST)		= P4_OPCODE_PACK(0x03, 0x00),
+	/*
+	 * MSR_P4_BPU_ESCR0:	0, 1
+	 * MSR_P4_BPU_ESCR1:	2, 3
+	 */
+
+	P4_OPCODE(P4_EVENT_ITLB_REFERENCE)		= P4_OPCODE_PACK(0x18, 0x03),
+	/*
+	 * MSR_P4_ITLB_ESCR0:	0, 1
+	 * MSR_P4_ITLB_ESCR1:	2, 3
+	 */
+
+	P4_OPCODE(P4_EVENT_MEMORY_CANCEL)		= P4_OPCODE_PACK(0x02, 0x05),
+	/*
+	 * MSR_P4_DAC_ESCR0:	8, 9
+	 * MSR_P4_DAC_ESCR1:	10, 11
+	 */
+
+	P4_OPCODE(P4_EVENT_MEMORY_COMPLETE)		= P4_OPCODE_PACK(0x08, 0x02),
+	/*
+	 * MSR_P4_SAAT_ESCR0:	8, 9
+	 * MSR_P4_SAAT_ESCR1:	10, 11
+	 */
+
+	P4_OPCODE(P4_EVENT_LOAD_PORT_REPLAY)		= P4_OPCODE_PACK(0x04, 0x02),
+	/*
+	 * MSR_P4_SAAT_ESCR0:	8, 9
+	 * MSR_P4_SAAT_ESCR1:	10, 11
+	 */
+
+	P4_OPCODE(P4_EVENT_STORE_PORT_REPLAY)		= P4_OPCODE_PACK(0x05, 0x02),
+	/*
+	 * MSR_P4_SAAT_ESCR0:	8, 9
+	 * MSR_P4_SAAT_ESCR1:	10, 11
+	 */
+
+	P4_OPCODE(P4_EVENT_MOB_LOAD_REPLAY)		= P4_OPCODE_PACK(0x03, 0x02),
+	/*
+	 * MSR_P4_MOB_ESCR0:	0, 1
+	 * MSR_P4_MOB_ESCR1:	2, 3
+	 */
+
+	P4_OPCODE(P4_EVENT_PAGE_WALK_TYPE)		= P4_OPCODE_PACK(0x01, 0x04),
+	/*
+	 * MSR_P4_PMH_ESCR0:	0, 1
+	 * MSR_P4_PMH_ESCR1:	2, 3
+	 */
+
+	P4_OPCODE(P4_EVENT_BSQ_CACHE_REFERENCE)		= P4_OPCODE_PACK(0x0c, 0x07),
+	/*
+	 * MSR_P4_BSU_ESCR0:	0, 1
+	 * MSR_P4_BSU_ESCR1:	2, 3
+	 */
+
+	P4_OPCODE(P4_EVENT_IOQ_ALLOCATION)		= P4_OPCODE_PACK(0x03, 0x06),
+	/*
+	 * MSR_P4_FSB_ESCR0:	0, 1
+	 * MSR_P4_FSB_ESCR1:	2, 3
+	 */
+
+	P4_OPCODE(P4_EVENT_IOQ_ACTIVE_ENTRIES)		= P4_OPCODE_PACK(0x1a, 0x06),
+	/*
+	 * MSR_P4_FSB_ESCR1:	2, 3
+	 */
+
+	P4_OPCODE(P4_EVENT_FSB_DATA_ACTIVITY)		= P4_OPCODE_PACK(0x17, 0x06),
+	/*
+	 * MSR_P4_FSB_ESCR0:	0, 1
+	 * MSR_P4_FSB_ESCR1:	2, 3
+	 */
+
+	P4_OPCODE(P4_EVENT_BSQ_ALLOCATION)		= P4_OPCODE_PACK(0x05, 0x07),
+	/*
+	 * MSR_P4_BSU_ESCR0:	0, 1
+	 */
+
+	P4_OPCODE(P4_EVENT_BSQ_ACTIVE_ENTRIES)		= P4_OPCODE_PACK(0x06, 0x07),
+	/*
+	 * NOTE: no ESCR name in docs, it's guessed
+	 * MSR_P4_BSU_ESCR1:	2, 3
+	 */
+
+	P4_OPCODE(P4_EVENT_SSE_INPUT_ASSIST)		= P4_OPCODE_PACK(0x34, 0x01),
+	/*
+	 * MSR_P4_FIRM_ESCR0:	8, 9
+	 * MSR_P4_FIRM_ESCR1:	10, 11
+	 */
+
+	P4_OPCODE(P4_EVENT_PACKED_SP_UOP)		= P4_OPCODE_PACK(0x08, 0x01),
+	/*
+	 * MSR_P4_FIRM_ESCR0:	8, 9
+	 * MSR_P4_FIRM_ESCR1:	10, 11
+	 */
+
+	P4_OPCODE(P4_EVENT_PACKED_DP_UOP)		= P4_OPCODE_PACK(0x0c, 0x01),
+	/*
+	 * MSR_P4_FIRM_ESCR0:	8, 9
+	 * MSR_P4_FIRM_ESCR1:	10, 11
+	 */
+
+	P4_OPCODE(P4_EVENT_SCALAR_SP_UOP)		= P4_OPCODE_PACK(0x0a, 0x01),
+	/*
+	 * MSR_P4_FIRM_ESCR0:	8, 9
+	 * MSR_P4_FIRM_ESCR1:	10, 11
+	 */
+
+	P4_OPCODE(P4_EVENT_SCALAR_DP_UOP)		= P4_OPCODE_PACK(0x0e, 0x01),
+	/*
+	 * MSR_P4_FIRM_ESCR0:	8, 9
+	 * MSR_P4_FIRM_ESCR1:	10, 11
+	 */
+
+	P4_OPCODE(P4_EVENT_64BIT_MMX_UOP)		= P4_OPCODE_PACK(0x02, 0x01),
+	/*
+	 * MSR_P4_FIRM_ESCR0:	8, 9
+	 * MSR_P4_FIRM_ESCR1:	10, 11
+	 */
+
+	P4_OPCODE(P4_EVENT_128BIT_MMX_UOP)		= P4_OPCODE_PACK(0x1a, 0x01),
+	/*
+	 * MSR_P4_FIRM_ESCR0:	8, 9
+	 * MSR_P4_FIRM_ESCR1:	10, 11
+	 */
+
+	P4_OPCODE(P4_EVENT_X87_FP_UOP)			= P4_OPCODE_PACK(0x04, 0x01),
+	/*
+	 * MSR_P4_FIRM_ESCR0:	8, 9
+	 * MSR_P4_FIRM_ESCR1:	10, 11
+	 */
+
+	P4_OPCODE(P4_EVENT_TC_MISC)			= P4_OPCODE_PACK(0x06, 0x01),
+	/*
+	 * MSR_P4_TC_ESCR0:	4, 5
+	 * MSR_P4_TC_ESCR1:	6, 7
+	 */
+
+	P4_OPCODE(P4_EVENT_GLOBAL_POWER_EVENTS)		= P4_OPCODE_PACK(0x13, 0x06),
+	/*
+	 * MSR_P4_FSB_ESCR0:	0, 1
+	 * MSR_P4_FSB_ESCR1:	2, 3
+	 */
+
+	P4_OPCODE(P4_EVENT_TC_MS_XFER)			= P4_OPCODE_PACK(0x05, 0x00),
+	/*
+	 * MSR_P4_MS_ESCR0:	4, 5
+	 * MSR_P4_MS_ESCR1:	6, 7
+	 */
+
+	P4_OPCODE(P4_EVENT_UOP_QUEUE_WRITES)		= P4_OPCODE_PACK(0x09, 0x00),
+	/*
+	 * MSR_P4_MS_ESCR0:	4, 5
+	 * MSR_P4_MS_ESCR1:	6, 7
+	 */
+
+	P4_OPCODE(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE)	= P4_OPCODE_PACK(0x05, 0x02),
+	/*
+	 * MSR_P4_TBPU_ESCR0:	4, 5
+	 * MSR_P4_TBPU_ESCR1:	6, 7
+	 */
+
+	P4_OPCODE(P4_EVENT_RETIRED_BRANCH_TYPE)		= P4_OPCODE_PACK(0x04, 0x02),
+	/*
+	 * MSR_P4_TBPU_ESCR0:	4, 5
+	 * MSR_P4_TBPU_ESCR1:	6, 7
+	 */
+
+	P4_OPCODE(P4_EVENT_RESOURCE_STALL)		= P4_OPCODE_PACK(0x01, 0x01),
+	/*
+	 * MSR_P4_ALF_ESCR0:	12, 13, 16
+	 * MSR_P4_ALF_ESCR1:	14, 15, 17
+	 */
+
+	P4_OPCODE(P4_EVENT_WC_BUFFER)			= P4_OPCODE_PACK(0x05, 0x05),
+	/*
+	 * MSR_P4_DAC_ESCR0:	8, 9
+	 * MSR_P4_DAC_ESCR1:	10, 11
+	 */
+
+	P4_OPCODE(P4_EVENT_B2B_CYCLES)			= P4_OPCODE_PACK(0x16, 0x03),
+	/*
+	 * MSR_P4_FSB_ESCR0:	0, 1
+	 * MSR_P4_FSB_ESCR1:	2, 3
+	 */
+
+	P4_OPCODE(P4_EVENT_BNR)				= P4_OPCODE_PACK(0x08, 0x03),
+	/*
+	 * MSR_P4_FSB_ESCR0:	0, 1
+	 * MSR_P4_FSB_ESCR1:	2, 3
+	 */
+
+	P4_OPCODE(P4_EVENT_SNOOP)			= P4_OPCODE_PACK(0x06, 0x03),
+	/*
+	 * MSR_P4_FSB_ESCR0:	0, 1
+	 * MSR_P4_FSB_ESCR1:	2, 3
+	 */
+
+	P4_OPCODE(P4_EVENT_RESPONSE)			= P4_OPCODE_PACK(0x04, 0x03),
+	/*
+	 * MSR_P4_FSB_ESCR0:	0, 1
+	 * MSR_P4_FSB_ESCR1:	2, 3
+	 */
+
+	P4_OPCODE(P4_EVENT_FRONT_END_EVENT)		= P4_OPCODE_PACK(0x08, 0x05),
+	/*
+	 * MSR_P4_CRU_ESCR2:	12, 13, 16
+	 * MSR_P4_CRU_ESCR3:	14, 15, 17
+	 */
+
+	P4_OPCODE(P4_EVENT_EXECUTION_EVENT)		= P4_OPCODE_PACK(0x0c, 0x05),
+	/*
+	 * MSR_P4_CRU_ESCR2:	12, 13, 16
+	 * MSR_P4_CRU_ESCR3:	14, 15, 17
+	 */
+
+	P4_OPCODE(P4_EVENT_REPLAY_EVENT)		= P4_OPCODE_PACK(0x09, 0x05),
+	/*
+	 * MSR_P4_CRU_ESCR2:	12, 13, 16
+	 * MSR_P4_CRU_ESCR3:	14, 15, 17
+	 */
+
+	P4_OPCODE(P4_EVENT_INSTR_RETIRED)		= P4_OPCODE_PACK(0x02, 0x04),
+	/*
+	 * MSR_P4_CRU_ESCR0:	12, 13, 16
+	 * MSR_P4_CRU_ESCR1:	14, 15, 17
+	 */
+
+	P4_OPCODE(P4_EVENT_UOPS_RETIRED)		= P4_OPCODE_PACK(0x01, 0x04),
+	/*
+	 * MSR_P4_CRU_ESCR0:	12, 13, 16
+	 * MSR_P4_CRU_ESCR1:	14, 15, 17
+	 */
+
+	P4_OPCODE(P4_EVENT_UOP_TYPE)			= P4_OPCODE_PACK(0x02, 0x02),
+	/*
+	 * MSR_P4_RAT_ESCR0:	12, 13, 16
+	 * MSR_P4_RAT_ESCR1:	14, 15, 17
+	 */
+
+	P4_OPCODE(P4_EVENT_BRANCH_RETIRED)		= P4_OPCODE_PACK(0x06, 0x05),
+	/*
+	 * MSR_P4_CRU_ESCR2:	12, 13, 16
+	 * MSR_P4_CRU_ESCR3:	14, 15, 17
+	 */
+
+	P4_OPCODE(P4_EVENT_MISPRED_BRANCH_RETIRED)	= P4_OPCODE_PACK(0x03, 0x04),
+	/*
+	 * MSR_P4_CRU_ESCR0:	12, 13, 16
+	 * MSR_P4_CRU_ESCR1:	14, 15, 17
+	 */
+
+	P4_OPCODE(P4_EVENT_X87_ASSIST)			= P4_OPCODE_PACK(0x03, 0x05),
+	/*
+	 * MSR_P4_CRU_ESCR2:	12, 13, 16
+	 * MSR_P4_CRU_ESCR3:	14, 15, 17
+	 */
+
+	P4_OPCODE(P4_EVENT_MACHINE_CLEAR)		= P4_OPCODE_PACK(0x02, 0x05),
+	/*
+	 * MSR_P4_CRU_ESCR2:	12, 13, 16
+	 * MSR_P4_CRU_ESCR3:	14, 15, 17
+	 */
+
+	P4_OPCODE(P4_EVENT_INSTR_COMPLETED)		= P4_OPCODE_PACK(0x07, 0x04),
+	/*
+	 * MSR_P4_CRU_ESCR0:	12, 13, 16
+	 * MSR_P4_CRU_ESCR1:	14, 15, 17
+	 */
+};
+
+/*
+ * a caller should use P4_ESCR_EMASK_NAME helper to
+ * pick the EventMask needed, for example
+ *
+ *	P4_ESCR_EMASK_NAME(P4_EVENT_TC_DELIVER_MODE, DD)
+ */
+enum P4_ESCR_EMASKS {
+	P4_GEN_ESCR_EMASK(P4_EVENT_TC_DELIVER_MODE, DD, 0),
+	P4_GEN_ESCR_EMASK(P4_EVENT_TC_DELIVER_MODE, DB, 1),
+	P4_GEN_ESCR_EMASK(P4_EVENT_TC_DELIVER_MODE, DI, 2),
+	P4_GEN_ESCR_EMASK(P4_EVENT_TC_DELIVER_MODE, BD, 3),
+	P4_GEN_ESCR_EMASK(P4_EVENT_TC_DELIVER_MODE, BB, 4),
+	P4_GEN_ESCR_EMASK(P4_EVENT_TC_DELIVER_MODE, BI, 5),
+	P4_GEN_ESCR_EMASK(P4_EVENT_TC_DELIVER_MODE, ID, 6),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_BPU_FETCH_REQUEST, TCMISS, 0),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_ITLB_REFERENCE, HIT, 0),
+	P4_GEN_ESCR_EMASK(P4_EVENT_ITLB_REFERENCE, MISS, 1),
+	P4_GEN_ESCR_EMASK(P4_EVENT_ITLB_REFERENCE, HIT_UK, 2),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_MEMORY_CANCEL, ST_RB_FULL, 2),
+	P4_GEN_ESCR_EMASK(P4_EVENT_MEMORY_CANCEL, 64K_CONF, 3),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_MEMORY_COMPLETE, LSC, 0),
+	P4_GEN_ESCR_EMASK(P4_EVENT_MEMORY_COMPLETE, SSC, 1),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_LOAD_PORT_REPLAY, SPLIT_LD, 1),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_STORE_PORT_REPLAY, SPLIT_ST, 1),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_MOB_LOAD_REPLAY, NO_STA, 1),
+	P4_GEN_ESCR_EMASK(P4_EVENT_MOB_LOAD_REPLAY, NO_STD, 3),
+	P4_GEN_ESCR_EMASK(P4_EVENT_MOB_LOAD_REPLAY, PARTIAL_DATA, 4),
+	P4_GEN_ESCR_EMASK(P4_EVENT_MOB_LOAD_REPLAY, UNALGN_ADDR, 5),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_PAGE_WALK_TYPE, DTMISS, 0),
+	P4_GEN_ESCR_EMASK(P4_EVENT_PAGE_WALK_TYPE, ITMISS, 1),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITS, 0),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITE, 1),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITM, 2),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITS, 3),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITE, 4),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITM, 5),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_MISS, 8),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_MISS, 9),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_CACHE_REFERENCE, WR_2ndL_MISS, 10),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ALLOCATION, DEFAULT, 0),
+	P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ALLOCATION, ALL_READ, 5),
+	P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ALLOCATION, ALL_WRITE, 6),
+	P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ALLOCATION, MEM_UC, 7),
+	P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ALLOCATION, MEM_WC, 8),
+	P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ALLOCATION, MEM_WT, 9),
+	P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ALLOCATION, MEM_WP, 10),
+	P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ALLOCATION, MEM_WB, 11),
+	P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ALLOCATION, OWN, 13),
+	P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ALLOCATION, OTHER, 14),
+	P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ALLOCATION, PREFETCH, 15),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ACTIVE_ENTRIES, DEFAULT, 0),
+	P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ACTIVE_ENTRIES, ALL_READ, 5),
+	P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ACTIVE_ENTRIES, ALL_WRITE, 6),
+	P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_UC, 7),
+	P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WC, 8),
+	P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WT, 9),
+	P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WP, 10),
+	P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WB, 11),
+	P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ACTIVE_ENTRIES, OWN, 13),
+	P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ACTIVE_ENTRIES, OTHER, 14),
+	P4_GEN_ESCR_EMASK(P4_EVENT_IOQ_ACTIVE_ENTRIES, PREFETCH, 15),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_DRV, 0),
+	P4_GEN_ESCR_EMASK(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_OWN, 1),
+	P4_GEN_ESCR_EMASK(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_OTHER, 2),
+	P4_GEN_ESCR_EMASK(P4_EVENT_FSB_DATA_ACTIVITY, DBSY_DRV, 3),
+	P4_GEN_ESCR_EMASK(P4_EVENT_FSB_DATA_ACTIVITY, DBSY_OWN, 4),
+	P4_GEN_ESCR_EMASK(P4_EVENT_FSB_DATA_ACTIVITY, DBSY_OTHER, 5),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, REQ_TYPE0, 0),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, REQ_TYPE1, 1),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, REQ_LEN0, 2),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, REQ_LEN1, 3),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, REQ_IO_TYPE, 5),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, REQ_LOCK_TYPE, 6),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, REQ_CACHE_TYPE, 7),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, REQ_SPLIT_TYPE, 8),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, REQ_DEM_TYPE, 9),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, REQ_ORD_TYPE, 10),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, MEM_TYPE0, 11),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, MEM_TYPE1, 12),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ALLOCATION, MEM_TYPE2, 13),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_TYPE0, 0),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_TYPE1, 1),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_LEN0, 2),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_LEN1, 3),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_IO_TYPE, 5),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_LOCK_TYPE, 6),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_CACHE_TYPE, 7),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_SPLIT_TYPE, 8),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_DEM_TYPE, 9),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_ORD_TYPE, 10),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, MEM_TYPE0, 11),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, MEM_TYPE1, 12),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BSQ_ACTIVE_ENTRIES, MEM_TYPE2, 13),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_SSE_INPUT_ASSIST, ALL, 15),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_PACKED_SP_UOP, ALL, 15),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_PACKED_DP_UOP, ALL, 15),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_SCALAR_SP_UOP, ALL, 15),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_SCALAR_DP_UOP, ALL, 15),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_64BIT_MMX_UOP, ALL, 15),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_128BIT_MMX_UOP, ALL, 15),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_X87_FP_UOP, ALL, 15),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_TC_MISC, FLUSH, 4),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_GLOBAL_POWER_EVENTS, RUNNING, 0),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_TC_MS_XFER, CISC, 0),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_UOP_QUEUE_WRITES, FROM_TC_BUILD, 0),
+	P4_GEN_ESCR_EMASK(P4_EVENT_UOP_QUEUE_WRITES, FROM_TC_DELIVER, 1),
+	P4_GEN_ESCR_EMASK(P4_EVENT_UOP_QUEUE_WRITES, FROM_ROM, 2),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, CONDITIONAL, 1),
+	P4_GEN_ESCR_EMASK(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, CALL, 2),
+	P4_GEN_ESCR_EMASK(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, RETURN, 3),
+	P4_GEN_ESCR_EMASK(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, INDIRECT, 4),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_RETIRED_BRANCH_TYPE, CONDITIONAL, 1),
+	P4_GEN_ESCR_EMASK(P4_EVENT_RETIRED_BRANCH_TYPE, CALL, 2),
+	P4_GEN_ESCR_EMASK(P4_EVENT_RETIRED_BRANCH_TYPE, RETURN, 3),
+	P4_GEN_ESCR_EMASK(P4_EVENT_RETIRED_BRANCH_TYPE, INDIRECT, 4),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_RESOURCE_STALL, SBFULL, 5),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_WC_BUFFER, WCB_EVICTS, 0),
+	P4_GEN_ESCR_EMASK(P4_EVENT_WC_BUFFER, WCB_FULL_EVICTS, 1),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_FRONT_END_EVENT, NBOGUS, 0),
+	P4_GEN_ESCR_EMASK(P4_EVENT_FRONT_END_EVENT, BOGUS, 1),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_EXECUTION_EVENT, NBOGUS0, 0),
+	P4_GEN_ESCR_EMASK(P4_EVENT_EXECUTION_EVENT, NBOGUS1, 1),
+	P4_GEN_ESCR_EMASK(P4_EVENT_EXECUTION_EVENT, NBOGUS2, 2),
+	P4_GEN_ESCR_EMASK(P4_EVENT_EXECUTION_EVENT, NBOGUS3, 3),
+	P4_GEN_ESCR_EMASK(P4_EVENT_EXECUTION_EVENT, BOGUS0, 4),
+	P4_GEN_ESCR_EMASK(P4_EVENT_EXECUTION_EVENT, BOGUS1, 5),
+	P4_GEN_ESCR_EMASK(P4_EVENT_EXECUTION_EVENT, BOGUS2, 6),
+	P4_GEN_ESCR_EMASK(P4_EVENT_EXECUTION_EVENT, BOGUS3, 7),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_REPLAY_EVENT, NBOGUS, 0),
+	P4_GEN_ESCR_EMASK(P4_EVENT_REPLAY_EVENT, BOGUS, 1),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_INSTR_RETIRED, NBOGUSNTAG, 0),
+	P4_GEN_ESCR_EMASK(P4_EVENT_INSTR_RETIRED, NBOGUSTAG, 1),
+	P4_GEN_ESCR_EMASK(P4_EVENT_INSTR_RETIRED, BOGUSNTAG, 2),
+	P4_GEN_ESCR_EMASK(P4_EVENT_INSTR_RETIRED, BOGUSTAG, 3),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_UOPS_RETIRED, NBOGUS, 0),
+	P4_GEN_ESCR_EMASK(P4_EVENT_UOPS_RETIRED, BOGUS, 1),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_UOP_TYPE, TAGLOADS, 1),
+	P4_GEN_ESCR_EMASK(P4_EVENT_UOP_TYPE, TAGSTORES, 2),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_BRANCH_RETIRED, MMNP, 0),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BRANCH_RETIRED, MMNM, 1),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BRANCH_RETIRED, MMTP, 2),
+	P4_GEN_ESCR_EMASK(P4_EVENT_BRANCH_RETIRED, MMTM, 3),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_MISPRED_BRANCH_RETIRED, NBOGUS, 0),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_X87_ASSIST, FPSU, 0),
+	P4_GEN_ESCR_EMASK(P4_EVENT_X87_ASSIST, FPSO, 1),
+	P4_GEN_ESCR_EMASK(P4_EVENT_X87_ASSIST, POAO, 2),
+	P4_GEN_ESCR_EMASK(P4_EVENT_X87_ASSIST, POAU, 3),
+	P4_GEN_ESCR_EMASK(P4_EVENT_X87_ASSIST, PREA, 4),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_MACHINE_CLEAR, CLEAR, 0),
+	P4_GEN_ESCR_EMASK(P4_EVENT_MACHINE_CLEAR, MOCLEAR, 1),
+	P4_GEN_ESCR_EMASK(P4_EVENT_MACHINE_CLEAR, SMCLEAR, 2),
+
+	P4_GEN_ESCR_EMASK(P4_EVENT_INSTR_COMPLETED, NBOGUS, 0),
+	P4_GEN_ESCR_EMASK(P4_EVENT_INSTR_COMPLETED, BOGUS, 1),
+};
+
+/* P4 PEBS: stale for a while */
+#define P4_PEBS_METRIC_MASK	0x00001fffU
+#define P4_PEBS_UOB_TAG		0x01000000U
+#define P4_PEBS_ENABLE		0x02000000U
+
+/* Replay metrics for MSR_IA32_PEBS_ENABLE and MSR_P4_PEBS_MATRIX_VERT */
+#define P4_PEBS__1stl_cache_load_miss_retired	0x3000001
+#define P4_PEBS__2ndl_cache_load_miss_retired	0x3000002
+#define P4_PEBS__dtlb_load_miss_retired		0x3000004
+#define P4_PEBS__dtlb_store_miss_retired	0x3000004
+#define P4_PEBS__dtlb_all_miss_retired		0x3000004
+#define P4_PEBS__tagged_mispred_branch		0x3018000
+#define P4_PEBS__mob_load_replay_retired	0x3000200
+#define P4_PEBS__split_load_retired		0x3000400
+#define P4_PEBS__split_store_retired		0x3000400
+
+#define P4_VERT__1stl_cache_load_miss_retired	0x0000001
+#define P4_VERT__2ndl_cache_load_miss_retired	0x0000001
+#define P4_VERT__dtlb_load_miss_retired		0x0000001
+#define P4_VERT__dtlb_store_miss_retired	0x0000002
+#define P4_VERT__dtlb_all_miss_retired		0x0000003
+#define P4_VERT__tagged_mispred_branch		0x0000010
+#define P4_VERT__mob_load_replay_retired	0x0000001
+#define P4_VERT__split_load_retired		0x0000001
+#define P4_VERT__split_store_retired		0x0000002
+
+enum P4_CACHE_EVENTS {
+	P4_CACHE__NONE,
+
+	P4_CACHE__1stl_cache_load_miss_retired,
+	P4_CACHE__2ndl_cache_load_miss_retired,
+	P4_CACHE__dtlb_load_miss_retired,
+	P4_CACHE__dtlb_store_miss_retired,
+	P4_CACHE__itlb_reference_hit,
+	P4_CACHE__itlb_reference_miss,
+
+	P4_CACHE__MAX
+};
+
+#endif /* PERF_EVENT_P4_H */

arch/x86/include/asm/processor.h

@@ -21,7 +21,6 @@ struct mm_struct;
 #include <asm/msr.h>
 #include <asm/desc_defs.h>
 #include <asm/nops.h>
-#include <asm/ds.h>
 
 #include <linux/personality.h>
 #include <linux/cpumask.h>
@@ -29,6 +28,7 @@ struct mm_struct;
 #include <linux/threads.h>
 #include <linux/math64.h>
 #include <linux/init.h>
+#include <linux/err.h>
 
 #define HBP_NUM 4
 /*
@@ -473,10 +473,6 @@ struct thread_struct {
 	unsigned long		iopl;
 	/* Max allowed port in the bitmap, in bytes: */
 	unsigned		io_bitmap_max;
-/* MSR_IA32_DEBUGCTLMSR value to switch in if TIF_DEBUGCTLMSR is set.  */
-	unsigned long	debugctlmsr;
-	/* Debug Store context; see asm/ds.h */
-	struct ds_context	*ds_ctx;
 };
 
 static inline unsigned long native_get_debugreg(int regno)
@@ -803,7 +799,7 @@ extern void cpu_init(void);
 
 static inline unsigned long get_debugctlmsr(void)
 {
-    unsigned long debugctlmsr = 0;
+	unsigned long debugctlmsr = 0;
 
 #ifndef CONFIG_X86_DEBUGCTLMSR
 	if (boot_cpu_data.x86 < 6)
@@ -811,21 +807,6 @@ static inline unsigned long get_debugctlmsr(void)
 #endif
 	rdmsrl(MSR_IA32_DEBUGCTLMSR, debugctlmsr);
 
-    return debugctlmsr;
-}
-
-static inline unsigned long get_debugctlmsr_on_cpu(int cpu)
-{
-	u64 debugctlmsr = 0;
-	u32 val1, val2;
-
-#ifndef CONFIG_X86_DEBUGCTLMSR
-	if (boot_cpu_data.x86 < 6)
-		return 0;
-#endif
-	rdmsr_on_cpu(cpu, MSR_IA32_DEBUGCTLMSR, &val1, &val2);
-	debugctlmsr = val1 | ((u64)val2 << 32);
-
 	return debugctlmsr;
 }
 
@@ -838,18 +819,6 @@ static inline void update_debugctlmsr(unsigned long debugctlmsr)
 	wrmsrl(MSR_IA32_DEBUGCTLMSR, debugctlmsr);
 }
 
-static inline void update_debugctlmsr_on_cpu(int cpu,
-					     unsigned long debugctlmsr)
-{
-#ifndef CONFIG_X86_DEBUGCTLMSR
-	if (boot_cpu_data.x86 < 6)
-		return;
-#endif
-	wrmsr_on_cpu(cpu, MSR_IA32_DEBUGCTLMSR,
-		     (u32)((u64)debugctlmsr),
-		     (u32)((u64)debugctlmsr >> 32));
-}
-
 /*
  * from system description table in BIOS. Mostly for MCA use, but
  * others may find it useful:

arch/x86/include/asm/ptrace-abi.h

@@ -82,61 +82,6 @@
 
 #ifndef __ASSEMBLY__
 #include <linux/types.h>
-
-/* configuration/status structure used in PTRACE_BTS_CONFIG and
-   PTRACE_BTS_STATUS commands.
-*/
-struct ptrace_bts_config {
-	/* requested or actual size of BTS buffer in bytes */
-	__u32 size;
-	/* bitmask of below flags */
-	__u32 flags;
-	/* buffer overflow signal */
-	__u32 signal;
-	/* actual size of bts_struct in bytes */
-	__u32 bts_size;
-};
-#endif /* __ASSEMBLY__ */
-
-#define PTRACE_BTS_O_TRACE	0x1 /* branch trace */
-#define PTRACE_BTS_O_SCHED	0x2 /* scheduling events w/ jiffies */
-#define PTRACE_BTS_O_SIGNAL     0x4 /* send SIG<signal> on buffer overflow
-				       instead of wrapping around */
-#define PTRACE_BTS_O_ALLOC	0x8 /* (re)allocate buffer */
-
-#define PTRACE_BTS_CONFIG	40
-/* Configure branch trace recording.
-   ADDR points to a struct ptrace_bts_config.
-   DATA gives the size of that buffer.
-   A new buffer is allocated, if requested in the flags.
-   An overflow signal may only be requested for new buffers.
-   Returns the number of bytes read.
-*/
-#define PTRACE_BTS_STATUS	41
-/* Return the current configuration in a struct ptrace_bts_config
-   pointed to by ADDR; DATA gives the size of that buffer.
-   Returns the number of bytes written.
-*/
-#define PTRACE_BTS_SIZE		42
-/* Return the number of available BTS records for draining.
-   DATA and ADDR are ignored.
-*/
-#define PTRACE_BTS_GET		43
-/* Get a single BTS record.
-   DATA defines the index into the BTS array, where 0 is the newest
-   entry, and higher indices refer to older entries.
-   ADDR is pointing to struct bts_struct (see asm/ds.h).
-*/
-#define PTRACE_BTS_CLEAR	44
-/* Clear the BTS buffer.
-   DATA and ADDR are ignored.
-*/
-#define PTRACE_BTS_DRAIN	45
-/* Read all available BTS records and clear the buffer.
-   ADDR points to an array of struct bts_struct.
-   DATA gives the size of that buffer.
-   BTS records are read from oldest to newest.
-   Returns number of BTS records drained.
-*/
+#endif
 
 #endif /* _ASM_X86_PTRACE_ABI_H */

arch/x86/include/asm/ptrace.h

@@ -289,12 +289,6 @@ extern int do_get_thread_area(struct task_struct *p, int idx,
 extern int do_set_thread_area(struct task_struct *p, int idx,
 			      struct user_desc __user *info, int can_allocate);
 
-#ifdef CONFIG_X86_PTRACE_BTS
-extern void ptrace_bts_untrace(struct task_struct *tsk);
-
-#define arch_ptrace_untrace(tsk)	ptrace_bts_untrace(tsk)
-#endif /* CONFIG_X86_PTRACE_BTS */
-
 #endif /* __KERNEL__ */
 
 #endif /* !__ASSEMBLY__ */

arch/x86/include/asm/thread_info.h

@@ -92,8 +92,7 @@ struct thread_info {
 #define TIF_IO_BITMAP		22	/* uses I/O bitmap */
 #define TIF_FREEZE		23	/* is freezing for suspend */
 #define TIF_FORCED_TF		24	/* true if TF in eflags artificially */
-#define TIF_DEBUGCTLMSR		25	/* uses thread_struct.debugctlmsr */
-#define TIF_DS_AREA_MSR		26      /* uses thread_struct.ds_area_msr */
+#define TIF_BLOCKSTEP		25	/* set when we want DEBUGCTLMSR_BTF */
 #define TIF_LAZY_MMU_UPDATES	27	/* task is updating the mmu lazily */
 #define TIF_SYSCALL_TRACEPOINT	28	/* syscall tracepoint instrumentation */
 
@@ -115,8 +114,7 @@ struct thread_info {
 #define _TIF_IO_BITMAP		(1 << TIF_IO_BITMAP)
 #define _TIF_FREEZE		(1 << TIF_FREEZE)
 #define _TIF_FORCED_TF		(1 << TIF_FORCED_TF)
-#define _TIF_DEBUGCTLMSR	(1 << TIF_DEBUGCTLMSR)
-#define _TIF_DS_AREA_MSR	(1 << TIF_DS_AREA_MSR)
+#define _TIF_BLOCKSTEP		(1 << TIF_BLOCKSTEP)
 #define _TIF_LAZY_MMU_UPDATES	(1 << TIF_LAZY_MMU_UPDATES)
 #define _TIF_SYSCALL_TRACEPOINT	(1 << TIF_SYSCALL_TRACEPOINT)
 
@@ -147,7 +145,7 @@ struct thread_info {
 
 /* flags to check in __switch_to() */
 #define _TIF_WORK_CTXSW							\
-	(_TIF_IO_BITMAP|_TIF_DEBUGCTLMSR|_TIF_DS_AREA_MSR|_TIF_NOTSC)
+	(_TIF_IO_BITMAP|_TIF_NOTSC|_TIF_BLOCKSTEP)
 
 #define _TIF_WORK_CTXSW_PREV (_TIF_WORK_CTXSW|_TIF_USER_RETURN_NOTIFY)
 #define _TIF_WORK_CTXSW_NEXT (_TIF_WORK_CTXSW|_TIF_DEBUG)

arch/x86/kernel/Makefile

@@ -47,8 +47,6 @@ obj-$(CONFIG_X86_TRAMPOLINE)	+= trampoline.o
 obj-y				+= process.o
 obj-y				+= i387.o xsave.o
 obj-y				+= ptrace.o
-obj-$(CONFIG_X86_DS)		+= ds.o
-obj-$(CONFIG_X86_DS_SELFTEST)		+= ds_selftest.o
 obj-$(CONFIG_X86_32)		+= tls.o
 obj-$(CONFIG_IA32_EMULATION)	+= tls.o
 obj-y				+= step.o

arch/x86/kernel/cpu/intel.c

@@ -12,7 +12,6 @@
 #include <asm/processor.h>
 #include <asm/pgtable.h>
 #include <asm/msr.h>
-#include <asm/ds.h>
 #include <asm/bugs.h>
 #include <asm/cpu.h>
 
@@ -367,7 +366,6 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 			set_cpu_cap(c, X86_FEATURE_BTS);
 		if (!(l1 & (1<<12)))
 			set_cpu_cap(c, X86_FEATURE_PEBS);
-		ds_init_intel(c);
 	}
 
 	if (c->x86 == 6 && c->x86_model == 29 && cpu_has_clflush)

arch/x86/kernel/cpu/perf_event.c

@@ -31,46 +31,51 @@
 #include <asm/nmi.h>
 #include <asm/compat.h>
 
-static u64 perf_event_mask __read_mostly;
+#if 0
+#undef wrmsrl
+#define wrmsrl(msr, val) 					\
+do {								\
+	trace_printk("wrmsrl(%lx, %lx)\n", (unsigned long)(msr),\
+			(unsigned long)(val));			\
+	native_write_msr((msr), (u32)((u64)(val)), 		\
+			(u32)((u64)(val) >> 32));		\
+} while (0)
+#endif
 
-/* The maximal number of PEBS events: */
-#define MAX_PEBS_EVENTS	4
+/*
+ * best effort, GUP based copy_from_user() that assumes IRQ or NMI context
+ */
+static unsigned long
+copy_from_user_nmi(void *to, const void __user *from, unsigned long n)
+{
+	unsigned long offset, addr = (unsigned long)from;
+	int type = in_nmi() ? KM_NMI : KM_IRQ0;
+	unsigned long size, len = 0;
+	struct page *page;
+	void *map;
+	int ret;
 
-/* The size of a BTS record in bytes: */
-#define BTS_RECORD_SIZE		24
+	do {
+		ret = __get_user_pages_fast(addr, 1, 0, &page);
+		if (!ret)
+			break;
 
-/* The size of a per-cpu BTS buffer in bytes: */
-#define BTS_BUFFER_SIZE		(BTS_RECORD_SIZE * 2048)
+		offset = addr & (PAGE_SIZE - 1);
+		size = min(PAGE_SIZE - offset, n - len);
 
-/* The BTS overflow threshold in bytes from the end of the buffer: */
-#define BTS_OVFL_TH		(BTS_RECORD_SIZE * 128)
+		map = kmap_atomic(page, type);
+		memcpy(to, map+offset, size);
+		kunmap_atomic(map, type);
+		put_page(page);
 
+		len  += size;
+		to   += size;
+		addr += size;
 
-/*
- * 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)
+	} while (len < n);
 
-/*
- * 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_event_reset[MAX_PEBS_EVENTS];
-};
+	return len;
+}
 
 struct event_constraint {
 	union {
@@ -89,18 +94,39 @@ struct amd_nb {
 	struct event_constraint event_constraints[X86_PMC_IDX_MAX];
 };
 
+#define MAX_LBR_ENTRIES		16
+
 struct cpu_hw_events {
+	/*
+	 * Generic x86 PMC bits
+	 */
 	struct perf_event	*events[X86_PMC_IDX_MAX]; /* in counter order */
 	unsigned long		active_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
-	unsigned long		interrupts;
 	int			enabled;
-	struct debug_store	*ds;
 
 	int			n_events;
 	int			n_added;
 	int			assign[X86_PMC_IDX_MAX]; /* event to counter assignment */
 	u64			tags[X86_PMC_IDX_MAX];
 	struct perf_event	*event_list[X86_PMC_IDX_MAX]; /* in enabled order */
+
+	/*
+	 * Intel DebugStore bits
+	 */
+	struct debug_store	*ds;
+	u64			pebs_enabled;
+
+	/*
+	 * Intel LBR bits
+	 */
+	int				lbr_users;
+	void				*lbr_context;
+	struct perf_branch_stack	lbr_stack;
+	struct perf_branch_entry	lbr_entries[MAX_LBR_ENTRIES];
+
+	/*
+	 * AMD specific bits
+	 */
 	struct amd_nb		*amd_nb;
 };
 
@@ -114,11 +140,31 @@ struct cpu_hw_events {
 #define EVENT_CONSTRAINT(c, n, m)	\
 	__EVENT_CONSTRAINT(c, n, m, HWEIGHT(n))
 
+/*
+ * Constraint on the Event code.
+ */
 #define INTEL_EVENT_CONSTRAINT(c, n)	\
-	EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVTSEL_MASK)
+	EVENT_CONSTRAINT(c, n, ARCH_PERFMON_EVENTSEL_EVENT)
 
+/*
+ * Constraint on the Event code + UMask + fixed-mask
+ *
+ * filter mask to validate fixed counter events.
+ * the following filters disqualify for fixed counters:
+ *  - inv
+ *  - edge
+ *  - cnt-mask
+ *  The other filters are supported by fixed counters.
+ *  The any-thread option is supported starting with v3.
+ */
 #define FIXED_EVENT_CONSTRAINT(c, n)	\
-	EVENT_CONSTRAINT(c, (1ULL << (32+n)), INTEL_ARCH_FIXED_MASK)
+	EVENT_CONSTRAINT(c, (1ULL << (32+n)), X86_RAW_EVENT_MASK)
+
+/*
+ * Constraint on the Event code + UMask
+ */
+#define PEBS_EVENT_CONSTRAINT(c, n)	\
+	EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVENT_MASK)
 
 #define EVENT_CONSTRAINT_END		\
 	EVENT_CONSTRAINT(0, 0, 0)
@@ -126,32 +172,43 @@ struct cpu_hw_events {
 #define for_each_event_constraint(e, c)	\
 	for ((e) = (c); (e)->cmask; (e)++)
 
+union perf_capabilities {
+	struct {
+		u64	lbr_format    : 6;
+		u64	pebs_trap     : 1;
+		u64	pebs_arch_reg : 1;
+		u64	pebs_format   : 4;
+		u64	smm_freeze    : 1;
+	};
+	u64	capabilities;
+};
+
 /*
  * struct x86_pmu - generic x86 pmu
  */
 struct x86_pmu {
+	/*
+	 * Generic x86 PMC bits
+	 */
 	const char	*name;
 	int		version;
 	int		(*handle_irq)(struct pt_regs *);
 	void		(*disable_all)(void);
-	void		(*enable_all)(void);
+	void		(*enable_all)(int added);
 	void		(*enable)(struct perf_event *);
 	void		(*disable)(struct perf_event *);
+	int		(*hw_config)(struct perf_event *event);
+	int		(*schedule_events)(struct cpu_hw_events *cpuc, int n, int *assign);
 	unsigned	eventsel;
 	unsigned	perfctr;
 	u64		(*event_map)(int);
-	u64		(*raw_event)(u64);
 	int		max_events;
-	int		num_events;
-	int		num_events_fixed;
-	int		event_bits;
-	u64		event_mask;
+	int		num_counters;
+	int		num_counters_fixed;
+	int		cntval_bits;
+	u64		cntval_mask;
 	int		apic;
 	u64		max_period;
-	u64		intel_ctrl;
-	void		(*enable_bts)(u64 config);
-	void		(*disable_bts)(void);
-
 	struct event_constraint *
 			(*get_event_constraints)(struct cpu_hw_events *cpuc,
 						 struct perf_event *event);
@@ -159,11 +216,32 @@ struct x86_pmu {
 	void		(*put_event_constraints)(struct cpu_hw_events *cpuc,
 						 struct perf_event *event);
 	struct event_constraint *event_constraints;
+	void		(*quirks)(void);
 
 	int		(*cpu_prepare)(int cpu);
 	void		(*cpu_starting)(int cpu);
 	void		(*cpu_dying)(int cpu);
 	void		(*cpu_dead)(int cpu);
+
+	/*
+	 * Intel Arch Perfmon v2+
+	 */
+	u64			intel_ctrl;
+	union perf_capabilities intel_cap;
+
+	/*
+	 * Intel DebugStore bits
+	 */
+	int		bts, pebs;
+	int		pebs_record_size;
+	void		(*drain_pebs)(struct pt_regs *regs);
+	struct event_constraint *pebs_constraints;
+
+	/*
+	 * Intel LBR
+	 */
+	unsigned long	lbr_tos, lbr_from, lbr_to; /* MSR base regs       */
+	int		lbr_nr;			   /* hardware stack size */
 };
 
 static struct x86_pmu x86_pmu __read_mostly;
@@ -198,7 +276,7 @@ static u64
 x86_perf_event_update(struct perf_event *event)
 {
 	struct hw_perf_event *hwc = &event->hw;
-	int shift = 64 - x86_pmu.event_bits;
+	int shift = 64 - x86_pmu.cntval_bits;
 	u64 prev_raw_count, new_raw_count;
 	int idx = hwc->idx;
 	s64 delta;
@@ -241,33 +319,32 @@ again:
 static atomic_t active_events;
 static DEFINE_MUTEX(pmc_reserve_mutex);
 
+#ifdef CONFIG_X86_LOCAL_APIC
+
 static bool reserve_pmc_hardware(void)
 {
-#ifdef CONFIG_X86_LOCAL_APIC
 	int i;
 
 	if (nmi_watchdog == NMI_LOCAL_APIC)
 		disable_lapic_nmi_watchdog();
 
-	for (i = 0; i < x86_pmu.num_events; i++) {
+	for (i = 0; i < x86_pmu.num_counters; i++) {
 		if (!reserve_perfctr_nmi(x86_pmu.perfctr + i))
 			goto perfctr_fail;
 	}
 
-	for (i = 0; i < x86_pmu.num_events; i++) {
+	for (i = 0; i < x86_pmu.num_counters; i++) {
 		if (!reserve_evntsel_nmi(x86_pmu.eventsel + i))
 			goto eventsel_fail;
 	}
-#endif
 
 	return true;
 
-#ifdef CONFIG_X86_LOCAL_APIC
 eventsel_fail:
 	for (i--; i >= 0; i--)
 		release_evntsel_nmi(x86_pmu.eventsel + i);
 
-	i = x86_pmu.num_events;
+	i = x86_pmu.num_counters;
 
 perfctr_fail:
 	for (i--; i >= 0; i--)
@@ -277,128 +354,36 @@ perfctr_fail:
 		enable_lapic_nmi_watchdog();
 
 	return false;
-#endif
 }
 
 static void release_pmc_hardware(void)
 {
-#ifdef CONFIG_X86_LOCAL_APIC
 	int i;
 
-	for (i = 0; i < x86_pmu.num_events; i++) {
+	for (i = 0; i < x86_pmu.num_counters; i++) {
 		release_perfctr_nmi(x86_pmu.perfctr + i);
 		release_evntsel_nmi(x86_pmu.eventsel + i);
 	}
 
 	if (nmi_watchdog == NMI_LOCAL_APIC)
 		enable_lapic_nmi_watchdog();
-#endif
-}
-
-static inline bool bts_available(void)
-{
-	return x86_pmu.enable_bts != NULL;
-}
-
-static void init_debug_store_on_cpu(int cpu)
-{
-	struct debug_store *ds = per_cpu(cpu_hw_events, cpu).ds;
-
-	if (!ds)
-		return;
-
-	wrmsr_on_cpu(cpu, MSR_IA32_DS_AREA,
-		     (u32)((u64)(unsigned long)ds),
-		     (u32)((u64)(unsigned long)ds >> 32));
-}
-
-static void fini_debug_store_on_cpu(int cpu)
-{
-	if (!per_cpu(cpu_hw_events, 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_events, cpu).ds;
-
-		if (!ds)
-			continue;
-
-		per_cpu(cpu_hw_events, cpu).ds = NULL;
-
-		kfree((void *)(unsigned long)ds->bts_buffer_base);
-		kfree(ds);
-	}
-
-	put_online_cpus();
 }
 
-static int reserve_bts_hardware(void)
-{
-	int cpu, err = 0;
-
-	if (!bts_available())
-		return 0;
-
-	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)(unsigned 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;
+#else
 
-		per_cpu(cpu_hw_events, cpu).ds = ds;
-		err = 0;
-	}
+static bool reserve_pmc_hardware(void) { return true; }
+static void release_pmc_hardware(void) {}
 
-	if (err)
-		release_bts_hardware();
-	else {
-		for_each_online_cpu(cpu)
-			init_debug_store_on_cpu(cpu);
-	}
-
-	put_online_cpus();
+#endif
 
-	return err;
-}
+static int reserve_ds_buffers(void);
+static void release_ds_buffers(void);
 
 static void hw_perf_event_destroy(struct perf_event *event)
 {
 	if (atomic_dec_and_mutex_lock(&active_events, &pmc_reserve_mutex)) {
 		release_pmc_hardware();
-		release_bts_hardware();
+		release_ds_buffers();
 		mutex_unlock(&pmc_reserve_mutex);
 	}
 }
@@ -441,6 +426,28 @@ set_ext_hw_attr(struct hw_perf_event *hwc, struct perf_event_attr *attr)
 	return 0;
 }
 
+static int x86_pmu_hw_config(struct perf_event *event)
+{
+	/*
+	 * Generate PMC IRQs:
+	 * (keep 'enabled' bit clear for now)
+	 */
+	event->hw.config = ARCH_PERFMON_EVENTSEL_INT;
+
+	/*
+	 * Count user and OS events unless requested not to
+	 */
+	if (!event->attr.exclude_user)
+		event->hw.config |= ARCH_PERFMON_EVENTSEL_USR;
+	if (!event->attr.exclude_kernel)
+		event->hw.config |= ARCH_PERFMON_EVENTSEL_OS;
+
+	if (event->attr.type == PERF_TYPE_RAW)
+		event->hw.config |= event->attr.config & X86_RAW_EVENT_MASK;
+
+	return 0;
+}
+
 /*
  * Setup the hardware configuration for a given attr_type
  */
@@ -460,8 +467,11 @@ static int __hw_perf_event_init(struct perf_event *event)
 		if (atomic_read(&active_events) == 0) {
 			if (!reserve_pmc_hardware())
 				err = -EBUSY;
-			else
-				err = reserve_bts_hardware();
+			else {
+				err = reserve_ds_buffers();
+				if (err)
+					release_pmc_hardware();
+			}
 		}
 		if (!err)
 			atomic_inc(&active_events);
@@ -472,23 +482,14 @@ static int __hw_perf_event_init(struct perf_event *event)
 
 	event->destroy = hw_perf_event_destroy;
 
-	/*
-	 * Generate PMC IRQs:
-	 * (keep 'enabled' bit clear for now)
-	 */
-	hwc->config = ARCH_PERFMON_EVENTSEL_INT;
-
 	hwc->idx = -1;
 	hwc->last_cpu = -1;
 	hwc->last_tag = ~0ULL;
 
-	/*
-	 * Count user and OS events unless requested not to.
-	 */
-	if (!attr->exclude_user)
-		hwc->config |= ARCH_PERFMON_EVENTSEL_USR;
-	if (!attr->exclude_kernel)
-		hwc->config |= ARCH_PERFMON_EVENTSEL_OS;
+	/* Processor specifics */
+	err = x86_pmu.hw_config(event);
+	if (err)
+		return err;
 
 	if (!hwc->sample_period) {
 		hwc->sample_period = x86_pmu.max_period;
@@ -505,16 +506,8 @@ static int __hw_perf_event_init(struct perf_event *event)
 			return -EOPNOTSUPP;
 	}
 
-	/*
-	 * Raw hw_event type provide the config in the hw_event structure
-	 */
-	if (attr->type == PERF_TYPE_RAW) {
-		hwc->config |= x86_pmu.raw_event(attr->config);
-		if ((hwc->config & ARCH_PERFMON_EVENTSEL_ANY) &&
-		    perf_paranoid_cpu() && !capable(CAP_SYS_ADMIN))
-			return -EACCES;
+	if (attr->type == PERF_TYPE_RAW)
 		return 0;
-	}
 
 	if (attr->type == PERF_TYPE_HW_CACHE)
 		return set_ext_hw_attr(hwc, attr);
@@ -539,11 +532,11 @@ static int __hw_perf_event_init(struct perf_event *event)
 	if ((attr->config == PERF_COUNT_HW_BRANCH_INSTRUCTIONS) &&
 	    (hwc->sample_period == 1)) {
 		/* BTS is not supported by this architecture. */
-		if (!bts_available())
+		if (!x86_pmu.bts)
 			return -EOPNOTSUPP;
 
 		/* BTS is currently only allowed for user-mode. */
-		if (hwc->config & ARCH_PERFMON_EVENTSEL_OS)
+		if (!attr->exclude_kernel)
 			return -EOPNOTSUPP;
 	}
 
@@ -557,7 +550,7 @@ static void x86_pmu_disable_all(void)
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	int idx;
 
-	for (idx = 0; idx < x86_pmu.num_events; idx++) {
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
 		u64 val;
 
 		if (!test_bit(idx, cpuc->active_mask))
@@ -587,12 +580,12 @@ void hw_perf_disable(void)
 	x86_pmu.disable_all();
 }
 
-static void x86_pmu_enable_all(void)
+static void x86_pmu_enable_all(int added)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	int idx;
 
-	for (idx = 0; idx < x86_pmu.num_events; idx++) {
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
 		struct perf_event *event = cpuc->events[idx];
 		u64 val;
 
@@ -667,14 +660,14 @@ static int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign)
 	 * assign events to counters starting with most
 	 * constrained events.
 	 */
-	wmax = x86_pmu.num_events;
+	wmax = x86_pmu.num_counters;
 
 	/*
 	 * when fixed event counters are present,
 	 * wmax is incremented by 1 to account
 	 * for one more choice
 	 */
-	if (x86_pmu.num_events_fixed)
+	if (x86_pmu.num_counters_fixed)
 		wmax++;
 
 	for (w = 1, num = n; num && w <= wmax; w++) {
@@ -724,7 +717,7 @@ static int collect_events(struct cpu_hw_events *cpuc, struct perf_event *leader,
 	struct perf_event *event;
 	int n, max_count;
 
-	max_count = x86_pmu.num_events + x86_pmu.num_events_fixed;
+	max_count = x86_pmu.num_counters + x86_pmu.num_counters_fixed;
 
 	/* current number of events already accepted */
 	n = cpuc->n_events;
@@ -795,7 +788,7 @@ void hw_perf_enable(void)
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct perf_event *event;
 	struct hw_perf_event *hwc;
-	int i;
+	int i, added = cpuc->n_added;
 
 	if (!x86_pmu_initialized())
 		return;
@@ -847,19 +840,20 @@ void hw_perf_enable(void)
 	cpuc->enabled = 1;
 	barrier();
 
-	x86_pmu.enable_all();
+	x86_pmu.enable_all(added);
 }
 
 static inline void __x86_pmu_enable_event(struct hw_perf_event *hwc)
 {
-	(void)checking_wrmsrl(hwc->config_base + hwc->idx,
+	wrmsrl(hwc->config_base + hwc->idx,
 			      hwc->config | ARCH_PERFMON_EVENTSEL_ENABLE);
 }
 
 static inline void x86_pmu_disable_event(struct perf_event *event)
 {
 	struct hw_perf_event *hwc = &event->hw;
-	(void)checking_wrmsrl(hwc->config_base + hwc->idx, hwc->config);
+
+	wrmsrl(hwc->config_base + hwc->idx, hwc->config);
 }
 
 static DEFINE_PER_CPU(u64 [X86_PMC_IDX_MAX], pmc_prev_left);
@@ -874,7 +868,7 @@ x86_perf_event_set_period(struct perf_event *event)
 	struct hw_perf_event *hwc = &event->hw;
 	s64 left = atomic64_read(&hwc->period_left);
 	s64 period = hwc->sample_period;
-	int err, ret = 0, idx = hwc->idx;
+	int ret = 0, idx = hwc->idx;
 
 	if (idx == X86_PMC_IDX_FIXED_BTS)
 		return 0;
@@ -912,8 +906,8 @@ x86_perf_event_set_period(struct perf_event *event)
 	 */
 	atomic64_set(&hwc->prev_count, (u64)-left);
 
-	err = checking_wrmsrl(hwc->event_base + idx,
-			     (u64)(-left) & x86_pmu.event_mask);
+	wrmsrl(hwc->event_base + idx,
+			(u64)(-left) & x86_pmu.cntval_mask);
 
 	perf_event_update_userpage(event);
 
@@ -950,7 +944,7 @@ static int x86_pmu_enable(struct perf_event *event)
 	if (n < 0)
 		return n;
 
-	ret = x86_schedule_events(cpuc, n, assign);
+	ret = x86_pmu.schedule_events(cpuc, n, assign);
 	if (ret)
 		return ret;
 	/*
@@ -991,11 +985,12 @@ static void x86_pmu_unthrottle(struct perf_event *event)
 void perf_event_print_debug(void)
 {
 	u64 ctrl, status, overflow, pmc_ctrl, pmc_count, prev_left, fixed;
+	u64 pebs;
 	struct cpu_hw_events *cpuc;
 	unsigned long flags;
 	int cpu, idx;
 
-	if (!x86_pmu.num_events)
+	if (!x86_pmu.num_counters)
 		return;
 
 	local_irq_save(flags);
@@ -1008,16 +1003,18 @@ void perf_event_print_debug(void)
 		rdmsrl(MSR_CORE_PERF_GLOBAL_STATUS, status);
 		rdmsrl(MSR_CORE_PERF_GLOBAL_OVF_CTRL, overflow);
 		rdmsrl(MSR_ARCH_PERFMON_FIXED_CTR_CTRL, fixed);
+		rdmsrl(MSR_IA32_PEBS_ENABLE, pebs);
 
 		pr_info("\n");
 		pr_info("CPU#%d: ctrl:       %016llx\n", cpu, ctrl);
 		pr_info("CPU#%d: status:     %016llx\n", cpu, status);
 		pr_info("CPU#%d: overflow:   %016llx\n", cpu, overflow);
 		pr_info("CPU#%d: fixed:      %016llx\n", cpu, fixed);
+		pr_info("CPU#%d: pebs:       %016llx\n", cpu, pebs);
 	}
-	pr_info("CPU#%d: active:       %016llx\n", cpu, *(u64 *)cpuc->active_mask);
+	pr_info("CPU#%d: active:     %016llx\n", cpu, *(u64 *)cpuc->active_mask);
 
-	for (idx = 0; idx < x86_pmu.num_events; idx++) {
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
 		rdmsrl(x86_pmu.eventsel + idx, pmc_ctrl);
 		rdmsrl(x86_pmu.perfctr  + idx, pmc_count);
 
@@ -1030,7 +1027,7 @@ void perf_event_print_debug(void)
 		pr_info("CPU#%d:   gen-PMC%d left:  %016llx\n",
 			cpu, idx, prev_left);
 	}
-	for (idx = 0; idx < x86_pmu.num_events_fixed; idx++) {
+	for (idx = 0; idx < x86_pmu.num_counters_fixed; idx++) {
 		rdmsrl(MSR_ARCH_PERFMON_FIXED_CTR0 + idx, pmc_count);
 
 		pr_info("CPU#%d: fixed-PMC%d count: %016llx\n",
@@ -1095,7 +1092,7 @@ static int x86_pmu_handle_irq(struct pt_regs *regs)
 
 	cpuc = &__get_cpu_var(cpu_hw_events);
 
-	for (idx = 0; idx < x86_pmu.num_events; idx++) {
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
 		if (!test_bit(idx, cpuc->active_mask))
 			continue;
 
@@ -1103,7 +1100,7 @@ static int x86_pmu_handle_irq(struct pt_regs *regs)
 		hwc = &event->hw;
 
 		val = x86_perf_event_update(event);
-		if (val & (1ULL << (x86_pmu.event_bits - 1)))
+		if (val & (1ULL << (x86_pmu.cntval_bits - 1)))
 			continue;
 
 		/*
@@ -1146,7 +1143,6 @@ void set_perf_event_pending(void)
 
 void perf_events_lapic_init(void)
 {
-#ifdef CONFIG_X86_LOCAL_APIC
 	if (!x86_pmu.apic || !x86_pmu_initialized())
 		return;
 
@@ -1154,7 +1150,6 @@ void perf_events_lapic_init(void)
 	 * Always use NMI for PMU
 	 */
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
-#endif
 }
 
 static int __kprobes
@@ -1178,9 +1173,7 @@ perf_event_nmi_handler(struct notifier_block *self,
 
 	regs = args->regs;
 
-#ifdef CONFIG_X86_LOCAL_APIC
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
-#endif
 	/*
 	 * Can't rely on the handled return value to say it was our NMI, two
 	 * events could trigger 'simultaneously' raising two back-to-back NMIs.
@@ -1274,12 +1267,15 @@ int hw_perf_group_sched_in(struct perf_event *leader,
 	int assign[X86_PMC_IDX_MAX];
 	int n0, n1, ret;
 
+	if (!x86_pmu_initialized())
+		return 0;
+
 	/* n0 = total number of events */
 	n0 = collect_events(cpuc, leader, true);
 	if (n0 < 0)
 		return n0;
 
-	ret = x86_schedule_events(cpuc, n0, assign);
+	ret = x86_pmu.schedule_events(cpuc, n0, assign);
 	if (ret)
 		return ret;
 
@@ -1329,6 +1325,9 @@ undo:
 
 #include "perf_event_amd.c"
 #include "perf_event_p6.c"
+#include "perf_event_p4.c"
+#include "perf_event_intel_lbr.c"
+#include "perf_event_intel_ds.c"
 #include "perf_event_intel.c"
 
 static int __cpuinit
@@ -1402,48 +1401,50 @@ void __init init_hw_perf_events(void)
 
 	pr_cont("%s PMU driver.\n", x86_pmu.name);
 
-	if (x86_pmu.num_events > X86_PMC_MAX_GENERIC) {
+	if (x86_pmu.quirks)
+		x86_pmu.quirks();
+
+	if (x86_pmu.num_counters > X86_PMC_MAX_GENERIC) {
 		WARN(1, KERN_ERR "hw perf events %d > max(%d), clipping!",
-		     x86_pmu.num_events, X86_PMC_MAX_GENERIC);
-		x86_pmu.num_events = X86_PMC_MAX_GENERIC;
+		     x86_pmu.num_counters, X86_PMC_MAX_GENERIC);
+		x86_pmu.num_counters = X86_PMC_MAX_GENERIC;
 	}
-	perf_event_mask = (1 << x86_pmu.num_events) - 1;
-	perf_max_events = x86_pmu.num_events;
+	x86_pmu.intel_ctrl = (1 << x86_pmu.num_counters) - 1;
+	perf_max_events = x86_pmu.num_counters;
 
-	if (x86_pmu.num_events_fixed > X86_PMC_MAX_FIXED) {
+	if (x86_pmu.num_counters_fixed > X86_PMC_MAX_FIXED) {
 		WARN(1, KERN_ERR "hw perf events fixed %d > max(%d), clipping!",
-		     x86_pmu.num_events_fixed, X86_PMC_MAX_FIXED);
-		x86_pmu.num_events_fixed = X86_PMC_MAX_FIXED;
+		     x86_pmu.num_counters_fixed, X86_PMC_MAX_FIXED);
+		x86_pmu.num_counters_fixed = X86_PMC_MAX_FIXED;
 	}
 
-	perf_event_mask |=
-		((1LL << x86_pmu.num_events_fixed)-1) << X86_PMC_IDX_FIXED;
-	x86_pmu.intel_ctrl = perf_event_mask;
+	x86_pmu.intel_ctrl |=
+		((1LL << x86_pmu.num_counters_fixed)-1) << X86_PMC_IDX_FIXED;
 
 	perf_events_lapic_init();
 	register_die_notifier(&perf_event_nmi_notifier);
 
 	unconstrained = (struct event_constraint)
-		__EVENT_CONSTRAINT(0, (1ULL << x86_pmu.num_events) - 1,
-				   0, x86_pmu.num_events);
+		__EVENT_CONSTRAINT(0, (1ULL << x86_pmu.num_counters) - 1,
+				   0, x86_pmu.num_counters);
 
 	if (x86_pmu.event_constraints) {
 		for_each_event_constraint(c, x86_pmu.event_constraints) {
-			if (c->cmask != INTEL_ARCH_FIXED_MASK)
+			if (c->cmask != X86_RAW_EVENT_MASK)
 				continue;
 
-			c->idxmsk64 |= (1ULL << x86_pmu.num_events) - 1;
-			c->weight += x86_pmu.num_events;
+			c->idxmsk64 |= (1ULL << x86_pmu.num_counters) - 1;
+			c->weight += x86_pmu.num_counters;
 		}
 	}
 
 	pr_info("... version:                %d\n",     x86_pmu.version);
-	pr_info("... bit width:              %d\n",     x86_pmu.event_bits);
-	pr_info("... generic registers:      %d\n",     x86_pmu.num_events);
-	pr_info("... value mask:             %016Lx\n", x86_pmu.event_mask);
+	pr_info("... bit width:              %d\n",     x86_pmu.cntval_bits);
+	pr_info("... generic registers:      %d\n",     x86_pmu.num_counters);
+	pr_info("... value mask:             %016Lx\n", x86_pmu.cntval_mask);
 	pr_info("... max period:             %016Lx\n", x86_pmu.max_period);
-	pr_info("... fixed-purpose events:   %d\n",     x86_pmu.num_events_fixed);
-	pr_info("... event mask:             %016Lx\n", perf_event_mask);
+	pr_info("... fixed-purpose events:   %d\n",     x86_pmu.num_counters_fixed);
+	pr_info("... event mask:             %016Lx\n", x86_pmu.intel_ctrl);
 
 	perf_cpu_notifier(x86_pmu_notifier);
 }
@@ -1462,6 +1463,32 @@ static const struct pmu pmu = {
 	.unthrottle	= x86_pmu_unthrottle,
 };
 
+/*
+ * validate that we can schedule this event
+ */
+static int validate_event(struct perf_event *event)
+{
+	struct cpu_hw_events *fake_cpuc;
+	struct event_constraint *c;
+	int ret = 0;
+
+	fake_cpuc = kmalloc(sizeof(*fake_cpuc), GFP_KERNEL | __GFP_ZERO);
+	if (!fake_cpuc)
+		return -ENOMEM;
+
+	c = x86_pmu.get_event_constraints(fake_cpuc, event);
+
+	if (!c || !c->weight)
+		ret = -ENOSPC;
+
+	if (x86_pmu.put_event_constraints)
+		x86_pmu.put_event_constraints(fake_cpuc, event);
+
+	kfree(fake_cpuc);
+
+	return ret;
+}
+
 /*
  * validate a single event group
  *
@@ -1502,7 +1529,7 @@ static int validate_group(struct perf_event *event)
 
 	fake_cpuc->n_events = n;
 
-	ret = x86_schedule_events(fake_cpuc, n, NULL);
+	ret = x86_pmu.schedule_events(fake_cpuc, n, NULL);
 
 out_free:
 	kfree(fake_cpuc);
@@ -1527,6 +1554,8 @@ const struct pmu *hw_perf_event_init(struct perf_event *event)
 
 		if (event->group_leader != event)
 			err = validate_group(event);
+		else
+			err = validate_event(event);
 
 		event->pmu = tmp;
 	}
@@ -1574,8 +1603,7 @@ static void backtrace_address(void *data, unsigned long addr, int reliable)
 {
 	struct perf_callchain_entry *entry = data;
 
-	if (reliable)
-		callchain_store(entry, addr);
+	callchain_store(entry, addr);
 }
 
 static const struct stacktrace_ops backtrace_ops = {
@@ -1597,41 +1625,6 @@ perf_callchain_kernel(struct pt_regs *regs, struct perf_callchain_entry *entry)
 	dump_trace(NULL, regs, NULL, regs->bp, &backtrace_ops, entry);
 }
 
-/*
- * best effort, GUP based copy_from_user() that assumes IRQ or NMI context
- */
-static unsigned long
-copy_from_user_nmi(void *to, const void __user *from, unsigned long n)
-{
-	unsigned long offset, addr = (unsigned long)from;
-	int type = in_nmi() ? KM_NMI : KM_IRQ0;
-	unsigned long size, len = 0;
-	struct page *page;
-	void *map;
-	int ret;
-
-	do {
-		ret = __get_user_pages_fast(addr, 1, 0, &page);
-		if (!ret)
-			break;
-
-		offset = addr & (PAGE_SIZE - 1);
-		size = min(PAGE_SIZE - offset, n - len);
-
-		map = kmap_atomic(page, type);
-		memcpy(to, map+offset, size);
-		kunmap_atomic(map, type);
-		put_page(page);
-
-		len  += size;
-		to   += size;
-		addr += size;
-
-	} while (len < n);
-
-	return len;
-}
-
 #ifdef CONFIG_COMPAT
 static inline int
 perf_callchain_user32(struct pt_regs *regs, struct perf_callchain_entry *entry)
@@ -1727,6 +1720,11 @@ struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
 {
 	struct perf_callchain_entry *entry;
 
+	if (perf_guest_cbs && perf_guest_cbs->is_in_guest()) {
+		/* TODO: We don't support guest os callchain now */
+		return NULL;
+	}
+
 	if (in_nmi())
 		entry = &__get_cpu_var(pmc_nmi_entry);
 	else
@@ -1750,3 +1748,29 @@ void perf_arch_fetch_caller_regs(struct pt_regs *regs, unsigned long ip, int ski
 	regs->cs = __KERNEL_CS;
 	local_save_flags(regs->flags);
 }
+
+unsigned long perf_instruction_pointer(struct pt_regs *regs)
+{
+	unsigned long ip;
+	if (perf_guest_cbs && perf_guest_cbs->is_in_guest())
+		ip = perf_guest_cbs->get_guest_ip();
+	else
+		ip = instruction_pointer(regs);
+	return ip;
+}
+
+unsigned long perf_misc_flags(struct pt_regs *regs)
+{
+	int misc = 0;
+	if (perf_guest_cbs && perf_guest_cbs->is_in_guest()) {
+		misc |= perf_guest_cbs->is_user_mode() ?
+			PERF_RECORD_MISC_GUEST_USER :
+			PERF_RECORD_MISC_GUEST_KERNEL;
+	} else
+		misc |= user_mode(regs) ? PERF_RECORD_MISC_USER :
+			PERF_RECORD_MISC_KERNEL;
+	if (regs->flags & PERF_EFLAGS_EXACT)
+		misc |= PERF_RECORD_MISC_EXACT;
+
+	return misc;
+}

arch/x86/kernel/cpu/perf_event_amd.c

@@ -2,7 +2,7 @@
 
 static DEFINE_RAW_SPINLOCK(amd_nb_lock);
 
-static __initconst u64 amd_hw_cache_event_ids
+static __initconst const u64 amd_hw_cache_event_ids
 				[PERF_COUNT_HW_CACHE_MAX]
 				[PERF_COUNT_HW_CACHE_OP_MAX]
 				[PERF_COUNT_HW_CACHE_RESULT_MAX] =
@@ -111,22 +111,19 @@ static u64 amd_pmu_event_map(int hw_event)
 	return amd_perfmon_event_map[hw_event];
 }
 
-static u64 amd_pmu_raw_event(u64 hw_event)
+static int amd_pmu_hw_config(struct perf_event *event)
 {
-#define K7_EVNTSEL_EVENT_MASK	0xF000000FFULL
-#define K7_EVNTSEL_UNIT_MASK	0x00000FF00ULL
-#define K7_EVNTSEL_EDGE_MASK	0x000040000ULL
-#define K7_EVNTSEL_INV_MASK	0x000800000ULL
-#define K7_EVNTSEL_REG_MASK	0x0FF000000ULL
-
-#define K7_EVNTSEL_MASK			\
-	(K7_EVNTSEL_EVENT_MASK |	\
-	 K7_EVNTSEL_UNIT_MASK  |	\
-	 K7_EVNTSEL_EDGE_MASK  |	\
-	 K7_EVNTSEL_INV_MASK   |	\
-	 K7_EVNTSEL_REG_MASK)
-
-	return hw_event & K7_EVNTSEL_MASK;
+	int ret = x86_pmu_hw_config(event);
+
+	if (ret)
+		return ret;
+
+	if (event->attr.type != PERF_TYPE_RAW)
+		return 0;
+
+	event->hw.config |= event->attr.config & AMD64_RAW_EVENT_MASK;
+
+	return 0;
 }
 
 /*
@@ -165,7 +162,7 @@ static void amd_put_event_constraints(struct cpu_hw_events *cpuc,
 	 * be removed on one CPU at a time AND PMU is disabled
 	 * when we come here
 	 */
-	for (i = 0; i < x86_pmu.num_events; i++) {
+	for (i = 0; i < x86_pmu.num_counters; i++) {
 		if (nb->owners[i] == event) {
 			cmpxchg(nb->owners+i, event, NULL);
 			break;
@@ -215,7 +212,7 @@ amd_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event)
 	struct hw_perf_event *hwc = &event->hw;
 	struct amd_nb *nb = cpuc->amd_nb;
 	struct perf_event *old = NULL;
-	int max = x86_pmu.num_events;
+	int max = x86_pmu.num_counters;
 	int i, j, k = -1;
 
 	/*
@@ -293,7 +290,7 @@ static struct amd_nb *amd_alloc_nb(int cpu, int nb_id)
 	/*
 	 * initialize all possible NB constraints
 	 */
-	for (i = 0; i < x86_pmu.num_events; i++) {
+	for (i = 0; i < x86_pmu.num_counters; i++) {
 		__set_bit(i, nb->event_constraints[i].idxmsk);
 		nb->event_constraints[i].weight = 1;
 	}
@@ -371,21 +368,22 @@ static void amd_pmu_cpu_dead(int cpu)
 	raw_spin_unlock(&amd_nb_lock);
 }
 
-static __initconst struct x86_pmu amd_pmu = {
+static __initconst const struct x86_pmu amd_pmu = {
 	.name			= "AMD",
 	.handle_irq		= x86_pmu_handle_irq,
 	.disable_all		= x86_pmu_disable_all,
 	.enable_all		= x86_pmu_enable_all,
 	.enable			= x86_pmu_enable_event,
 	.disable		= x86_pmu_disable_event,
+	.hw_config		= amd_pmu_hw_config,
+	.schedule_events	= x86_schedule_events,
 	.eventsel		= MSR_K7_EVNTSEL0,
 	.perfctr		= MSR_K7_PERFCTR0,
 	.event_map		= amd_pmu_event_map,
-	.raw_event		= amd_pmu_raw_event,
 	.max_events		= ARRAY_SIZE(amd_perfmon_event_map),
-	.num_events		= 4,
-	.event_bits		= 48,
-	.event_mask		= (1ULL << 48) - 1,
+	.num_counters		= 4,
+	.cntval_bits		= 48,
+	.cntval_mask		= (1ULL << 48) - 1,
 	.apic			= 1,
 	/* use highest bit to detect overflow */
 	.max_period		= (1ULL << 47) - 1,

arch/x86/kernel/cpu/perf_event_intel.c

@@ -88,7 +88,7 @@ static u64 intel_pmu_event_map(int hw_event)
 	return intel_perfmon_event_map[hw_event];
 }
 
-static __initconst u64 westmere_hw_cache_event_ids
+static __initconst const u64 westmere_hw_cache_event_ids
 				[PERF_COUNT_HW_CACHE_MAX]
 				[PERF_COUNT_HW_CACHE_OP_MAX]
 				[PERF_COUNT_HW_CACHE_RESULT_MAX] =
@@ -179,7 +179,7 @@ static __initconst u64 westmere_hw_cache_event_ids
  },
 };
 
-static __initconst u64 nehalem_hw_cache_event_ids
+static __initconst const u64 nehalem_hw_cache_event_ids
 				[PERF_COUNT_HW_CACHE_MAX]
 				[PERF_COUNT_HW_CACHE_OP_MAX]
 				[PERF_COUNT_HW_CACHE_RESULT_MAX] =
@@ -270,7 +270,7 @@ static __initconst u64 nehalem_hw_cache_event_ids
  },
 };
 
-static __initconst u64 core2_hw_cache_event_ids
+static __initconst const u64 core2_hw_cache_event_ids
 				[PERF_COUNT_HW_CACHE_MAX]
 				[PERF_COUNT_HW_CACHE_OP_MAX]
 				[PERF_COUNT_HW_CACHE_RESULT_MAX] =
@@ -361,7 +361,7 @@ static __initconst u64 core2_hw_cache_event_ids
  },
 };
 
-static __initconst u64 atom_hw_cache_event_ids
+static __initconst const u64 atom_hw_cache_event_ids
 				[PERF_COUNT_HW_CACHE_MAX]
 				[PERF_COUNT_HW_CACHE_OP_MAX]
 				[PERF_COUNT_HW_CACHE_RESULT_MAX] =
@@ -452,60 +452,6 @@ static __initconst u64 atom_hw_cache_event_ids
  },
 };
 
-static u64 intel_pmu_raw_event(u64 hw_event)
-{
-#define CORE_EVNTSEL_EVENT_MASK		0x000000FFULL
-#define CORE_EVNTSEL_UNIT_MASK		0x0000FF00ULL
-#define CORE_EVNTSEL_EDGE_MASK		0x00040000ULL
-#define CORE_EVNTSEL_INV_MASK		0x00800000ULL
-#define CORE_EVNTSEL_REG_MASK		0xFF000000ULL
-
-#define CORE_EVNTSEL_MASK		\
-	(INTEL_ARCH_EVTSEL_MASK |	\
-	 INTEL_ARCH_UNIT_MASK   |	\
-	 INTEL_ARCH_EDGE_MASK   |	\
-	 INTEL_ARCH_INV_MASK    |	\
-	 INTEL_ARCH_CNT_MASK)
-
-	return hw_event & CORE_EVNTSEL_MASK;
-}
-
-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_events *cpuc = &__get_cpu_var(cpu_hw_events);
-	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);
-}
-
 static void intel_pmu_disable_all(void)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
@@ -514,12 +460,17 @@ static void intel_pmu_disable_all(void)
 
 	if (test_bit(X86_PMC_IDX_FIXED_BTS, cpuc->active_mask))
 		intel_pmu_disable_bts();
+
+	intel_pmu_pebs_disable_all();
+	intel_pmu_lbr_disable_all();
 }
 
-static void intel_pmu_enable_all(void)
+static void intel_pmu_enable_all(int added)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
+	intel_pmu_pebs_enable_all();
+	intel_pmu_lbr_enable_all();
 	wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, x86_pmu.intel_ctrl);
 
 	if (test_bit(X86_PMC_IDX_FIXED_BTS, cpuc->active_mask)) {
@@ -533,6 +484,41 @@ static void intel_pmu_enable_all(void)
 	}
 }
 
+/*
+ * Workaround for:
+ *   Intel Errata AAK100 (model 26)
+ *   Intel Errata AAP53  (model 30)
+ *   Intel Errata BD53   (model 44)
+ *
+ * These chips need to be 'reset' when adding counters by programming
+ * the magic three (non counting) events 0x4300D2, 0x4300B1 and 0x4300B5
+ * either in sequence on the same PMC or on different PMCs.
+ */
+static void intel_pmu_nhm_enable_all(int added)
+{
+	if (added) {
+		struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+		int i;
+
+		wrmsrl(MSR_ARCH_PERFMON_EVENTSEL0 + 0, 0x4300D2);
+		wrmsrl(MSR_ARCH_PERFMON_EVENTSEL0 + 1, 0x4300B1);
+		wrmsrl(MSR_ARCH_PERFMON_EVENTSEL0 + 2, 0x4300B5);
+
+		wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0x3);
+		wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0x0);
+
+		for (i = 0; i < 3; i++) {
+			struct perf_event *event = cpuc->events[i];
+
+			if (!event)
+				continue;
+
+			__x86_pmu_enable_event(&event->hw);
+		}
+	}
+	intel_pmu_enable_all(added);
+}
+
 static inline u64 intel_pmu_get_status(void)
 {
 	u64 status;
@@ -547,8 +533,7 @@ static inline void intel_pmu_ack_status(u64 ack)
 	wrmsrl(MSR_CORE_PERF_GLOBAL_OVF_CTRL, ack);
 }
 
-static inline void
-intel_pmu_disable_fixed(struct hw_perf_event *hwc)
+static void intel_pmu_disable_fixed(struct hw_perf_event *hwc)
 {
 	int idx = hwc->idx - X86_PMC_IDX_FIXED;
 	u64 ctrl_val, mask;
@@ -557,71 +542,10 @@ intel_pmu_disable_fixed(struct hw_perf_event *hwc)
 
 	rdmsrl(hwc->config_base, ctrl_val);
 	ctrl_val &= ~mask;
-	(void)checking_wrmsrl(hwc->config_base, ctrl_val);
-}
-
-static void intel_pmu_drain_bts_buffer(void)
-{
-	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
-	struct debug_store *ds = cpuc->ds;
-	struct bts_record {
-		u64	from;
-		u64	to;
-		u64	flags;
-	};
-	struct perf_event *event = cpuc->events[X86_PMC_IDX_FIXED_BTS];
-	struct bts_record *at, *top;
-	struct perf_output_handle handle;
-	struct perf_event_header header;
-	struct perf_sample_data data;
-	struct pt_regs regs;
-
-	if (!event)
-		return;
-
-	if (!ds)
-		return;
-
-	at  = (struct bts_record *)(unsigned long)ds->bts_buffer_base;
-	top = (struct bts_record *)(unsigned long)ds->bts_index;
-
-	if (top <= at)
-		return;
-
-	ds->bts_index = ds->bts_buffer_base;
-
-	perf_sample_data_init(&data, 0);
-
-	data.period	= event->hw.last_period;
-	regs.ip		= 0;
-
-	/*
-	 * Prepare a generic sample, i.e. fill in the invariant fields.
-	 * We will overwrite the from and to address before we output
-	 * the sample.
-	 */
-	perf_prepare_sample(&header, &data, event, &regs);
-
-	if (perf_output_begin(&handle, event,
-			      header.size * (top - at), 1, 1))
-		return;
-
-	for (; at < top; at++) {
-		data.ip		= at->from;
-		data.addr	= at->to;
-
-		perf_output_sample(&handle, &header, &data, event);
-	}
-
-	perf_output_end(&handle);
-
-	/* There's new data available. */
-	event->hw.interrupts++;
-	event->pending_kill = POLL_IN;
+	wrmsrl(hwc->config_base, ctrl_val);
 }
 
-static inline void
-intel_pmu_disable_event(struct perf_event *event)
+static void intel_pmu_disable_event(struct perf_event *event)
 {
 	struct hw_perf_event *hwc = &event->hw;
 
@@ -637,14 +561,15 @@ intel_pmu_disable_event(struct perf_event *event)
 	}
 
 	x86_pmu_disable_event(event);
+
+	if (unlikely(event->attr.precise))
+		intel_pmu_pebs_disable(event);
 }
 
-static inline void
-intel_pmu_enable_fixed(struct hw_perf_event *hwc)
+static void intel_pmu_enable_fixed(struct hw_perf_event *hwc)
 {
 	int idx = hwc->idx - X86_PMC_IDX_FIXED;
 	u64 ctrl_val, bits, mask;
-	int err;
 
 	/*
 	 * Enable IRQ generation (0x8),
@@ -669,7 +594,7 @@ intel_pmu_enable_fixed(struct hw_perf_event *hwc)
 	rdmsrl(hwc->config_base, ctrl_val);
 	ctrl_val &= ~mask;
 	ctrl_val |= bits;
-	err = checking_wrmsrl(hwc->config_base, ctrl_val);
+	wrmsrl(hwc->config_base, ctrl_val);
 }
 
 static void intel_pmu_enable_event(struct perf_event *event)
@@ -689,6 +614,9 @@ static void intel_pmu_enable_event(struct perf_event *event)
 		return;
 	}
 
+	if (unlikely(event->attr.precise))
+		intel_pmu_pebs_enable(event);
+
 	__x86_pmu_enable_event(hwc);
 }
 
@@ -708,20 +636,20 @@ static void intel_pmu_reset(void)
 	unsigned long flags;
 	int idx;
 
-	if (!x86_pmu.num_events)
+	if (!x86_pmu.num_counters)
 		return;
 
 	local_irq_save(flags);
 
 	printk("clearing PMU state on CPU#%d\n", smp_processor_id());
 
-	for (idx = 0; idx < x86_pmu.num_events; idx++) {
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
 		checking_wrmsrl(x86_pmu.eventsel + idx, 0ull);
 		checking_wrmsrl(x86_pmu.perfctr  + idx, 0ull);
 	}
-	for (idx = 0; idx < x86_pmu.num_events_fixed; idx++) {
+	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;
 
@@ -747,7 +675,7 @@ static int intel_pmu_handle_irq(struct pt_regs *regs)
 	intel_pmu_drain_bts_buffer();
 	status = intel_pmu_get_status();
 	if (!status) {
-		intel_pmu_enable_all();
+		intel_pmu_enable_all(0);
 		return 0;
 	}
 
@@ -762,6 +690,15 @@ again:
 
 	inc_irq_stat(apic_perf_irqs);
 	ack = status;
+
+	intel_pmu_lbr_read();
+
+	/*
+	 * PEBS overflow sets bit 62 in the global status register
+	 */
+	if (__test_and_clear_bit(62, (unsigned long *)&status))
+		x86_pmu.drain_pebs(regs);
+
 	for_each_set_bit(bit, (unsigned long *)&status, X86_PMC_IDX_MAX) {
 		struct perf_event *event = cpuc->events[bit];
 
@@ -787,26 +724,22 @@ again:
 		goto again;
 
 done:
-	intel_pmu_enable_all();
+	intel_pmu_enable_all(0);
 	return 1;
 }
 
-static struct event_constraint bts_constraint =
-	EVENT_CONSTRAINT(0, 1ULL << X86_PMC_IDX_FIXED_BTS, 0);
-
 static struct event_constraint *
-intel_special_constraints(struct perf_event *event)
+intel_bts_constraints(struct perf_event *event)
 {
-	unsigned int hw_event;
-
-	hw_event = event->hw.config & INTEL_ARCH_EVENT_MASK;
+	struct hw_perf_event *hwc = &event->hw;
+	unsigned int hw_event, bts_event;
 
-	if (unlikely((hw_event ==
-		      x86_pmu.event_map(PERF_COUNT_HW_BRANCH_INSTRUCTIONS)) &&
-		     (event->hw.sample_period == 1))) {
+	hw_event = hwc->config & INTEL_ARCH_EVENT_MASK;
+	bts_event = x86_pmu.event_map(PERF_COUNT_HW_BRANCH_INSTRUCTIONS);
 
+	if (unlikely(hw_event == bts_event && hwc->sample_period == 1))
 		return &bts_constraint;
-	}
+
 	return NULL;
 }
 
@@ -815,24 +748,53 @@ intel_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event
 {
 	struct event_constraint *c;
 
-	c = intel_special_constraints(event);
+	c = intel_bts_constraints(event);
+	if (c)
+		return c;
+
+	c = intel_pebs_constraints(event);
 	if (c)
 		return c;
 
 	return x86_get_event_constraints(cpuc, event);
 }
 
-static __initconst struct x86_pmu core_pmu = {
+static int intel_pmu_hw_config(struct perf_event *event)
+{
+	int ret = x86_pmu_hw_config(event);
+
+	if (ret)
+		return ret;
+
+	if (event->attr.type != PERF_TYPE_RAW)
+		return 0;
+
+	if (!(event->attr.config & ARCH_PERFMON_EVENTSEL_ANY))
+		return 0;
+
+	if (x86_pmu.version < 3)
+		return -EINVAL;
+
+	if (perf_paranoid_cpu() && !capable(CAP_SYS_ADMIN))
+		return -EACCES;
+
+	event->hw.config |= ARCH_PERFMON_EVENTSEL_ANY;
+
+	return 0;
+}
+
+static __initconst const struct x86_pmu core_pmu = {
 	.name			= "core",
 	.handle_irq		= x86_pmu_handle_irq,
 	.disable_all		= x86_pmu_disable_all,
 	.enable_all		= x86_pmu_enable_all,
 	.enable			= x86_pmu_enable_event,
 	.disable		= x86_pmu_disable_event,
+	.hw_config		= x86_pmu_hw_config,
+	.schedule_events	= x86_schedule_events,
 	.eventsel		= MSR_ARCH_PERFMON_EVENTSEL0,
 	.perfctr		= MSR_ARCH_PERFMON_PERFCTR0,
 	.event_map		= intel_pmu_event_map,
-	.raw_event		= intel_pmu_raw_event,
 	.max_events		= ARRAY_SIZE(intel_perfmon_event_map),
 	.apic			= 1,
 	/*
@@ -845,17 +807,32 @@ static __initconst struct x86_pmu core_pmu = {
 	.event_constraints	= intel_core_event_constraints,
 };
 
-static __initconst struct x86_pmu intel_pmu = {
+static void intel_pmu_cpu_starting(int cpu)
+{
+	init_debug_store_on_cpu(cpu);
+	/*
+	 * Deal with CPUs that don't clear their LBRs on power-up.
+	 */
+	intel_pmu_lbr_reset();
+}
+
+static void intel_pmu_cpu_dying(int cpu)
+{
+	fini_debug_store_on_cpu(cpu);
+}
+
+static __initconst const struct x86_pmu intel_pmu = {
 	.name			= "Intel",
 	.handle_irq		= intel_pmu_handle_irq,
 	.disable_all		= intel_pmu_disable_all,
 	.enable_all		= intel_pmu_enable_all,
 	.enable			= intel_pmu_enable_event,
 	.disable		= intel_pmu_disable_event,
+	.hw_config		= intel_pmu_hw_config,
+	.schedule_events	= x86_schedule_events,
 	.eventsel		= MSR_ARCH_PERFMON_EVENTSEL0,
 	.perfctr		= MSR_ARCH_PERFMON_PERFCTR0,
 	.event_map		= intel_pmu_event_map,
-	.raw_event		= intel_pmu_raw_event,
 	.max_events		= ARRAY_SIZE(intel_perfmon_event_map),
 	.apic			= 1,
 	/*
@@ -864,14 +841,38 @@ static __initconst struct x86_pmu intel_pmu = {
 	 * the generic event period:
 	 */
 	.max_period		= (1ULL << 31) - 1,
-	.enable_bts		= intel_pmu_enable_bts,
-	.disable_bts		= intel_pmu_disable_bts,
 	.get_event_constraints	= intel_get_event_constraints,
 
-	.cpu_starting		= init_debug_store_on_cpu,
-	.cpu_dying		= fini_debug_store_on_cpu,
+	.cpu_starting		= intel_pmu_cpu_starting,
+	.cpu_dying		= intel_pmu_cpu_dying,
 };
 
+static void intel_clovertown_quirks(void)
+{
+	/*
+	 * PEBS is unreliable due to:
+	 *
+	 *   AJ67  - PEBS may experience CPL leaks
+	 *   AJ68  - PEBS PMI may be delayed by one event
+	 *   AJ69  - GLOBAL_STATUS[62] will only be set when DEBUGCTL[12]
+	 *   AJ106 - FREEZE_LBRS_ON_PMI doesn't work in combination with PEBS
+	 *
+	 * AJ67 could be worked around by restricting the OS/USR flags.
+	 * AJ69 could be worked around by setting PMU_FREEZE_ON_PMI.
+	 *
+	 * AJ106 could possibly be worked around by not allowing LBR
+	 *       usage from PEBS, including the fixup.
+	 * AJ68  could possibly be worked around by always programming
+	 * 	 a pebs_event_reset[0] value and coping with the lost events.
+	 *
+	 * But taken together it might just make sense to not enable PEBS on
+	 * these chips.
+	 */
+	printk(KERN_WARNING "PEBS disabled due to CPU errata.\n");
+	x86_pmu.pebs = 0;
+	x86_pmu.pebs_constraints = NULL;
+}
+
 static __init int intel_pmu_init(void)
 {
 	union cpuid10_edx edx;
@@ -881,12 +882,13 @@ static __init int intel_pmu_init(void)
 	int version;
 
 	if (!cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) {
-		/* check for P6 processor family */
-	   if (boot_cpu_data.x86 == 6) {
-		return p6_pmu_init();
-	   } else {
+		switch (boot_cpu_data.x86) {
+		case 0x6:
+			return p6_pmu_init();
+		case 0xf:
+			return p4_pmu_init();
+		}
 		return -ENODEV;
-	   }
 	}
 
 	/*
@@ -904,16 +906,28 @@ static __init int intel_pmu_init(void)
 		x86_pmu = intel_pmu;
 
 	x86_pmu.version			= version;
-	x86_pmu.num_events		= eax.split.num_events;
-	x86_pmu.event_bits		= eax.split.bit_width;
-	x86_pmu.event_mask		= (1ULL << eax.split.bit_width) - 1;
+	x86_pmu.num_counters		= eax.split.num_counters;
+	x86_pmu.cntval_bits		= eax.split.bit_width;
+	x86_pmu.cntval_mask		= (1ULL << eax.split.bit_width) - 1;
 
 	/*
 	 * Quirk: v2 perfmon does not report fixed-purpose events, so
 	 * assume at least 3 events:
 	 */
 	if (version > 1)
-		x86_pmu.num_events_fixed = max((int)edx.split.num_events_fixed, 3);
+		x86_pmu.num_counters_fixed = max((int)edx.split.num_counters_fixed, 3);
+
+	/*
+	 * v2 and above have a perf capabilities MSR
+	 */
+	if (version > 1) {
+		u64 capabilities;
+
+		rdmsrl(MSR_IA32_PERF_CAPABILITIES, capabilities);
+		x86_pmu.intel_cap.capabilities = capabilities;
+	}
+
+	intel_ds_init();
 
 	/*
 	 * Install the hw-cache-events table:
@@ -924,12 +938,15 @@ static __init int intel_pmu_init(void)
 		break;
 
 	case 15: /* original 65 nm celeron/pentium/core2/xeon, "Merom"/"Conroe" */
+		x86_pmu.quirks = intel_clovertown_quirks;
 	case 22: /* single-core 65 nm celeron/core2solo "Merom-L"/"Conroe-L" */
 	case 23: /* current 45 nm celeron/core2/xeon "Penryn"/"Wolfdale" */
 	case 29: /* six-core 45 nm xeon "Dunnington" */
 		memcpy(hw_cache_event_ids, core2_hw_cache_event_ids,
 		       sizeof(hw_cache_event_ids));
 
+		intel_pmu_lbr_init_core();
+
 		x86_pmu.event_constraints = intel_core2_event_constraints;
 		pr_cont("Core2 events, ");
 		break;
@@ -940,13 +957,19 @@ static __init int intel_pmu_init(void)
 		memcpy(hw_cache_event_ids, nehalem_hw_cache_event_ids,
 		       sizeof(hw_cache_event_ids));
 
+		intel_pmu_lbr_init_nhm();
+
 		x86_pmu.event_constraints = intel_nehalem_event_constraints;
-		pr_cont("Nehalem/Corei7 events, ");
+		x86_pmu.enable_all = intel_pmu_nhm_enable_all;
+		pr_cont("Nehalem events, ");
 		break;
+
 	case 28: /* Atom */
 		memcpy(hw_cache_event_ids, atom_hw_cache_event_ids,
 		       sizeof(hw_cache_event_ids));
 
+		intel_pmu_lbr_init_atom();
+
 		x86_pmu.event_constraints = intel_gen_event_constraints;
 		pr_cont("Atom events, ");
 		break;
@@ -956,7 +979,10 @@ static __init int intel_pmu_init(void)
 		memcpy(hw_cache_event_ids, westmere_hw_cache_event_ids,
 		       sizeof(hw_cache_event_ids));
 
+		intel_pmu_lbr_init_nhm();
+
 		x86_pmu.event_constraints = intel_westmere_event_constraints;
+		x86_pmu.enable_all = intel_pmu_nhm_enable_all;
 		pr_cont("Westmere events, ");
 		break;
 

arch/x86/kernel/cpu/perf_event_intel_ds.c

@@ -0,0 +1,664 @@
+#ifdef CONFIG_CPU_SUP_INTEL
+
+/* The maximal number of PEBS events: */
+#define MAX_PEBS_EVENTS		4
+
+/* The size of a BTS record in bytes: */
+#define BTS_RECORD_SIZE		24
+
+#define BTS_BUFFER_SIZE		(PAGE_SIZE << 4)
+#define PEBS_BUFFER_SIZE	PAGE_SIZE
+
+/*
+ * pebs_record_32 for p4 and core not supported
+
+struct pebs_record_32 {
+	u32 flags, ip;
+	u32 ax, bc, cx, dx;
+	u32 si, di, bp, sp;
+};
+
+ */
+
+struct pebs_record_core {
+	u64 flags, ip;
+	u64 ax, bx, cx, dx;
+	u64 si, di, bp, sp;
+	u64 r8,  r9,  r10, r11;
+	u64 r12, r13, r14, r15;
+};
+
+struct pebs_record_nhm {
+	u64 flags, ip;
+	u64 ax, bx, cx, dx;
+	u64 si, di, bp, sp;
+	u64 r8,  r9,  r10, r11;
+	u64 r12, r13, r14, r15;
+	u64 status, dla, dse, lat;
+};
+
+/*
+ * 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_event_reset[MAX_PEBS_EVENTS];
+};
+
+static void init_debug_store_on_cpu(int cpu)
+{
+	struct debug_store *ds = per_cpu(cpu_hw_events, cpu).ds;
+
+	if (!ds)
+		return;
+
+	wrmsr_on_cpu(cpu, MSR_IA32_DS_AREA,
+		     (u32)((u64)(unsigned long)ds),
+		     (u32)((u64)(unsigned long)ds >> 32));
+}
+
+static void fini_debug_store_on_cpu(int cpu)
+{
+	if (!per_cpu(cpu_hw_events, cpu).ds)
+		return;
+
+	wrmsr_on_cpu(cpu, MSR_IA32_DS_AREA, 0, 0);
+}
+
+static void release_ds_buffers(void)
+{
+	int cpu;
+
+	if (!x86_pmu.bts && !x86_pmu.pebs)
+		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_events, cpu).ds;
+
+		if (!ds)
+			continue;
+
+		per_cpu(cpu_hw_events, cpu).ds = NULL;
+
+		kfree((void *)(unsigned long)ds->pebs_buffer_base);
+		kfree((void *)(unsigned long)ds->bts_buffer_base);
+		kfree(ds);
+	}
+
+	put_online_cpus();
+}
+
+static int reserve_ds_buffers(void)
+{
+	int cpu, err = 0;
+
+	if (!x86_pmu.bts && !x86_pmu.pebs)
+		return 0;
+
+	get_online_cpus();
+
+	for_each_possible_cpu(cpu) {
+		struct debug_store *ds;
+		void *buffer;
+		int max, thresh;
+
+		err = -ENOMEM;
+		ds = kzalloc(sizeof(*ds), GFP_KERNEL);
+		if (unlikely(!ds))
+			break;
+		per_cpu(cpu_hw_events, cpu).ds = ds;
+
+		if (x86_pmu.bts) {
+			buffer = kzalloc(BTS_BUFFER_SIZE, GFP_KERNEL);
+			if (unlikely(!buffer))
+				break;
+
+			max = BTS_BUFFER_SIZE / BTS_RECORD_SIZE;
+			thresh = max / 16;
+
+			ds->bts_buffer_base = (u64)(unsigned long)buffer;
+			ds->bts_index = ds->bts_buffer_base;
+			ds->bts_absolute_maximum = ds->bts_buffer_base +
+				max * BTS_RECORD_SIZE;
+			ds->bts_interrupt_threshold = ds->bts_absolute_maximum -
+				thresh * BTS_RECORD_SIZE;
+		}
+
+		if (x86_pmu.pebs) {
+			buffer = kzalloc(PEBS_BUFFER_SIZE, GFP_KERNEL);
+			if (unlikely(!buffer))
+				break;
+
+			max = PEBS_BUFFER_SIZE / x86_pmu.pebs_record_size;
+
+			ds->pebs_buffer_base = (u64)(unsigned long)buffer;
+			ds->pebs_index = ds->pebs_buffer_base;
+			ds->pebs_absolute_maximum = ds->pebs_buffer_base +
+				max * x86_pmu.pebs_record_size;
+			/*
+			 * Always use single record PEBS
+			 */
+			ds->pebs_interrupt_threshold = ds->pebs_buffer_base +
+				x86_pmu.pebs_record_size;
+		}
+
+		err = 0;
+	}
+
+	if (err)
+		release_ds_buffers();
+	else {
+		for_each_online_cpu(cpu)
+			init_debug_store_on_cpu(cpu);
+	}
+
+	put_online_cpus();
+
+	return err;
+}
+
+/*
+ * BTS
+ */
+
+static struct event_constraint bts_constraint =
+	EVENT_CONSTRAINT(0, 1ULL << X86_PMC_IDX_FIXED_BTS, 0);
+
+static void intel_pmu_enable_bts(u64 config)
+{
+	unsigned long debugctlmsr;
+
+	debugctlmsr = get_debugctlmsr();
+
+	debugctlmsr |= DEBUGCTLMSR_TR;
+	debugctlmsr |= DEBUGCTLMSR_BTS;
+	debugctlmsr |= DEBUGCTLMSR_BTINT;
+
+	if (!(config & ARCH_PERFMON_EVENTSEL_OS))
+		debugctlmsr |= DEBUGCTLMSR_BTS_OFF_OS;
+
+	if (!(config & ARCH_PERFMON_EVENTSEL_USR))
+		debugctlmsr |= DEBUGCTLMSR_BTS_OFF_USR;
+
+	update_debugctlmsr(debugctlmsr);
+}
+
+static void intel_pmu_disable_bts(void)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	unsigned long debugctlmsr;
+
+	if (!cpuc->ds)
+		return;
+
+	debugctlmsr = get_debugctlmsr();
+
+	debugctlmsr &=
+		~(DEBUGCTLMSR_TR | DEBUGCTLMSR_BTS | DEBUGCTLMSR_BTINT |
+		  DEBUGCTLMSR_BTS_OFF_OS | DEBUGCTLMSR_BTS_OFF_USR);
+
+	update_debugctlmsr(debugctlmsr);
+}
+
+static void intel_pmu_drain_bts_buffer(void)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	struct debug_store *ds = cpuc->ds;
+	struct bts_record {
+		u64	from;
+		u64	to;
+		u64	flags;
+	};
+	struct perf_event *event = cpuc->events[X86_PMC_IDX_FIXED_BTS];
+	struct bts_record *at, *top;
+	struct perf_output_handle handle;
+	struct perf_event_header header;
+	struct perf_sample_data data;
+	struct pt_regs regs;
+
+	if (!event)
+		return;
+
+	if (!ds)
+		return;
+
+	at  = (struct bts_record *)(unsigned long)ds->bts_buffer_base;
+	top = (struct bts_record *)(unsigned long)ds->bts_index;
+
+	if (top <= at)
+		return;
+
+	ds->bts_index = ds->bts_buffer_base;
+
+	perf_sample_data_init(&data, 0);
+	data.period = event->hw.last_period;
+	regs.ip     = 0;
+
+	/*
+	 * Prepare a generic sample, i.e. fill in the invariant fields.
+	 * We will overwrite the from and to address before we output
+	 * the sample.
+	 */
+	perf_prepare_sample(&header, &data, event, &regs);
+
+	if (perf_output_begin(&handle, event, header.size * (top - at), 1, 1))
+		return;
+
+	for (; at < top; at++) {
+		data.ip		= at->from;
+		data.addr	= at->to;
+
+		perf_output_sample(&handle, &header, &data, event);
+	}
+
+	perf_output_end(&handle);
+
+	/* There's new data available. */
+	event->hw.interrupts++;
+	event->pending_kill = POLL_IN;
+}
+
+/*
+ * PEBS
+ */
+
+static struct event_constraint intel_core_pebs_events[] = {
+	PEBS_EVENT_CONSTRAINT(0x00c0, 0x1), /* INSTR_RETIRED.ANY */
+	PEBS_EVENT_CONSTRAINT(0xfec1, 0x1), /* X87_OPS_RETIRED.ANY */
+	PEBS_EVENT_CONSTRAINT(0x00c5, 0x1), /* BR_INST_RETIRED.MISPRED */
+	PEBS_EVENT_CONSTRAINT(0x1fc7, 0x1), /* SIMD_INST_RETURED.ANY */
+	PEBS_EVENT_CONSTRAINT(0x01cb, 0x1), /* MEM_LOAD_RETIRED.L1D_MISS */
+	PEBS_EVENT_CONSTRAINT(0x02cb, 0x1), /* MEM_LOAD_RETIRED.L1D_LINE_MISS */
+	PEBS_EVENT_CONSTRAINT(0x04cb, 0x1), /* MEM_LOAD_RETIRED.L2_MISS */
+	PEBS_EVENT_CONSTRAINT(0x08cb, 0x1), /* MEM_LOAD_RETIRED.L2_LINE_MISS */
+	PEBS_EVENT_CONSTRAINT(0x10cb, 0x1), /* MEM_LOAD_RETIRED.DTLB_MISS */
+	EVENT_CONSTRAINT_END
+};
+
+static struct event_constraint intel_nehalem_pebs_events[] = {
+	PEBS_EVENT_CONSTRAINT(0x00c0, 0xf), /* INSTR_RETIRED.ANY */
+	PEBS_EVENT_CONSTRAINT(0xfec1, 0xf), /* X87_OPS_RETIRED.ANY */
+	PEBS_EVENT_CONSTRAINT(0x00c5, 0xf), /* BR_INST_RETIRED.MISPRED */
+	PEBS_EVENT_CONSTRAINT(0x1fc7, 0xf), /* SIMD_INST_RETURED.ANY */
+	PEBS_EVENT_CONSTRAINT(0x01cb, 0xf), /* MEM_LOAD_RETIRED.L1D_MISS */
+	PEBS_EVENT_CONSTRAINT(0x02cb, 0xf), /* MEM_LOAD_RETIRED.L1D_LINE_MISS */
+	PEBS_EVENT_CONSTRAINT(0x04cb, 0xf), /* MEM_LOAD_RETIRED.L2_MISS */
+	PEBS_EVENT_CONSTRAINT(0x08cb, 0xf), /* MEM_LOAD_RETIRED.L2_LINE_MISS */
+	PEBS_EVENT_CONSTRAINT(0x10cb, 0xf), /* MEM_LOAD_RETIRED.DTLB_MISS */
+	EVENT_CONSTRAINT_END
+};
+
+static struct event_constraint *
+intel_pebs_constraints(struct perf_event *event)
+{
+	struct event_constraint *c;
+
+	if (!event->attr.precise)
+		return NULL;
+
+	if (x86_pmu.pebs_constraints) {
+		for_each_event_constraint(c, x86_pmu.pebs_constraints) {
+			if ((event->hw.config & c->cmask) == c->code)
+				return c;
+		}
+	}
+
+	return &emptyconstraint;
+}
+
+static void intel_pmu_pebs_enable(struct perf_event *event)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	struct hw_perf_event *hwc = &event->hw;
+
+	hwc->config &= ~ARCH_PERFMON_EVENTSEL_INT;
+
+	cpuc->pebs_enabled |= 1ULL << hwc->idx;
+	WARN_ON_ONCE(cpuc->enabled);
+
+	if (x86_pmu.intel_cap.pebs_trap)
+		intel_pmu_lbr_enable(event);
+}
+
+static void intel_pmu_pebs_disable(struct perf_event *event)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	struct hw_perf_event *hwc = &event->hw;
+
+	cpuc->pebs_enabled &= ~(1ULL << hwc->idx);
+	if (cpuc->enabled)
+		wrmsrl(MSR_IA32_PEBS_ENABLE, cpuc->pebs_enabled);
+
+	hwc->config |= ARCH_PERFMON_EVENTSEL_INT;
+
+	if (x86_pmu.intel_cap.pebs_trap)
+		intel_pmu_lbr_disable(event);
+}
+
+static void intel_pmu_pebs_enable_all(void)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+
+	if (cpuc->pebs_enabled)
+		wrmsrl(MSR_IA32_PEBS_ENABLE, cpuc->pebs_enabled);
+}
+
+static void intel_pmu_pebs_disable_all(void)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+
+	if (cpuc->pebs_enabled)
+		wrmsrl(MSR_IA32_PEBS_ENABLE, 0);
+}
+
+#include <asm/insn.h>
+
+static inline bool kernel_ip(unsigned long ip)
+{
+#ifdef CONFIG_X86_32
+	return ip > PAGE_OFFSET;
+#else
+	return (long)ip < 0;
+#endif
+}
+
+static int intel_pmu_pebs_fixup_ip(struct pt_regs *regs)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	unsigned long from = cpuc->lbr_entries[0].from;
+	unsigned long old_to, to = cpuc->lbr_entries[0].to;
+	unsigned long ip = regs->ip;
+
+	/*
+	 * We don't need to fixup if the PEBS assist is fault like
+	 */
+	if (!x86_pmu.intel_cap.pebs_trap)
+		return 1;
+
+	/*
+	 * No LBR entry, no basic block, no rewinding
+	 */
+	if (!cpuc->lbr_stack.nr || !from || !to)
+		return 0;
+
+	/*
+	 * Basic blocks should never cross user/kernel boundaries
+	 */
+	if (kernel_ip(ip) != kernel_ip(to))
+		return 0;
+
+	/*
+	 * unsigned math, either ip is before the start (impossible) or
+	 * the basic block is larger than 1 page (sanity)
+	 */
+	if ((ip - to) > PAGE_SIZE)
+		return 0;
+
+	/*
+	 * We sampled a branch insn, rewind using the LBR stack
+	 */
+	if (ip == to) {
+		regs->ip = from;
+		return 1;
+	}
+
+	do {
+		struct insn insn;
+		u8 buf[MAX_INSN_SIZE];
+		void *kaddr;
+
+		old_to = to;
+		if (!kernel_ip(ip)) {
+			int bytes, size = MAX_INSN_SIZE;
+
+			bytes = copy_from_user_nmi(buf, (void __user *)to, size);
+			if (bytes != size)
+				return 0;
+
+			kaddr = buf;
+		} else
+			kaddr = (void *)to;
+
+		kernel_insn_init(&insn, kaddr);
+		insn_get_length(&insn);
+		to += insn.length;
+	} while (to < ip);
+
+	if (to == ip) {
+		regs->ip = old_to;
+		return 1;
+	}
+
+	/*
+	 * Even though we decoded the basic block, the instruction stream
+	 * never matched the given IP, either the TO or the IP got corrupted.
+	 */
+	return 0;
+}
+
+static int intel_pmu_save_and_restart(struct perf_event *event);
+
+static void intel_pmu_drain_pebs_core(struct pt_regs *iregs)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	struct debug_store *ds = cpuc->ds;
+	struct perf_event *event = cpuc->events[0]; /* PMC0 only */
+	struct pebs_record_core *at, *top;
+	struct perf_sample_data data;
+	struct perf_raw_record raw;
+	struct pt_regs regs;
+	int n;
+
+	if (!ds || !x86_pmu.pebs)
+		return;
+
+	at  = (struct pebs_record_core *)(unsigned long)ds->pebs_buffer_base;
+	top = (struct pebs_record_core *)(unsigned long)ds->pebs_index;
+
+	/*
+	 * Whatever else happens, drain the thing
+	 */
+	ds->pebs_index = ds->pebs_buffer_base;
+
+	if (!test_bit(0, cpuc->active_mask))
+		return;
+
+	WARN_ON_ONCE(!event);
+
+	if (!event->attr.precise)
+		return;
+
+	n = top - at;
+	if (n <= 0)
+		return;
+
+	if (!intel_pmu_save_and_restart(event))
+		return;
+
+	/*
+	 * Should not happen, we program the threshold at 1 and do not
+	 * set a reset value.
+	 */
+	WARN_ON_ONCE(n > 1);
+	at += n - 1;
+
+	perf_sample_data_init(&data, 0);
+	data.period = event->hw.last_period;
+
+	if (event->attr.sample_type & PERF_SAMPLE_RAW) {
+		raw.size = x86_pmu.pebs_record_size;
+		raw.data = at;
+		data.raw = &raw;
+	}
+
+	/*
+	 * We use the interrupt regs as a base because the PEBS record
+	 * does not contain a full regs set, specifically it seems to
+	 * lack segment descriptors, which get used by things like
+	 * user_mode().
+	 *
+	 * In the simple case fix up only the IP and BP,SP regs, for
+	 * PERF_SAMPLE_IP and PERF_SAMPLE_CALLCHAIN to function properly.
+	 * A possible PERF_SAMPLE_REGS will have to transfer all regs.
+	 */
+	regs = *iregs;
+	regs.ip = at->ip;
+	regs.bp = at->bp;
+	regs.sp = at->sp;
+
+	if (intel_pmu_pebs_fixup_ip(&regs))
+		regs.flags |= PERF_EFLAGS_EXACT;
+	else
+		regs.flags &= ~PERF_EFLAGS_EXACT;
+
+	if (perf_event_overflow(event, 1, &data, &regs))
+		x86_pmu_stop(event);
+}
+
+static void intel_pmu_drain_pebs_nhm(struct pt_regs *iregs)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	struct debug_store *ds = cpuc->ds;
+	struct pebs_record_nhm *at, *top;
+	struct perf_sample_data data;
+	struct perf_event *event = NULL;
+	struct perf_raw_record raw;
+	struct pt_regs regs;
+	u64 status = 0;
+	int bit, n;
+
+	if (!ds || !x86_pmu.pebs)
+		return;
+
+	at  = (struct pebs_record_nhm *)(unsigned long)ds->pebs_buffer_base;
+	top = (struct pebs_record_nhm *)(unsigned long)ds->pebs_index;
+
+	ds->pebs_index = ds->pebs_buffer_base;
+
+	n = top - at;
+	if (n <= 0)
+		return;
+
+	/*
+	 * Should not happen, we program the threshold at 1 and do not
+	 * set a reset value.
+	 */
+	WARN_ON_ONCE(n > MAX_PEBS_EVENTS);
+
+	for ( ; at < top; at++) {
+		for_each_set_bit(bit, (unsigned long *)&at->status, MAX_PEBS_EVENTS) {
+			event = cpuc->events[bit];
+			if (!test_bit(bit, cpuc->active_mask))
+				continue;
+
+			WARN_ON_ONCE(!event);
+
+			if (!event->attr.precise)
+				continue;
+
+			if (__test_and_set_bit(bit, (unsigned long *)&status))
+				continue;
+
+			break;
+		}
+
+		if (!event || bit >= MAX_PEBS_EVENTS)
+			continue;
+
+		if (!intel_pmu_save_and_restart(event))
+			continue;
+
+		perf_sample_data_init(&data, 0);
+		data.period = event->hw.last_period;
+
+		if (event->attr.sample_type & PERF_SAMPLE_RAW) {
+			raw.size = x86_pmu.pebs_record_size;
+			raw.data = at;
+			data.raw = &raw;
+		}
+
+		/*
+		 * See the comment in intel_pmu_drain_pebs_core()
+		 */
+		regs = *iregs;
+		regs.ip = at->ip;
+		regs.bp = at->bp;
+		regs.sp = at->sp;
+
+		if (intel_pmu_pebs_fixup_ip(&regs))
+			regs.flags |= PERF_EFLAGS_EXACT;
+		else
+			regs.flags &= ~PERF_EFLAGS_EXACT;
+
+		if (perf_event_overflow(event, 1, &data, &regs))
+			x86_pmu_stop(event);
+	}
+}
+
+/*
+ * BTS, PEBS probe and setup
+ */
+
+static void intel_ds_init(void)
+{
+	/*
+	 * No support for 32bit formats
+	 */
+	if (!boot_cpu_has(X86_FEATURE_DTES64))
+		return;
+
+	x86_pmu.bts  = boot_cpu_has(X86_FEATURE_BTS);
+	x86_pmu.pebs = boot_cpu_has(X86_FEATURE_PEBS);
+	if (x86_pmu.pebs) {
+		char pebs_type = x86_pmu.intel_cap.pebs_trap ?  '+' : '-';
+		int format = x86_pmu.intel_cap.pebs_format;
+
+		switch (format) {
+		case 0:
+			printk(KERN_CONT "PEBS fmt0%c, ", pebs_type);
+			x86_pmu.pebs_record_size = sizeof(struct pebs_record_core);
+			x86_pmu.drain_pebs = intel_pmu_drain_pebs_core;
+			x86_pmu.pebs_constraints = intel_core_pebs_events;
+			break;
+
+		case 1:
+			printk(KERN_CONT "PEBS fmt1%c, ", pebs_type);
+			x86_pmu.pebs_record_size = sizeof(struct pebs_record_nhm);
+			x86_pmu.drain_pebs = intel_pmu_drain_pebs_nhm;
+			x86_pmu.pebs_constraints = intel_nehalem_pebs_events;
+			break;
+
+		default:
+			printk(KERN_CONT "no PEBS fmt%d%c, ", format, pebs_type);
+			x86_pmu.pebs = 0;
+			break;
+		}
+	}
+}
+
+#else /* CONFIG_CPU_SUP_INTEL */
+
+static int reserve_ds_buffers(void)
+{
+	return 0;
+}
+
+static void release_ds_buffers(void)
+{
+}
+
+#endif /* CONFIG_CPU_SUP_INTEL */

arch/x86/kernel/cpu/perf_event_intel_lbr.c

@@ -0,0 +1,218 @@
+#ifdef CONFIG_CPU_SUP_INTEL
+
+enum {
+	LBR_FORMAT_32		= 0x00,
+	LBR_FORMAT_LIP		= 0x01,
+	LBR_FORMAT_EIP		= 0x02,
+	LBR_FORMAT_EIP_FLAGS	= 0x03,
+};
+
+/*
+ * We only support LBR implementations that have FREEZE_LBRS_ON_PMI
+ * otherwise it becomes near impossible to get a reliable stack.
+ */
+
+static void __intel_pmu_lbr_enable(void)
+{
+	u64 debugctl;
+
+	rdmsrl(MSR_IA32_DEBUGCTLMSR, debugctl);
+	debugctl |= (DEBUGCTLMSR_LBR | DEBUGCTLMSR_FREEZE_LBRS_ON_PMI);
+	wrmsrl(MSR_IA32_DEBUGCTLMSR, debugctl);
+}
+
+static void __intel_pmu_lbr_disable(void)
+{
+	u64 debugctl;
+
+	rdmsrl(MSR_IA32_DEBUGCTLMSR, debugctl);
+	debugctl &= ~(DEBUGCTLMSR_LBR | DEBUGCTLMSR_FREEZE_LBRS_ON_PMI);
+	wrmsrl(MSR_IA32_DEBUGCTLMSR, debugctl);
+}
+
+static void intel_pmu_lbr_reset_32(void)
+{
+	int i;
+
+	for (i = 0; i < x86_pmu.lbr_nr; i++)
+		wrmsrl(x86_pmu.lbr_from + i, 0);
+}
+
+static void intel_pmu_lbr_reset_64(void)
+{
+	int i;
+
+	for (i = 0; i < x86_pmu.lbr_nr; i++) {
+		wrmsrl(x86_pmu.lbr_from + i, 0);
+		wrmsrl(x86_pmu.lbr_to   + i, 0);
+	}
+}
+
+static void intel_pmu_lbr_reset(void)
+{
+	if (!x86_pmu.lbr_nr)
+		return;
+
+	if (x86_pmu.intel_cap.lbr_format == LBR_FORMAT_32)
+		intel_pmu_lbr_reset_32();
+	else
+		intel_pmu_lbr_reset_64();
+}
+
+static void intel_pmu_lbr_enable(struct perf_event *event)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+
+	if (!x86_pmu.lbr_nr)
+		return;
+
+	WARN_ON_ONCE(cpuc->enabled);
+
+	/*
+	 * Reset the LBR stack if we changed task context to
+	 * avoid data leaks.
+	 */
+
+	if (event->ctx->task && cpuc->lbr_context != event->ctx) {
+		intel_pmu_lbr_reset();
+		cpuc->lbr_context = event->ctx;
+	}
+
+	cpuc->lbr_users++;
+}
+
+static void intel_pmu_lbr_disable(struct perf_event *event)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+
+	if (!x86_pmu.lbr_nr)
+		return;
+
+	cpuc->lbr_users--;
+	WARN_ON_ONCE(cpuc->lbr_users < 0);
+
+	if (cpuc->enabled && !cpuc->lbr_users)
+		__intel_pmu_lbr_disable();
+}
+
+static void intel_pmu_lbr_enable_all(void)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+
+	if (cpuc->lbr_users)
+		__intel_pmu_lbr_enable();
+}
+
+static void intel_pmu_lbr_disable_all(void)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+
+	if (cpuc->lbr_users)
+		__intel_pmu_lbr_disable();
+}
+
+static inline u64 intel_pmu_lbr_tos(void)
+{
+	u64 tos;
+
+	rdmsrl(x86_pmu.lbr_tos, tos);
+
+	return tos;
+}
+
+static void intel_pmu_lbr_read_32(struct cpu_hw_events *cpuc)
+{
+	unsigned long mask = x86_pmu.lbr_nr - 1;
+	u64 tos = intel_pmu_lbr_tos();
+	int i;
+
+	for (i = 0; i < x86_pmu.lbr_nr; i++) {
+		unsigned long lbr_idx = (tos - i) & mask;
+		union {
+			struct {
+				u32 from;
+				u32 to;
+			};
+			u64     lbr;
+		} msr_lastbranch;
+
+		rdmsrl(x86_pmu.lbr_from + lbr_idx, msr_lastbranch.lbr);
+
+		cpuc->lbr_entries[i].from  = msr_lastbranch.from;
+		cpuc->lbr_entries[i].to    = msr_lastbranch.to;
+		cpuc->lbr_entries[i].flags = 0;
+	}
+	cpuc->lbr_stack.nr = i;
+}
+
+#define LBR_FROM_FLAG_MISPRED  (1ULL << 63)
+
+/*
+ * Due to lack of segmentation in Linux the effective address (offset)
+ * is the same as the linear address, allowing us to merge the LIP and EIP
+ * LBR formats.
+ */
+static void intel_pmu_lbr_read_64(struct cpu_hw_events *cpuc)
+{
+	unsigned long mask = x86_pmu.lbr_nr - 1;
+	int lbr_format = x86_pmu.intel_cap.lbr_format;
+	u64 tos = intel_pmu_lbr_tos();
+	int i;
+
+	for (i = 0; i < x86_pmu.lbr_nr; i++) {
+		unsigned long lbr_idx = (tos - i) & mask;
+		u64 from, to, flags = 0;
+
+		rdmsrl(x86_pmu.lbr_from + lbr_idx, from);
+		rdmsrl(x86_pmu.lbr_to   + lbr_idx, to);
+
+		if (lbr_format == LBR_FORMAT_EIP_FLAGS) {
+			flags = !!(from & LBR_FROM_FLAG_MISPRED);
+			from = (u64)((((s64)from) << 1) >> 1);
+		}
+
+		cpuc->lbr_entries[i].from  = from;
+		cpuc->lbr_entries[i].to    = to;
+		cpuc->lbr_entries[i].flags = flags;
+	}
+	cpuc->lbr_stack.nr = i;
+}
+
+static void intel_pmu_lbr_read(void)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+
+	if (!cpuc->lbr_users)
+		return;
+
+	if (x86_pmu.intel_cap.lbr_format == LBR_FORMAT_32)
+		intel_pmu_lbr_read_32(cpuc);
+	else
+		intel_pmu_lbr_read_64(cpuc);
+}
+
+static void intel_pmu_lbr_init_core(void)
+{
+	x86_pmu.lbr_nr     = 4;
+	x86_pmu.lbr_tos    = 0x01c9;
+	x86_pmu.lbr_from   = 0x40;
+	x86_pmu.lbr_to     = 0x60;
+}
+
+static void intel_pmu_lbr_init_nhm(void)
+{
+	x86_pmu.lbr_nr     = 16;
+	x86_pmu.lbr_tos    = 0x01c9;
+	x86_pmu.lbr_from   = 0x680;
+	x86_pmu.lbr_to     = 0x6c0;
+}
+
+static void intel_pmu_lbr_init_atom(void)
+{
+	x86_pmu.lbr_nr	   = 8;
+	x86_pmu.lbr_tos    = 0x01c9;
+	x86_pmu.lbr_from   = 0x40;
+	x86_pmu.lbr_to     = 0x60;
+}
+
+#endif /* CONFIG_CPU_SUP_INTEL */

arch/x86/kernel/cpu/perf_event_p4.c

@@ -0,0 +1,834 @@
+/*
+ * Netburst Perfomance Events (P4, old Xeon)
+ *
+ *  Copyright (C) 2010 Parallels, Inc., Cyrill Gorcunov <gorcunov@openvz.org>
+ *  Copyright (C) 2010 Intel Corporation, Lin Ming <ming.m.lin@intel.com>
+ *
+ *  For licencing details see kernel-base/COPYING
+ */
+
+#ifdef CONFIG_CPU_SUP_INTEL
+
+#include <asm/perf_event_p4.h>
+
+#define P4_CNTR_LIMIT 3
+/*
+ * array indices: 0,1 - HT threads, used with HT enabled cpu
+ */
+struct p4_event_bind {
+	unsigned int opcode;			/* Event code and ESCR selector */
+	unsigned int escr_msr[2];		/* ESCR MSR for this event */
+	unsigned char cntr[2][P4_CNTR_LIMIT];	/* counter index (offset), -1 on abscence */
+};
+
+struct p4_cache_event_bind {
+	unsigned int metric_pebs;
+	unsigned int metric_vert;
+};
+
+#define P4_GEN_CACHE_EVENT_BIND(name)		\
+	[P4_CACHE__##name] = {			\
+		.metric_pebs = P4_PEBS__##name,	\
+		.metric_vert = P4_VERT__##name,	\
+	}
+
+static struct p4_cache_event_bind p4_cache_event_bind_map[] = {
+	P4_GEN_CACHE_EVENT_BIND(1stl_cache_load_miss_retired),
+	P4_GEN_CACHE_EVENT_BIND(2ndl_cache_load_miss_retired),
+	P4_GEN_CACHE_EVENT_BIND(dtlb_load_miss_retired),
+	P4_GEN_CACHE_EVENT_BIND(dtlb_store_miss_retired),
+};
+
+/*
+ * Note that we don't use CCCR1 here, there is an
+ * exception for P4_BSQ_ALLOCATION but we just have
+ * no workaround
+ *
+ * consider this binding as resources which particular
+ * event may borrow, it doesn't contain EventMask,
+ * Tags and friends -- they are left to a caller
+ */
+static struct p4_event_bind p4_event_bind_map[] = {
+	[P4_EVENT_TC_DELIVER_MODE] = {
+		.opcode		= P4_OPCODE(P4_EVENT_TC_DELIVER_MODE),
+		.escr_msr	= { MSR_P4_TC_ESCR0, MSR_P4_TC_ESCR1 },
+		.cntr		= { {4, 5, -1}, {6, 7, -1} },
+	},
+	[P4_EVENT_BPU_FETCH_REQUEST] = {
+		.opcode		= P4_OPCODE(P4_EVENT_BPU_FETCH_REQUEST),
+		.escr_msr	= { MSR_P4_BPU_ESCR0, MSR_P4_BPU_ESCR1 },
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_ITLB_REFERENCE] = {
+		.opcode		= P4_OPCODE(P4_EVENT_ITLB_REFERENCE),
+		.escr_msr	= { MSR_P4_ITLB_ESCR0, MSR_P4_ITLB_ESCR1 },
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_MEMORY_CANCEL] = {
+		.opcode		= P4_OPCODE(P4_EVENT_MEMORY_CANCEL),
+		.escr_msr	= { MSR_P4_DAC_ESCR0, MSR_P4_DAC_ESCR1 },
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_MEMORY_COMPLETE] = {
+		.opcode		= P4_OPCODE(P4_EVENT_MEMORY_COMPLETE),
+		.escr_msr	= { MSR_P4_SAAT_ESCR0 , MSR_P4_SAAT_ESCR1 },
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_LOAD_PORT_REPLAY] = {
+		.opcode		= P4_OPCODE(P4_EVENT_LOAD_PORT_REPLAY),
+		.escr_msr	= { MSR_P4_SAAT_ESCR0, MSR_P4_SAAT_ESCR1 },
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_STORE_PORT_REPLAY] = {
+		.opcode		= P4_OPCODE(P4_EVENT_STORE_PORT_REPLAY),
+		.escr_msr	= { MSR_P4_SAAT_ESCR0 ,  MSR_P4_SAAT_ESCR1 },
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_MOB_LOAD_REPLAY] = {
+		.opcode		= P4_OPCODE(P4_EVENT_MOB_LOAD_REPLAY),
+		.escr_msr	= { MSR_P4_MOB_ESCR0, MSR_P4_MOB_ESCR1 },
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_PAGE_WALK_TYPE] = {
+		.opcode		= P4_OPCODE(P4_EVENT_PAGE_WALK_TYPE),
+		.escr_msr	= { MSR_P4_PMH_ESCR0, MSR_P4_PMH_ESCR1 },
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_BSQ_CACHE_REFERENCE] = {
+		.opcode		= P4_OPCODE(P4_EVENT_BSQ_CACHE_REFERENCE),
+		.escr_msr	= { MSR_P4_BSU_ESCR0, MSR_P4_BSU_ESCR1 },
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_IOQ_ALLOCATION] = {
+		.opcode		= P4_OPCODE(P4_EVENT_IOQ_ALLOCATION),
+		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_IOQ_ACTIVE_ENTRIES] = {	/* shared ESCR */
+		.opcode		= P4_OPCODE(P4_EVENT_IOQ_ACTIVE_ENTRIES),
+		.escr_msr	= { MSR_P4_FSB_ESCR1,  MSR_P4_FSB_ESCR1 },
+		.cntr		= { {2, -1, -1}, {3, -1, -1} },
+	},
+	[P4_EVENT_FSB_DATA_ACTIVITY] = {
+		.opcode		= P4_OPCODE(P4_EVENT_FSB_DATA_ACTIVITY),
+		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_BSQ_ALLOCATION] = {		/* shared ESCR, broken CCCR1 */
+		.opcode		= P4_OPCODE(P4_EVENT_BSQ_ALLOCATION),
+		.escr_msr	= { MSR_P4_BSU_ESCR0, MSR_P4_BSU_ESCR0 },
+		.cntr		= { {0, -1, -1}, {1, -1, -1} },
+	},
+	[P4_EVENT_BSQ_ACTIVE_ENTRIES] = {	/* shared ESCR */
+		.opcode		= P4_OPCODE(P4_EVENT_BSQ_ACTIVE_ENTRIES),
+		.escr_msr	= { MSR_P4_BSU_ESCR1 , MSR_P4_BSU_ESCR1 },
+		.cntr		= { {2, -1, -1}, {3, -1, -1} },
+	},
+	[P4_EVENT_SSE_INPUT_ASSIST] = {
+		.opcode		= P4_OPCODE(P4_EVENT_SSE_INPUT_ASSIST),
+		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_PACKED_SP_UOP] = {
+		.opcode		= P4_OPCODE(P4_EVENT_PACKED_SP_UOP),
+		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_PACKED_DP_UOP] = {
+		.opcode		= P4_OPCODE(P4_EVENT_PACKED_DP_UOP),
+		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_SCALAR_SP_UOP] = {
+		.opcode		= P4_OPCODE(P4_EVENT_SCALAR_SP_UOP),
+		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_SCALAR_DP_UOP] = {
+		.opcode		= P4_OPCODE(P4_EVENT_SCALAR_DP_UOP),
+		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_64BIT_MMX_UOP] = {
+		.opcode		= P4_OPCODE(P4_EVENT_64BIT_MMX_UOP),
+		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_128BIT_MMX_UOP] = {
+		.opcode		= P4_OPCODE(P4_EVENT_128BIT_MMX_UOP),
+		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_X87_FP_UOP] = {
+		.opcode		= P4_OPCODE(P4_EVENT_X87_FP_UOP),
+		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_TC_MISC] = {
+		.opcode		= P4_OPCODE(P4_EVENT_TC_MISC),
+		.escr_msr	= { MSR_P4_TC_ESCR0, MSR_P4_TC_ESCR1 },
+		.cntr		= { {4, 5, -1}, {6, 7, -1} },
+	},
+	[P4_EVENT_GLOBAL_POWER_EVENTS] = {
+		.opcode		= P4_OPCODE(P4_EVENT_GLOBAL_POWER_EVENTS),
+		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_TC_MS_XFER] = {
+		.opcode		= P4_OPCODE(P4_EVENT_TC_MS_XFER),
+		.escr_msr	= { MSR_P4_MS_ESCR0, MSR_P4_MS_ESCR1 },
+		.cntr		= { {4, 5, -1}, {6, 7, -1} },
+	},
+	[P4_EVENT_UOP_QUEUE_WRITES] = {
+		.opcode		= P4_OPCODE(P4_EVENT_UOP_QUEUE_WRITES),
+		.escr_msr	= { MSR_P4_MS_ESCR0, MSR_P4_MS_ESCR1 },
+		.cntr		= { {4, 5, -1}, {6, 7, -1} },
+	},
+	[P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE] = {
+		.opcode		= P4_OPCODE(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE),
+		.escr_msr	= { MSR_P4_TBPU_ESCR0 , MSR_P4_TBPU_ESCR0 },
+		.cntr		= { {4, 5, -1}, {6, 7, -1} },
+	},
+	[P4_EVENT_RETIRED_BRANCH_TYPE] = {
+		.opcode		= P4_OPCODE(P4_EVENT_RETIRED_BRANCH_TYPE),
+		.escr_msr	= { MSR_P4_TBPU_ESCR0 , MSR_P4_TBPU_ESCR1 },
+		.cntr		= { {4, 5, -1}, {6, 7, -1} },
+	},
+	[P4_EVENT_RESOURCE_STALL] = {
+		.opcode		= P4_OPCODE(P4_EVENT_RESOURCE_STALL),
+		.escr_msr	= { MSR_P4_ALF_ESCR0, MSR_P4_ALF_ESCR1 },
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+	[P4_EVENT_WC_BUFFER] = {
+		.opcode		= P4_OPCODE(P4_EVENT_WC_BUFFER),
+		.escr_msr	= { MSR_P4_DAC_ESCR0, MSR_P4_DAC_ESCR1 },
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_B2B_CYCLES] = {
+		.opcode		= P4_OPCODE(P4_EVENT_B2B_CYCLES),
+		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_BNR] = {
+		.opcode		= P4_OPCODE(P4_EVENT_BNR),
+		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_SNOOP] = {
+		.opcode		= P4_OPCODE(P4_EVENT_SNOOP),
+		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_RESPONSE] = {
+		.opcode		= P4_OPCODE(P4_EVENT_RESPONSE),
+		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_FRONT_END_EVENT] = {
+		.opcode		= P4_OPCODE(P4_EVENT_FRONT_END_EVENT),
+		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+	[P4_EVENT_EXECUTION_EVENT] = {
+		.opcode		= P4_OPCODE(P4_EVENT_EXECUTION_EVENT),
+		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+	[P4_EVENT_REPLAY_EVENT] = {
+		.opcode		= P4_OPCODE(P4_EVENT_REPLAY_EVENT),
+		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+	[P4_EVENT_INSTR_RETIRED] = {
+		.opcode		= P4_OPCODE(P4_EVENT_INSTR_RETIRED),
+		.escr_msr	= { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 },
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+	[P4_EVENT_UOPS_RETIRED] = {
+		.opcode		= P4_OPCODE(P4_EVENT_UOPS_RETIRED),
+		.escr_msr	= { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 },
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+	[P4_EVENT_UOP_TYPE] = {
+		.opcode		= P4_OPCODE(P4_EVENT_UOP_TYPE),
+		.escr_msr	= { MSR_P4_RAT_ESCR0, MSR_P4_RAT_ESCR1 },
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+	[P4_EVENT_BRANCH_RETIRED] = {
+		.opcode		= P4_OPCODE(P4_EVENT_BRANCH_RETIRED),
+		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+	[P4_EVENT_MISPRED_BRANCH_RETIRED] = {
+		.opcode		= P4_OPCODE(P4_EVENT_MISPRED_BRANCH_RETIRED),
+		.escr_msr	= { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 },
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+	[P4_EVENT_X87_ASSIST] = {
+		.opcode		= P4_OPCODE(P4_EVENT_X87_ASSIST),
+		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+	[P4_EVENT_MACHINE_CLEAR] = {
+		.opcode		= P4_OPCODE(P4_EVENT_MACHINE_CLEAR),
+		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+	[P4_EVENT_INSTR_COMPLETED] = {
+		.opcode		= P4_OPCODE(P4_EVENT_INSTR_COMPLETED),
+		.escr_msr	= { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 },
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+};
+
+#define P4_GEN_CACHE_EVENT(event, bit, cache_event)			  \
+	p4_config_pack_escr(P4_ESCR_EVENT(event)			| \
+			    P4_ESCR_EMASK_BIT(event, bit))		| \
+	p4_config_pack_cccr(cache_event					| \
+			    P4_CCCR_ESEL(P4_OPCODE_ESEL(P4_OPCODE(event))))
+
+static __initconst const u64 p4_hw_cache_event_ids
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] =
+{
+ [ C(L1D ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = P4_GEN_CACHE_EVENT(P4_EVENT_REPLAY_EVENT, NBOGUS,
+						P4_CACHE__1stl_cache_load_miss_retired),
+	},
+ },
+ [ C(LL  ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = P4_GEN_CACHE_EVENT(P4_EVENT_REPLAY_EVENT, NBOGUS,
+						P4_CACHE__2ndl_cache_load_miss_retired),
+	},
+},
+ [ C(DTLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = P4_GEN_CACHE_EVENT(P4_EVENT_REPLAY_EVENT, NBOGUS,
+						P4_CACHE__dtlb_load_miss_retired),
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = P4_GEN_CACHE_EVENT(P4_EVENT_REPLAY_EVENT, NBOGUS,
+						P4_CACHE__dtlb_store_miss_retired),
+	},
+ },
+ [ C(ITLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = P4_GEN_CACHE_EVENT(P4_EVENT_ITLB_REFERENCE, HIT,
+						P4_CACHE__itlb_reference_hit),
+		[ C(RESULT_MISS)   ] = P4_GEN_CACHE_EVENT(P4_EVENT_ITLB_REFERENCE, MISS,
+						P4_CACHE__itlb_reference_miss),
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+};
+
+static u64 p4_general_events[PERF_COUNT_HW_MAX] = {
+  /* non-halted CPU clocks */
+  [PERF_COUNT_HW_CPU_CYCLES] =
+	p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_GLOBAL_POWER_EVENTS)		|
+		P4_ESCR_EMASK_BIT(P4_EVENT_GLOBAL_POWER_EVENTS, RUNNING)),
+
+  /*
+   * retired instructions
+   * in a sake of simplicity we don't use the FSB tagging
+   */
+  [PERF_COUNT_HW_INSTRUCTIONS] =
+	p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_INSTR_RETIRED)		|
+		P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, NBOGUSNTAG)		|
+		P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, BOGUSNTAG)),
+
+  /* cache hits */
+  [PERF_COUNT_HW_CACHE_REFERENCES] =
+	p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_BSQ_CACHE_REFERENCE)		|
+		P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITS)	|
+		P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITE)	|
+		P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITM)	|
+		P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITS)	|
+		P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITE)	|
+		P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITM)),
+
+  /* cache misses */
+  [PERF_COUNT_HW_CACHE_MISSES] =
+	p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_BSQ_CACHE_REFERENCE)		|
+		P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_MISS)	|
+		P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_MISS)	|
+		P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, WR_2ndL_MISS)),
+
+  /* branch instructions retired */
+  [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] =
+	p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_RETIRED_BRANCH_TYPE)		|
+		P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, CONDITIONAL)	|
+		P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, CALL)		|
+		P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, RETURN)		|
+		P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, INDIRECT)),
+
+  /* mispredicted branches retired */
+  [PERF_COUNT_HW_BRANCH_MISSES]	=
+	p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_MISPRED_BRANCH_RETIRED)	|
+		P4_ESCR_EMASK_BIT(P4_EVENT_MISPRED_BRANCH_RETIRED, NBOGUS)),
+
+  /* bus ready clocks (cpu is driving #DRDY_DRV\#DRDY_OWN):  */
+  [PERF_COUNT_HW_BUS_CYCLES] =
+	p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_FSB_DATA_ACTIVITY)		|
+		P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_DRV)		|
+		P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_OWN))	|
+	p4_config_pack_cccr(P4_CCCR_EDGE | P4_CCCR_COMPARE),
+};
+
+static struct p4_event_bind *p4_config_get_bind(u64 config)
+{
+	unsigned int evnt = p4_config_unpack_event(config);
+	struct p4_event_bind *bind = NULL;
+
+	if (evnt < ARRAY_SIZE(p4_event_bind_map))
+		bind = &p4_event_bind_map[evnt];
+
+	return bind;
+}
+
+static u64 p4_pmu_event_map(int hw_event)
+{
+	struct p4_event_bind *bind;
+	unsigned int esel;
+	u64 config;
+
+	if (hw_event > ARRAY_SIZE(p4_general_events)) {
+		printk_once(KERN_ERR "P4 PMU: Bad index: %i\n", hw_event);
+		return 0;
+	}
+
+	config = p4_general_events[hw_event];
+	bind = p4_config_get_bind(config);
+	esel = P4_OPCODE_ESEL(bind->opcode);
+	config |= p4_config_pack_cccr(P4_CCCR_ESEL(esel));
+
+	return config;
+}
+
+static int p4_hw_config(struct perf_event *event)
+{
+	int cpu = raw_smp_processor_id();
+	u32 escr, cccr;
+
+	/*
+	 * the reason we use cpu that early is that: if we get scheduled
+	 * first time on the same cpu -- we will not need swap thread
+	 * specific flags in config (and will save some cpu cycles)
+	 */
+
+	cccr = p4_default_cccr_conf(cpu);
+	escr = p4_default_escr_conf(cpu, event->attr.exclude_kernel,
+					 event->attr.exclude_user);
+	event->hw.config = p4_config_pack_escr(escr) |
+			   p4_config_pack_cccr(cccr);
+
+	if (p4_ht_active() && p4_ht_thread(cpu))
+		event->hw.config = p4_set_ht_bit(event->hw.config);
+
+	if (event->attr.type != PERF_TYPE_RAW)
+		return 0;
+
+	/*
+	 * We don't control raw events so it's up to the caller
+	 * to pass sane values (and we don't count the thread number
+	 * on HT machine but allow HT-compatible specifics to be
+	 * passed on)
+	 *
+	 * XXX: HT wide things should check perf_paranoid_cpu() &&
+	 *      CAP_SYS_ADMIN
+	 */
+	event->hw.config |= event->attr.config &
+		(p4_config_pack_escr(P4_ESCR_MASK_HT) |
+		 p4_config_pack_cccr(P4_CCCR_MASK_HT));
+
+	return 0;
+}
+
+static inline void p4_pmu_clear_cccr_ovf(struct hw_perf_event *hwc)
+{
+	unsigned long dummy;
+
+	rdmsrl(hwc->config_base + hwc->idx, dummy);
+	if (dummy & P4_CCCR_OVF) {
+		(void)checking_wrmsrl(hwc->config_base + hwc->idx,
+			((u64)dummy) & ~P4_CCCR_OVF);
+	}
+}
+
+static inline void p4_pmu_disable_event(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	/*
+	 * If event gets disabled while counter is in overflowed
+	 * state we need to clear P4_CCCR_OVF, otherwise interrupt get
+	 * asserted again and again
+	 */
+	(void)checking_wrmsrl(hwc->config_base + hwc->idx,
+		(u64)(p4_config_unpack_cccr(hwc->config)) &
+			~P4_CCCR_ENABLE & ~P4_CCCR_OVF & ~P4_CCCR_RESERVED);
+}
+
+static void p4_pmu_disable_all(void)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	int idx;
+
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
+		struct perf_event *event = cpuc->events[idx];
+		if (!test_bit(idx, cpuc->active_mask))
+			continue;
+		p4_pmu_disable_event(event);
+	}
+}
+
+static void p4_pmu_enable_event(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	int thread = p4_ht_config_thread(hwc->config);
+	u64 escr_conf = p4_config_unpack_escr(p4_clear_ht_bit(hwc->config));
+	unsigned int idx = p4_config_unpack_event(hwc->config);
+	unsigned int idx_cache = p4_config_unpack_cache_event(hwc->config);
+	struct p4_event_bind *bind;
+	struct p4_cache_event_bind *bind_cache;
+	u64 escr_addr, cccr;
+
+	bind = &p4_event_bind_map[idx];
+	escr_addr = (u64)bind->escr_msr[thread];
+
+	/*
+	 * - we dont support cascaded counters yet
+	 * - and counter 1 is broken (erratum)
+	 */
+	WARN_ON_ONCE(p4_is_event_cascaded(hwc->config));
+	WARN_ON_ONCE(hwc->idx == 1);
+
+	/* we need a real Event value */
+	escr_conf &= ~P4_ESCR_EVENT_MASK;
+	escr_conf |= P4_ESCR_EVENT(P4_OPCODE_EVNT(bind->opcode));
+
+	cccr = p4_config_unpack_cccr(hwc->config);
+
+	/*
+	 * it could be Cache event so that we need to
+	 * set metrics into additional MSRs
+	 */
+	BUILD_BUG_ON(P4_CACHE__MAX > P4_CCCR_CACHE_OPS_MASK);
+	if (idx_cache > P4_CACHE__NONE &&
+		idx_cache < ARRAY_SIZE(p4_cache_event_bind_map)) {
+		bind_cache = &p4_cache_event_bind_map[idx_cache];
+		(void)checking_wrmsrl(MSR_IA32_PEBS_ENABLE, (u64)bind_cache->metric_pebs);
+		(void)checking_wrmsrl(MSR_P4_PEBS_MATRIX_VERT, (u64)bind_cache->metric_vert);
+	}
+
+	(void)checking_wrmsrl(escr_addr, escr_conf);
+	(void)checking_wrmsrl(hwc->config_base + hwc->idx,
+				(cccr & ~P4_CCCR_RESERVED) | P4_CCCR_ENABLE);
+}
+
+static void p4_pmu_enable_all(int added)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	int idx;
+
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
+		struct perf_event *event = cpuc->events[idx];
+		if (!test_bit(idx, cpuc->active_mask))
+			continue;
+		p4_pmu_enable_event(event);
+	}
+}
+
+static int p4_pmu_handle_irq(struct pt_regs *regs)
+{
+	struct perf_sample_data data;
+	struct cpu_hw_events *cpuc;
+	struct perf_event *event;
+	struct hw_perf_event *hwc;
+	int idx, handled = 0;
+	u64 val;
+
+	data.addr = 0;
+	data.raw = NULL;
+
+	cpuc = &__get_cpu_var(cpu_hw_events);
+
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
+
+		if (!test_bit(idx, cpuc->active_mask))
+			continue;
+
+		event = cpuc->events[idx];
+		hwc = &event->hw;
+
+		WARN_ON_ONCE(hwc->idx != idx);
+
+		/*
+		 * FIXME: Redundant call, actually not needed
+		 * but just to check if we're screwed
+		 */
+		p4_pmu_clear_cccr_ovf(hwc);
+
+		val = x86_perf_event_update(event);
+		if (val & (1ULL << (x86_pmu.cntval_bits - 1)))
+			continue;
+
+		/*
+		 * event overflow
+		 */
+		handled		= 1;
+		data.period	= event->hw.last_period;
+
+		if (!x86_perf_event_set_period(event))
+			continue;
+		if (perf_event_overflow(event, 1, &data, regs))
+			p4_pmu_disable_event(event);
+	}
+
+	if (handled) {
+		/* p4 quirk: unmask it again */
+		apic_write(APIC_LVTPC, apic_read(APIC_LVTPC) & ~APIC_LVT_MASKED);
+		inc_irq_stat(apic_perf_irqs);
+	}
+
+	return handled;
+}
+
+/*
+ * swap thread specific fields according to a thread
+ * we are going to run on
+ */
+static void p4_pmu_swap_config_ts(struct hw_perf_event *hwc, int cpu)
+{
+	u32 escr, cccr;
+
+	/*
+	 * we either lucky and continue on same cpu or no HT support
+	 */
+	if (!p4_should_swap_ts(hwc->config, cpu))
+		return;
+
+	/*
+	 * the event is migrated from an another logical
+	 * cpu, so we need to swap thread specific flags
+	 */
+
+	escr = p4_config_unpack_escr(hwc->config);
+	cccr = p4_config_unpack_cccr(hwc->config);
+
+	if (p4_ht_thread(cpu)) {
+		cccr &= ~P4_CCCR_OVF_PMI_T0;
+		cccr |= P4_CCCR_OVF_PMI_T1;
+		if (escr & P4_ESCR_T0_OS) {
+			escr &= ~P4_ESCR_T0_OS;
+			escr |= P4_ESCR_T1_OS;
+		}
+		if (escr & P4_ESCR_T0_USR) {
+			escr &= ~P4_ESCR_T0_USR;
+			escr |= P4_ESCR_T1_USR;
+		}
+		hwc->config  = p4_config_pack_escr(escr);
+		hwc->config |= p4_config_pack_cccr(cccr);
+		hwc->config |= P4_CONFIG_HT;
+	} else {
+		cccr &= ~P4_CCCR_OVF_PMI_T1;
+		cccr |= P4_CCCR_OVF_PMI_T0;
+		if (escr & P4_ESCR_T1_OS) {
+			escr &= ~P4_ESCR_T1_OS;
+			escr |= P4_ESCR_T0_OS;
+		}
+		if (escr & P4_ESCR_T1_USR) {
+			escr &= ~P4_ESCR_T1_USR;
+			escr |= P4_ESCR_T0_USR;
+		}
+		hwc->config  = p4_config_pack_escr(escr);
+		hwc->config |= p4_config_pack_cccr(cccr);
+		hwc->config &= ~P4_CONFIG_HT;
+	}
+}
+
+/* ESCRs are not sequential in memory so we need a map */
+static const unsigned int p4_escr_map[ARCH_P4_TOTAL_ESCR] = {
+	MSR_P4_ALF_ESCR0,	/*  0 */
+	MSR_P4_ALF_ESCR1,	/*  1 */
+	MSR_P4_BPU_ESCR0,	/*  2 */
+	MSR_P4_BPU_ESCR1,	/*  3 */
+	MSR_P4_BSU_ESCR0,	/*  4 */
+	MSR_P4_BSU_ESCR1,	/*  5 */
+	MSR_P4_CRU_ESCR0,	/*  6 */
+	MSR_P4_CRU_ESCR1,	/*  7 */
+	MSR_P4_CRU_ESCR2,	/*  8 */
+	MSR_P4_CRU_ESCR3,	/*  9 */
+	MSR_P4_CRU_ESCR4,	/* 10 */
+	MSR_P4_CRU_ESCR5,	/* 11 */
+	MSR_P4_DAC_ESCR0,	/* 12 */
+	MSR_P4_DAC_ESCR1,	/* 13 */
+	MSR_P4_FIRM_ESCR0,	/* 14 */
+	MSR_P4_FIRM_ESCR1,	/* 15 */
+	MSR_P4_FLAME_ESCR0,	/* 16 */
+	MSR_P4_FLAME_ESCR1,	/* 17 */
+	MSR_P4_FSB_ESCR0,	/* 18 */
+	MSR_P4_FSB_ESCR1,	/* 19 */
+	MSR_P4_IQ_ESCR0,	/* 20 */
+	MSR_P4_IQ_ESCR1,	/* 21 */
+	MSR_P4_IS_ESCR0,	/* 22 */
+	MSR_P4_IS_ESCR1,	/* 23 */
+	MSR_P4_ITLB_ESCR0,	/* 24 */
+	MSR_P4_ITLB_ESCR1,	/* 25 */
+	MSR_P4_IX_ESCR0,	/* 26 */
+	MSR_P4_IX_ESCR1,	/* 27 */
+	MSR_P4_MOB_ESCR0,	/* 28 */
+	MSR_P4_MOB_ESCR1,	/* 29 */
+	MSR_P4_MS_ESCR0,	/* 30 */
+	MSR_P4_MS_ESCR1,	/* 31 */
+	MSR_P4_PMH_ESCR0,	/* 32 */
+	MSR_P4_PMH_ESCR1,	/* 33 */
+	MSR_P4_RAT_ESCR0,	/* 34 */
+	MSR_P4_RAT_ESCR1,	/* 35 */
+	MSR_P4_SAAT_ESCR0,	/* 36 */
+	MSR_P4_SAAT_ESCR1,	/* 37 */
+	MSR_P4_SSU_ESCR0,	/* 38 */
+	MSR_P4_SSU_ESCR1,	/* 39 */
+	MSR_P4_TBPU_ESCR0,	/* 40 */
+	MSR_P4_TBPU_ESCR1,	/* 41 */
+	MSR_P4_TC_ESCR0,	/* 42 */
+	MSR_P4_TC_ESCR1,	/* 43 */
+	MSR_P4_U2L_ESCR0,	/* 44 */
+	MSR_P4_U2L_ESCR1,	/* 45 */
+};
+
+static int p4_get_escr_idx(unsigned int addr)
+{
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(p4_escr_map); i++) {
+		if (addr == p4_escr_map[i])
+			return i;
+	}
+
+	return -1;
+}
+
+static int p4_next_cntr(int thread, unsigned long *used_mask,
+			struct p4_event_bind *bind)
+{
+	int i = 0, j;
+
+	for (i = 0; i < P4_CNTR_LIMIT; i++) {
+		j = bind->cntr[thread][i++];
+		if (j == -1 || !test_bit(j, used_mask))
+			return j;
+	}
+
+	return -1;
+}
+
+static int p4_pmu_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign)
+{
+	unsigned long used_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
+	unsigned long escr_mask[BITS_TO_LONGS(ARCH_P4_TOTAL_ESCR)];
+	int cpu = raw_smp_processor_id();
+	struct hw_perf_event *hwc;
+	struct p4_event_bind *bind;
+	unsigned int i, thread, num;
+	int cntr_idx, escr_idx;
+
+	bitmap_zero(used_mask, X86_PMC_IDX_MAX);
+	bitmap_zero(escr_mask, ARCH_P4_TOTAL_ESCR);
+
+	for (i = 0, num = n; i < n; i++, num--) {
+
+		hwc = &cpuc->event_list[i]->hw;
+		thread = p4_ht_thread(cpu);
+		bind = p4_config_get_bind(hwc->config);
+		escr_idx = p4_get_escr_idx(bind->escr_msr[thread]);
+
+		if (hwc->idx != -1 && !p4_should_swap_ts(hwc->config, cpu)) {
+			cntr_idx = hwc->idx;
+			if (assign)
+				assign[i] = hwc->idx;
+			goto reserve;
+		}
+
+		cntr_idx = p4_next_cntr(thread, used_mask, bind);
+		if (cntr_idx == -1 || test_bit(escr_idx, escr_mask))
+			goto done;
+
+		p4_pmu_swap_config_ts(hwc, cpu);
+		if (assign)
+			assign[i] = cntr_idx;
+reserve:
+		set_bit(cntr_idx, used_mask);
+		set_bit(escr_idx, escr_mask);
+	}
+
+done:
+	return num ? -ENOSPC : 0;
+}
+
+static __initconst const struct x86_pmu p4_pmu = {
+	.name			= "Netburst P4/Xeon",
+	.handle_irq		= p4_pmu_handle_irq,
+	.disable_all		= p4_pmu_disable_all,
+	.enable_all		= p4_pmu_enable_all,
+	.enable			= p4_pmu_enable_event,
+	.disable		= p4_pmu_disable_event,
+	.eventsel		= MSR_P4_BPU_CCCR0,
+	.perfctr		= MSR_P4_BPU_PERFCTR0,
+	.event_map		= p4_pmu_event_map,
+	.max_events		= ARRAY_SIZE(p4_general_events),
+	.get_event_constraints	= x86_get_event_constraints,
+	/*
+	 * IF HT disabled we may need to use all
+	 * ARCH_P4_MAX_CCCR counters simulaneously
+	 * though leave it restricted at moment assuming
+	 * HT is on
+	 */
+	.num_counters		= ARCH_P4_MAX_CCCR,
+	.apic			= 1,
+	.cntval_bits		= 40,
+	.cntval_mask		= (1ULL << 40) - 1,
+	.max_period		= (1ULL << 39) - 1,
+	.hw_config		= p4_hw_config,
+	.schedule_events	= p4_pmu_schedule_events,
+};
+
+static __init int p4_pmu_init(void)
+{
+	unsigned int low, high;
+
+	/* If we get stripped -- indexig fails */
+	BUILD_BUG_ON(ARCH_P4_MAX_CCCR > X86_PMC_MAX_GENERIC);
+
+	rdmsr(MSR_IA32_MISC_ENABLE, low, high);
+	if (!(low & (1 << 7))) {
+		pr_cont("unsupported Netburst CPU model %d ",
+			boot_cpu_data.x86_model);
+		return -ENODEV;
+	}
+
+	memcpy(hw_cache_event_ids, p4_hw_cache_event_ids,
+		sizeof(hw_cache_event_ids));
+
+	pr_cont("Netburst events, ");
+
+	x86_pmu = p4_pmu;
+
+	return 0;
+}
+
+#endif /* CONFIG_CPU_SUP_INTEL */

arch/x86/kernel/cpu/perf_event_p6.c

@@ -27,24 +27,6 @@ static u64 p6_pmu_event_map(int hw_event)
  */
 #define P6_NOP_EVENT			0x0000002EULL
 
-static u64 p6_pmu_raw_event(u64 hw_event)
-{
-#define P6_EVNTSEL_EVENT_MASK		0x000000FFULL
-#define P6_EVNTSEL_UNIT_MASK		0x0000FF00ULL
-#define P6_EVNTSEL_EDGE_MASK		0x00040000ULL
-#define P6_EVNTSEL_INV_MASK		0x00800000ULL
-#define P6_EVNTSEL_REG_MASK		0xFF000000ULL
-
-#define P6_EVNTSEL_MASK			\
-	(P6_EVNTSEL_EVENT_MASK |	\
-	 P6_EVNTSEL_UNIT_MASK  |	\
-	 P6_EVNTSEL_EDGE_MASK  |	\
-	 P6_EVNTSEL_INV_MASK   |	\
-	 P6_EVNTSEL_REG_MASK)
-
-	return hw_event & P6_EVNTSEL_MASK;
-}
-
 static struct event_constraint p6_event_constraints[] =
 {
 	INTEL_EVENT_CONSTRAINT(0xc1, 0x1),	/* FLOPS */
@@ -66,7 +48,7 @@ static void p6_pmu_disable_all(void)
 	wrmsrl(MSR_P6_EVNTSEL0, val);
 }
 
-static void p6_pmu_enable_all(void)
+static void p6_pmu_enable_all(int added)
 {
 	unsigned long val;
 
@@ -102,22 +84,23 @@ static void p6_pmu_enable_event(struct perf_event *event)
 	(void)checking_wrmsrl(hwc->config_base + hwc->idx, val);
 }
 
-static __initconst struct x86_pmu p6_pmu = {
+static __initconst const struct x86_pmu p6_pmu = {
 	.name			= "p6",
 	.handle_irq		= x86_pmu_handle_irq,
 	.disable_all		= p6_pmu_disable_all,
 	.enable_all		= p6_pmu_enable_all,
 	.enable			= p6_pmu_enable_event,
 	.disable		= p6_pmu_disable_event,
+	.hw_config		= x86_pmu_hw_config,
+	.schedule_events	= x86_schedule_events,
 	.eventsel		= MSR_P6_EVNTSEL0,
 	.perfctr		= MSR_P6_PERFCTR0,
 	.event_map		= p6_pmu_event_map,
-	.raw_event		= p6_pmu_raw_event,
 	.max_events		= ARRAY_SIZE(p6_perfmon_event_map),
 	.apic			= 1,
 	.max_period		= (1ULL << 31) - 1,
 	.version		= 0,
-	.num_events		= 2,
+	.num_counters		= 2,
 	/*
 	 * Events have 40 bits implemented. However they are designed such
 	 * that bits [32-39] are sign extensions of bit 31. As such the
@@ -125,8 +108,8 @@ static __initconst struct x86_pmu p6_pmu = {
 	 *
 	 * See IA-32 Intel Architecture Software developer manual Vol 3B
 	 */
-	.event_bits		= 32,
-	.event_mask		= (1ULL << 32) - 1,
+	.cntval_bits		= 32,
+	.cntval_mask		= (1ULL << 32) - 1,
 	.get_event_constraints	= x86_get_event_constraints,
 	.event_constraints	= p6_event_constraints,
 };

arch/x86/kernel/ds.c

@@ -1,1437 +0,0 @@
-/*
- * Debug Store support
- *
- * This provides a low-level interface to the hardware's Debug Store
- * feature that is used for branch trace store (BTS) and
- * precise-event based sampling (PEBS).
- *
- * It manages:
- * - DS and BTS hardware configuration
- * - buffer overflow handling (to be done)
- * - buffer access
- *
- * It does not do:
- * - security checking (is the caller allowed to trace the task)
- * - buffer allocation (memory accounting)
- *
- *
- * Copyright (C) 2007-2009 Intel Corporation.
- * Markus Metzger <markus.t.metzger@intel.com>, 2007-2009
- */
-
-#include <linux/kernel.h>
-#include <linux/string.h>
-#include <linux/errno.h>
-#include <linux/sched.h>
-#include <linux/slab.h>
-#include <linux/mm.h>
-#include <linux/trace_clock.h>
-
-#include <asm/ds.h>
-
-#include "ds_selftest.h"
-
-/*
- * The configuration for a particular DS hardware implementation:
- */
-struct ds_configuration {
-	/* The name of the configuration: */
-	const char		*name;
-
-	/* The size of pointer-typed fields in DS, BTS, and PEBS: */
-	unsigned char		sizeof_ptr_field;
-
-	/* The size of a BTS/PEBS record in bytes: */
-	unsigned char		sizeof_rec[2];
-
-	/* The number of pebs counter reset values in the DS structure. */
-	unsigned char		nr_counter_reset;
-
-	/* Control bit-masks indexed by enum ds_feature: */
-	unsigned long		ctl[dsf_ctl_max];
-};
-static struct ds_configuration ds_cfg __read_mostly;
-
-
-/* Maximal size of a DS configuration: */
-#define MAX_SIZEOF_DS		0x80
-
-/* Maximal size of a BTS record: */
-#define MAX_SIZEOF_BTS		(3 * 8)
-
-/* BTS and PEBS buffer alignment: */
-#define DS_ALIGNMENT		(1 << 3)
-
-/* Number of buffer pointers in DS: */
-#define NUM_DS_PTR_FIELDS	8
-
-/* Size of a pebs reset value in DS: */
-#define PEBS_RESET_FIELD_SIZE	8
-
-/* Mask of control bits in the DS MSR register: */
-#define BTS_CONTROL				  \
-	( ds_cfg.ctl[dsf_bts]			| \
-	  ds_cfg.ctl[dsf_bts_kernel]		| \
-	  ds_cfg.ctl[dsf_bts_user]		| \
-	  ds_cfg.ctl[dsf_bts_overflow] )
-
-/*
- * A BTS or PEBS tracer.
- *
- * This holds the configuration of the tracer and serves as a handle
- * to identify tracers.
- */
-struct ds_tracer {
-	/* The DS context (partially) owned by this tracer. */
-	struct ds_context	*context;
-	/* The buffer provided on ds_request() and its size in bytes. */
-	void			*buffer;
-	size_t			size;
-};
-
-struct bts_tracer {
-	/* The common DS part: */
-	struct ds_tracer	ds;
-
-	/* The trace including the DS configuration: */
-	struct bts_trace	trace;
-
-	/* Buffer overflow notification function: */
-	bts_ovfl_callback_t	ovfl;
-
-	/* Active flags affecting trace collection. */
-	unsigned int		flags;
-};
-
-struct pebs_tracer {
-	/* The common DS part: */
-	struct ds_tracer	ds;
-
-	/* The trace including the DS configuration: */
-	struct pebs_trace	trace;
-
-	/* Buffer overflow notification function: */
-	pebs_ovfl_callback_t	ovfl;
-};
-
-/*
- * Debug Store (DS) save area configuration (see Intel64 and IA32
- * Architectures Software Developer's Manual, section 18.5)
- *
- * The DS configuration consists of the following fields; different
- * architetures vary in the size of those fields.
- *
- * - double-word aligned base linear address of the BTS buffer
- * - write pointer into the BTS buffer
- * - end linear address of the BTS buffer (one byte beyond the end of
- *   the buffer)
- * - interrupt pointer into BTS buffer
- *   (interrupt occurs when write pointer passes interrupt pointer)
- * - double-word aligned base linear address of the PEBS buffer
- * - write pointer into the PEBS buffer
- * - end linear address of the PEBS buffer (one byte beyond the end of
- *   the buffer)
- * - interrupt pointer into PEBS buffer
- *   (interrupt occurs when write pointer passes interrupt pointer)
- * - value to which counter is reset following counter overflow
- *
- * Later architectures use 64bit pointers throughout, whereas earlier
- * architectures use 32bit pointers in 32bit mode.
- *
- *
- * We compute the base address for the first 8 fields based on:
- * - the field size stored in the DS configuration
- * - the relative field position
- * - an offset giving the start of the respective region
- *
- * This offset is further used to index various arrays holding
- * information for BTS and PEBS at the respective index.
- *
- * On later 32bit processors, we only access the lower 32bit of the
- * 64bit pointer fields. The upper halves will be zeroed out.
- */
-
-enum ds_field {
-	ds_buffer_base = 0,
-	ds_index,
-	ds_absolute_maximum,
-	ds_interrupt_threshold,
-};
-
-enum ds_qualifier {
-	ds_bts = 0,
-	ds_pebs
-};
-
-static inline unsigned long
-ds_get(const unsigned char *base, enum ds_qualifier qual, enum ds_field field)
-{
-	base += (ds_cfg.sizeof_ptr_field * (field + (4 * qual)));
-	return *(unsigned long *)base;
-}
-
-static inline void
-ds_set(unsigned char *base, enum ds_qualifier qual, enum ds_field field,
-       unsigned long value)
-{
-	base += (ds_cfg.sizeof_ptr_field * (field + (4 * qual)));
-	(*(unsigned long *)base) = value;
-}
-
-
-/*
- * Locking is done only for allocating BTS or PEBS resources.
- */
-static DEFINE_SPINLOCK(ds_lock);
-
-/*
- * We either support (system-wide) per-cpu or per-thread allocation.
- * We distinguish the two based on the task_struct pointer, where a
- * NULL pointer indicates per-cpu allocation for the current cpu.
- *
- * Allocations are use-counted. As soon as resources are allocated,
- * further allocations must be of the same type (per-cpu or
- * per-thread). We model this by counting allocations (i.e. the number
- * of tracers of a certain type) for one type negatively:
- *   =0  no tracers
- *   >0  number of per-thread tracers
- *   <0  number of per-cpu tracers
- *
- * Tracers essentially gives the number of ds contexts for a certain
- * type of allocation.
- */
-static atomic_t tracers = ATOMIC_INIT(0);
-
-static inline int get_tracer(struct task_struct *task)
-{
-	int error;
-
-	spin_lock_irq(&ds_lock);
-
-	if (task) {
-		error = -EPERM;
-		if (atomic_read(&tracers) < 0)
-			goto out;
-		atomic_inc(&tracers);
-	} else {
-		error = -EPERM;
-		if (atomic_read(&tracers) > 0)
-			goto out;
-		atomic_dec(&tracers);
-	}
-
-	error = 0;
-out:
-	spin_unlock_irq(&ds_lock);
-	return error;
-}
-
-static inline void put_tracer(struct task_struct *task)
-{
-	if (task)
-		atomic_dec(&tracers);
-	else
-		atomic_inc(&tracers);
-}
-
-/*
- * The DS context is either attached to a thread or to a cpu:
- * - in the former case, the thread_struct contains a pointer to the
- *   attached context.
- * - in the latter case, we use a static array of per-cpu context
- *   pointers.
- *
- * Contexts are use-counted. They are allocated on first access and
- * deallocated when the last user puts the context.
- */
-struct ds_context {
-	/* The DS configuration; goes into MSR_IA32_DS_AREA: */
-	unsigned char		ds[MAX_SIZEOF_DS];
-
-	/* The owner of the BTS and PEBS configuration, respectively: */
-	struct bts_tracer	*bts_master;
-	struct pebs_tracer	*pebs_master;
-
-	/* Use count: */
-	unsigned long		count;
-
-	/* Pointer to the context pointer field: */
-	struct ds_context	**this;
-
-	/* The traced task; NULL for cpu tracing: */
-	struct task_struct	*task;
-
-	/* The traced cpu; only valid if task is NULL: */
-	int			cpu;
-};
-
-static DEFINE_PER_CPU(struct ds_context *, cpu_ds_context);
-
-
-static struct ds_context *ds_get_context(struct task_struct *task, int cpu)
-{
-	struct ds_context **p_context =
-		(task ? &task->thread.ds_ctx : &per_cpu(cpu_ds_context, cpu));
-	struct ds_context *context = NULL;
-	struct ds_context *new_context = NULL;
-
-	/* Chances are small that we already have a context. */
-	new_context = kzalloc(sizeof(*new_context), GFP_KERNEL);
-	if (!new_context)
-		return NULL;
-
-	spin_lock_irq(&ds_lock);
-
-	context = *p_context;
-	if (likely(!context)) {
-		context = new_context;
-
-		context->this = p_context;
-		context->task = task;
-		context->cpu = cpu;
-		context->count = 0;
-
-		*p_context = context;
-	}
-
-	context->count++;
-
-	spin_unlock_irq(&ds_lock);
-
-	if (context != new_context)
-		kfree(new_context);
-
-	return context;
-}
-
-static void ds_put_context(struct ds_context *context)
-{
-	struct task_struct *task;
-	unsigned long irq;
-
-	if (!context)
-		return;
-
-	spin_lock_irqsave(&ds_lock, irq);
-
-	if (--context->count) {
-		spin_unlock_irqrestore(&ds_lock, irq);
-		return;
-	}
-
-	*(context->this) = NULL;
-
-	task = context->task;
-
-	if (task)
-		clear_tsk_thread_flag(task, TIF_DS_AREA_MSR);
-
-	/*
-	 * We leave the (now dangling) pointer to the DS configuration in
-	 * the DS_AREA msr. This is as good or as bad as replacing it with
-	 * NULL - the hardware would crash if we enabled tracing.
-	 *
-	 * This saves us some problems with having to write an msr on a
-	 * different cpu while preventing others from doing the same for the
-	 * next context for that same cpu.
-	 */
-
-	spin_unlock_irqrestore(&ds_lock, irq);
-
-	/* The context might still be in use for context switching. */
-	if (task && (task != current))
-		wait_task_context_switch(task);
-
-	kfree(context);
-}
-
-static void ds_install_ds_area(struct ds_context *context)
-{
-	unsigned long ds;
-
-	ds = (unsigned long)context->ds;
-
-	/*
-	 * There is a race between the bts master and the pebs master.
-	 *
-	 * The thread/cpu access is synchronized via get/put_cpu() for
-	 * task tracing and via wrmsr_on_cpu for cpu tracing.
-	 *
-	 * If bts and pebs are collected for the same task or same cpu,
-	 * the same confiuration is written twice.
-	 */
-	if (context->task) {
-		get_cpu();
-		if (context->task == current)
-			wrmsrl(MSR_IA32_DS_AREA, ds);
-		set_tsk_thread_flag(context->task, TIF_DS_AREA_MSR);
-		put_cpu();
-	} else
-		wrmsr_on_cpu(context->cpu, MSR_IA32_DS_AREA,
-			     (u32)((u64)ds), (u32)((u64)ds >> 32));
-}
-
-/*
- * Call the tracer's callback on a buffer overflow.
- *
- * context: the ds context
- * qual: the buffer type
- */
-static void ds_overflow(struct ds_context *context, enum ds_qualifier qual)
-{
-	switch (qual) {
-	case ds_bts:
-		if (context->bts_master &&
-		    context->bts_master->ovfl)
-			context->bts_master->ovfl(context->bts_master);
-		break;
-	case ds_pebs:
-		if (context->pebs_master &&
-		    context->pebs_master->ovfl)
-			context->pebs_master->ovfl(context->pebs_master);
-		break;
-	}
-}
-
-
-/*
- * Write raw data into the BTS or PEBS buffer.
- *
- * The remainder of any partially written record is zeroed out.
- *
- * context: the DS context
- * qual:    the buffer type
- * record:  the data to write
- * size:    the size of the data
- */
-static int ds_write(struct ds_context *context, enum ds_qualifier qual,
-		    const void *record, size_t size)
-{
-	int bytes_written = 0;
-
-	if (!record)
-		return -EINVAL;
-
-	while (size) {
-		unsigned long base, index, end, write_end, int_th;
-		unsigned long write_size, adj_write_size;
-
-		/*
-		 * Write as much as possible without producing an
-		 * overflow interrupt.
-		 *
-		 * Interrupt_threshold must either be
-		 * - bigger than absolute_maximum or
-		 * - point to a record between buffer_base and absolute_maximum
-		 *
-		 * Index points to a valid record.
-		 */
-		base   = ds_get(context->ds, qual, ds_buffer_base);
-		index  = ds_get(context->ds, qual, ds_index);
-		end    = ds_get(context->ds, qual, ds_absolute_maximum);
-		int_th = ds_get(context->ds, qual, ds_interrupt_threshold);
-
-		write_end = min(end, int_th);
-
-		/*
-		 * If we are already beyond the interrupt threshold,
-		 * we fill the entire buffer.
-		 */
-		if (write_end <= index)
-			write_end = end;
-
-		if (write_end <= index)
-			break;
-
-		write_size = min((unsigned long) size, write_end - index);
-		memcpy((void *)index, record, write_size);
-
-		record = (const char *)record + write_size;
-		size -= write_size;
-		bytes_written += write_size;
-
-		adj_write_size = write_size / ds_cfg.sizeof_rec[qual];
-		adj_write_size *= ds_cfg.sizeof_rec[qual];
-
-		/* Zero out trailing bytes. */
-		memset((char *)index + write_size, 0,
-		       adj_write_size - write_size);
-		index += adj_write_size;
-
-		if (index >= end)
-			index = base;
-		ds_set(context->ds, qual, ds_index, index);
-
-		if (index >= int_th)
-			ds_overflow(context, qual);
-	}
-
-	return bytes_written;
-}
-
-
-/*
- * Branch Trace Store (BTS) uses the following format. Different
- * architectures vary in the size of those fields.
- * - source linear address
- * - destination linear address
- * - flags
- *
- * Later architectures use 64bit pointers throughout, whereas earlier
- * architectures use 32bit pointers in 32bit mode.
- *
- * We compute the base address for the fields based on:
- * - the field size stored in the DS configuration
- * - the relative field position
- *
- * In order to store additional information in the BTS buffer, we use
- * a special source address to indicate that the record requires
- * special interpretation.
- *
- * Netburst indicated via a bit in the flags field whether the branch
- * was predicted; this is ignored.
- *
- * We use two levels of abstraction:
- * - the raw data level defined here
- * - an arch-independent level defined in ds.h
- */
-
-enum bts_field {
-	bts_from,
-	bts_to,
-	bts_flags,
-
-	bts_qual		= bts_from,
-	bts_clock		= bts_to,
-	bts_pid			= bts_flags,
-
-	bts_qual_mask		= (bts_qual_max - 1),
-	bts_escape		= ((unsigned long)-1 & ~bts_qual_mask)
-};
-
-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, unsigned long field, unsigned long val)
-{
-	base += (ds_cfg.sizeof_ptr_field * field);
-	(*(unsigned long *)base) = val;
-}
-
-
-/*
- * The raw BTS data is architecture dependent.
- *
- * For higher-level users, we give an arch-independent view.
- * - ds.h defines struct bts_struct
- * - bts_read translates one raw bts record into a bts_struct
- * - bts_write translates one bts_struct into the raw format and
- *   writes it into the top of the parameter tracer's buffer.
- *
- * return: bytes read/written on success; -Eerrno, otherwise
- */
-static int
-bts_read(struct bts_tracer *tracer, const void *at, struct bts_struct *out)
-{
-	if (!tracer)
-		return -EINVAL;
-
-	if (at < tracer->trace.ds.begin)
-		return -EINVAL;
-
-	if (tracer->trace.ds.end < (at + tracer->trace.ds.size))
-		return -EINVAL;
-
-	memset(out, 0, sizeof(*out));
-	if ((bts_get(at, bts_qual) & ~bts_qual_mask) == bts_escape) {
-		out->qualifier = (bts_get(at, bts_qual) & bts_qual_mask);
-		out->variant.event.clock = bts_get(at, bts_clock);
-		out->variant.event.pid = bts_get(at, bts_pid);
-	} else {
-		out->qualifier = bts_branch;
-		out->variant.lbr.from = bts_get(at, bts_from);
-		out->variant.lbr.to   = bts_get(at, bts_to);
-
-		if (!out->variant.lbr.from && !out->variant.lbr.to)
-			out->qualifier = bts_invalid;
-	}
-
-	return ds_cfg.sizeof_rec[ds_bts];
-}
-
-static int bts_write(struct bts_tracer *tracer, const struct bts_struct *in)
-{
-	unsigned char raw[MAX_SIZEOF_BTS];
-
-	if (!tracer)
-		return -EINVAL;
-
-	if (MAX_SIZEOF_BTS < ds_cfg.sizeof_rec[ds_bts])
-		return -EOVERFLOW;
-
-	switch (in->qualifier) {
-	case bts_invalid:
-		bts_set(raw, bts_from, 0);
-		bts_set(raw, bts_to, 0);
-		bts_set(raw, bts_flags, 0);
-		break;
-	case bts_branch:
-		bts_set(raw, bts_from, in->variant.lbr.from);
-		bts_set(raw, bts_to,   in->variant.lbr.to);
-		bts_set(raw, bts_flags, 0);
-		break;
-	case bts_task_arrives:
-	case bts_task_departs:
-		bts_set(raw, bts_qual, (bts_escape | in->qualifier));
-		bts_set(raw, bts_clock, in->variant.event.clock);
-		bts_set(raw, bts_pid, in->variant.event.pid);
-		break;
-	default:
-		return -EINVAL;
-	}
-
-	return ds_write(tracer->ds.context, ds_bts, raw,
-			ds_cfg.sizeof_rec[ds_bts]);
-}
-
-
-static void ds_write_config(struct ds_context *context,
-			    struct ds_trace *cfg, enum ds_qualifier qual)
-{
-	unsigned char *ds = context->ds;
-
-	ds_set(ds, qual, ds_buffer_base, (unsigned long)cfg->begin);
-	ds_set(ds, qual, ds_index, (unsigned long)cfg->top);
-	ds_set(ds, qual, ds_absolute_maximum, (unsigned long)cfg->end);
-	ds_set(ds, qual, ds_interrupt_threshold, (unsigned long)cfg->ith);
-}
-
-static void ds_read_config(struct ds_context *context,
-			   struct ds_trace *cfg, enum ds_qualifier qual)
-{
-	unsigned char *ds = context->ds;
-
-	cfg->begin = (void *)ds_get(ds, qual, ds_buffer_base);
-	cfg->top = (void *)ds_get(ds, qual, ds_index);
-	cfg->end = (void *)ds_get(ds, qual, ds_absolute_maximum);
-	cfg->ith = (void *)ds_get(ds, qual, ds_interrupt_threshold);
-}
-
-static void ds_init_ds_trace(struct ds_trace *trace, enum ds_qualifier qual,
-			     void *base, size_t size, size_t ith,
-			     unsigned int flags) {
-	unsigned long buffer, adj;
-
-	/*
-	 * Adjust the buffer address and size to meet alignment
-	 * constraints:
-	 * - buffer is double-word aligned
-	 * - size is multiple of record size
-	 *
-	 * We checked the size at the very beginning; we have enough
-	 * space to do the adjustment.
-	 */
-	buffer = (unsigned long)base;
-
-	adj = ALIGN(buffer, DS_ALIGNMENT) - buffer;
-	buffer += adj;
-	size   -= adj;
-
-	trace->n = size / ds_cfg.sizeof_rec[qual];
-	trace->size = ds_cfg.sizeof_rec[qual];
-
-	size = (trace->n * trace->size);
-
-	trace->begin = (void *)buffer;
-	trace->top = trace->begin;
-	trace->end = (void *)(buffer + size);
-	/*
-	 * The value for 'no threshold' is -1, which will set the
-	 * threshold outside of the buffer, just like we want it.
-	 */
-	ith *= ds_cfg.sizeof_rec[qual];
-	trace->ith = (void *)(buffer + size - ith);
-
-	trace->flags = flags;
-}
-
-
-static int ds_request(struct ds_tracer *tracer, struct ds_trace *trace,
-		      enum ds_qualifier qual, struct task_struct *task,
-		      int cpu, void *base, size_t size, size_t th)
-{
-	struct ds_context *context;
-	int error;
-	size_t req_size;
-
-	error = -EOPNOTSUPP;
-	if (!ds_cfg.sizeof_rec[qual])
-		goto out;
-
-	error = -EINVAL;
-	if (!base)
-		goto out;
-
-	req_size = ds_cfg.sizeof_rec[qual];
-	/* We might need space for alignment adjustments. */
-	if (!IS_ALIGNED((unsigned long)base, DS_ALIGNMENT))
-		req_size += DS_ALIGNMENT;
-
-	error = -EINVAL;
-	if (size < req_size)
-		goto out;
-
-	if (th != (size_t)-1) {
-		th *= ds_cfg.sizeof_rec[qual];
-
-		error = -EINVAL;
-		if (size <= th)
-			goto out;
-	}
-
-	tracer->buffer = base;
-	tracer->size = size;
-
-	error = -ENOMEM;
-	context = ds_get_context(task, cpu);
-	if (!context)
-		goto out;
-	tracer->context = context;
-
-	/*
-	 * Defer any tracer-specific initialization work for the context until
-	 * context ownership has been clarified.
-	 */
-
-	error = 0;
- out:
-	return error;
-}
-
-static struct bts_tracer *ds_request_bts(struct task_struct *task, int cpu,
-					 void *base, size_t size,
-					 bts_ovfl_callback_t ovfl, size_t th,
-					 unsigned int flags)
-{
-	struct bts_tracer *tracer;
-	int error;
-
-	/* Buffer overflow notification is not yet implemented. */
-	error = -EOPNOTSUPP;
-	if (ovfl)
-		goto out;
-
-	error = get_tracer(task);
-	if (error < 0)
-		goto out;
-
-	error = -ENOMEM;
-	tracer = kzalloc(sizeof(*tracer), GFP_KERNEL);
-	if (!tracer)
-		goto out_put_tracer;
-	tracer->ovfl = ovfl;
-
-	/* Do some more error checking and acquire a tracing context. */
-	error = ds_request(&tracer->ds, &tracer->trace.ds,
-			   ds_bts, task, cpu, base, size, th);
-	if (error < 0)
-		goto out_tracer;
-
-	/* Claim the bts part of the tracing context we acquired above. */
-	spin_lock_irq(&ds_lock);
-
-	error = -EPERM;
-	if (tracer->ds.context->bts_master)
-		goto out_unlock;
-	tracer->ds.context->bts_master = tracer;
-
-	spin_unlock_irq(&ds_lock);
-
-	/*
-	 * Now that we own the bts part of the context, let's complete the
-	 * initialization for that part.
-	 */
-	ds_init_ds_trace(&tracer->trace.ds, ds_bts, base, size, th, flags);
-	ds_write_config(tracer->ds.context, &tracer->trace.ds, ds_bts);
-	ds_install_ds_area(tracer->ds.context);
-
-	tracer->trace.read  = bts_read;
-	tracer->trace.write = bts_write;
-
-	/* Start tracing. */
-	ds_resume_bts(tracer);
-
-	return tracer;
-
- out_unlock:
-	spin_unlock_irq(&ds_lock);
-	ds_put_context(tracer->ds.context);
- out_tracer:
-	kfree(tracer);
- out_put_tracer:
-	put_tracer(task);
- out:
-	return ERR_PTR(error);
-}
-
-struct bts_tracer *ds_request_bts_task(struct task_struct *task,
-				       void *base, size_t size,
-				       bts_ovfl_callback_t ovfl,
-				       size_t th, unsigned int flags)
-{
-	return ds_request_bts(task, 0, base, size, ovfl, th, flags);
-}
-
-struct bts_tracer *ds_request_bts_cpu(int cpu, void *base, size_t size,
-				      bts_ovfl_callback_t ovfl,
-				      size_t th, unsigned int flags)
-{
-	return ds_request_bts(NULL, cpu, base, size, ovfl, th, flags);
-}
-
-static struct pebs_tracer *ds_request_pebs(struct task_struct *task, int cpu,
-					   void *base, size_t size,
-					   pebs_ovfl_callback_t ovfl, size_t th,
-					   unsigned int flags)
-{
-	struct pebs_tracer *tracer;
-	int error;
-
-	/* Buffer overflow notification is not yet implemented. */
-	error = -EOPNOTSUPP;
-	if (ovfl)
-		goto out;
-
-	error = get_tracer(task);
-	if (error < 0)
-		goto out;
-
-	error = -ENOMEM;
-	tracer = kzalloc(sizeof(*tracer), GFP_KERNEL);
-	if (!tracer)
-		goto out_put_tracer;
-	tracer->ovfl = ovfl;
-
-	/* Do some more error checking and acquire a tracing context. */
-	error = ds_request(&tracer->ds, &tracer->trace.ds,
-			   ds_pebs, task, cpu, base, size, th);
-	if (error < 0)
-		goto out_tracer;
-
-	/* Claim the pebs part of the tracing context we acquired above. */
-	spin_lock_irq(&ds_lock);
-
-	error = -EPERM;
-	if (tracer->ds.context->pebs_master)
-		goto out_unlock;
-	tracer->ds.context->pebs_master = tracer;
-
-	spin_unlock_irq(&ds_lock);
-
-	/*
-	 * Now that we own the pebs part of the context, let's complete the
-	 * initialization for that part.
-	 */
-	ds_init_ds_trace(&tracer->trace.ds, ds_pebs, base, size, th, flags);
-	ds_write_config(tracer->ds.context, &tracer->trace.ds, ds_pebs);
-	ds_install_ds_area(tracer->ds.context);
-
-	/* Start tracing. */
-	ds_resume_pebs(tracer);
-
-	return tracer;
-
- out_unlock:
-	spin_unlock_irq(&ds_lock);
-	ds_put_context(tracer->ds.context);
- out_tracer:
-	kfree(tracer);
- out_put_tracer:
-	put_tracer(task);
- out:
-	return ERR_PTR(error);
-}
-
-struct pebs_tracer *ds_request_pebs_task(struct task_struct *task,
-					 void *base, size_t size,
-					 pebs_ovfl_callback_t ovfl,
-					 size_t th, unsigned int flags)
-{
-	return ds_request_pebs(task, 0, base, size, ovfl, th, flags);
-}
-
-struct pebs_tracer *ds_request_pebs_cpu(int cpu, void *base, size_t size,
-					pebs_ovfl_callback_t ovfl,
-					size_t th, unsigned int flags)
-{
-	return ds_request_pebs(NULL, cpu, base, size, ovfl, th, flags);
-}
-
-static void ds_free_bts(struct bts_tracer *tracer)
-{
-	struct task_struct *task;
-
-	task = tracer->ds.context->task;
-
-	WARN_ON_ONCE(tracer->ds.context->bts_master != tracer);
-	tracer->ds.context->bts_master = NULL;
-
-	/* Make sure tracing stopped and the tracer is not in use. */
-	if (task && (task != current))
-		wait_task_context_switch(task);
-
-	ds_put_context(tracer->ds.context);
-	put_tracer(task);
-
-	kfree(tracer);
-}
-
-void ds_release_bts(struct bts_tracer *tracer)
-{
-	might_sleep();
-
-	if (!tracer)
-		return;
-
-	ds_suspend_bts(tracer);
-	ds_free_bts(tracer);
-}
-
-int ds_release_bts_noirq(struct bts_tracer *tracer)
-{
-	struct task_struct *task;
-	unsigned long irq;
-	int error;
-
-	if (!tracer)
-		return 0;
-
-	task = tracer->ds.context->task;
-
-	local_irq_save(irq);
-
-	error = -EPERM;
-	if (!task &&
-	    (tracer->ds.context->cpu != smp_processor_id()))
-		goto out;
-
-	error = -EPERM;
-	if (task && (task != current))
-		goto out;
-
-	ds_suspend_bts_noirq(tracer);
-	ds_free_bts(tracer);
-
-	error = 0;
- out:
-	local_irq_restore(irq);
-	return error;
-}
-
-static void update_task_debugctlmsr(struct task_struct *task,
-				    unsigned long debugctlmsr)
-{
-	task->thread.debugctlmsr = debugctlmsr;
-
-	get_cpu();
-	if (task == current)
-		update_debugctlmsr(debugctlmsr);
-	put_cpu();
-}
-
-void ds_suspend_bts(struct bts_tracer *tracer)
-{
-	struct task_struct *task;
-	unsigned long debugctlmsr;
-	int cpu;
-
-	if (!tracer)
-		return;
-
-	tracer->flags = 0;
-
-	task = tracer->ds.context->task;
-	cpu  = tracer->ds.context->cpu;
-
-	WARN_ON(!task && irqs_disabled());
-
-	debugctlmsr = (task ?
-		       task->thread.debugctlmsr :
-		       get_debugctlmsr_on_cpu(cpu));
-	debugctlmsr &= ~BTS_CONTROL;
-
-	if (task)
-		update_task_debugctlmsr(task, debugctlmsr);
-	else
-		update_debugctlmsr_on_cpu(cpu, debugctlmsr);
-}
-
-int ds_suspend_bts_noirq(struct bts_tracer *tracer)
-{
-	struct task_struct *task;
-	unsigned long debugctlmsr, irq;
-	int cpu, error = 0;
-
-	if (!tracer)
-		return 0;
-
-	tracer->flags = 0;
-
-	task = tracer->ds.context->task;
-	cpu  = tracer->ds.context->cpu;
-
-	local_irq_save(irq);
-
-	error = -EPERM;
-	if (!task && (cpu != smp_processor_id()))
-		goto out;
-
-	debugctlmsr = (task ?
-		       task->thread.debugctlmsr :
-		       get_debugctlmsr());
-	debugctlmsr &= ~BTS_CONTROL;
-
-	if (task)
-		update_task_debugctlmsr(task, debugctlmsr);
-	else
-		update_debugctlmsr(debugctlmsr);
-
-	error = 0;
- out:
-	local_irq_restore(irq);
-	return error;
-}
-
-static unsigned long ds_bts_control(struct bts_tracer *tracer)
-{
-	unsigned long control;
-
-	control = ds_cfg.ctl[dsf_bts];
-	if (!(tracer->trace.ds.flags & BTS_KERNEL))
-		control |= ds_cfg.ctl[dsf_bts_kernel];
-	if (!(tracer->trace.ds.flags & BTS_USER))
-		control |= ds_cfg.ctl[dsf_bts_user];
-
-	return control;
-}
-
-void ds_resume_bts(struct bts_tracer *tracer)
-{
-	struct task_struct *task;
-	unsigned long debugctlmsr;
-	int cpu;
-
-	if (!tracer)
-		return;
-
-	tracer->flags = tracer->trace.ds.flags;
-
-	task = tracer->ds.context->task;
-	cpu  = tracer->ds.context->cpu;
-
-	WARN_ON(!task && irqs_disabled());
-
-	debugctlmsr = (task ?
-		       task->thread.debugctlmsr :
-		       get_debugctlmsr_on_cpu(cpu));
-	debugctlmsr |= ds_bts_control(tracer);
-
-	if (task)
-		update_task_debugctlmsr(task, debugctlmsr);
-	else
-		update_debugctlmsr_on_cpu(cpu, debugctlmsr);
-}
-
-int ds_resume_bts_noirq(struct bts_tracer *tracer)
-{
-	struct task_struct *task;
-	unsigned long debugctlmsr, irq;
-	int cpu, error = 0;
-
-	if (!tracer)
-		return 0;
-
-	tracer->flags = tracer->trace.ds.flags;
-
-	task = tracer->ds.context->task;
-	cpu  = tracer->ds.context->cpu;
-
-	local_irq_save(irq);
-
-	error = -EPERM;
-	if (!task && (cpu != smp_processor_id()))
-		goto out;
-
-	debugctlmsr = (task ?
-		       task->thread.debugctlmsr :
-		       get_debugctlmsr());
-	debugctlmsr |= ds_bts_control(tracer);
-
-	if (task)
-		update_task_debugctlmsr(task, debugctlmsr);
-	else
-		update_debugctlmsr(debugctlmsr);
-
-	error = 0;
- out:
-	local_irq_restore(irq);
-	return error;
-}
-
-static void ds_free_pebs(struct pebs_tracer *tracer)
-{
-	struct task_struct *task;
-
-	task = tracer->ds.context->task;
-
-	WARN_ON_ONCE(tracer->ds.context->pebs_master != tracer);
-	tracer->ds.context->pebs_master = NULL;
-
-	ds_put_context(tracer->ds.context);
-	put_tracer(task);
-
-	kfree(tracer);
-}
-
-void ds_release_pebs(struct pebs_tracer *tracer)
-{
-	might_sleep();
-
-	if (!tracer)
-		return;
-
-	ds_suspend_pebs(tracer);
-	ds_free_pebs(tracer);
-}
-
-int ds_release_pebs_noirq(struct pebs_tracer *tracer)
-{
-	struct task_struct *task;
-	unsigned long irq;
-	int error;
-
-	if (!tracer)
-		return 0;
-
-	task = tracer->ds.context->task;
-
-	local_irq_save(irq);
-
-	error = -EPERM;
-	if (!task &&
-	    (tracer->ds.context->cpu != smp_processor_id()))
-		goto out;
-
-	error = -EPERM;
-	if (task && (task != current))
-		goto out;
-
-	ds_suspend_pebs_noirq(tracer);
-	ds_free_pebs(tracer);
-
-	error = 0;
- out:
-	local_irq_restore(irq);
-	return error;
-}
-
-void ds_suspend_pebs(struct pebs_tracer *tracer)
-{
-
-}
-
-int ds_suspend_pebs_noirq(struct pebs_tracer *tracer)
-{
-	return 0;
-}
-
-void ds_resume_pebs(struct pebs_tracer *tracer)
-{
-
-}
-
-int ds_resume_pebs_noirq(struct pebs_tracer *tracer)
-{
-	return 0;
-}
-
-const struct bts_trace *ds_read_bts(struct bts_tracer *tracer)
-{
-	if (!tracer)
-		return NULL;
-
-	ds_read_config(tracer->ds.context, &tracer->trace.ds, ds_bts);
-	return &tracer->trace;
-}
-
-const struct pebs_trace *ds_read_pebs(struct pebs_tracer *tracer)
-{
-	if (!tracer)
-		return NULL;
-
-	ds_read_config(tracer->ds.context, &tracer->trace.ds, ds_pebs);
-
-	tracer->trace.counters = ds_cfg.nr_counter_reset;
-	memcpy(tracer->trace.counter_reset,
-	       tracer->ds.context->ds +
-	       (NUM_DS_PTR_FIELDS * ds_cfg.sizeof_ptr_field),
-	       ds_cfg.nr_counter_reset * PEBS_RESET_FIELD_SIZE);
-
-	return &tracer->trace;
-}
-
-int ds_reset_bts(struct bts_tracer *tracer)
-{
-	if (!tracer)
-		return -EINVAL;
-
-	tracer->trace.ds.top = tracer->trace.ds.begin;
-
-	ds_set(tracer->ds.context->ds, ds_bts, ds_index,
-	       (unsigned long)tracer->trace.ds.top);
-
-	return 0;
-}
-
-int ds_reset_pebs(struct pebs_tracer *tracer)
-{
-	if (!tracer)
-		return -EINVAL;
-
-	tracer->trace.ds.top = tracer->trace.ds.begin;
-
-	ds_set(tracer->ds.context->ds, ds_pebs, ds_index,
-	       (unsigned long)tracer->trace.ds.top);
-
-	return 0;
-}
-
-int ds_set_pebs_reset(struct pebs_tracer *tracer,
-		      unsigned int counter, u64 value)
-{
-	if (!tracer)
-		return -EINVAL;
-
-	if (ds_cfg.nr_counter_reset < counter)
-		return -EINVAL;
-
-	*(u64 *)(tracer->ds.context->ds +
-		 (NUM_DS_PTR_FIELDS * ds_cfg.sizeof_ptr_field) +
-		 (counter * PEBS_RESET_FIELD_SIZE)) = value;
-
-	return 0;
-}
-
-static const struct ds_configuration ds_cfg_netburst = {
-	.name = "Netburst",
-	.ctl[dsf_bts]		= (1 << 2) | (1 << 3),
-	.ctl[dsf_bts_kernel]	= (1 << 5),
-	.ctl[dsf_bts_user]	= (1 << 6),
-	.nr_counter_reset	= 1,
-};
-static const struct ds_configuration ds_cfg_pentium_m = {
-	.name = "Pentium M",
-	.ctl[dsf_bts]		= (1 << 6) | (1 << 7),
-	.nr_counter_reset	= 1,
-};
-static const struct ds_configuration ds_cfg_core2_atom = {
-	.name = "Core 2/Atom",
-	.ctl[dsf_bts]		= (1 << 6) | (1 << 7),
-	.ctl[dsf_bts_kernel]	= (1 << 9),
-	.ctl[dsf_bts_user]	= (1 << 10),
-	.nr_counter_reset	= 1,
-};
-static const struct ds_configuration ds_cfg_core_i7 = {
-	.name = "Core i7",
-	.ctl[dsf_bts]		= (1 << 6) | (1 << 7),
-	.ctl[dsf_bts_kernel]	= (1 << 9),
-	.ctl[dsf_bts_user]	= (1 << 10),
-	.nr_counter_reset	= 4,
-};
-
-static void
-ds_configure(const struct ds_configuration *cfg,
-	     struct cpuinfo_x86 *cpu)
-{
-	unsigned long nr_pebs_fields = 0;
-
-	printk(KERN_INFO "[ds] using %s configuration\n", cfg->name);
-
-#ifdef __i386__
-	nr_pebs_fields = 10;
-#else
-	nr_pebs_fields = 18;
-#endif
-
-	/*
-	 * Starting with version 2, architectural performance
-	 * monitoring supports a format specifier.
-	 */
-	if ((cpuid_eax(0xa) & 0xff) > 1) {
-		unsigned long perf_capabilities, format;
-
-		rdmsrl(MSR_IA32_PERF_CAPABILITIES, perf_capabilities);
-
-		format = (perf_capabilities >> 8) & 0xf;
-
-		switch (format) {
-		case 0:
-			nr_pebs_fields = 18;
-			break;
-		case 1:
-			nr_pebs_fields = 22;
-			break;
-		default:
-			printk(KERN_INFO
-			       "[ds] unknown PEBS format: %lu\n", format);
-			nr_pebs_fields = 0;
-			break;
-		}
-	}
-
-	memset(&ds_cfg, 0, sizeof(ds_cfg));
-	ds_cfg = *cfg;
-
-	ds_cfg.sizeof_ptr_field =
-		(cpu_has(cpu, X86_FEATURE_DTES64) ? 8 : 4);
-
-	ds_cfg.sizeof_rec[ds_bts]  = ds_cfg.sizeof_ptr_field * 3;
-	ds_cfg.sizeof_rec[ds_pebs] = ds_cfg.sizeof_ptr_field * nr_pebs_fields;
-
-	if (!cpu_has(cpu, X86_FEATURE_BTS)) {
-		ds_cfg.sizeof_rec[ds_bts] = 0;
-		printk(KERN_INFO "[ds] bts not available\n");
-	}
-	if (!cpu_has(cpu, X86_FEATURE_PEBS)) {
-		ds_cfg.sizeof_rec[ds_pebs] = 0;
-		printk(KERN_INFO "[ds] pebs not available\n");
-	}
-
-	printk(KERN_INFO "[ds] sizes: address: %u bit, ",
-	       8 * ds_cfg.sizeof_ptr_field);
-	printk("bts/pebs record: %u/%u bytes\n",
-	       ds_cfg.sizeof_rec[ds_bts], ds_cfg.sizeof_rec[ds_pebs]);
-
-	WARN_ON_ONCE(MAX_PEBS_COUNTERS < ds_cfg.nr_counter_reset);
-}
-
-void __cpuinit ds_init_intel(struct cpuinfo_x86 *c)
-{
-	/* Only configure the first cpu. Others are identical. */
-	if (ds_cfg.name)
-		return;
-
-	switch (c->x86) {
-	case 0x6:
-		switch (c->x86_model) {
-		case 0x9:
-		case 0xd: /* Pentium M */
-			ds_configure(&ds_cfg_pentium_m, c);
-			break;
-		case 0xf:
-		case 0x17: /* Core2 */
-		case 0x1c: /* Atom */
-			ds_configure(&ds_cfg_core2_atom, c);
-			break;
-		case 0x1a: /* Core i7 */
-			ds_configure(&ds_cfg_core_i7, c);
-			break;
-		default:
-			/* Sorry, don't know about them. */
-			break;
-		}
-		break;
-	case 0xf:
-		switch (c->x86_model) {
-		case 0x0:
-		case 0x1:
-		case 0x2: /* Netburst */
-			ds_configure(&ds_cfg_netburst, c);
-			break;
-		default:
-			/* Sorry, don't know about them. */
-			break;
-		}
-		break;
-	default:
-		/* Sorry, don't know about them. */
-		break;
-	}
-}
-
-static inline void ds_take_timestamp(struct ds_context *context,
-				     enum bts_qualifier qualifier,
-				     struct task_struct *task)
-{
-	struct bts_tracer *tracer = context->bts_master;
-	struct bts_struct ts;
-
-	/* Prevent compilers from reading the tracer pointer twice. */
-	barrier();
-
-	if (!tracer || !(tracer->flags & BTS_TIMESTAMPS))
-		return;
-
-	memset(&ts, 0, sizeof(ts));
-	ts.qualifier		= qualifier;
-	ts.variant.event.clock	= trace_clock_global();
-	ts.variant.event.pid	= task->pid;
-
-	bts_write(tracer, &ts);
-}
-
-/*
- * Change the DS configuration from tracing prev to tracing next.
- */
-void ds_switch_to(struct task_struct *prev, struct task_struct *next)
-{
-	struct ds_context *prev_ctx	= prev->thread.ds_ctx;
-	struct ds_context *next_ctx	= next->thread.ds_ctx;
-	unsigned long debugctlmsr	= next->thread.debugctlmsr;
-
-	/* Make sure all data is read before we start. */
-	barrier();
-
-	if (prev_ctx) {
-		update_debugctlmsr(0);
-
-		ds_take_timestamp(prev_ctx, bts_task_departs, prev);
-	}
-
-	if (next_ctx) {
-		ds_take_timestamp(next_ctx, bts_task_arrives, next);
-
-		wrmsrl(MSR_IA32_DS_AREA, (unsigned long)next_ctx->ds);
-	}
-
-	update_debugctlmsr(debugctlmsr);
-}
-
-static __init int ds_selftest(void)
-{
-	if (ds_cfg.sizeof_rec[ds_bts]) {
-		int error;
-
-		error = ds_selftest_bts();
-		if (error) {
-			WARN(1, "[ds] selftest failed. disabling bts.\n");
-			ds_cfg.sizeof_rec[ds_bts] = 0;
-		}
-	}
-
-	if (ds_cfg.sizeof_rec[ds_pebs]) {
-		int error;
-
-		error = ds_selftest_pebs();
-		if (error) {
-			WARN(1, "[ds] selftest failed. disabling pebs.\n");
-			ds_cfg.sizeof_rec[ds_pebs] = 0;
-		}
-	}
-
-	return 0;
-}
-device_initcall(ds_selftest);

arch/x86/kernel/ds_selftest.c

@@ -1,408 +0,0 @@
-/*
- * Debug Store support - selftest
- *
- *
- * Copyright (C) 2009 Intel Corporation.
- * Markus Metzger <markus.t.metzger@intel.com>, 2009
- */
-
-#include "ds_selftest.h"
-
-#include <linux/kernel.h>
-#include <linux/string.h>
-#include <linux/smp.h>
-#include <linux/cpu.h>
-
-#include <asm/ds.h>
-
-
-#define BUFFER_SIZE		521	/* Intentionally chose an odd size. */
-#define SMALL_BUFFER_SIZE	24	/* A single bts entry. */
-
-struct ds_selftest_bts_conf {
-	struct bts_tracer *tracer;
-	int error;
-	int (*suspend)(struct bts_tracer *);
-	int (*resume)(struct bts_tracer *);
-};
-
-static int ds_selftest_bts_consistency(const struct bts_trace *trace)
-{
-	int error = 0;
-
-	if (!trace) {
-		printk(KERN_CONT "failed to access trace...");
-		/* Bail out. Other tests are pointless. */
-		return -1;
-	}
-
-	if (!trace->read) {
-		printk(KERN_CONT "bts read not available...");
-		error = -1;
-	}
-
-	/* Do some sanity checks on the trace configuration. */
-	if (!trace->ds.n) {
-		printk(KERN_CONT "empty bts buffer...");
-		error = -1;
-	}
-	if (!trace->ds.size) {
-		printk(KERN_CONT "bad bts trace setup...");
-		error = -1;
-	}
-	if (trace->ds.end !=
-	    (char *)trace->ds.begin + (trace->ds.n * trace->ds.size)) {
-		printk(KERN_CONT "bad bts buffer setup...");
-		error = -1;
-	}
-	/*
-	 * We allow top in [begin; end], since its not clear when the
-	 * overflow adjustment happens: after the increment or before the
-	 * write.
-	 */
-	if ((trace->ds.top < trace->ds.begin) ||
-	    (trace->ds.end < trace->ds.top)) {
-		printk(KERN_CONT "bts top out of bounds...");
-		error = -1;
-	}
-
-	return error;
-}
-
-static int ds_selftest_bts_read(struct bts_tracer *tracer,
-				const struct bts_trace *trace,
-				const void *from, const void *to)
-{
-	const unsigned char *at;
-
-	/*
-	 * Check a few things which do not belong to this test.
-	 * They should be covered by other tests.
-	 */
-	if (!trace)
-		return -1;
-
-	if (!trace->read)
-		return -1;
-
-	if (to < from)
-		return -1;
-
-	if (from < trace->ds.begin)
-		return -1;
-
-	if (trace->ds.end < to)
-		return -1;
-
-	if (!trace->ds.size)
-		return -1;
-
-	/* Now to the test itself. */
-	for (at = from; (void *)at < to; at += trace->ds.size) {
-		struct bts_struct bts;
-		unsigned long index;
-		int error;
-
-		if (((void *)at - trace->ds.begin) % trace->ds.size) {
-			printk(KERN_CONT
-			       "read from non-integer index...");
-			return -1;
-		}
-		index = ((void *)at - trace->ds.begin) / trace->ds.size;
-
-		memset(&bts, 0, sizeof(bts));
-		error = trace->read(tracer, at, &bts);
-		if (error < 0) {
-			printk(KERN_CONT
-			       "error reading bts trace at [%lu] (0x%p)...",
-			       index, at);
-			return error;
-		}
-
-		switch (bts.qualifier) {
-		case BTS_BRANCH:
-			break;
-		default:
-			printk(KERN_CONT
-			       "unexpected bts entry %llu at [%lu] (0x%p)...",
-			       bts.qualifier, index, at);
-			return -1;
-		}
-	}
-
-	return 0;
-}
-
-static void ds_selftest_bts_cpu(void *arg)
-{
-	struct ds_selftest_bts_conf *conf = arg;
-	const struct bts_trace *trace;
-	void *top;
-
-	if (IS_ERR(conf->tracer)) {
-		conf->error = PTR_ERR(conf->tracer);
-		conf->tracer = NULL;
-
-		printk(KERN_CONT
-		       "initialization failed (err: %d)...", conf->error);
-		return;
-	}
-
-	/* We should meanwhile have enough trace. */
-	conf->error = conf->suspend(conf->tracer);
-	if (conf->error < 0)
-		return;
-
-	/* Let's see if we can access the trace. */
-	trace = ds_read_bts(conf->tracer);
-
-	conf->error = ds_selftest_bts_consistency(trace);
-	if (conf->error < 0)
-		return;
-
-	/* If everything went well, we should have a few trace entries. */
-	if (trace->ds.top == trace->ds.begin) {
-		/*
-		 * It is possible but highly unlikely that we got a
-		 * buffer overflow and end up at exactly the same
-		 * position we started from.
-		 * Let's issue a warning, but continue.
-		 */
-		printk(KERN_CONT "no trace/overflow...");
-	}
-
-	/* Let's try to read the trace we collected. */
-	conf->error =
-		ds_selftest_bts_read(conf->tracer, trace,
-				     trace->ds.begin, trace->ds.top);
-	if (conf->error < 0)
-		return;
-
-	/*
-	 * Let's read the trace again.
-	 * Since we suspended tracing, we should get the same result.
-	 */
-	top = trace->ds.top;
-
-	trace = ds_read_bts(conf->tracer);
-	conf->error = ds_selftest_bts_consistency(trace);
-	if (conf->error < 0)
-		return;
-
-	if (top != trace->ds.top) {
-		printk(KERN_CONT "suspend not working...");
-		conf->error = -1;
-		return;
-	}
-
-	/* Let's collect some more trace - see if resume is working. */
-	conf->error = conf->resume(conf->tracer);
-	if (conf->error < 0)
-		return;
-
-	conf->error = conf->suspend(conf->tracer);
-	if (conf->error < 0)
-		return;
-
-	trace = ds_read_bts(conf->tracer);
-
-	conf->error = ds_selftest_bts_consistency(trace);
-	if (conf->error < 0)
-		return;
-
-	if (trace->ds.top == top) {
-		/*
-		 * It is possible but highly unlikely that we got a
-		 * buffer overflow and end up at exactly the same
-		 * position we started from.
-		 * Let's issue a warning and check the full trace.
-		 */
-		printk(KERN_CONT
-		       "no resume progress/overflow...");
-
-		conf->error =
-			ds_selftest_bts_read(conf->tracer, trace,
-					     trace->ds.begin, trace->ds.end);
-	} else if (trace->ds.top < top) {
-		/*
-		 * We had a buffer overflow - the entire buffer should
-		 * contain trace records.
-		 */
-		conf->error =
-			ds_selftest_bts_read(conf->tracer, trace,
-					     trace->ds.begin, trace->ds.end);
-	} else {
-		/*
-		 * It is quite likely that the buffer did not overflow.
-		 * Let's just check the delta trace.
-		 */
-		conf->error =
-			ds_selftest_bts_read(conf->tracer, trace, top,
-					     trace->ds.top);
-	}
-	if (conf->error < 0)
-		return;
-
-	conf->error = 0;
-}
-
-static int ds_suspend_bts_wrap(struct bts_tracer *tracer)
-{
-	ds_suspend_bts(tracer);
-	return 0;
-}
-
-static int ds_resume_bts_wrap(struct bts_tracer *tracer)
-{
-	ds_resume_bts(tracer);
-	return 0;
-}
-
-static void ds_release_bts_noirq_wrap(void *tracer)
-{
-	(void)ds_release_bts_noirq(tracer);
-}
-
-static int ds_selftest_bts_bad_release_noirq(int cpu,
-					     struct bts_tracer *tracer)
-{
-	int error = -EPERM;
-
-	/* Try to release the tracer on the wrong cpu. */
-	get_cpu();
-	if (cpu != smp_processor_id()) {
-		error = ds_release_bts_noirq(tracer);
-		if (error != -EPERM)
-			printk(KERN_CONT "release on wrong cpu...");
-	}
-	put_cpu();
-
-	return error ? 0 : -1;
-}
-
-static int ds_selftest_bts_bad_request_cpu(int cpu, void *buffer)
-{
-	struct bts_tracer *tracer;
-	int error;
-
-	/* Try to request cpu tracing while task tracing is active. */
-	tracer = ds_request_bts_cpu(cpu, buffer, BUFFER_SIZE, NULL,
-				    (size_t)-1, BTS_KERNEL);
-	error = PTR_ERR(tracer);
-	if (!IS_ERR(tracer)) {
-		ds_release_bts(tracer);
-		error = 0;
-	}
-
-	if (error != -EPERM)
-		printk(KERN_CONT "cpu/task tracing overlap...");
-
-	return error ? 0 : -1;
-}
-
-static int ds_selftest_bts_bad_request_task(void *buffer)
-{
-	struct bts_tracer *tracer;
-	int error;
-
-	/* Try to request cpu tracing while task tracing is active. */
-	tracer = ds_request_bts_task(current, buffer, BUFFER_SIZE, NULL,
-				    (size_t)-1, BTS_KERNEL);
-	error = PTR_ERR(tracer);
-	if (!IS_ERR(tracer)) {
-		error = 0;
-		ds_release_bts(tracer);
-	}
-
-	if (error != -EPERM)
-		printk(KERN_CONT "task/cpu tracing overlap...");
-
-	return error ? 0 : -1;
-}
-
-int ds_selftest_bts(void)
-{
-	struct ds_selftest_bts_conf conf;
-	unsigned char buffer[BUFFER_SIZE], *small_buffer;
-	unsigned long irq;
-	int cpu;
-
-	printk(KERN_INFO "[ds] bts selftest...");
-	conf.error = 0;
-
-	small_buffer = (unsigned char *)ALIGN((unsigned long)buffer, 8) + 8;
-
-	get_online_cpus();
-	for_each_online_cpu(cpu) {
-		conf.suspend = ds_suspend_bts_wrap;
-		conf.resume = ds_resume_bts_wrap;
-		conf.tracer =
-			ds_request_bts_cpu(cpu, buffer, BUFFER_SIZE,
-					   NULL, (size_t)-1, BTS_KERNEL);
-		ds_selftest_bts_cpu(&conf);
-		if (conf.error >= 0)
-			conf.error = ds_selftest_bts_bad_request_task(buffer);
-		ds_release_bts(conf.tracer);
-		if (conf.error < 0)
-			goto out;
-
-		conf.suspend = ds_suspend_bts_noirq;
-		conf.resume = ds_resume_bts_noirq;
-		conf.tracer =
-			ds_request_bts_cpu(cpu, buffer, BUFFER_SIZE,
-					   NULL, (size_t)-1, BTS_KERNEL);
-		smp_call_function_single(cpu, ds_selftest_bts_cpu, &conf, 1);
-		if (conf.error >= 0) {
-			conf.error =
-				ds_selftest_bts_bad_release_noirq(cpu,
-								  conf.tracer);
-			/* We must not release the tracer twice. */
-			if (conf.error < 0)
-				conf.tracer = NULL;
-		}
-		if (conf.error >= 0)
-			conf.error = ds_selftest_bts_bad_request_task(buffer);
-		smp_call_function_single(cpu, ds_release_bts_noirq_wrap,
-					 conf.tracer, 1);
-		if (conf.error < 0)
-			goto out;
-	}
-
-	conf.suspend = ds_suspend_bts_wrap;
-	conf.resume = ds_resume_bts_wrap;
-	conf.tracer =
-		ds_request_bts_task(current, buffer, BUFFER_SIZE,
-				    NULL, (size_t)-1, BTS_KERNEL);
-	ds_selftest_bts_cpu(&conf);
-	if (conf.error >= 0)
-		conf.error = ds_selftest_bts_bad_request_cpu(0, buffer);
-	ds_release_bts(conf.tracer);
-	if (conf.error < 0)
-		goto out;
-
-	conf.suspend = ds_suspend_bts_noirq;
-	conf.resume = ds_resume_bts_noirq;
-	conf.tracer =
-		ds_request_bts_task(current, small_buffer, SMALL_BUFFER_SIZE,
-				   NULL, (size_t)-1, BTS_KERNEL);
-	local_irq_save(irq);
-	ds_selftest_bts_cpu(&conf);
-	if (conf.error >= 0)
-		conf.error = ds_selftest_bts_bad_request_cpu(0, buffer);
-	ds_release_bts_noirq(conf.tracer);
-	local_irq_restore(irq);
-	if (conf.error < 0)
-		goto out;
-
-	conf.error = 0;
- out:
-	put_online_cpus();
-	printk(KERN_CONT "%s.\n", (conf.error ? "failed" : "passed"));
-
-	return conf.error;
-}
-
-int ds_selftest_pebs(void)
-{
-	return 0;
-}

arch/x86/kernel/ds_selftest.h

@@ -1,15 +0,0 @@
-/*
- * Debug Store support - selftest
- *
- *
- * Copyright (C) 2009 Intel Corporation.
- * Markus Metzger <markus.t.metzger@intel.com>, 2009
- */
-
-#ifdef CONFIG_X86_DS_SELFTEST
-extern int ds_selftest_bts(void);
-extern int ds_selftest_pebs(void);
-#else
-static inline int ds_selftest_bts(void) { return 0; }
-static inline int ds_selftest_pebs(void) { return 0; }
-#endif

arch/x86/kernel/dumpstack.c

@@ -224,11 +224,6 @@ unsigned __kprobes long oops_begin(void)
 	int cpu;
 	unsigned long flags;
 
-	/* notify the hw-branch tracer so it may disable tracing and
-	   add the last trace to the trace buffer -
-	   the earlier this happens, the more useful the trace. */
-	trace_hw_branch_oops();
-
 	oops_enter();
 
 	/* racy, but better than risking deadlock. */

arch/x86/kernel/kprobes.c

@@ -422,14 +422,22 @@ static void __kprobes set_current_kprobe(struct kprobe *p, struct pt_regs *regs,
 
 static void __kprobes clear_btf(void)
 {
-	if (test_thread_flag(TIF_DEBUGCTLMSR))
-		update_debugctlmsr(0);
+	if (test_thread_flag(TIF_BLOCKSTEP)) {
+		unsigned long debugctl = get_debugctlmsr();
+
+		debugctl &= ~DEBUGCTLMSR_BTF;
+		update_debugctlmsr(debugctl);
+	}
 }
 
 static void __kprobes restore_btf(void)
 {
-	if (test_thread_flag(TIF_DEBUGCTLMSR))
-		update_debugctlmsr(current->thread.debugctlmsr);
+	if (test_thread_flag(TIF_BLOCKSTEP)) {
+		unsigned long debugctl = get_debugctlmsr();
+
+		debugctl |= DEBUGCTLMSR_BTF;
+		update_debugctlmsr(debugctl);
+	}
 }
 
 void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri,

arch/x86/kernel/process.c

@@ -20,7 +20,6 @@
 #include <asm/idle.h>
 #include <asm/uaccess.h>
 #include <asm/i387.h>
-#include <asm/ds.h>
 #include <asm/debugreg.h>
 
 unsigned long idle_halt;
@@ -50,8 +49,6 @@ void free_thread_xstate(struct task_struct *tsk)
 		kmem_cache_free(task_xstate_cachep, tsk->thread.xstate);
 		tsk->thread.xstate = NULL;
 	}
-
-	WARN(tsk->thread.ds_ctx, "leaking DS context\n");
 }
 
 void free_thread_info(struct thread_info *ti)
@@ -198,11 +195,16 @@ void __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
 	prev = &prev_p->thread;
 	next = &next_p->thread;
 
-	if (test_tsk_thread_flag(next_p, TIF_DS_AREA_MSR) ||
-	    test_tsk_thread_flag(prev_p, TIF_DS_AREA_MSR))
-		ds_switch_to(prev_p, next_p);
-	else if (next->debugctlmsr != prev->debugctlmsr)
-		update_debugctlmsr(next->debugctlmsr);
+	if (test_tsk_thread_flag(prev_p, TIF_BLOCKSTEP) ^
+	    test_tsk_thread_flag(next_p, TIF_BLOCKSTEP)) {
+		unsigned long debugctl = get_debugctlmsr();
+
+		debugctl &= ~DEBUGCTLMSR_BTF;
+		if (test_tsk_thread_flag(next_p, TIF_BLOCKSTEP))
+			debugctl |= DEBUGCTLMSR_BTF;
+
+		update_debugctlmsr(debugctl);
+	}
 
 	if (test_tsk_thread_flag(prev_p, TIF_NOTSC) ^
 	    test_tsk_thread_flag(next_p, TIF_NOTSC)) {

arch/x86/kernel/process_32.c

@@ -55,7 +55,6 @@
 #include <asm/cpu.h>
 #include <asm/idle.h>
 #include <asm/syscalls.h>
-#include <asm/ds.h>
 #include <asm/debugreg.h>
 
 asmlinkage void ret_from_fork(void) __asm__("ret_from_fork");
@@ -238,13 +237,6 @@ int copy_thread(unsigned long clone_flags, unsigned long sp,
 		kfree(p->thread.io_bitmap_ptr);
 		p->thread.io_bitmap_max = 0;
 	}
-
-	clear_tsk_thread_flag(p, TIF_DS_AREA_MSR);
-	p->thread.ds_ctx = NULL;
-
-	clear_tsk_thread_flag(p, TIF_DEBUGCTLMSR);
-	p->thread.debugctlmsr = 0;
-
 	return err;
 }
 

arch/x86/kernel/process_64.c

@@ -49,7 +49,6 @@
 #include <asm/ia32.h>
 #include <asm/idle.h>
 #include <asm/syscalls.h>
-#include <asm/ds.h>
 #include <asm/debugreg.h>
 
 asmlinkage extern void ret_from_fork(void);
@@ -313,13 +312,6 @@ int copy_thread(unsigned long clone_flags, unsigned long sp,
 		if (err)
 			goto out;
 	}
-
-	clear_tsk_thread_flag(p, TIF_DS_AREA_MSR);
-	p->thread.ds_ctx = NULL;
-
-	clear_tsk_thread_flag(p, TIF_DEBUGCTLMSR);
-	p->thread.debugctlmsr = 0;
-
 	err = 0;
 out:
 	if (err && p->thread.io_bitmap_ptr) {

arch/x86/kernel/ptrace.c

@@ -2,9 +2,6 @@
 /*
  * Pentium III FXSR, SSE support
  *	Gareth Hughes <gareth@valinux.com>, May 2000
- *
- * BTS tracing
- *	Markus Metzger <markus.t.metzger@intel.com>, Dec 2007
  */
 
 #include <linux/kernel.h>
@@ -22,7 +19,6 @@
 #include <linux/audit.h>
 #include <linux/seccomp.h>
 #include <linux/signal.h>
-#include <linux/workqueue.h>
 #include <linux/perf_event.h>
 #include <linux/hw_breakpoint.h>
 
@@ -36,7 +32,6 @@
 #include <asm/desc.h>
 #include <asm/prctl.h>
 #include <asm/proto.h>
-#include <asm/ds.h>
 #include <asm/hw_breakpoint.h>
 
 #include "tls.h"
@@ -789,342 +784,6 @@ static int ioperm_get(struct task_struct *target,
 				   0, IO_BITMAP_BYTES);
 }
 
-#ifdef CONFIG_X86_PTRACE_BTS
-/*
- * A branch trace store context.
- *
- * Contexts may only be installed by ptrace_bts_config() and only for
- * ptraced tasks.
- *
- * Contexts are destroyed when the tracee is detached from the tracer.
- * The actual destruction work requires interrupts enabled, so the
- * work is deferred and will be scheduled during __ptrace_unlink().
- *
- * Contexts hold an additional task_struct reference on the traced
- * task, as well as a reference on the tracer's mm.
- *
- * Ptrace already holds a task_struct for the duration of ptrace operations,
- * but since destruction is deferred, it may be executed after both
- * tracer and tracee exited.
- */
-struct bts_context {
-	/* The branch trace handle. */
-	struct bts_tracer	*tracer;
-
-	/* The buffer used to store the branch trace and its size. */
-	void			*buffer;
-	unsigned int		size;
-
-	/* The mm that paid for the above buffer. */
-	struct mm_struct	*mm;
-
-	/* The task this context belongs to. */
-	struct task_struct	*task;
-
-	/* The signal to send on a bts buffer overflow. */
-	unsigned int		bts_ovfl_signal;
-
-	/* The work struct to destroy a context. */
-	struct work_struct	work;
-};
-
-static int alloc_bts_buffer(struct bts_context *context, unsigned int size)
-{
-	void *buffer = NULL;
-	int err = -ENOMEM;
-
-	err = account_locked_memory(current->mm, current->signal->rlim, size);
-	if (err < 0)
-		return err;
-
-	buffer = kzalloc(size, GFP_KERNEL);
-	if (!buffer)
-		goto out_refund;
-
-	context->buffer = buffer;
-	context->size = size;
-	context->mm = get_task_mm(current);
-
-	return 0;
-
- out_refund:
-	refund_locked_memory(current->mm, size);
-	return err;
-}
-
-static inline void free_bts_buffer(struct bts_context *context)
-{
-	if (!context->buffer)
-		return;
-
-	kfree(context->buffer);
-	context->buffer = NULL;
-
-	refund_locked_memory(context->mm, context->size);
-	context->size = 0;
-
-	mmput(context->mm);
-	context->mm = NULL;
-}
-
-static void free_bts_context_work(struct work_struct *w)
-{
-	struct bts_context *context;
-
-	context = container_of(w, struct bts_context, work);
-
-	ds_release_bts(context->tracer);
-	put_task_struct(context->task);
-	free_bts_buffer(context);
-	kfree(context);
-}
-
-static inline void free_bts_context(struct bts_context *context)
-{
-	INIT_WORK(&context->work, free_bts_context_work);
-	schedule_work(&context->work);
-}
-
-static inline struct bts_context *alloc_bts_context(struct task_struct *task)
-{
-	struct bts_context *context = kzalloc(sizeof(*context), GFP_KERNEL);
-	if (context) {
-		context->task = task;
-		task->bts = context;
-
-		get_task_struct(task);
-	}
-
-	return context;
-}
-
-static int ptrace_bts_read_record(struct task_struct *child, size_t index,
-				  struct bts_struct __user *out)
-{
-	struct bts_context *context;
-	const struct bts_trace *trace;
-	struct bts_struct bts;
-	const unsigned char *at;
-	int error;
-
-	context = child->bts;
-	if (!context)
-		return -ESRCH;
-
-	trace = ds_read_bts(context->tracer);
-	if (!trace)
-		return -ESRCH;
-
-	at = trace->ds.top - ((index + 1) * trace->ds.size);
-	if ((void *)at < trace->ds.begin)
-		at += (trace->ds.n * trace->ds.size);
-
-	if (!trace->read)
-		return -EOPNOTSUPP;
-
-	error = trace->read(context->tracer, at, &bts);
-	if (error < 0)
-		return error;
-
-	if (copy_to_user(out, &bts, sizeof(bts)))
-		return -EFAULT;
-
-	return sizeof(bts);
-}
-
-static int ptrace_bts_drain(struct task_struct *child,
-			    long size,
-			    struct bts_struct __user *out)
-{
-	struct bts_context *context;
-	const struct bts_trace *trace;
-	const unsigned char *at;
-	int error, drained = 0;
-
-	context = child->bts;
-	if (!context)
-		return -ESRCH;
-
-	trace = ds_read_bts(context->tracer);
-	if (!trace)
-		return -ESRCH;
-
-	if (!trace->read)
-		return -EOPNOTSUPP;
-
-	if (size < (trace->ds.top - trace->ds.begin))
-		return -EIO;
-
-	for (at = trace->ds.begin; (void *)at < trace->ds.top;
-	     out++, drained++, at += trace->ds.size) {
-		struct bts_struct bts;
-
-		error = trace->read(context->tracer, at, &bts);
-		if (error < 0)
-			return error;
-
-		if (copy_to_user(out, &bts, sizeof(bts)))
-			return -EFAULT;
-	}
-
-	memset(trace->ds.begin, 0, trace->ds.n * trace->ds.size);
-
-	error = ds_reset_bts(context->tracer);
-	if (error < 0)
-		return error;
-
-	return drained;
-}
-
-static int ptrace_bts_config(struct task_struct *child,
-			     long cfg_size,
-			     const struct ptrace_bts_config __user *ucfg)
-{
-	struct bts_context *context;
-	struct ptrace_bts_config cfg;
-	unsigned int flags = 0;
-
-	if (cfg_size < sizeof(cfg))
-		return -EIO;
-
-	if (copy_from_user(&cfg, ucfg, sizeof(cfg)))
-		return -EFAULT;
-
-	context = child->bts;
-	if (!context)
-		context = alloc_bts_context(child);
-	if (!context)
-		return -ENOMEM;
-
-	if (cfg.flags & PTRACE_BTS_O_SIGNAL) {
-		if (!cfg.signal)
-			return -EINVAL;
-
-		return -EOPNOTSUPP;
-		context->bts_ovfl_signal = cfg.signal;
-	}
-
-	ds_release_bts(context->tracer);
-	context->tracer = NULL;
-
-	if ((cfg.flags & PTRACE_BTS_O_ALLOC) && (cfg.size != context->size)) {
-		int err;
-
-		free_bts_buffer(context);
-		if (!cfg.size)
-			return 0;
-
-		err = alloc_bts_buffer(context, cfg.size);
-		if (err < 0)
-			return err;
-	}
-
-	if (cfg.flags & PTRACE_BTS_O_TRACE)
-		flags |= BTS_USER;
-
-	if (cfg.flags & PTRACE_BTS_O_SCHED)
-		flags |= BTS_TIMESTAMPS;
-
-	context->tracer =
-		ds_request_bts_task(child, context->buffer, context->size,
-				    NULL, (size_t)-1, flags);
-	if (unlikely(IS_ERR(context->tracer))) {
-		int error = PTR_ERR(context->tracer);
-
-		free_bts_buffer(context);
-		context->tracer = NULL;
-		return error;
-	}
-
-	return sizeof(cfg);
-}
-
-static int ptrace_bts_status(struct task_struct *child,
-			     long cfg_size,
-			     struct ptrace_bts_config __user *ucfg)
-{
-	struct bts_context *context;
-	const struct bts_trace *trace;
-	struct ptrace_bts_config cfg;
-
-	context = child->bts;
-	if (!context)
-		return -ESRCH;
-
-	if (cfg_size < sizeof(cfg))
-		return -EIO;
-
-	trace = ds_read_bts(context->tracer);
-	if (!trace)
-		return -ESRCH;
-
-	memset(&cfg, 0, sizeof(cfg));
-	cfg.size	= trace->ds.end - trace->ds.begin;
-	cfg.signal	= context->bts_ovfl_signal;
-	cfg.bts_size	= sizeof(struct bts_struct);
-
-	if (cfg.signal)
-		cfg.flags |= PTRACE_BTS_O_SIGNAL;
-
-	if (trace->ds.flags & BTS_USER)
-		cfg.flags |= PTRACE_BTS_O_TRACE;
-
-	if (trace->ds.flags & BTS_TIMESTAMPS)
-		cfg.flags |= PTRACE_BTS_O_SCHED;
-
-	if (copy_to_user(ucfg, &cfg, sizeof(cfg)))
-		return -EFAULT;
-
-	return sizeof(cfg);
-}
-
-static int ptrace_bts_clear(struct task_struct *child)
-{
-	struct bts_context *context;
-	const struct bts_trace *trace;
-
-	context = child->bts;
-	if (!context)
-		return -ESRCH;
-
-	trace = ds_read_bts(context->tracer);
-	if (!trace)
-		return -ESRCH;
-
-	memset(trace->ds.begin, 0, trace->ds.n * trace->ds.size);
-
-	return ds_reset_bts(context->tracer);
-}
-
-static int ptrace_bts_size(struct task_struct *child)
-{
-	struct bts_context *context;
-	const struct bts_trace *trace;
-
-	context = child->bts;
-	if (!context)
-		return -ESRCH;
-
-	trace = ds_read_bts(context->tracer);
-	if (!trace)
-		return -ESRCH;
-
-	return (trace->ds.top - trace->ds.begin) / trace->ds.size;
-}
-
-/*
- * Called from __ptrace_unlink() after the child has been moved back
- * to its original parent.
- */
-void ptrace_bts_untrace(struct task_struct *child)
-{
-	if (unlikely(child->bts)) {
-		free_bts_context(child->bts);
-		child->bts = NULL;
-	}
-}
-#endif /* CONFIG_X86_PTRACE_BTS */
-
 /*
  * Called by kernel/ptrace.c when detaching..
  *
@@ -1252,39 +911,6 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 		break;
 #endif
 
-	/*
-	 * These bits need more cooking - not enabled yet:
-	 */
-#ifdef CONFIG_X86_PTRACE_BTS
-	case PTRACE_BTS_CONFIG:
-		ret = ptrace_bts_config
-			(child, data, (struct ptrace_bts_config __user *)addr);
-		break;
-
-	case PTRACE_BTS_STATUS:
-		ret = ptrace_bts_status
-			(child, data, (struct ptrace_bts_config __user *)addr);
-		break;
-
-	case PTRACE_BTS_SIZE:
-		ret = ptrace_bts_size(child);
-		break;
-
-	case PTRACE_BTS_GET:
-		ret = ptrace_bts_read_record
-			(child, data, (struct bts_struct __user *) addr);
-		break;
-
-	case PTRACE_BTS_CLEAR:
-		ret = ptrace_bts_clear(child);
-		break;
-
-	case PTRACE_BTS_DRAIN:
-		ret = ptrace_bts_drain
-			(child, data, (struct bts_struct __user *) addr);
-		break;
-#endif /* CONFIG_X86_PTRACE_BTS */
-
 	default:
 		ret = ptrace_request(child, request, addr, data);
 		break;
@@ -1544,14 +1170,6 @@ long compat_arch_ptrace(struct task_struct *child, compat_long_t request,
 
 	case PTRACE_GET_THREAD_AREA:
 	case PTRACE_SET_THREAD_AREA:
-#ifdef CONFIG_X86_PTRACE_BTS
-	case PTRACE_BTS_CONFIG:
-	case PTRACE_BTS_STATUS:
-	case PTRACE_BTS_SIZE:
-	case PTRACE_BTS_GET:
-	case PTRACE_BTS_CLEAR:
-	case PTRACE_BTS_DRAIN:
-#endif /* CONFIG_X86_PTRACE_BTS */
 		return arch_ptrace(child, request, addr, data);
 
 	default:

arch/x86/kernel/step.c

@@ -157,22 +157,6 @@ static int enable_single_step(struct task_struct *child)
 	return 1;
 }
 
-/*
- * Install this value in MSR_IA32_DEBUGCTLMSR whenever child is running.
- */
-static void write_debugctlmsr(struct task_struct *child, unsigned long val)
-{
-	if (child->thread.debugctlmsr == val)
-		return;
-
-	child->thread.debugctlmsr = val;
-
-	if (child != current)
-		return;
-
-	update_debugctlmsr(val);
-}
-
 /*
  * Enable single or block step.
  */
@@ -186,15 +170,17 @@ static void enable_step(struct task_struct *child, bool block)
 	 * that uses user-mode single stepping itself.
 	 */
 	if (enable_single_step(child) && block) {
-		set_tsk_thread_flag(child, TIF_DEBUGCTLMSR);
-		write_debugctlmsr(child,
-				  child->thread.debugctlmsr | DEBUGCTLMSR_BTF);
-	} else {
-		write_debugctlmsr(child,
-				  child->thread.debugctlmsr & ~DEBUGCTLMSR_BTF);
-
-		if (!child->thread.debugctlmsr)
-			clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR);
+		unsigned long debugctl = get_debugctlmsr();
+
+		debugctl |= DEBUGCTLMSR_BTF;
+		update_debugctlmsr(debugctl);
+		set_tsk_thread_flag(child, TIF_BLOCKSTEP);
+	} else if (test_tsk_thread_flag(child, TIF_BLOCKSTEP)) {
+		unsigned long debugctl = get_debugctlmsr();
+
+		debugctl &= ~DEBUGCTLMSR_BTF;
+		update_debugctlmsr(debugctl);
+		clear_tsk_thread_flag(child, TIF_BLOCKSTEP);
 	}
 }
 
@@ -213,11 +199,13 @@ void user_disable_single_step(struct task_struct *child)
 	/*
 	 * Make sure block stepping (BTF) is disabled.
 	 */
-	write_debugctlmsr(child,
-			  child->thread.debugctlmsr & ~DEBUGCTLMSR_BTF);
+	if (test_tsk_thread_flag(child, TIF_BLOCKSTEP)) {
+		unsigned long debugctl = get_debugctlmsr();
 
-	if (!child->thread.debugctlmsr)
-		clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR);
+		debugctl &= ~DEBUGCTLMSR_BTF;
+		update_debugctlmsr(debugctl);
+		clear_tsk_thread_flag(child, TIF_BLOCKSTEP);
+	}
 
 	/* Always clear TIF_SINGLESTEP... */
 	clear_tsk_thread_flag(child, TIF_SINGLESTEP);

arch/x86/kernel/traps.c

@@ -543,11 +543,11 @@ dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code)
 
 	/* DR6 may or may not be cleared by the CPU */
 	set_debugreg(0, 6);
+
 	/*
 	 * The processor cleared BTF, so don't mark that we need it set.
 	 */
-	clear_tsk_thread_flag(tsk, TIF_DEBUGCTLMSR);
-	tsk->thread.debugctlmsr = 0;
+	clear_tsk_thread_flag(tsk, TIF_BLOCKSTEP);
 
 	/* Store the virtualized DR6 value */
 	tsk->thread.debugreg6 = dr6;

arch/x86/kvm/vmx.c

@@ -3652,8 +3652,11 @@ static void vmx_complete_interrupts(struct vcpu_vmx *vmx)
 
 	/* We need to handle NMIs before interrupts are enabled */
 	if ((exit_intr_info & INTR_INFO_INTR_TYPE_MASK) == INTR_TYPE_NMI_INTR &&
-	    (exit_intr_info & INTR_INFO_VALID_MASK))
+	    (exit_intr_info & INTR_INFO_VALID_MASK)) {
+		kvm_before_handle_nmi(&vmx->vcpu);
 		asm("int $2");
+		kvm_after_handle_nmi(&vmx->vcpu);
+	}
 
 	idtv_info_valid = idt_vectoring_info & VECTORING_INFO_VALID_MASK;
 

arch/x86/kvm/x86.c

@@ -40,6 +40,7 @@
 #include <linux/user-return-notifier.h>
 #include <linux/srcu.h>
 #include <linux/slab.h>
+#include <linux/perf_event.h>
 #include <trace/events/kvm.h>
 
 #define CREATE_TRACE_POINTS
@@ -3955,6 +3956,47 @@ static void kvm_timer_init(void)
 	}
 }
 
+static DEFINE_PER_CPU(struct kvm_vcpu *, current_vcpu);
+
+static int kvm_is_in_guest(void)
+{
+	return percpu_read(current_vcpu) != NULL;
+}
+
+static int kvm_is_user_mode(void)
+{
+	int user_mode = 3;
+	if (percpu_read(current_vcpu))
+		user_mode = kvm_x86_ops->get_cpl(percpu_read(current_vcpu));
+	return user_mode != 0;
+}
+
+static unsigned long kvm_get_guest_ip(void)
+{
+	unsigned long ip = 0;
+	if (percpu_read(current_vcpu))
+		ip = kvm_rip_read(percpu_read(current_vcpu));
+	return ip;
+}
+
+static struct perf_guest_info_callbacks kvm_guest_cbs = {
+	.is_in_guest		= kvm_is_in_guest,
+	.is_user_mode		= kvm_is_user_mode,
+	.get_guest_ip		= kvm_get_guest_ip,
+};
+
+void kvm_before_handle_nmi(struct kvm_vcpu *vcpu)
+{
+	percpu_write(current_vcpu, vcpu);
+}
+EXPORT_SYMBOL_GPL(kvm_before_handle_nmi);
+
+void kvm_after_handle_nmi(struct kvm_vcpu *vcpu)
+{
+	percpu_write(current_vcpu, NULL);
+}
+EXPORT_SYMBOL_GPL(kvm_after_handle_nmi);
+
 int kvm_arch_init(void *opaque)
 {
 	int r;
@@ -3991,6 +4033,8 @@ int kvm_arch_init(void *opaque)
 
 	kvm_timer_init();
 
+	perf_register_guest_info_callbacks(&kvm_guest_cbs);
+
 	return 0;
 
 out:
@@ -3999,6 +4043,8 @@ out:
 
 void kvm_arch_exit(void)
 {
+	perf_unregister_guest_info_callbacks(&kvm_guest_cbs);
+
 	if (!boot_cpu_has(X86_FEATURE_CONSTANT_TSC))
 		cpufreq_unregister_notifier(&kvmclock_cpufreq_notifier_block,
 					    CPUFREQ_TRANSITION_NOTIFIER);

arch/x86/kvm/x86.h

@@ -65,4 +65,7 @@ static inline int is_paging(struct kvm_vcpu *vcpu)
 	return kvm_read_cr0_bits(vcpu, X86_CR0_PG);
 }
 
+void kvm_before_handle_nmi(struct kvm_vcpu *vcpu);
+void kvm_after_handle_nmi(struct kvm_vcpu *vcpu);
+
 #endif

arch/x86/lib/Makefile

@@ -20,7 +20,7 @@ lib-y := delay.o
 lib-y += thunk_$(BITS).o
 lib-y += usercopy_$(BITS).o getuser.o putuser.o
 lib-y += memcpy_$(BITS).o
-lib-$(CONFIG_KPROBES) += insn.o inat.o
+lib-$(CONFIG_INSTRUCTION_DECODER) += insn.o inat.o
 
 obj-y += msr.o msr-reg.o msr-reg-export.o
 

arch/x86/oprofile/op_model_ppro.c

@@ -239,11 +239,11 @@ static void arch_perfmon_setup_counters(void)
 	if (eax.split.version_id == 0 && current_cpu_data.x86 == 6 &&
 		current_cpu_data.x86_model == 15) {
 		eax.split.version_id = 2;
-		eax.split.num_events = 2;
+		eax.split.num_counters = 2;
 		eax.split.bit_width = 40;
 	}
 
-	num_counters = eax.split.num_events;
+	num_counters = eax.split.num_counters;
 
 	op_arch_perfmon_spec.num_counters = num_counters;
 	op_arch_perfmon_spec.num_controls = num_counters;

include/linux/ftrace.h

@@ -504,18 +504,6 @@ extern int ftrace_dump_on_oops;
 #define INIT_TRACE_RECURSION
 #endif
 
-#ifdef CONFIG_HW_BRANCH_TRACER
-
-void trace_hw_branch(u64 from, u64 to);
-void trace_hw_branch_oops(void);
-
-#else /* CONFIG_HW_BRANCH_TRACER */
-
-static inline void trace_hw_branch(u64 from, u64 to) {}
-static inline void trace_hw_branch_oops(void) {}
-
-#endif /* CONFIG_HW_BRANCH_TRACER */
-
 #ifdef CONFIG_FTRACE_SYSCALLS
 
 unsigned long arch_syscall_addr(int nr);

include/linux/mm.h

@@ -19,7 +19,6 @@ struct anon_vma;
 struct file_ra_state;
 struct user_struct;
 struct writeback_control;
-struct rlimit;
 
 #ifndef CONFIG_DISCONTIGMEM          /* Don't use mapnrs, do it properly */
 extern unsigned long max_mapnr;
@@ -1449,9 +1448,6 @@ int vmemmap_populate_basepages(struct page *start_page,
 int vmemmap_populate(struct page *start_page, unsigned long pages, int node);
 void vmemmap_populate_print_last(void);
 
-extern int account_locked_memory(struct mm_struct *mm, struct rlimit *rlim,
-				 size_t size);
-extern void refund_locked_memory(struct mm_struct *mm, size_t size);
 
 enum mf_flags {
 	MF_COUNT_INCREASED = 1 << 0,

include/linux/perf_event.h

@@ -203,8 +203,9 @@ struct perf_event_attr {
 				enable_on_exec :  1, /* next exec enables     */
 				task           :  1, /* trace fork/exit       */
 				watermark      :  1, /* wakeup_watermark      */
+				precise        :  1, /* OoO invariant counter */
 
-				__reserved_1   : 49;
+				__reserved_1   : 48;
 
 	union {
 		__u32		wakeup_events;	  /* wakeup every n events */
@@ -287,11 +288,19 @@ struct perf_event_mmap_page {
 	__u64	data_tail;		/* user-space written tail */
 };
 
-#define PERF_RECORD_MISC_CPUMODE_MASK		(3 << 0)
+#define PERF_RECORD_MISC_CPUMODE_MASK		(7 << 0)
 #define PERF_RECORD_MISC_CPUMODE_UNKNOWN	(0 << 0)
 #define PERF_RECORD_MISC_KERNEL			(1 << 0)
 #define PERF_RECORD_MISC_USER			(2 << 0)
 #define PERF_RECORD_MISC_HYPERVISOR		(3 << 0)
+#define PERF_RECORD_MISC_GUEST_KERNEL		(4 << 0)
+#define PERF_RECORD_MISC_GUEST_USER		(5 << 0)
+
+#define PERF_RECORD_MISC_EXACT			(1 << 14)
+/*
+ * Reserve the last bit to indicate some extended misc field
+ */
+#define PERF_RECORD_MISC_EXT_RESERVED		(1 << 15)
 
 struct perf_event_header {
 	__u32	type;
@@ -439,6 +448,12 @@ enum perf_callchain_context {
 # include <asm/perf_event.h>
 #endif
 
+struct perf_guest_info_callbacks {
+	int (*is_in_guest) (void);
+	int (*is_user_mode) (void);
+	unsigned long (*get_guest_ip) (void);
+};
+
 #ifdef CONFIG_HAVE_HW_BREAKPOINT
 #include <asm/hw_breakpoint.h>
 #endif
@@ -468,6 +483,17 @@ struct perf_raw_record {
 	void				*data;
 };
 
+struct perf_branch_entry {
+	__u64				from;
+	__u64				to;
+	__u64				flags;
+};
+
+struct perf_branch_stack {
+	__u64				nr;
+	struct perf_branch_entry	entries[0];
+};
+
 struct task_struct;
 
 /**
@@ -571,6 +597,14 @@ enum perf_group_flag {
 	PERF_GROUP_SOFTWARE = 0x1,
 };
 
+#define SWEVENT_HLIST_BITS	8
+#define SWEVENT_HLIST_SIZE	(1 << SWEVENT_HLIST_BITS)
+
+struct swevent_hlist {
+	struct hlist_head	heads[SWEVENT_HLIST_SIZE];
+	struct rcu_head		rcu_head;
+};
+
 /**
  * struct perf_event - performance event kernel representation:
  */
@@ -579,6 +613,7 @@ struct perf_event {
 	struct list_head		group_entry;
 	struct list_head		event_entry;
 	struct list_head		sibling_list;
+	struct hlist_node		hlist_entry;
 	int				nr_siblings;
 	int				group_flags;
 	struct perf_event		*group_leader;
@@ -726,6 +761,9 @@ struct perf_cpu_context {
 	int				active_oncpu;
 	int				max_pertask;
 	int				exclusive;
+	struct swevent_hlist		*swevent_hlist;
+	struct mutex			hlist_mutex;
+	int				hlist_refcount;
 
 	/*
 	 * Recursion avoidance:
@@ -902,6 +940,12 @@ static inline void perf_event_mmap(struct vm_area_struct *vma)
 		__perf_event_mmap(vma);
 }
 
+extern struct perf_guest_info_callbacks *perf_guest_cbs;
+extern int perf_register_guest_info_callbacks(
+		struct perf_guest_info_callbacks *);
+extern int perf_unregister_guest_info_callbacks(
+		struct perf_guest_info_callbacks *);
+
 extern void perf_event_comm(struct task_struct *tsk);
 extern void perf_event_fork(struct task_struct *tsk);
 
@@ -971,6 +1015,11 @@ perf_sw_event(u32 event_id, u64 nr, int nmi,
 static inline void
 perf_bp_event(struct perf_event *event, void *data)			{ }
 
+static inline int perf_register_guest_info_callbacks
+(struct perf_guest_info_callbacks *) {return 0; }
+static inline int perf_unregister_guest_info_callbacks
+(struct perf_guest_info_callbacks *) {return 0; }
+
 static inline void perf_event_mmap(struct vm_area_struct *vma)		{ }
 static inline void perf_event_comm(struct task_struct *tsk)		{ }
 static inline void perf_event_fork(struct task_struct *tsk)		{ }

include/linux/ptrace.h

@@ -345,18 +345,6 @@ static inline void user_single_step_siginfo(struct task_struct *tsk,
 #define arch_ptrace_stop(code, info)		do { } while (0)
 #endif
 
-#ifndef arch_ptrace_untrace
-/*
- * Do machine-specific work before untracing child.
- *
- * This is called for a normal detach as well as from ptrace_exit()
- * when the tracing task dies.
- *
- * Called with write_lock(&tasklist_lock) held.
- */
-#define arch_ptrace_untrace(task)		do { } while (0)
-#endif
-
 extern int task_current_syscall(struct task_struct *target, long *callno,
 				unsigned long args[6], unsigned int maxargs,
 				unsigned long *sp, unsigned long *pc);

include/linux/sched.h

@@ -99,7 +99,6 @@ struct futex_pi_state;
 struct robust_list_head;
 struct bio_list;
 struct fs_struct;
-struct bts_context;
 struct perf_event_context;
 
 /*
@@ -1272,12 +1271,6 @@ struct task_struct {
 	struct list_head ptraced;
 	struct list_head ptrace_entry;
 
-	/*
-	 * This is the tracer handle for the ptrace BTS extension.
-	 * This field actually belongs to the ptracer task.
-	 */
-	struct bts_context *bts;
-
 	/* PID/PID hash table linkage. */
 	struct pid_link pids[PIDTYPE_MAX];
 	struct list_head thread_group;
@@ -2123,10 +2116,8 @@ extern void set_task_comm(struct task_struct *tsk, char *from);
 extern char *get_task_comm(char *to, struct task_struct *tsk);
 
 #ifdef CONFIG_SMP
-extern void wait_task_context_switch(struct task_struct *p);
 extern unsigned long wait_task_inactive(struct task_struct *, long match_state);
 #else
-static inline void wait_task_context_switch(struct task_struct *p) {}
 static inline unsigned long wait_task_inactive(struct task_struct *p,
 					       long match_state)
 {

include/trace/ftrace.h

@@ -758,13 +758,12 @@ __attribute__((section("_ftrace_events"))) event_##call = {		\
 #define DECLARE_EVENT_CLASS(call, proto, args, tstruct, assign, print)	\
 static notrace void							\
 perf_trace_templ_##call(struct ftrace_event_call *event_call,		\
-			    proto)					\
+			struct pt_regs *__regs, proto)			\
 {									\
 	struct ftrace_data_offsets_##call __maybe_unused __data_offsets;\
 	struct ftrace_raw_##call *entry;				\
 	u64 __addr = 0, __count = 1;					\
 	unsigned long irq_flags;					\
-	struct pt_regs *__regs;						\
 	int __entry_size;						\
 	int __data_size;						\
 	int rctx;							\
@@ -785,20 +784,22 @@ perf_trace_templ_##call(struct ftrace_event_call *event_call,		\
 									\
 	{ assign; }							\
 									\
-	__regs = &__get_cpu_var(perf_trace_regs);			\
-	perf_fetch_caller_regs(__regs, 2);				\
-									\
 	perf_trace_buf_submit(entry, __entry_size, rctx, __addr,	\
 			       __count, irq_flags, __regs);		\
 }
 
 #undef DEFINE_EVENT
-#define DEFINE_EVENT(template, call, proto, args)		\
-static notrace void perf_trace_##call(proto)			\
-{								\
-	struct ftrace_event_call *event_call = &event_##call;	\
-								\
-	perf_trace_templ_##template(event_call, args);		\
+#define DEFINE_EVENT(template, call, proto, args)			\
+static notrace void perf_trace_##call(proto)				\
+{									\
+	struct ftrace_event_call *event_call = &event_##call;		\
+	struct pt_regs *__regs = &get_cpu_var(perf_trace_regs);		\
+									\
+	perf_fetch_caller_regs(__regs, 1);				\
+									\
+	perf_trace_templ_##template(event_call, __regs, args);		\
+									\
+	put_cpu_var(perf_trace_regs);					\
 }
 
 #undef DEFINE_EVENT_PRINT

kernel/fork.c

@@ -1111,9 +1111,6 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 	p->memcg_batch.do_batch = 0;
 	p->memcg_batch.memcg = NULL;
 #endif
-
-	p->bts = NULL;
-
 	p->stack_start = stack_start;
 
 	/* Perform scheduler related setup. Assign this task to a CPU. */

kernel/perf_event.c

@@ -16,6 +16,7 @@
 #include <linux/file.h>
 #include <linux/poll.h>
 #include <linux/slab.h>
+#include <linux/hash.h>
 #include <linux/sysfs.h>
 #include <linux/dcache.h>
 #include <linux/percpu.h>
@@ -1367,6 +1368,8 @@ void perf_event_task_sched_in(struct task_struct *task)
 	if (cpuctx->task_ctx == ctx)
 		return;
 
+	perf_disable();
+
 	/*
 	 * We want to keep the following priority order:
 	 * cpu pinned (that don't need to move), task pinned,
@@ -1379,6 +1382,8 @@ void perf_event_task_sched_in(struct task_struct *task)
 	ctx_sched_in(ctx, cpuctx, EVENT_FLEXIBLE);
 
 	cpuctx->task_ctx = ctx;
+
+	perf_enable();
 }
 
 #define MAX_INTERRUPTS (~0ULL)
@@ -2642,6 +2647,7 @@ static int perf_fasync(int fd, struct file *filp, int on)
 }
 
 static const struct file_operations perf_fops = {
+	.llseek			= no_llseek,
 	.release		= perf_release,
 	.read			= perf_read,
 	.poll			= perf_poll,
@@ -2791,6 +2797,27 @@ void perf_arch_fetch_caller_regs(struct pt_regs *regs, unsigned long ip, int ski
 }
 
 
+/*
+ * We assume there is only KVM supporting the callbacks.
+ * Later on, we might change it to a list if there is
+ * another virtualization implementation supporting the callbacks.
+ */
+struct perf_guest_info_callbacks *perf_guest_cbs;
+
+int perf_register_guest_info_callbacks(struct perf_guest_info_callbacks *cbs)
+{
+	perf_guest_cbs = cbs;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(perf_register_guest_info_callbacks);
+
+int perf_unregister_guest_info_callbacks(struct perf_guest_info_callbacks *cbs)
+{
+	perf_guest_cbs = NULL;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(perf_unregister_guest_info_callbacks);
+
 /*
  * Output
  */
@@ -3743,7 +3770,7 @@ void __perf_event_mmap(struct vm_area_struct *vma)
 		.event_id  = {
 			.header = {
 				.type = PERF_RECORD_MMAP,
-				.misc = 0,
+				.misc = PERF_RECORD_MISC_USER,
 				/* .size */
 			},
 			/* .pid */
@@ -3961,36 +3988,6 @@ static void perf_swevent_add(struct perf_event *event, u64 nr,
 	perf_swevent_overflow(event, 0, nmi, data, regs);
 }
 
-static int perf_swevent_is_counting(struct perf_event *event)
-{
-	/*
-	 * The event is active, we're good!
-	 */
-	if (event->state == PERF_EVENT_STATE_ACTIVE)
-		return 1;
-
-	/*
-	 * The event is off/error, not counting.
-	 */
-	if (event->state != PERF_EVENT_STATE_INACTIVE)
-		return 0;
-
-	/*
-	 * The event is inactive, if the context is active
-	 * we're part of a group that didn't make it on the 'pmu',
-	 * not counting.
-	 */
-	if (event->ctx->is_active)
-		return 0;
-
-	/*
-	 * We're inactive and the context is too, this means the
-	 * task is scheduled out, we're counting events that happen
-	 * to us, like migration events.
-	 */
-	return 1;
-}
-
 static int perf_tp_event_match(struct perf_event *event,
 				struct perf_sample_data *data);
 
@@ -4014,12 +4011,6 @@ static int perf_swevent_match(struct perf_event *event,
 				struct perf_sample_data *data,
 				struct pt_regs *regs)
 {
-	if (event->cpu != -1 && event->cpu != smp_processor_id())
-		return 0;
-
-	if (!perf_swevent_is_counting(event))
-		return 0;
-
 	if (event->attr.type != type)
 		return 0;
 
@@ -4036,18 +4027,53 @@ static int perf_swevent_match(struct perf_event *event,
 	return 1;
 }
 
-static void perf_swevent_ctx_event(struct perf_event_context *ctx,
-				     enum perf_type_id type,
-				     u32 event_id, u64 nr, int nmi,
-				     struct perf_sample_data *data,
-				     struct pt_regs *regs)
+static inline u64 swevent_hash(u64 type, u32 event_id)
+{
+	u64 val = event_id | (type << 32);
+
+	return hash_64(val, SWEVENT_HLIST_BITS);
+}
+
+static struct hlist_head *
+find_swevent_head(struct perf_cpu_context *ctx, u64 type, u32 event_id)
 {
+	u64 hash;
+	struct swevent_hlist *hlist;
+
+	hash = swevent_hash(type, event_id);
+
+	hlist = rcu_dereference(ctx->swevent_hlist);
+	if (!hlist)
+		return NULL;
+
+	return &hlist->heads[hash];
+}
+
+static void do_perf_sw_event(enum perf_type_id type, u32 event_id,
+				    u64 nr, int nmi,
+				    struct perf_sample_data *data,
+				    struct pt_regs *regs)
+{
+	struct perf_cpu_context *cpuctx;
 	struct perf_event *event;
+	struct hlist_node *node;
+	struct hlist_head *head;
 
-	list_for_each_entry_rcu(event, &ctx->event_list, event_entry) {
+	cpuctx = &__get_cpu_var(perf_cpu_context);
+
+	rcu_read_lock();
+
+	head = find_swevent_head(cpuctx, type, event_id);
+
+	if (!head)
+		goto end;
+
+	hlist_for_each_entry_rcu(event, node, head, hlist_entry) {
 		if (perf_swevent_match(event, type, event_id, data, regs))
 			perf_swevent_add(event, nr, nmi, data, regs);
 	}
+end:
+	rcu_read_unlock();
 }
 
 int perf_swevent_get_recursion_context(void)
@@ -4085,27 +4111,6 @@ void perf_swevent_put_recursion_context(int rctx)
 }
 EXPORT_SYMBOL_GPL(perf_swevent_put_recursion_context);
 
-static void do_perf_sw_event(enum perf_type_id type, u32 event_id,
-				    u64 nr, int nmi,
-				    struct perf_sample_data *data,
-				    struct pt_regs *regs)
-{
-	struct perf_cpu_context *cpuctx;
-	struct perf_event_context *ctx;
-
-	cpuctx = &__get_cpu_var(perf_cpu_context);
-	rcu_read_lock();
-	perf_swevent_ctx_event(&cpuctx->ctx, type, event_id,
-				 nr, nmi, data, regs);
-	/*
-	 * doesn't really matter which of the child contexts the
-	 * events ends up in.
-	 */
-	ctx = rcu_dereference(current->perf_event_ctxp);
-	if (ctx)
-		perf_swevent_ctx_event(ctx, type, event_id, nr, nmi, data, regs);
-	rcu_read_unlock();
-}
 
 void __perf_sw_event(u32 event_id, u64 nr, int nmi,
 			    struct pt_regs *regs, u64 addr)
@@ -4131,16 +4136,28 @@ static void perf_swevent_read(struct perf_event *event)
 static int perf_swevent_enable(struct perf_event *event)
 {
 	struct hw_perf_event *hwc = &event->hw;
+	struct perf_cpu_context *cpuctx;
+	struct hlist_head *head;
+
+	cpuctx = &__get_cpu_var(perf_cpu_context);
 
 	if (hwc->sample_period) {
 		hwc->last_period = hwc->sample_period;
 		perf_swevent_set_period(event);
 	}
+
+	head = find_swevent_head(cpuctx, event->attr.type, event->attr.config);
+	if (WARN_ON_ONCE(!head))
+		return -EINVAL;
+
+	hlist_add_head_rcu(&event->hlist_entry, head);
+
 	return 0;
 }
 
 static void perf_swevent_disable(struct perf_event *event)
 {
+	hlist_del_rcu(&event->hlist_entry);
 }
 
 static const struct pmu perf_ops_generic = {
@@ -4168,15 +4185,8 @@ static enum hrtimer_restart perf_swevent_hrtimer(struct hrtimer *hrtimer)
 	perf_sample_data_init(&data, 0);
 	data.period = event->hw.last_period;
 	regs = get_irq_regs();
-	/*
-	 * In case we exclude kernel IPs or are somehow not in interrupt
-	 * context, provide the next best thing, the user IP.
-	 */
-	if ((event->attr.exclude_kernel || !regs) &&
-			!event->attr.exclude_user)
-		regs = task_pt_regs(current);
 
-	if (regs) {
+	if (regs && !perf_exclude_event(event, regs)) {
 		if (!(event->attr.exclude_idle && current->pid == 0))
 			if (perf_event_overflow(event, 0, &data, regs))
 				ret = HRTIMER_NORESTART;
@@ -4324,6 +4334,105 @@ static const struct pmu perf_ops_task_clock = {
 	.read		= task_clock_perf_event_read,
 };
 
+static void swevent_hlist_release_rcu(struct rcu_head *rcu_head)
+{
+	struct swevent_hlist *hlist;
+
+	hlist = container_of(rcu_head, struct swevent_hlist, rcu_head);
+	kfree(hlist);
+}
+
+static void swevent_hlist_release(struct perf_cpu_context *cpuctx)
+{
+	struct swevent_hlist *hlist;
+
+	if (!cpuctx->swevent_hlist)
+		return;
+
+	hlist = cpuctx->swevent_hlist;
+	rcu_assign_pointer(cpuctx->swevent_hlist, NULL);
+	call_rcu(&hlist->rcu_head, swevent_hlist_release_rcu);
+}
+
+static void swevent_hlist_put_cpu(struct perf_event *event, int cpu)
+{
+	struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu);
+
+	mutex_lock(&cpuctx->hlist_mutex);
+
+	if (!--cpuctx->hlist_refcount)
+		swevent_hlist_release(cpuctx);
+
+	mutex_unlock(&cpuctx->hlist_mutex);
+}
+
+static void swevent_hlist_put(struct perf_event *event)
+{
+	int cpu;
+
+	if (event->cpu != -1) {
+		swevent_hlist_put_cpu(event, event->cpu);
+		return;
+	}
+
+	for_each_possible_cpu(cpu)
+		swevent_hlist_put_cpu(event, cpu);
+}
+
+static int swevent_hlist_get_cpu(struct perf_event *event, int cpu)
+{
+	struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu);
+	int err = 0;
+
+	mutex_lock(&cpuctx->hlist_mutex);
+
+	if (!cpuctx->swevent_hlist && cpu_online(cpu)) {
+		struct swevent_hlist *hlist;
+
+		hlist = kzalloc(sizeof(*hlist), GFP_KERNEL);
+		if (!hlist) {
+			err = -ENOMEM;
+			goto exit;
+		}
+		rcu_assign_pointer(cpuctx->swevent_hlist, hlist);
+	}
+	cpuctx->hlist_refcount++;
+ exit:
+	mutex_unlock(&cpuctx->hlist_mutex);
+
+	return err;
+}
+
+static int swevent_hlist_get(struct perf_event *event)
+{
+	int err;
+	int cpu, failed_cpu;
+
+	if (event->cpu != -1)
+		return swevent_hlist_get_cpu(event, event->cpu);
+
+	get_online_cpus();
+	for_each_possible_cpu(cpu) {
+		err = swevent_hlist_get_cpu(event, cpu);
+		if (err) {
+			failed_cpu = cpu;
+			goto fail;
+		}
+	}
+	put_online_cpus();
+
+	return 0;
+ fail:
+	for_each_possible_cpu(cpu) {
+		if (cpu == failed_cpu)
+			break;
+		swevent_hlist_put_cpu(event, cpu);
+	}
+
+	put_online_cpus();
+	return err;
+}
+
 #ifdef CONFIG_EVENT_TRACING
 
 void perf_tp_event(int event_id, u64 addr, u64 count, void *record,
@@ -4357,10 +4466,13 @@ static int perf_tp_event_match(struct perf_event *event,
 static void tp_perf_event_destroy(struct perf_event *event)
 {
 	perf_trace_disable(event->attr.config);
+	swevent_hlist_put(event);
 }
 
 static const struct pmu *tp_perf_event_init(struct perf_event *event)
 {
+	int err;
+
 	/*
 	 * Raw tracepoint data is a severe data leak, only allow root to
 	 * have these.
@@ -4374,6 +4486,11 @@ static const struct pmu *tp_perf_event_init(struct perf_event *event)
 		return NULL;
 
 	event->destroy = tp_perf_event_destroy;
+	err = swevent_hlist_get(event);
+	if (err) {
+		perf_trace_disable(event->attr.config);
+		return ERR_PTR(err);
+	}
 
 	return &perf_ops_generic;
 }
@@ -4474,6 +4591,7 @@ static void sw_perf_event_destroy(struct perf_event *event)
 	WARN_ON(event->parent);
 
 	atomic_dec(&perf_swevent_enabled[event_id]);
+	swevent_hlist_put(event);
 }
 
 static const struct pmu *sw_perf_event_init(struct perf_event *event)
@@ -4512,6 +4630,12 @@ static const struct pmu *sw_perf_event_init(struct perf_event *event)
 	case PERF_COUNT_SW_ALIGNMENT_FAULTS:
 	case PERF_COUNT_SW_EMULATION_FAULTS:
 		if (!event->parent) {
+			int err;
+
+			err = swevent_hlist_get(event);
+			if (err)
+				return ERR_PTR(err);
+
 			atomic_inc(&perf_swevent_enabled[event_id]);
 			event->destroy = sw_perf_event_destroy;
 		}
@@ -5384,6 +5508,7 @@ static void __init perf_event_init_all_cpus(void)
 
 	for_each_possible_cpu(cpu) {
 		cpuctx = &per_cpu(perf_cpu_context, cpu);
+		mutex_init(&cpuctx->hlist_mutex);
 		__perf_event_init_context(&cpuctx->ctx, NULL);
 	}
 }
@@ -5397,6 +5522,16 @@ static void __cpuinit perf_event_init_cpu(int cpu)
 	spin_lock(&perf_resource_lock);
 	cpuctx->max_pertask = perf_max_events - perf_reserved_percpu;
 	spin_unlock(&perf_resource_lock);
+
+	mutex_lock(&cpuctx->hlist_mutex);
+	if (cpuctx->hlist_refcount > 0) {
+		struct swevent_hlist *hlist;
+
+		hlist = kzalloc(sizeof(*hlist), GFP_KERNEL);
+		WARN_ON_ONCE(!hlist);
+		rcu_assign_pointer(cpuctx->swevent_hlist, hlist);
+	}
+	mutex_unlock(&cpuctx->hlist_mutex);
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
@@ -5416,6 +5551,10 @@ static void perf_event_exit_cpu(int cpu)
 	struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu);
 	struct perf_event_context *ctx = &cpuctx->ctx;
 
+	mutex_lock(&cpuctx->hlist_mutex);
+	swevent_hlist_release(cpuctx);
+	mutex_unlock(&cpuctx->hlist_mutex);
+
 	mutex_lock(&ctx->mutex);
 	smp_call_function_single(cpu, __perf_event_exit_cpu, NULL, 1);
 	mutex_unlock(&ctx->mutex);

kernel/ptrace.c

@@ -76,7 +76,6 @@ void __ptrace_unlink(struct task_struct *child)
 	child->parent = child->real_parent;
 	list_del_init(&child->ptrace_entry);
 
-	arch_ptrace_untrace(child);
 	if (task_is_traced(child))
 		ptrace_untrace(child);
 }

kernel/sched.c

@@ -2077,49 +2077,6 @@ migrate_task(struct task_struct *p, int dest_cpu, struct migration_req *req)
 	return 1;
 }
 
-/*
- * wait_task_context_switch -	wait for a thread to complete at least one
- *				context switch.
- *
- * @p must not be current.
- */
-void wait_task_context_switch(struct task_struct *p)
-{
-	unsigned long nvcsw, nivcsw, flags;
-	int running;
-	struct rq *rq;
-
-	nvcsw	= p->nvcsw;
-	nivcsw	= p->nivcsw;
-	for (;;) {
-		/*
-		 * The runqueue is assigned before the actual context
-		 * switch. We need to take the runqueue lock.
-		 *
-		 * We could check initially without the lock but it is
-		 * very likely that we need to take the lock in every
-		 * iteration.
-		 */
-		rq = task_rq_lock(p, &flags);
-		running = task_running(rq, p);
-		task_rq_unlock(rq, &flags);
-
-		if (likely(!running))
-			break;
-		/*
-		 * The switch count is incremented before the actual
-		 * context switch. We thus wait for two switches to be
-		 * sure at least one completed.
-		 */
-		if ((p->nvcsw - nvcsw) > 1)
-			break;
-		if ((p->nivcsw - nivcsw) > 1)
-			break;
-
-		cpu_relax();
-	}
-}
-
 /*
  * wait_task_inactive - wait for a thread to unschedule.
  *

kernel/trace/Kconfig

@@ -44,9 +44,6 @@ config HAVE_FTRACE_MCOUNT_RECORD
 	help
 	  See Documentation/trace/ftrace-design.txt
 
-config HAVE_HW_BRANCH_TRACER
-	bool
-
 config HAVE_SYSCALL_TRACEPOINTS
 	bool
 	help
@@ -374,14 +371,6 @@ config STACK_TRACER
 
 	  Say N if unsure.
 
-config HW_BRANCH_TRACER
-	depends on HAVE_HW_BRANCH_TRACER
-	bool "Trace hw branches"
-	select GENERIC_TRACER
-	help
-	  This tracer records all branches on the system in a circular
-	  buffer, giving access to the last N branches for each cpu.
-
 config KMEMTRACE
 	bool "Trace SLAB allocations"
 	select GENERIC_TRACER

kernel/trace/Makefile

@@ -41,7 +41,6 @@ obj-$(CONFIG_MMIOTRACE) += trace_mmiotrace.o
 obj-$(CONFIG_BOOT_TRACER) += trace_boot.o
 obj-$(CONFIG_FUNCTION_GRAPH_TRACER) += trace_functions_graph.o
 obj-$(CONFIG_TRACE_BRANCH_PROFILING) += trace_branch.o
-obj-$(CONFIG_HW_BRANCH_TRACER) += trace_hw_branches.o
 obj-$(CONFIG_KMEMTRACE) += kmemtrace.o
 obj-$(CONFIG_WORKQUEUE_TRACER) += trace_workqueue.o
 obj-$(CONFIG_BLK_DEV_IO_TRACE) += blktrace.o

kernel/trace/trace.h

@@ -34,7 +34,6 @@ enum trace_type {
 	TRACE_GRAPH_RET,
 	TRACE_GRAPH_ENT,
 	TRACE_USER_STACK,
-	TRACE_HW_BRANCHES,
 	TRACE_KMEM_ALLOC,
 	TRACE_KMEM_FREE,
 	TRACE_BLK,
@@ -103,29 +102,17 @@ struct syscall_trace_exit {
 	long			ret;
 };
 
-struct kprobe_trace_entry {
+struct kprobe_trace_entry_head {
 	struct trace_entry	ent;
 	unsigned long		ip;
-	int			nargs;
-	unsigned long		args[];
 };
 
-#define SIZEOF_KPROBE_TRACE_ENTRY(n)			\
-	(offsetof(struct kprobe_trace_entry, args) +	\
-	(sizeof(unsigned long) * (n)))
-
-struct kretprobe_trace_entry {
+struct kretprobe_trace_entry_head {
 	struct trace_entry	ent;
 	unsigned long		func;
 	unsigned long		ret_ip;
-	int			nargs;
-	unsigned long		args[];
 };
 
-#define SIZEOF_KRETPROBE_TRACE_ENTRY(n)			\
-	(offsetof(struct kretprobe_trace_entry, args) +	\
-	(sizeof(unsigned long) * (n)))
-
 /*
  * trace_flag_type is an enumeration that holds different
  * states when a trace occurs. These are:
@@ -229,7 +216,6 @@ extern void __ftrace_bad_type(void);
 			  TRACE_GRAPH_ENT);		\
 		IF_ASSIGN(var, ent, struct ftrace_graph_ret_entry,	\
 			  TRACE_GRAPH_RET);		\
-		IF_ASSIGN(var, ent, struct hw_branch_entry, TRACE_HW_BRANCHES);\
 		IF_ASSIGN(var, ent, struct kmemtrace_alloc_entry,	\
 			  TRACE_KMEM_ALLOC);	\
 		IF_ASSIGN(var, ent, struct kmemtrace_free_entry,	\
@@ -467,8 +453,6 @@ extern int trace_selftest_startup_sysprof(struct tracer *trace,
 					       struct trace_array *tr);
 extern int trace_selftest_startup_branch(struct tracer *trace,
 					 struct trace_array *tr);
-extern int trace_selftest_startup_hw_branches(struct tracer *trace,
-					      struct trace_array *tr);
 extern int trace_selftest_startup_ksym(struct tracer *trace,
 					 struct trace_array *tr);
 #endif /* CONFIG_FTRACE_STARTUP_TEST */

kernel/trace/trace_entries.h

@@ -318,18 +318,6 @@ FTRACE_ENTRY(branch, trace_branch,
 		 __entry->func, __entry->file, __entry->correct)
 );
 
-FTRACE_ENTRY(hw_branch, hw_branch_entry,
-
-	TRACE_HW_BRANCHES,
-
-	F_STRUCT(
-		__field(	u64,	from	)
-		__field(	u64,	to	)
-	),
-
-	F_printk("from: %llx to: %llx", __entry->from, __entry->to)
-);
-
 FTRACE_ENTRY(kmem_alloc, kmemtrace_alloc_entry,
 
 	TRACE_KMEM_ALLOC,

kernel/trace/trace_hw_branches.c

@@ -1,312 +0,0 @@
-/*
- * h/w branch tracer for x86 based on BTS
- *
- * Copyright (C) 2008-2009 Intel Corporation.
- * Markus Metzger <markus.t.metzger@gmail.com>, 2008-2009
- */
-#include <linux/kallsyms.h>
-#include <linux/debugfs.h>
-#include <linux/ftrace.h>
-#include <linux/module.h>
-#include <linux/cpu.h>
-#include <linux/smp.h>
-#include <linux/fs.h>
-
-#include <asm/ds.h>
-
-#include "trace_output.h"
-#include "trace.h"
-
-
-#define BTS_BUFFER_SIZE (1 << 13)
-
-static DEFINE_PER_CPU(struct bts_tracer *, hwb_tracer);
-static DEFINE_PER_CPU(unsigned char[BTS_BUFFER_SIZE], hwb_buffer);
-
-#define this_tracer per_cpu(hwb_tracer, smp_processor_id())
-
-static int trace_hw_branches_enabled __read_mostly;
-static int trace_hw_branches_suspended __read_mostly;
-static struct trace_array *hw_branch_trace __read_mostly;
-
-
-static void bts_trace_init_cpu(int cpu)
-{
-	per_cpu(hwb_tracer, cpu) =
-		ds_request_bts_cpu(cpu, per_cpu(hwb_buffer, cpu),
-				   BTS_BUFFER_SIZE, NULL, (size_t)-1,
-				   BTS_KERNEL);
-
-	if (IS_ERR(per_cpu(hwb_tracer, cpu)))
-		per_cpu(hwb_tracer, cpu) = NULL;
-}
-
-static int bts_trace_init(struct trace_array *tr)
-{
-	int cpu;
-
-	hw_branch_trace = tr;
-	trace_hw_branches_enabled = 0;
-
-	get_online_cpus();
-	for_each_online_cpu(cpu) {
-		bts_trace_init_cpu(cpu);
-
-		if (likely(per_cpu(hwb_tracer, cpu)))
-			trace_hw_branches_enabled = 1;
-	}
-	trace_hw_branches_suspended = 0;
-	put_online_cpus();
-
-	/* If we could not enable tracing on a single cpu, we fail. */
-	return trace_hw_branches_enabled ? 0 : -EOPNOTSUPP;
-}
-
-static void bts_trace_reset(struct trace_array *tr)
-{
-	int cpu;
-
-	get_online_cpus();
-	for_each_online_cpu(cpu) {
-		if (likely(per_cpu(hwb_tracer, cpu))) {
-			ds_release_bts(per_cpu(hwb_tracer, cpu));
-			per_cpu(hwb_tracer, cpu) = NULL;
-		}
-	}
-	trace_hw_branches_enabled = 0;
-	trace_hw_branches_suspended = 0;
-	put_online_cpus();
-}
-
-static void bts_trace_start(struct trace_array *tr)
-{
-	int cpu;
-
-	get_online_cpus();
-	for_each_online_cpu(cpu)
-		if (likely(per_cpu(hwb_tracer, cpu)))
-			ds_resume_bts(per_cpu(hwb_tracer, cpu));
-	trace_hw_branches_suspended = 0;
-	put_online_cpus();
-}
-
-static void bts_trace_stop(struct trace_array *tr)
-{
-	int cpu;
-
-	get_online_cpus();
-	for_each_online_cpu(cpu)
-		if (likely(per_cpu(hwb_tracer, cpu)))
-			ds_suspend_bts(per_cpu(hwb_tracer, cpu));
-	trace_hw_branches_suspended = 1;
-	put_online_cpus();
-}
-
-static int __cpuinit bts_hotcpu_handler(struct notifier_block *nfb,
-				     unsigned long action, void *hcpu)
-{
-	int cpu = (long)hcpu;
-
-	switch (action) {
-	case CPU_ONLINE:
-	case CPU_DOWN_FAILED:
-		/* The notification is sent with interrupts enabled. */
-		if (trace_hw_branches_enabled) {
-			bts_trace_init_cpu(cpu);
-
-			if (trace_hw_branches_suspended &&
-			    likely(per_cpu(hwb_tracer, cpu)))
-				ds_suspend_bts(per_cpu(hwb_tracer, cpu));
-		}
-		break;
-
-	case CPU_DOWN_PREPARE:
-		/* The notification is sent with interrupts enabled. */
-		if (likely(per_cpu(hwb_tracer, cpu))) {
-			ds_release_bts(per_cpu(hwb_tracer, cpu));
-			per_cpu(hwb_tracer, cpu) = NULL;
-		}
-	}
-
-	return NOTIFY_DONE;
-}
-
-static struct notifier_block bts_hotcpu_notifier __cpuinitdata = {
-	.notifier_call = bts_hotcpu_handler
-};
-
-static void bts_trace_print_header(struct seq_file *m)
-{
-	seq_puts(m, "# CPU#        TO  <-  FROM\n");
-}
-
-static enum print_line_t bts_trace_print_line(struct trace_iterator *iter)
-{
-	unsigned long symflags = TRACE_ITER_SYM_OFFSET;
-	struct trace_entry *entry = iter->ent;
-	struct trace_seq *seq = &iter->seq;
-	struct hw_branch_entry *it;
-
-	trace_assign_type(it, entry);
-
-	if (entry->type == TRACE_HW_BRANCHES) {
-		if (trace_seq_printf(seq, "%4d  ", iter->cpu) &&
-		    seq_print_ip_sym(seq, it->to, symflags) &&
-		    trace_seq_printf(seq, "\t  <-  ") &&
-		    seq_print_ip_sym(seq, it->from, symflags) &&
-		    trace_seq_printf(seq, "\n"))
-			return TRACE_TYPE_HANDLED;
-		return TRACE_TYPE_PARTIAL_LINE;
-	}
-	return TRACE_TYPE_UNHANDLED;
-}
-
-void trace_hw_branch(u64 from, u64 to)
-{
-	struct ftrace_event_call *call = &event_hw_branch;
-	struct trace_array *tr = hw_branch_trace;
-	struct ring_buffer_event *event;
-	struct ring_buffer *buf;
-	struct hw_branch_entry *entry;
-	unsigned long irq1;
-	int cpu;
-
-	if (unlikely(!tr))
-		return;
-
-	if (unlikely(!trace_hw_branches_enabled))
-		return;
-
-	local_irq_save(irq1);
-	cpu = raw_smp_processor_id();
-	if (atomic_inc_return(&tr->data[cpu]->disabled) != 1)
-		goto out;
-
-	buf = tr->buffer;
-	event = trace_buffer_lock_reserve(buf, TRACE_HW_BRANCHES,
-					  sizeof(*entry), 0, 0);
-	if (!event)
-		goto out;
-	entry	= ring_buffer_event_data(event);
-	tracing_generic_entry_update(&entry->ent, 0, from);
-	entry->ent.type = TRACE_HW_BRANCHES;
-	entry->from = from;
-	entry->to   = to;
-	if (!filter_check_discard(call, entry, buf, event))
-		trace_buffer_unlock_commit(buf, event, 0, 0);
-
- out:
-	atomic_dec(&tr->data[cpu]->disabled);
-	local_irq_restore(irq1);
-}
-
-static void trace_bts_at(const struct bts_trace *trace, void *at)
-{
-	struct bts_struct bts;
-	int err = 0;
-
-	WARN_ON_ONCE(!trace->read);
-	if (!trace->read)
-		return;
-
-	err = trace->read(this_tracer, at, &bts);
-	if (err < 0)
-		return;
-
-	switch (bts.qualifier) {
-	case BTS_BRANCH:
-		trace_hw_branch(bts.variant.lbr.from, bts.variant.lbr.to);
-		break;
-	}
-}
-
-/*
- * Collect the trace on the current cpu and write it into the ftrace buffer.
- *
- * pre: tracing must be suspended on the current cpu
- */
-static void trace_bts_cpu(void *arg)
-{
-	struct trace_array *tr = (struct trace_array *)arg;
-	const struct bts_trace *trace;
-	unsigned char *at;
-
-	if (unlikely(!tr))
-		return;
-
-	if (unlikely(atomic_read(&tr->data[raw_smp_processor_id()]->disabled)))
-		return;
-
-	if (unlikely(!this_tracer))
-		return;
-
-	trace = ds_read_bts(this_tracer);
-	if (!trace)
-		return;
-
-	for (at = trace->ds.top; (void *)at < trace->ds.end;
-	     at += trace->ds.size)
-		trace_bts_at(trace, at);
-
-	for (at = trace->ds.begin; (void *)at < trace->ds.top;
-	     at += trace->ds.size)
-		trace_bts_at(trace, at);
-}
-
-static void trace_bts_prepare(struct trace_iterator *iter)
-{
-	int cpu;
-
-	get_online_cpus();
-	for_each_online_cpu(cpu)
-		if (likely(per_cpu(hwb_tracer, cpu)))
-			ds_suspend_bts(per_cpu(hwb_tracer, cpu));
-	/*
-	 * We need to collect the trace on the respective cpu since ftrace
-	 * implicitly adds the record for the current cpu.
-	 * Once that is more flexible, we could collect the data from any cpu.
-	 */
-	on_each_cpu(trace_bts_cpu, iter->tr, 1);
-
-	for_each_online_cpu(cpu)
-		if (likely(per_cpu(hwb_tracer, cpu)))
-			ds_resume_bts(per_cpu(hwb_tracer, cpu));
-	put_online_cpus();
-}
-
-static void trace_bts_close(struct trace_iterator *iter)
-{
-	tracing_reset_online_cpus(iter->tr);
-}
-
-void trace_hw_branch_oops(void)
-{
-	if (this_tracer) {
-		ds_suspend_bts_noirq(this_tracer);
-		trace_bts_cpu(hw_branch_trace);
-		ds_resume_bts_noirq(this_tracer);
-	}
-}
-
-struct tracer bts_tracer __read_mostly =
-{
-	.name		= "hw-branch-tracer",
-	.init		= bts_trace_init,
-	.reset		= bts_trace_reset,
-	.print_header	= bts_trace_print_header,
-	.print_line	= bts_trace_print_line,
-	.start		= bts_trace_start,
-	.stop		= bts_trace_stop,
-	.open		= trace_bts_prepare,
-	.close		= trace_bts_close,
-#ifdef CONFIG_FTRACE_SELFTEST
-	.selftest	= trace_selftest_startup_hw_branches,
-#endif /* CONFIG_FTRACE_SELFTEST */
-};
-
-__init static int init_bts_trace(void)
-{
-	register_hotcpu_notifier(&bts_hotcpu_notifier);
-	return register_tracer(&bts_tracer);
-}
-device_initcall(init_bts_trace);

kernel/trace/trace_kprobe.c

@@ -29,6 +29,8 @@
 #include <linux/ctype.h>
 #include <linux/ptrace.h>
 #include <linux/perf_event.h>
+#include <linux/stringify.h>
+#include <asm/bitsperlong.h>
 
 #include "trace.h"
 #include "trace_output.h"
@@ -40,7 +42,6 @@
 
 /* Reserved field names */
 #define FIELD_STRING_IP "__probe_ip"
-#define FIELD_STRING_NARGS "__probe_nargs"
 #define FIELD_STRING_RETIP "__probe_ret_ip"
 #define FIELD_STRING_FUNC "__probe_func"
 
@@ -52,56 +53,102 @@ const char *reserved_field_names[] = {
 	"common_tgid",
 	"common_lock_depth",
 	FIELD_STRING_IP,
-	FIELD_STRING_NARGS,
 	FIELD_STRING_RETIP,
 	FIELD_STRING_FUNC,
 };
 
-struct fetch_func {
-	unsigned long (*func)(struct pt_regs *, void *);
+/* Printing function type */
+typedef int (*print_type_func_t)(struct trace_seq *, const char *, void *);
+#define PRINT_TYPE_FUNC_NAME(type)	print_type_##type
+#define PRINT_TYPE_FMT_NAME(type)	print_type_format_##type
+
+/* Printing  in basic type function template */
+#define DEFINE_BASIC_PRINT_TYPE_FUNC(type, fmt, cast)			\
+static __kprobes int PRINT_TYPE_FUNC_NAME(type)(struct trace_seq *s,	\
+						const char *name, void *data)\
+{									\
+	return trace_seq_printf(s, " %s=" fmt, name, (cast)*(type *)data);\
+}									\
+static const char PRINT_TYPE_FMT_NAME(type)[] = fmt;
+
+DEFINE_BASIC_PRINT_TYPE_FUNC(u8, "%x", unsigned int)
+DEFINE_BASIC_PRINT_TYPE_FUNC(u16, "%x", unsigned int)
+DEFINE_BASIC_PRINT_TYPE_FUNC(u32, "%lx", unsigned long)
+DEFINE_BASIC_PRINT_TYPE_FUNC(u64, "%llx", unsigned long long)
+DEFINE_BASIC_PRINT_TYPE_FUNC(s8, "%d", int)
+DEFINE_BASIC_PRINT_TYPE_FUNC(s16, "%d", int)
+DEFINE_BASIC_PRINT_TYPE_FUNC(s32, "%ld", long)
+DEFINE_BASIC_PRINT_TYPE_FUNC(s64, "%lld", long long)
+
+/* Data fetch function type */
+typedef	void (*fetch_func_t)(struct pt_regs *, void *, void *);
+
+struct fetch_param {
+	fetch_func_t	fn;
 	void *data;
 };
 
-static __kprobes unsigned long call_fetch(struct fetch_func *f,
-					  struct pt_regs *regs)
+static __kprobes void call_fetch(struct fetch_param *fprm,
+				 struct pt_regs *regs, void *dest)
 {
-	return f->func(regs, f->data);
+	return fprm->fn(regs, fprm->data, dest);
 }
 
-/* fetch handlers */
-static __kprobes unsigned long fetch_register(struct pt_regs *regs,
-					      void *offset)
-{
-	return regs_get_register(regs, (unsigned int)((unsigned long)offset));
+#define FETCH_FUNC_NAME(kind, type)	fetch_##kind##_##type
+/*
+ * Define macro for basic types - we don't need to define s* types, because
+ * we have to care only about bitwidth at recording time.
+ */
+#define DEFINE_BASIC_FETCH_FUNCS(kind)  \
+DEFINE_FETCH_##kind(u8)			\
+DEFINE_FETCH_##kind(u16)		\
+DEFINE_FETCH_##kind(u32)		\
+DEFINE_FETCH_##kind(u64)
+
+#define CHECK_BASIC_FETCH_FUNCS(kind, fn)	\
+	((FETCH_FUNC_NAME(kind, u8) == fn) ||	\
+	 (FETCH_FUNC_NAME(kind, u16) == fn) ||	\
+	 (FETCH_FUNC_NAME(kind, u32) == fn) ||	\
+	 (FETCH_FUNC_NAME(kind, u64) == fn))
+
+/* Data fetch function templates */
+#define DEFINE_FETCH_reg(type)						\
+static __kprobes void FETCH_FUNC_NAME(reg, type)(struct pt_regs *regs,	\
+					  void *offset, void *dest)	\
+{									\
+	*(type *)dest = (type)regs_get_register(regs,			\
+				(unsigned int)((unsigned long)offset));	\
 }
-
-static __kprobes unsigned long fetch_stack(struct pt_regs *regs,
-					   void *num)
-{
-	return regs_get_kernel_stack_nth(regs,
-					 (unsigned int)((unsigned long)num));
+DEFINE_BASIC_FETCH_FUNCS(reg)
+
+#define DEFINE_FETCH_stack(type)					\
+static __kprobes void FETCH_FUNC_NAME(stack, type)(struct pt_regs *regs,\
+					  void *offset, void *dest)	\
+{									\
+	*(type *)dest = (type)regs_get_kernel_stack_nth(regs,		\
+				(unsigned int)((unsigned long)offset));	\
 }
+DEFINE_BASIC_FETCH_FUNCS(stack)
 
-static __kprobes unsigned long fetch_memory(struct pt_regs *regs, void *addr)
-{
-	unsigned long retval;
-
-	if (probe_kernel_address(addr, retval))
-		return 0;
-	return retval;
+#define DEFINE_FETCH_retval(type)					\
+static __kprobes void FETCH_FUNC_NAME(retval, type)(struct pt_regs *regs,\
+					  void *dummy, void *dest)	\
+{									\
+	*(type *)dest = (type)regs_return_value(regs);			\
 }
-
-static __kprobes unsigned long fetch_retvalue(struct pt_regs *regs,
-					      void *dummy)
-{
-	return regs_return_value(regs);
-}
-
-static __kprobes unsigned long fetch_stack_address(struct pt_regs *regs,
-						   void *dummy)
-{
-	return kernel_stack_pointer(regs);
+DEFINE_BASIC_FETCH_FUNCS(retval)
+
+#define DEFINE_FETCH_memory(type)					\
+static __kprobes void FETCH_FUNC_NAME(memory, type)(struct pt_regs *regs,\
+					  void *addr, void *dest)	\
+{									\
+	type retval;							\
+	if (probe_kernel_address(addr, retval))				\
+		*(type *)dest = 0;					\
+	else								\
+		*(type *)dest = retval;					\
 }
+DEFINE_BASIC_FETCH_FUNCS(memory)
 
 /* Memory fetching by symbol */
 struct symbol_cache {
@@ -145,51 +192,126 @@ static struct symbol_cache *alloc_symbol_cache(const char *sym, long offset)
 	return sc;
 }
 
-static __kprobes unsigned long fetch_symbol(struct pt_regs *regs, void *data)
-{
-	struct symbol_cache *sc = data;
-
-	if (sc->addr)
-		return fetch_memory(regs, (void *)sc->addr);
-	else
-		return 0;
+#define DEFINE_FETCH_symbol(type)					\
+static __kprobes void FETCH_FUNC_NAME(symbol, type)(struct pt_regs *regs,\
+					  void *data, void *dest)	\
+{									\
+	struct symbol_cache *sc = data;					\
+	if (sc->addr)							\
+		fetch_memory_##type(regs, (void *)sc->addr, dest);	\
+	else								\
+		*(type *)dest = 0;					\
 }
+DEFINE_BASIC_FETCH_FUNCS(symbol)
 
-/* Special indirect memory access interface */
-struct indirect_fetch_data {
-	struct fetch_func orig;
+/* Dereference memory access function */
+struct deref_fetch_param {
+	struct fetch_param orig;
 	long offset;
 };
 
-static __kprobes unsigned long fetch_indirect(struct pt_regs *regs, void *data)
-{
-	struct indirect_fetch_data *ind = data;
-	unsigned long addr;
-
-	addr = call_fetch(&ind->orig, regs);
-	if (addr) {
-		addr += ind->offset;
-		return fetch_memory(regs, (void *)addr);
-	} else
-		return 0;
+#define DEFINE_FETCH_deref(type)					\
+static __kprobes void FETCH_FUNC_NAME(deref, type)(struct pt_regs *regs,\
+					    void *data, void *dest)	\
+{									\
+	struct deref_fetch_param *dprm = data;				\
+	unsigned long addr;						\
+	call_fetch(&dprm->orig, regs, &addr);				\
+	if (addr) {							\
+		addr += dprm->offset;					\
+		fetch_memory_##type(regs, (void *)addr, dest);		\
+	} else								\
+		*(type *)dest = 0;					\
 }
+DEFINE_BASIC_FETCH_FUNCS(deref)
 
-static __kprobes void free_indirect_fetch_data(struct indirect_fetch_data *data)
+static __kprobes void free_deref_fetch_param(struct deref_fetch_param *data)
 {
-	if (data->orig.func == fetch_indirect)
-		free_indirect_fetch_data(data->orig.data);
-	else if (data->orig.func == fetch_symbol)
+	if (CHECK_BASIC_FETCH_FUNCS(deref, data->orig.fn))
+		free_deref_fetch_param(data->orig.data);
+	else if (CHECK_BASIC_FETCH_FUNCS(symbol, data->orig.fn))
 		free_symbol_cache(data->orig.data);
 	kfree(data);
 }
 
+/* Default (unsigned long) fetch type */
+#define __DEFAULT_FETCH_TYPE(t) u##t
+#define _DEFAULT_FETCH_TYPE(t) __DEFAULT_FETCH_TYPE(t)
+#define DEFAULT_FETCH_TYPE _DEFAULT_FETCH_TYPE(BITS_PER_LONG)
+#define DEFAULT_FETCH_TYPE_STR __stringify(DEFAULT_FETCH_TYPE)
+
+#define ASSIGN_FETCH_FUNC(kind, type)	\
+	.kind = FETCH_FUNC_NAME(kind, type)
+
+#define ASSIGN_FETCH_TYPE(ptype, ftype, sign)	\
+	{.name = #ptype,			\
+	 .size = sizeof(ftype),			\
+	 .is_signed = sign,			\
+	 .print = PRINT_TYPE_FUNC_NAME(ptype),	\
+	 .fmt = PRINT_TYPE_FMT_NAME(ptype),	\
+ASSIGN_FETCH_FUNC(reg, ftype),			\
+ASSIGN_FETCH_FUNC(stack, ftype),		\
+ASSIGN_FETCH_FUNC(retval, ftype),		\
+ASSIGN_FETCH_FUNC(memory, ftype),		\
+ASSIGN_FETCH_FUNC(symbol, ftype),		\
+ASSIGN_FETCH_FUNC(deref, ftype),		\
+	}
+
+/* Fetch type information table */
+static const struct fetch_type {
+	const char	*name;		/* Name of type */
+	size_t		size;		/* Byte size of type */
+	int		is_signed;	/* Signed flag */
+	print_type_func_t	print;	/* Print functions */
+	const char	*fmt;		/* Fromat string */
+	/* Fetch functions */
+	fetch_func_t	reg;
+	fetch_func_t	stack;
+	fetch_func_t	retval;
+	fetch_func_t	memory;
+	fetch_func_t	symbol;
+	fetch_func_t	deref;
+} fetch_type_table[] = {
+	ASSIGN_FETCH_TYPE(u8,  u8,  0),
+	ASSIGN_FETCH_TYPE(u16, u16, 0),
+	ASSIGN_FETCH_TYPE(u32, u32, 0),
+	ASSIGN_FETCH_TYPE(u64, u64, 0),
+	ASSIGN_FETCH_TYPE(s8,  u8,  1),
+	ASSIGN_FETCH_TYPE(s16, u16, 1),
+	ASSIGN_FETCH_TYPE(s32, u32, 1),
+	ASSIGN_FETCH_TYPE(s64, u64, 1),
+};
+
+static const struct fetch_type *find_fetch_type(const char *type)
+{
+	int i;
+
+	if (!type)
+		type = DEFAULT_FETCH_TYPE_STR;
+
+	for (i = 0; i < ARRAY_SIZE(fetch_type_table); i++)
+		if (strcmp(type, fetch_type_table[i].name) == 0)
+			return &fetch_type_table[i];
+	return NULL;
+}
+
+/* Special function : only accept unsigned long */
+static __kprobes void fetch_stack_address(struct pt_regs *regs,
+					  void *dummy, void *dest)
+{
+	*(unsigned long *)dest = kernel_stack_pointer(regs);
+}
+
 /**
  * Kprobe event core functions
  */
 
 struct probe_arg {
-	struct fetch_func	fetch;
-	const char		*name;
+	struct fetch_param	fetch;
+	unsigned int		offset;	/* Offset from argument entry */
+	const char		*name;	/* Name of this argument */
+	const char		*comm;	/* Command of this argument */
+	const struct fetch_type	*type;	/* Type of this argument */
 };
 
 /* Flags for trace_probe */
@@ -204,6 +326,7 @@ struct trace_probe {
 	const char		*symbol;	/* symbol name */
 	struct ftrace_event_call	call;
 	struct trace_event		event;
+	ssize_t			size;		/* trace entry size */
 	unsigned int		nr_args;
 	struct probe_arg	args[];
 };
@@ -212,6 +335,7 @@ struct trace_probe {
 	(offsetof(struct trace_probe, args) +	\
 	(sizeof(struct probe_arg) * (n)))
 
+
 static __kprobes int probe_is_return(struct trace_probe *tp)
 {
 	return tp->rp.handler != NULL;
@@ -222,49 +346,6 @@ static __kprobes const char *probe_symbol(struct trace_probe *tp)
 	return tp->symbol ? tp->symbol : "unknown";
 }
 
-static int probe_arg_string(char *buf, size_t n, struct fetch_func *ff)
-{
-	int ret = -EINVAL;
-
-	if (ff->func == fetch_register) {
-		const char *name;
-		name = regs_query_register_name((unsigned int)((long)ff->data));
-		ret = snprintf(buf, n, "%%%s", name);
-	} else if (ff->func == fetch_stack)
-		ret = snprintf(buf, n, "$stack%lu", (unsigned long)ff->data);
-	else if (ff->func == fetch_memory)
-		ret = snprintf(buf, n, "@0x%p", ff->data);
-	else if (ff->func == fetch_symbol) {
-		struct symbol_cache *sc = ff->data;
-		if (sc->offset)
-			ret = snprintf(buf, n, "@%s%+ld", sc->symbol,
-					sc->offset);
-		else
-			ret = snprintf(buf, n, "@%s", sc->symbol);
-	} else if (ff->func == fetch_retvalue)
-		ret = snprintf(buf, n, "$retval");
-	else if (ff->func == fetch_stack_address)
-		ret = snprintf(buf, n, "$stack");
-	else if (ff->func == fetch_indirect) {
-		struct indirect_fetch_data *id = ff->data;
-		size_t l = 0;
-		ret = snprintf(buf, n, "%+ld(", id->offset);
-		if (ret >= n)
-			goto end;
-		l += ret;
-		ret = probe_arg_string(buf + l, n - l, &id->orig);
-		if (ret < 0)
-			goto end;
-		l += ret;
-		ret = snprintf(buf + l, n - l, ")");
-		ret += l;
-	}
-end:
-	if (ret >= n)
-		return -ENOSPC;
-	return ret;
-}
-
 static int register_probe_event(struct trace_probe *tp);
 static void unregister_probe_event(struct trace_probe *tp);
 
@@ -347,11 +428,12 @@ error:
 
 static void free_probe_arg(struct probe_arg *arg)
 {
-	if (arg->fetch.func == fetch_symbol)
+	if (CHECK_BASIC_FETCH_FUNCS(deref, arg->fetch.fn))
+		free_deref_fetch_param(arg->fetch.data);
+	else if (CHECK_BASIC_FETCH_FUNCS(symbol, arg->fetch.fn))
 		free_symbol_cache(arg->fetch.data);
-	else if (arg->fetch.func == fetch_indirect)
-		free_indirect_fetch_data(arg->fetch.data);
 	kfree(arg->name);
+	kfree(arg->comm);
 }
 
 static void free_trace_probe(struct trace_probe *tp)
@@ -457,28 +539,30 @@ static int split_symbol_offset(char *symbol, unsigned long *offset)
 #define PARAM_MAX_ARGS 16
 #define PARAM_MAX_STACK (THREAD_SIZE / sizeof(unsigned long))
 
-static int parse_probe_vars(char *arg, struct fetch_func *ff, int is_return)
+static int parse_probe_vars(char *arg, const struct fetch_type *t,
+			    struct fetch_param *f, int is_return)
 {
 	int ret = 0;
 	unsigned long param;
 
 	if (strcmp(arg, "retval") == 0) {
-		if (is_return) {
-			ff->func = fetch_retvalue;
-			ff->data = NULL;
-		} else
+		if (is_return)
+			f->fn = t->retval;
+		else
 			ret = -EINVAL;
 	} else if (strncmp(arg, "stack", 5) == 0) {
 		if (arg[5] == '\0') {
-			ff->func = fetch_stack_address;
-			ff->data = NULL;
+			if (strcmp(t->name, DEFAULT_FETCH_TYPE_STR) == 0)
+				f->fn = fetch_stack_address;
+			else
+				ret = -EINVAL;
 		} else if (isdigit(arg[5])) {
 			ret = strict_strtoul(arg + 5, 10, &param);
 			if (ret || param > PARAM_MAX_STACK)
 				ret = -EINVAL;
 			else {
-				ff->func = fetch_stack;
-				ff->data = (void *)param;
+				f->fn = t->stack;
+				f->data = (void *)param;
 			}
 		} else
 			ret = -EINVAL;
@@ -488,7 +572,8 @@ static int parse_probe_vars(char *arg, struct fetch_func *ff, int is_return)
 }
 
 /* Recursive argument parser */
-static int __parse_probe_arg(char *arg, struct fetch_func *ff, int is_return)
+static int __parse_probe_arg(char *arg, const struct fetch_type *t,
+			     struct fetch_param *f, int is_return)
 {
 	int ret = 0;
 	unsigned long param;
@@ -497,13 +582,13 @@ static int __parse_probe_arg(char *arg, struct fetch_func *ff, int is_return)
 
 	switch (arg[0]) {
 	case '$':
-		ret = parse_probe_vars(arg + 1, ff, is_return);
+		ret = parse_probe_vars(arg + 1, t, f, is_return);
 		break;
 	case '%':	/* named register */
 		ret = regs_query_register_offset(arg + 1);
 		if (ret >= 0) {
-			ff->func = fetch_register;
-			ff->data = (void *)(unsigned long)ret;
+			f->fn = t->reg;
+			f->data = (void *)(unsigned long)ret;
 			ret = 0;
 		}
 		break;
@@ -512,26 +597,22 @@ static int __parse_probe_arg(char *arg, struct fetch_func *ff, int is_return)
 			ret = strict_strtoul(arg + 1, 0, &param);
 			if (ret)
 				break;
-			ff->func = fetch_memory;
-			ff->data = (void *)param;
+			f->fn = t->memory;
+			f->data = (void *)param;
 		} else {
 			ret = split_symbol_offset(arg + 1, &offset);
 			if (ret)
 				break;
-			ff->data = alloc_symbol_cache(arg + 1, offset);
-			if (ff->data)
-				ff->func = fetch_symbol;
-			else
-				ret = -EINVAL;
+			f->data = alloc_symbol_cache(arg + 1, offset);
+			if (f->data)
+				f->fn = t->symbol;
 		}
 		break;
-	case '+':	/* indirect memory */
+	case '+':	/* deref memory */
 	case '-':
 		tmp = strchr(arg, '(');
-		if (!tmp) {
-			ret = -EINVAL;
+		if (!tmp)
 			break;
-		}
 		*tmp = '\0';
 		ret = strict_strtol(arg + 1, 0, &offset);
 		if (ret)
@@ -541,38 +622,58 @@ static int __parse_probe_arg(char *arg, struct fetch_func *ff, int is_return)
 		arg = tmp + 1;
 		tmp = strrchr(arg, ')');
 		if (tmp) {
-			struct indirect_fetch_data *id;
+			struct deref_fetch_param *dprm;
+			const struct fetch_type *t2 = find_fetch_type(NULL);
 			*tmp = '\0';
-			id = kzalloc(sizeof(struct indirect_fetch_data),
-				     GFP_KERNEL);
-			if (!id)
+			dprm = kzalloc(sizeof(struct deref_fetch_param),
+				       GFP_KERNEL);
+			if (!dprm)
 				return -ENOMEM;
-			id->offset = offset;
-			ret = __parse_probe_arg(arg, &id->orig, is_return);
+			dprm->offset = offset;
+			ret = __parse_probe_arg(arg, t2, &dprm->orig,
+						is_return);
 			if (ret)
-				kfree(id);
+				kfree(dprm);
 			else {
-				ff->func = fetch_indirect;
-				ff->data = (void *)id;
+				f->fn = t->deref;
+				f->data = (void *)dprm;
 			}
-		} else
-			ret = -EINVAL;
+		}
 		break;
-	default:
-		/* TODO: support custom handler */
-		ret = -EINVAL;
 	}
+	if (!ret && !f->fn)
+		ret = -EINVAL;
 	return ret;
 }
 
 /* String length checking wrapper */
-static int parse_probe_arg(char *arg, struct fetch_func *ff, int is_return)
+static int parse_probe_arg(char *arg, struct trace_probe *tp,
+			   struct probe_arg *parg, int is_return)
 {
+	const char *t;
+
 	if (strlen(arg) > MAX_ARGSTR_LEN) {
 		pr_info("Argument is too long.: %s\n",  arg);
 		return -ENOSPC;
 	}
-	return __parse_probe_arg(arg, ff, is_return);
+	parg->comm = kstrdup(arg, GFP_KERNEL);
+	if (!parg->comm) {
+		pr_info("Failed to allocate memory for command '%s'.\n", arg);
+		return -ENOMEM;
+	}
+	t = strchr(parg->comm, ':');
+	if (t) {
+		arg[t - parg->comm] = '\0';
+		t++;
+	}
+	parg->type = find_fetch_type(t);
+	if (!parg->type) {
+		pr_info("Unsupported type: %s\n", t);
+		return -EINVAL;
+	}
+	parg->offset = tp->size;
+	tp->size += parg->type->size;
+	return __parse_probe_arg(arg, parg->type, &parg->fetch, is_return);
 }
 
 /* Return 1 if name is reserved or already used by another argument */
@@ -602,15 +703,18 @@ static int create_trace_probe(int argc, char **argv)
 	 *  @ADDR	: fetch memory at ADDR (ADDR should be in kernel)
 	 *  @SYM[+|-offs] : fetch memory at SYM +|- offs (SYM is a data symbol)
 	 *  %REG	: fetch register REG
-	 * Indirect memory fetch:
+	 * Dereferencing memory fetch:
 	 *  +|-offs(ARG) : fetch memory at ARG +|- offs address.
 	 * Alias name of args:
 	 *  NAME=FETCHARG : set NAME as alias of FETCHARG.
+	 * Type of args:
+	 *  FETCHARG:TYPE : use TYPE instead of unsigned long.
 	 */
 	struct trace_probe *tp;
 	int i, ret = 0;
 	int is_return = 0, is_delete = 0;
-	char *symbol = NULL, *event = NULL, *arg = NULL, *group = NULL;
+	char *symbol = NULL, *event = NULL, *group = NULL;
+	char *arg, *tmp;
 	unsigned long offset = 0;
 	void *addr = NULL;
 	char buf[MAX_EVENT_NAME_LEN];
@@ -723,13 +827,6 @@ static int create_trace_probe(int argc, char **argv)
 		else
 			arg = argv[i];
 
-		if (conflict_field_name(argv[i], tp->args, i)) {
-			pr_info("Argument%d name '%s' conflicts with "
-				"another field.\n", i, argv[i]);
-			ret = -EINVAL;
-			goto error;
-		}
-
 		tp->args[i].name = kstrdup(argv[i], GFP_KERNEL);
 		if (!tp->args[i].name) {
 			pr_info("Failed to allocate argument%d name '%s'.\n",
@@ -737,9 +834,19 @@ static int create_trace_probe(int argc, char **argv)
 			ret = -ENOMEM;
 			goto error;
 		}
+		tmp = strchr(tp->args[i].name, ':');
+		if (tmp)
+			*tmp = '_';	/* convert : to _ */
+
+		if (conflict_field_name(tp->args[i].name, tp->args, i)) {
+			pr_info("Argument%d name '%s' conflicts with "
+				"another field.\n", i, argv[i]);
+			ret = -EINVAL;
+			goto error;
+		}
 
 		/* Parse fetch argument */
-		ret = parse_probe_arg(arg, &tp->args[i].fetch, is_return);
+		ret = parse_probe_arg(arg, tp, &tp->args[i], is_return);
 		if (ret) {
 			pr_info("Parse error at argument%d. (%d)\n", i, ret);
 			kfree(tp->args[i].name);
@@ -794,8 +901,7 @@ static void probes_seq_stop(struct seq_file *m, void *v)
 static int probes_seq_show(struct seq_file *m, void *v)
 {
 	struct trace_probe *tp = v;
-	int i, ret;
-	char buf[MAX_ARGSTR_LEN + 1];
+	int i;
 
 	seq_printf(m, "%c", probe_is_return(tp) ? 'r' : 'p');
 	seq_printf(m, ":%s/%s", tp->call.system, tp->call.name);
@@ -807,15 +913,10 @@ static int probes_seq_show(struct seq_file *m, void *v)
 	else
 		seq_printf(m, " %s", probe_symbol(tp));
 
-	for (i = 0; i < tp->nr_args; i++) {
-		ret = probe_arg_string(buf, MAX_ARGSTR_LEN, &tp->args[i].fetch);
-		if (ret < 0) {
-			pr_warning("Argument%d decoding error(%d).\n", i, ret);
-			return ret;
-		}
-		seq_printf(m, " %s=%s", tp->args[i].name, buf);
-	}
+	for (i = 0; i < tp->nr_args; i++)
+		seq_printf(m, " %s=%s", tp->args[i].name, tp->args[i].comm);
 	seq_printf(m, "\n");
+
 	return 0;
 }
 
@@ -945,9 +1046,10 @@ static const struct file_operations kprobe_profile_ops = {
 static __kprobes void kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs)
 {
 	struct trace_probe *tp = container_of(kp, struct trace_probe, rp.kp);
-	struct kprobe_trace_entry *entry;
+	struct kprobe_trace_entry_head *entry;
 	struct ring_buffer_event *event;
 	struct ring_buffer *buffer;
+	u8 *data;
 	int size, i, pc;
 	unsigned long irq_flags;
 	struct ftrace_event_call *call = &tp->call;
@@ -957,7 +1059,7 @@ static __kprobes void kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs)
 	local_save_flags(irq_flags);
 	pc = preempt_count();
 
-	size = SIZEOF_KPROBE_TRACE_ENTRY(tp->nr_args);
+	size = sizeof(*entry) + tp->size;
 
 	event = trace_current_buffer_lock_reserve(&buffer, call->id, size,
 						  irq_flags, pc);
@@ -965,10 +1067,10 @@ static __kprobes void kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs)
 		return;
 
 	entry = ring_buffer_event_data(event);
-	entry->nargs = tp->nr_args;
 	entry->ip = (unsigned long)kp->addr;
+	data = (u8 *)&entry[1];
 	for (i = 0; i < tp->nr_args; i++)
-		entry->args[i] = call_fetch(&tp->args[i].fetch, regs);
+		call_fetch(&tp->args[i].fetch, regs, data + tp->args[i].offset);
 
 	if (!filter_current_check_discard(buffer, call, entry, event))
 		trace_nowake_buffer_unlock_commit(buffer, event, irq_flags, pc);
@@ -979,9 +1081,10 @@ static __kprobes void kretprobe_trace_func(struct kretprobe_instance *ri,
 					  struct pt_regs *regs)
 {
 	struct trace_probe *tp = container_of(ri->rp, struct trace_probe, rp);
-	struct kretprobe_trace_entry *entry;
+	struct kretprobe_trace_entry_head *entry;
 	struct ring_buffer_event *event;
 	struct ring_buffer *buffer;
+	u8 *data;
 	int size, i, pc;
 	unsigned long irq_flags;
 	struct ftrace_event_call *call = &tp->call;
@@ -989,7 +1092,7 @@ static __kprobes void kretprobe_trace_func(struct kretprobe_instance *ri,
 	local_save_flags(irq_flags);
 	pc = preempt_count();
 
-	size = SIZEOF_KRETPROBE_TRACE_ENTRY(tp->nr_args);
+	size = sizeof(*entry) + tp->size;
 
 	event = trace_current_buffer_lock_reserve(&buffer, call->id, size,
 						  irq_flags, pc);
@@ -997,11 +1100,11 @@ static __kprobes void kretprobe_trace_func(struct kretprobe_instance *ri,
 		return;
 
 	entry = ring_buffer_event_data(event);
-	entry->nargs = tp->nr_args;
 	entry->func = (unsigned long)tp->rp.kp.addr;
 	entry->ret_ip = (unsigned long)ri->ret_addr;
+	data = (u8 *)&entry[1];
 	for (i = 0; i < tp->nr_args; i++)
-		entry->args[i] = call_fetch(&tp->args[i].fetch, regs);
+		call_fetch(&tp->args[i].fetch, regs, data + tp->args[i].offset);
 
 	if (!filter_current_check_discard(buffer, call, entry, event))
 		trace_nowake_buffer_unlock_commit(buffer, event, irq_flags, pc);
@@ -1011,13 +1114,14 @@ static __kprobes void kretprobe_trace_func(struct kretprobe_instance *ri,
 enum print_line_t
 print_kprobe_event(struct trace_iterator *iter, int flags)
 {
-	struct kprobe_trace_entry *field;
+	struct kprobe_trace_entry_head *field;
 	struct trace_seq *s = &iter->seq;
 	struct trace_event *event;
 	struct trace_probe *tp;
+	u8 *data;
 	int i;
 
-	field = (struct kprobe_trace_entry *)iter->ent;
+	field = (struct kprobe_trace_entry_head *)iter->ent;
 	event = ftrace_find_event(field->ent.type);
 	tp = container_of(event, struct trace_probe, event);
 
@@ -1030,9 +1134,10 @@ print_kprobe_event(struct trace_iterator *iter, int flags)
 	if (!trace_seq_puts(s, ")"))
 		goto partial;
 
-	for (i = 0; i < field->nargs; i++)
-		if (!trace_seq_printf(s, " %s=%lx",
-				      tp->args[i].name, field->args[i]))
+	data = (u8 *)&field[1];
+	for (i = 0; i < tp->nr_args; i++)
+		if (!tp->args[i].type->print(s, tp->args[i].name,
+					     data + tp->args[i].offset))
 			goto partial;
 
 	if (!trace_seq_puts(s, "\n"))
@@ -1046,13 +1151,14 @@ partial:
 enum print_line_t
 print_kretprobe_event(struct trace_iterator *iter, int flags)
 {
-	struct kretprobe_trace_entry *field;
+	struct kretprobe_trace_entry_head *field;
 	struct trace_seq *s = &iter->seq;
 	struct trace_event *event;
 	struct trace_probe *tp;
+	u8 *data;
 	int i;
 
-	field = (struct kretprobe_trace_entry *)iter->ent;
+	field = (struct kretprobe_trace_entry_head *)iter->ent;
 	event = ftrace_find_event(field->ent.type);
 	tp = container_of(event, struct trace_probe, event);
 
@@ -1071,9 +1177,10 @@ print_kretprobe_event(struct trace_iterator *iter, int flags)
 	if (!trace_seq_puts(s, ")"))
 		goto partial;
 
-	for (i = 0; i < field->nargs; i++)
-		if (!trace_seq_printf(s, " %s=%lx",
-				      tp->args[i].name, field->args[i]))
+	data = (u8 *)&field[1];
+	for (i = 0; i < tp->nr_args; i++)
+		if (!tp->args[i].type->print(s, tp->args[i].name,
+					     data + tp->args[i].offset))
 			goto partial;
 
 	if (!trace_seq_puts(s, "\n"))
@@ -1129,29 +1236,43 @@ static int probe_event_raw_init(struct ftrace_event_call *event_call)
 static int kprobe_event_define_fields(struct ftrace_event_call *event_call)
 {
 	int ret, i;
-	struct kprobe_trace_entry field;
+	struct kprobe_trace_entry_head field;
 	struct trace_probe *tp = (struct trace_probe *)event_call->data;
 
 	DEFINE_FIELD(unsigned long, ip, FIELD_STRING_IP, 0);
-	DEFINE_FIELD(int, nargs, FIELD_STRING_NARGS, 1);
 	/* Set argument names as fields */
-	for (i = 0; i < tp->nr_args; i++)
-		DEFINE_FIELD(unsigned long, args[i], tp->args[i].name, 0);
+	for (i = 0; i < tp->nr_args; i++) {
+		ret = trace_define_field(event_call, tp->args[i].type->name,
+					 tp->args[i].name,
+					 sizeof(field) + tp->args[i].offset,
+					 tp->args[i].type->size,
+					 tp->args[i].type->is_signed,
+					 FILTER_OTHER);
+		if (ret)
+			return ret;
+	}
 	return 0;
 }
 
 static int kretprobe_event_define_fields(struct ftrace_event_call *event_call)
 {
 	int ret, i;
-	struct kretprobe_trace_entry field;
+	struct kretprobe_trace_entry_head field;
 	struct trace_probe *tp = (struct trace_probe *)event_call->data;
 
 	DEFINE_FIELD(unsigned long, func, FIELD_STRING_FUNC, 0);
 	DEFINE_FIELD(unsigned long, ret_ip, FIELD_STRING_RETIP, 0);
-	DEFINE_FIELD(int, nargs, FIELD_STRING_NARGS, 1);
 	/* Set argument names as fields */
-	for (i = 0; i < tp->nr_args; i++)
-		DEFINE_FIELD(unsigned long, args[i], tp->args[i].name, 0);
+	for (i = 0; i < tp->nr_args; i++) {
+		ret = trace_define_field(event_call, tp->args[i].type->name,
+					 tp->args[i].name,
+					 sizeof(field) + tp->args[i].offset,
+					 tp->args[i].type->size,
+					 tp->args[i].type->is_signed,
+					 FILTER_OTHER);
+		if (ret)
+			return ret;
+	}
 	return 0;
 }
 
@@ -1176,8 +1297,8 @@ static int __set_print_fmt(struct trace_probe *tp, char *buf, int len)
 	pos += snprintf(buf + pos, LEN_OR_ZERO, "\"%s", fmt);
 
 	for (i = 0; i < tp->nr_args; i++) {
-		pos += snprintf(buf + pos, LEN_OR_ZERO, " %s=%%lx",
-				tp->args[i].name);
+		pos += snprintf(buf + pos, LEN_OR_ZERO, " %s=%s",
+				tp->args[i].name, tp->args[i].type->fmt);
 	}
 
 	pos += snprintf(buf + pos, LEN_OR_ZERO, "\", %s", arg);
@@ -1219,12 +1340,13 @@ static __kprobes void kprobe_perf_func(struct kprobe *kp,
 {
 	struct trace_probe *tp = container_of(kp, struct trace_probe, rp.kp);
 	struct ftrace_event_call *call = &tp->call;
-	struct kprobe_trace_entry *entry;
+	struct kprobe_trace_entry_head *entry;
+	u8 *data;
 	int size, __size, i;
 	unsigned long irq_flags;
 	int rctx;
 
-	__size = SIZEOF_KPROBE_TRACE_ENTRY(tp->nr_args);
+	__size = sizeof(*entry) + tp->size;
 	size = ALIGN(__size + sizeof(u32), sizeof(u64));
 	size -= sizeof(u32);
 	if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE,
@@ -1235,10 +1357,10 @@ static __kprobes void kprobe_perf_func(struct kprobe *kp,
 	if (!entry)
 		return;
 
-	entry->nargs = tp->nr_args;
 	entry->ip = (unsigned long)kp->addr;
+	data = (u8 *)&entry[1];
 	for (i = 0; i < tp->nr_args; i++)
-		entry->args[i] = call_fetch(&tp->args[i].fetch, regs);
+		call_fetch(&tp->args[i].fetch, regs, data + tp->args[i].offset);
 
 	perf_trace_buf_submit(entry, size, rctx, entry->ip, 1, irq_flags, regs);
 }
@@ -1249,12 +1371,13 @@ static __kprobes void kretprobe_perf_func(struct kretprobe_instance *ri,
 {
 	struct trace_probe *tp = container_of(ri->rp, struct trace_probe, rp);
 	struct ftrace_event_call *call = &tp->call;
-	struct kretprobe_trace_entry *entry;
+	struct kretprobe_trace_entry_head *entry;
+	u8 *data;
 	int size, __size, i;
 	unsigned long irq_flags;
 	int rctx;
 
-	__size = SIZEOF_KRETPROBE_TRACE_ENTRY(tp->nr_args);
+	__size = sizeof(*entry) + tp->size;
 	size = ALIGN(__size + sizeof(u32), sizeof(u64));
 	size -= sizeof(u32);
 	if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE,
@@ -1265,11 +1388,11 @@ static __kprobes void kretprobe_perf_func(struct kretprobe_instance *ri,
 	if (!entry)
 		return;
 
-	entry->nargs = tp->nr_args;
 	entry->func = (unsigned long)tp->rp.kp.addr;
 	entry->ret_ip = (unsigned long)ri->ret_addr;
+	data = (u8 *)&entry[1];
 	for (i = 0; i < tp->nr_args; i++)
-		entry->args[i] = call_fetch(&tp->args[i].fetch, regs);
+		call_fetch(&tp->args[i].fetch, regs, data + tp->args[i].offset);
 
 	perf_trace_buf_submit(entry, size, rctx, entry->ret_ip, 1,
 			       irq_flags, regs);

kernel/trace/trace_selftest.c

@@ -17,7 +17,6 @@ static inline int trace_valid_entry(struct trace_entry *entry)
 	case TRACE_BRANCH:
 	case TRACE_GRAPH_ENT:
 	case TRACE_GRAPH_RET:
-	case TRACE_HW_BRANCHES:
 	case TRACE_KSYM:
 		return 1;
 	}
@@ -755,62 +754,6 @@ trace_selftest_startup_branch(struct tracer *trace, struct trace_array *tr)
 }
 #endif /* CONFIG_BRANCH_TRACER */
 
-#ifdef CONFIG_HW_BRANCH_TRACER
-int
-trace_selftest_startup_hw_branches(struct tracer *trace,
-				   struct trace_array *tr)
-{
-	struct trace_iterator *iter;
-	struct tracer tracer;
-	unsigned long count;
-	int ret;
-
-	if (!trace->open) {
-		printk(KERN_CONT "missing open function...");
-		return -1;
-	}
-
-	ret = tracer_init(trace, tr);
-	if (ret) {
-		warn_failed_init_tracer(trace, ret);
-		return ret;
-	}
-
-	/*
-	 * The hw-branch tracer needs to collect the trace from the various
-	 * cpu trace buffers - before tracing is stopped.
-	 */
-	iter = kzalloc(sizeof(*iter), GFP_KERNEL);
-	if (!iter)
-		return -ENOMEM;
-
-	memcpy(&tracer, trace, sizeof(tracer));
-
-	iter->trace = &tracer;
-	iter->tr = tr;
-	iter->pos = -1;
-	mutex_init(&iter->mutex);
-
-	trace->open(iter);
-
-	mutex_destroy(&iter->mutex);
-	kfree(iter);
-
-	tracing_stop();
-
-	ret = trace_test_buffer(tr, &count);
-	trace->reset(tr);
-	tracing_start();
-
-	if (!ret && !count) {
-		printk(KERN_CONT "no entries found..");
-		ret = -1;
-	}
-
-	return ret;
-}
-#endif /* CONFIG_HW_BRANCH_TRACER */
-
 #ifdef CONFIG_KSYM_TRACER
 static int ksym_selftest_dummy;
 

mm/mlock.c

@@ -607,44 +607,3 @@ void user_shm_unlock(size_t size, struct user_struct *user)
 	spin_unlock(&shmlock_user_lock);
 	free_uid(user);
 }
-
-int account_locked_memory(struct mm_struct *mm, struct rlimit *rlim,
-			  size_t size)
-{
-	unsigned long lim, vm, pgsz;
-	int error = -ENOMEM;
-
-	pgsz = PAGE_ALIGN(size) >> PAGE_SHIFT;
-
-	down_write(&mm->mmap_sem);
-
-	lim = ACCESS_ONCE(rlim[RLIMIT_AS].rlim_cur) >> PAGE_SHIFT;
-	vm   = mm->total_vm + pgsz;
-	if (lim < vm)
-		goto out;
-
-	lim = ACCESS_ONCE(rlim[RLIMIT_MEMLOCK].rlim_cur) >> PAGE_SHIFT;
-	vm   = mm->locked_vm + pgsz;
-	if (lim < vm)
-		goto out;
-
-	mm->total_vm  += pgsz;
-	mm->locked_vm += pgsz;
-
-	error = 0;
- out:
-	up_write(&mm->mmap_sem);
-	return error;
-}
-
-void refund_locked_memory(struct mm_struct *mm, size_t size)
-{
-	unsigned long pgsz = PAGE_ALIGN(size) >> PAGE_SHIFT;
-
-	down_write(&mm->mmap_sem);
-
-	mm->total_vm  -= pgsz;
-	mm->locked_vm -= pgsz;
-
-	up_write(&mm->mmap_sem);
-}

tools/perf/Documentation/perf-bench.txt

@@ -19,12 +19,12 @@ COMMON OPTIONS
 -f::
 --format=::
 Specify format style.
-Current available format styles are,
+Current available format styles are:
 
 'default'::
 Default style. This is mainly for human reading.
 ---------------------
-% perf bench sched pipe                      # with no style specify
+% perf bench sched pipe                      # with no style specified
 (executing 1000000 pipe operations between two tasks)
         Total time:5.855 sec
                 5.855061 usecs/op
@@ -79,7 +79,7 @@ options (20 sender and receiver processes per group)
 
       Total time:0.308 sec
 
-% perf bench sched messaging -t -g 20        # be multi-thread,with 20 groups
+% perf bench sched messaging -t -g 20        # be multi-thread, with 20 groups
 (20 sender and receiver threads per group)
 (20 groups == 800 threads run)
 

tools/perf/Documentation/perf-kvm.txt

@@ -0,0 +1,67 @@
+perf-kvm(1)
+==============
+
+NAME
+----
+perf-kvm - Tool to trace/measure kvm guest os
+
+SYNOPSIS
+--------
+[verse]
+'perf kvm' [--host] [--guest] [--guestmount=<path>
+	[--guestkallsyms=<path> --guestmodules=<path> | --guestvmlinux=<path>]]
+	{top|record|report|diff|buildid-list}
+'perf kvm' [--host] [--guest] [--guestkallsyms=<path> --guestmodules=<path>
+	| --guestvmlinux=<path>] {top|record|report|diff|buildid-list}
+
+DESCRIPTION
+-----------
+There are a couple of variants of perf kvm:
+
+  'perf kvm [options] top <command>' to generates and displays
+  a performance counter profile of guest os in realtime
+  of an arbitrary workload.
+
+  'perf kvm record <command>' to record the performance couinter profile
+  of an arbitrary workload and save it into a perf data file. If both
+  --host and --guest are input, the perf data file name is perf.data.kvm.
+  If there is  no --host but --guest, the file name is perf.data.guest.
+  If there is no --guest but --host, the file name is perf.data.host.
+
+  'perf kvm report' to display the performance counter profile information
+  recorded via perf kvm record.
+
+  'perf kvm diff' to displays the performance difference amongst two perf.data
+  files captured via perf record.
+
+  'perf kvm buildid-list' to  display the buildids found in a perf data file,
+  so that other tools can be used to fetch packages with matching symbol tables
+  for use by perf report.
+
+OPTIONS
+-------
+--host=::
+        Collect host side perforamnce profile.
+--guest=::
+        Collect guest side perforamnce profile.
+--guestmount=<path>::
+	Guest os root file system mount directory. Users mounts guest os
+        root directories under <path> by a specific filesystem access method,
+	typically, sshfs. For example, start 2 guest os. The one's pid is 8888
+	and the other's is 9999.
+        #mkdir ~/guestmount; cd ~/guestmount
+        #sshfs -o allow_other,direct_io -p 5551 localhost:/ 8888/
+        #sshfs -o allow_other,direct_io -p 5552 localhost:/ 9999/
+        #perf kvm --host --guest --guestmount=~/guestmount top
+--guestkallsyms=<path>::
+        Guest os /proc/kallsyms file copy. 'perf' kvm' reads it to get guest
+	kernel symbols. Users copy it out from guest os.
+--guestmodules=<path>::
+	Guest os /proc/modules file copy. 'perf' kvm' reads it to get guest
+	kernel module information. Users copy it out from guest os.
+--guestvmlinux=<path>::
+	Guest os kernel vmlinux.
+
+SEE ALSO
+--------
+linkperf:perf-top[1] perf-record[1] perf-report[1] perf-diff[1] perf-buildid-list[1]

tools/perf/Documentation/perf-probe.txt

@@ -57,6 +57,11 @@ OPTIONS
 --force::
 	Forcibly add events with existing name.
 
+-n::
+--dry-run::
+	Dry run. With this option, --add and --del doesn't execute actual
+	adding and removal operations.
+
 PROBE SYNTAX
 ------------
 Probe points are defined by following syntax.
@@ -74,13 +79,22 @@ Probe points are defined by following syntax.
 'EVENT' specifies the name of new event, if omitted, it will be set the name of the probed function. Currently, event group name is set as 'probe'.
 'FUNC' specifies a probed function name, and it may have one of the following options; '+OFFS' is the offset from function entry address in bytes, ':RLN' is the relative-line number from function entry line, and '%return' means that it probes function return. And ';PTN' means lazy matching pattern (see LAZY MATCHING). Note that ';PTN' must be the end of the probe point definition.  In addition, '@SRC' specifies a source file which has that function.
 It is also possible to specify a probe point by the source line number or lazy matching by using 'SRC:ALN' or 'SRC;PTN' syntax, where 'SRC' is the source file path, ':ALN' is the line number and ';PTN' is the lazy matching pattern.
-'ARG' specifies the arguments of this probe point. You can use the name of local variable, or kprobe-tracer argument format (e.g. $retval, %ax, etc).
+'ARG' specifies the arguments of this probe point, (see PROBE ARGUMENT).
+
+PROBE ARGUMENT
+--------------
+Each probe argument follows below syntax.
+
+ [NAME=]LOCALVAR|$retval|%REG|@SYMBOL[:TYPE]
+
+'NAME' specifies the name of this argument (optional). You can use the name of local variable, local data structure member (e.g. var->field, var.field2), or kprobe-tracer argument format (e.g. $retval, %ax, etc). Note that the name of this argument will be set as the last member name if you specify a local data structure member (e.g. field2 for 'var->field1.field2'.)
+'TYPE' casts the type of this argument (optional). If omitted, perf probe automatically set the type based on debuginfo.
 
 LINE SYNTAX
 -----------
 Line range is descripted by following syntax.
 
- "FUNC[:RLN[+NUM|:RLN2]]|SRC:ALN[+NUM|:ALN2]"
+ "FUNC[:RLN[+NUM|-RLN2]]|SRC:ALN[+NUM|-ALN2]"
 
 FUNC specifies the function name of showing lines. 'RLN' is the start line
 number from function entry line, and 'RLN2' is the end line number. As same as

tools/perf/Documentation/perf-record.txt

@@ -58,7 +58,7 @@ OPTIONS
 
 -f::
 --force::
-	Overwrite existing data file.
+	Overwrite existing data file. (deprecated)
 
 -c::
 --count=::
@@ -101,7 +101,7 @@ OPTIONS
 
 -R::
 --raw-samples::
-Collect raw sample records from all opened counters (typically for tracepoint counters).
+Collect raw sample records from all opened counters (default for tracepoint counters).
 
 SEE ALSO
 --------

tools/perf/Documentation/perf-sched.txt

@@ -12,7 +12,7 @@ SYNOPSIS
 
 DESCRIPTION
 -----------
-There's four variants of perf sched:
+There are four variants of perf sched:
 
   'perf sched record <command>' to record the scheduling events
   of an arbitrary workload.
@@ -27,7 +27,7 @@ There's four variants of perf sched:
   via perf sched record. (this is done by starting up mockup threads
   that mimic the workload based on the events in the trace. These
   threads can then replay the timings (CPU runtime and sleep patterns)
-  of the workload as it occured when it was recorded - and can repeat
+  of the workload as it occurred when it was recorded - and can repeat
   it a number of times, measuring its performance.)
 
 OPTIONS

tools/perf/Makefile

@@ -1,3 +1,7 @@
+ifeq ("$(origin O)", "command line")
+	OUTPUT := $(O)/
+endif
+
 # The default target of this Makefile is...
 all::
 
@@ -150,10 +154,17 @@ all::
 # Define LDFLAGS=-static to build a static binary.
 #
 # Define EXTRA_CFLAGS=-m64 or EXTRA_CFLAGS=-m32 as appropriate for cross-builds.
+#
+# Define NO_DWARF if you do not want debug-info analysis feature at all.
 
-PERF-VERSION-FILE: .FORCE-PERF-VERSION-FILE
-	@$(SHELL_PATH) util/PERF-VERSION-GEN
--include PERF-VERSION-FILE
+$(shell sh -c 'mkdir -p $(OUTPUT)scripts/python/Perf-Trace-Util/' 2> /dev/null)
+$(shell sh -c 'mkdir -p $(OUTPUT)scripts/perl/Perf-Trace-Util/' 2> /dev/null)
+$(shell sh -c 'mkdir -p $(OUTPUT)util/scripting-engines/' 2> /dev/null)
+$(shell sh -c 'mkdir $(OUTPUT)bench' 2> /dev/null)
+
+$(OUTPUT)PERF-VERSION-FILE: .FORCE-PERF-VERSION-FILE
+	@$(SHELL_PATH) util/PERF-VERSION-GEN $(OUTPUT)
+-include $(OUTPUT)PERF-VERSION-FILE
 
 uname_S := $(shell sh -c 'uname -s 2>/dev/null || echo not')
 uname_M := $(shell sh -c 'uname -m 2>/dev/null || echo not')
@@ -308,7 +319,7 @@ PROGRAMS += $(EXTRA_PROGRAMS)
 #
 # Single 'perf' binary right now:
 #
-PROGRAMS += perf
+PROGRAMS += $(OUTPUT)perf
 
 # List built-in command $C whose implementation cmd_$C() is not in
 # builtin-$C.o but is linked in as part of some other command.
@@ -318,7 +329,7 @@ PROGRAMS += perf
 ALL_PROGRAMS = $(PROGRAMS) $(SCRIPTS)
 
 # what 'all' will build but not install in perfexecdir
-OTHER_PROGRAMS = perf$X
+OTHER_PROGRAMS = $(OUTPUT)perf$X
 
 # Set paths to tools early so that they can be used for version tests.
 ifndef SHELL_PATH
@@ -330,7 +341,7 @@ endif
 
 export PERL_PATH
 
-LIB_FILE=libperf.a
+LIB_FILE=$(OUTPUT)libperf.a
 
 LIB_H += ../../include/linux/perf_event.h
 LIB_H += ../../include/linux/rbtree.h
@@ -375,7 +386,6 @@ LIB_H += util/header.h
 LIB_H += util/help.h
 LIB_H += util/session.h
 LIB_H += util/strbuf.h
-LIB_H += util/string.h
 LIB_H += util/strlist.h
 LIB_H += util/svghelper.h
 LIB_H += util/run-command.h
@@ -391,77 +401,78 @@ LIB_H += util/probe-finder.h
 LIB_H += util/probe-event.h
 LIB_H += util/cpumap.h
 
-LIB_OBJS += util/abspath.o
-LIB_OBJS += util/alias.o
-LIB_OBJS += util/build-id.o
-LIB_OBJS += util/config.o
-LIB_OBJS += util/ctype.o
-LIB_OBJS += util/debugfs.o
-LIB_OBJS += util/environment.o
-LIB_OBJS += util/event.o
-LIB_OBJS += util/exec_cmd.o
-LIB_OBJS += util/help.o
-LIB_OBJS += util/levenshtein.o
-LIB_OBJS += util/parse-options.o
-LIB_OBJS += util/parse-events.o
-LIB_OBJS += util/path.o
-LIB_OBJS += util/rbtree.o
-LIB_OBJS += util/bitmap.o
-LIB_OBJS += util/hweight.o
-LIB_OBJS += util/find_next_bit.o
-LIB_OBJS += util/run-command.o
-LIB_OBJS += util/quote.o
-LIB_OBJS += util/strbuf.o
-LIB_OBJS += util/string.o
-LIB_OBJS += util/strlist.o
-LIB_OBJS += util/usage.o
-LIB_OBJS += util/wrapper.o
-LIB_OBJS += util/sigchain.o
-LIB_OBJS += util/symbol.o
-LIB_OBJS += util/color.o
-LIB_OBJS += util/pager.o
-LIB_OBJS += util/header.o
-LIB_OBJS += util/callchain.o
-LIB_OBJS += util/values.o
-LIB_OBJS += util/debug.o
-LIB_OBJS += util/map.o
-LIB_OBJS += util/session.o
-LIB_OBJS += util/thread.o
-LIB_OBJS += util/trace-event-parse.o
-LIB_OBJS += util/trace-event-read.o
-LIB_OBJS += util/trace-event-info.o
-LIB_OBJS += util/trace-event-scripting.o
-LIB_OBJS += util/svghelper.o
-LIB_OBJS += util/sort.o
-LIB_OBJS += util/hist.o
-LIB_OBJS += util/probe-event.o
-LIB_OBJS += util/util.o
-LIB_OBJS += util/cpumap.o
-
-BUILTIN_OBJS += builtin-annotate.o
-
-BUILTIN_OBJS += builtin-bench.o
+LIB_OBJS += $(OUTPUT)util/abspath.o
+LIB_OBJS += $(OUTPUT)util/alias.o
+LIB_OBJS += $(OUTPUT)util/build-id.o
+LIB_OBJS += $(OUTPUT)util/config.o
+LIB_OBJS += $(OUTPUT)util/ctype.o
+LIB_OBJS += $(OUTPUT)util/debugfs.o
+LIB_OBJS += $(OUTPUT)util/environment.o
+LIB_OBJS += $(OUTPUT)util/event.o
+LIB_OBJS += $(OUTPUT)util/exec_cmd.o
+LIB_OBJS += $(OUTPUT)util/help.o
+LIB_OBJS += $(OUTPUT)util/levenshtein.o
+LIB_OBJS += $(OUTPUT)util/parse-options.o
+LIB_OBJS += $(OUTPUT)util/parse-events.o
+LIB_OBJS += $(OUTPUT)util/path.o
+LIB_OBJS += $(OUTPUT)util/rbtree.o
+LIB_OBJS += $(OUTPUT)util/bitmap.o
+LIB_OBJS += $(OUTPUT)util/hweight.o
+LIB_OBJS += $(OUTPUT)util/find_next_bit.o
+LIB_OBJS += $(OUTPUT)util/run-command.o
+LIB_OBJS += $(OUTPUT)util/quote.o
+LIB_OBJS += $(OUTPUT)util/strbuf.o
+LIB_OBJS += $(OUTPUT)util/string.o
+LIB_OBJS += $(OUTPUT)util/strlist.o
+LIB_OBJS += $(OUTPUT)util/usage.o
+LIB_OBJS += $(OUTPUT)util/wrapper.o
+LIB_OBJS += $(OUTPUT)util/sigchain.o
+LIB_OBJS += $(OUTPUT)util/symbol.o
+LIB_OBJS += $(OUTPUT)util/color.o
+LIB_OBJS += $(OUTPUT)util/pager.o
+LIB_OBJS += $(OUTPUT)util/header.o
+LIB_OBJS += $(OUTPUT)util/callchain.o
+LIB_OBJS += $(OUTPUT)util/values.o
+LIB_OBJS += $(OUTPUT)util/debug.o
+LIB_OBJS += $(OUTPUT)util/map.o
+LIB_OBJS += $(OUTPUT)util/session.o
+LIB_OBJS += $(OUTPUT)util/thread.o
+LIB_OBJS += $(OUTPUT)util/trace-event-parse.o
+LIB_OBJS += $(OUTPUT)util/trace-event-read.o
+LIB_OBJS += $(OUTPUT)util/trace-event-info.o
+LIB_OBJS += $(OUTPUT)util/trace-event-scripting.o
+LIB_OBJS += $(OUTPUT)util/svghelper.o
+LIB_OBJS += $(OUTPUT)util/sort.o
+LIB_OBJS += $(OUTPUT)util/hist.o
+LIB_OBJS += $(OUTPUT)util/probe-event.o
+LIB_OBJS += $(OUTPUT)util/util.o
+LIB_OBJS += $(OUTPUT)util/cpumap.o
+
+BUILTIN_OBJS += $(OUTPUT)builtin-annotate.o
+
+BUILTIN_OBJS += $(OUTPUT)builtin-bench.o
 
 # Benchmark modules
-BUILTIN_OBJS += bench/sched-messaging.o
-BUILTIN_OBJS += bench/sched-pipe.o
-BUILTIN_OBJS += bench/mem-memcpy.o
-
-BUILTIN_OBJS += builtin-diff.o
-BUILTIN_OBJS += builtin-help.o
-BUILTIN_OBJS += builtin-sched.o
-BUILTIN_OBJS += builtin-buildid-list.o
-BUILTIN_OBJS += builtin-buildid-cache.o
-BUILTIN_OBJS += builtin-list.o
-BUILTIN_OBJS += builtin-record.o
-BUILTIN_OBJS += builtin-report.o
-BUILTIN_OBJS += builtin-stat.o
-BUILTIN_OBJS += builtin-timechart.o
-BUILTIN_OBJS += builtin-top.o
-BUILTIN_OBJS += builtin-trace.o
-BUILTIN_OBJS += builtin-probe.o
-BUILTIN_OBJS += builtin-kmem.o
-BUILTIN_OBJS += builtin-lock.o
+BUILTIN_OBJS += $(OUTPUT)bench/sched-messaging.o
+BUILTIN_OBJS += $(OUTPUT)bench/sched-pipe.o
+BUILTIN_OBJS += $(OUTPUT)bench/mem-memcpy.o
+
+BUILTIN_OBJS += $(OUTPUT)builtin-diff.o
+BUILTIN_OBJS += $(OUTPUT)builtin-help.o
+BUILTIN_OBJS += $(OUTPUT)builtin-sched.o
+BUILTIN_OBJS += $(OUTPUT)builtin-buildid-list.o
+BUILTIN_OBJS += $(OUTPUT)builtin-buildid-cache.o
+BUILTIN_OBJS += $(OUTPUT)builtin-list.o
+BUILTIN_OBJS += $(OUTPUT)builtin-record.o
+BUILTIN_OBJS += $(OUTPUT)builtin-report.o
+BUILTIN_OBJS += $(OUTPUT)builtin-stat.o
+BUILTIN_OBJS += $(OUTPUT)builtin-timechart.o
+BUILTIN_OBJS += $(OUTPUT)builtin-top.o
+BUILTIN_OBJS += $(OUTPUT)builtin-trace.o
+BUILTIN_OBJS += $(OUTPUT)builtin-probe.o
+BUILTIN_OBJS += $(OUTPUT)builtin-kmem.o
+BUILTIN_OBJS += $(OUTPUT)builtin-lock.o
+BUILTIN_OBJS += $(OUTPUT)builtin-kvm.o
 
 PERFLIBS = $(LIB_FILE)
 
@@ -492,6 +503,10 @@ ifeq ($(uname_S),Darwin)
 	PTHREAD_LIBS =
 endif
 
+ifneq ($(OUTPUT),)
+	BASIC_CFLAGS += -I$(OUTPUT)
+endif
+
 ifeq ($(shell sh -c "(echo '\#include <libelf.h>'; echo 'int main(void) { Elf * elf = elf_begin(0, ELF_C_READ, 0); return (long)elf; }') | $(CC) -x c - $(ALL_CFLAGS) -o $(BITBUCKET) $(ALL_LDFLAGS) $(EXTLIBS) "$(QUIET_STDERR)" && echo y"), y)
 ifneq ($(shell sh -c "(echo '\#include <gnu/libc-version.h>'; echo 'int main(void) { const char * version = gnu_get_libc_version(); return (long)version; }') | $(CC) -x c - $(ALL_CFLAGS) -o $(BITBUCKET) $(ALL_LDFLAGS) $(EXTLIBS) "$(QUIET_STDERR)" && echo y"), y)
 	msg := $(error No gnu/libc-version.h found, please install glibc-dev[el]/glibc-static);
@@ -506,11 +521,20 @@ endif
 
 ifneq ($(shell sh -c "(echo '\#include <dwarf.h>'; echo '\#include <libdw.h>'; echo 'int main(void) { Dwarf *dbg; dbg = dwarf_begin(0, DWARF_C_READ); return (long)dbg; }') | $(CC) -x c - $(ALL_CFLAGS) -I/usr/include/elfutils -ldw -lelf -o $(BITBUCKET) $(ALL_LDFLAGS) $(EXTLIBS) "$(QUIET_STDERR)" && echo y"), y)
 	msg := $(warning No libdw.h found or old libdw.h found, disables dwarf support. Please install elfutils-devel/elfutils-dev);
-	BASIC_CFLAGS += -DNO_DWARF_SUPPORT
 else
-	BASIC_CFLAGS += -I/usr/include/elfutils
+ifndef NO_DWARF
+	BASIC_CFLAGS += -I/usr/include/elfutils -DDWARF_SUPPORT
 	EXTLIBS += -lelf -ldw
-	LIB_OBJS += util/probe-finder.o
+	LIB_OBJS += $(OUTPUT)util/probe-finder.o
+endif
+endif
+
+ifneq ($(shell sh -c "(echo '\#include <newt.h>'; echo 'int main(void) { newtInit(); newtCls(); return newtFinished(); }') | $(CC) -x c - $(ALL_CFLAGS) -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -lnewt -o $(BITBUCKET) $(ALL_LDFLAGS) $(EXTLIBS) "$(QUIET_STDERR)" && echo y"), y)
+	msg := $(warning newt not found, disables TUI support. Please install newt-devel or libnewt-dev);
+	BASIC_CFLAGS += -DNO_NEWT_SUPPORT
+else
+	EXTLIBS += -lnewt
+	LIB_OBJS += $(OUTPUT)util/newt.o
 endif
 
 ifndef NO_LIBPERL
@@ -522,8 +546,8 @@ ifneq ($(shell sh -c "(echo '\#include <EXTERN.h>'; echo '\#include <perl.h>'; e
 	BASIC_CFLAGS += -DNO_LIBPERL
 else
 	ALL_LDFLAGS += $(PERL_EMBED_LDOPTS)
-	LIB_OBJS += util/scripting-engines/trace-event-perl.o
-	LIB_OBJS += scripts/perl/Perf-Trace-Util/Context.o
+	LIB_OBJS += $(OUTPUT)util/scripting-engines/trace-event-perl.o
+	LIB_OBJS += $(OUTPUT)scripts/perl/Perf-Trace-Util/Context.o
 endif
 
 ifndef NO_LIBPYTHON
@@ -531,12 +555,12 @@ PYTHON_EMBED_LDOPTS = `python-config --ldflags 2>/dev/null`
 PYTHON_EMBED_CCOPTS = `python-config --cflags 2>/dev/null`
 endif
 
-ifneq ($(shell sh -c "(echo '\#include <Python.h>'; echo 'int main(void) { Py_Initialize(); return 0; }') | $(CC) -x c - $(PYTHON_EMBED_CCOPTS) -o /dev/null $(PYTHON_EMBED_LDOPTS) > /dev/null 2>&1 && echo y"), y)
+ifneq ($(shell sh -c "(echo '\#include <Python.h>'; echo 'int main(void) { Py_Initialize(); return 0; }') | $(CC) -x c - $(PYTHON_EMBED_CCOPTS) -o $(BITBUCKET) $(PYTHON_EMBED_LDOPTS) > /dev/null 2>&1 && echo y"), y)
 	BASIC_CFLAGS += -DNO_LIBPYTHON
 else
 	ALL_LDFLAGS += $(PYTHON_EMBED_LDOPTS)
-	LIB_OBJS += util/scripting-engines/trace-event-python.o
-	LIB_OBJS += scripts/python/Perf-Trace-Util/Context.o
+	LIB_OBJS += $(OUTPUT)util/scripting-engines/trace-event-python.o
+	LIB_OBJS += $(OUTPUT)scripts/python/Perf-Trace-Util/Context.o
 endif
 
 ifdef NO_DEMANGLE
@@ -607,53 +631,53 @@ ifdef NO_C99_FORMAT
 endif
 ifdef SNPRINTF_RETURNS_BOGUS
 	COMPAT_CFLAGS += -DSNPRINTF_RETURNS_BOGUS
-	COMPAT_OBJS += compat/snprintf.o
+	COMPAT_OBJS += $(OUTPUT)compat/snprintf.o
 endif
 ifdef FREAD_READS_DIRECTORIES
 	COMPAT_CFLAGS += -DFREAD_READS_DIRECTORIES
-	COMPAT_OBJS += compat/fopen.o
+	COMPAT_OBJS += $(OUTPUT)compat/fopen.o
 endif
 ifdef NO_SYMLINK_HEAD
 	BASIC_CFLAGS += -DNO_SYMLINK_HEAD
 endif
 ifdef NO_STRCASESTR
 	COMPAT_CFLAGS += -DNO_STRCASESTR
-	COMPAT_OBJS += compat/strcasestr.o
+	COMPAT_OBJS += $(OUTPUT)compat/strcasestr.o
 endif
 ifdef NO_STRTOUMAX
 	COMPAT_CFLAGS += -DNO_STRTOUMAX
-	COMPAT_OBJS += compat/strtoumax.o
+	COMPAT_OBJS += $(OUTPUT)compat/strtoumax.o
 endif
 ifdef NO_STRTOULL
 	COMPAT_CFLAGS += -DNO_STRTOULL
 endif
 ifdef NO_SETENV
 	COMPAT_CFLAGS += -DNO_SETENV
-	COMPAT_OBJS += compat/setenv.o
+	COMPAT_OBJS += $(OUTPUT)compat/setenv.o
 endif
 ifdef NO_MKDTEMP
 	COMPAT_CFLAGS += -DNO_MKDTEMP
-	COMPAT_OBJS += compat/mkdtemp.o
+	COMPAT_OBJS += $(OUTPUT)compat/mkdtemp.o
 endif
 ifdef NO_UNSETENV
 	COMPAT_CFLAGS += -DNO_UNSETENV
-	COMPAT_OBJS += compat/unsetenv.o
+	COMPAT_OBJS += $(OUTPUT)compat/unsetenv.o
 endif
 ifdef NO_SYS_SELECT_H
 	BASIC_CFLAGS += -DNO_SYS_SELECT_H
 endif
 ifdef NO_MMAP
 	COMPAT_CFLAGS += -DNO_MMAP
-	COMPAT_OBJS += compat/mmap.o
+	COMPAT_OBJS += $(OUTPUT)compat/mmap.o
 else
 	ifdef USE_WIN32_MMAP
 		COMPAT_CFLAGS += -DUSE_WIN32_MMAP
-		COMPAT_OBJS += compat/win32mmap.o
+		COMPAT_OBJS += $(OUTPUT)compat/win32mmap.o
 	endif
 endif
 ifdef NO_PREAD
 	COMPAT_CFLAGS += -DNO_PREAD
-	COMPAT_OBJS += compat/pread.o
+	COMPAT_OBJS += $(OUTPUT)compat/pread.o
 endif
 ifdef NO_FAST_WORKING_DIRECTORY
 	BASIC_CFLAGS += -DNO_FAST_WORKING_DIRECTORY
@@ -675,10 +699,10 @@ else
 endif
 endif
 ifdef NO_INET_NTOP
-	LIB_OBJS += compat/inet_ntop.o
+	LIB_OBJS += $(OUTPUT)compat/inet_ntop.o
 endif
 ifdef NO_INET_PTON
-	LIB_OBJS += compat/inet_pton.o
+	LIB_OBJS += $(OUTPUT)compat/inet_pton.o
 endif
 
 ifdef NO_ICONV
@@ -695,15 +719,15 @@ endif
 
 ifdef PPC_SHA1
 	SHA1_HEADER = "ppc/sha1.h"
-	LIB_OBJS += ppc/sha1.o ppc/sha1ppc.o
+	LIB_OBJS += $(OUTPUT)ppc/sha1.o ppc/sha1ppc.o
 else
 ifdef ARM_SHA1
 	SHA1_HEADER = "arm/sha1.h"
-	LIB_OBJS += arm/sha1.o arm/sha1_arm.o
+	LIB_OBJS += $(OUTPUT)arm/sha1.o $(OUTPUT)arm/sha1_arm.o
 else
 ifdef MOZILLA_SHA1
 	SHA1_HEADER = "mozilla-sha1/sha1.h"
-	LIB_OBJS += mozilla-sha1/sha1.o
+	LIB_OBJS += $(OUTPUT)mozilla-sha1/sha1.o
 else
 	SHA1_HEADER = <openssl/sha.h>
 	EXTLIBS += $(LIB_4_CRYPTO)
@@ -715,15 +739,15 @@ ifdef NO_PERL_MAKEMAKER
 endif
 ifdef NO_HSTRERROR
 	COMPAT_CFLAGS += -DNO_HSTRERROR
-	COMPAT_OBJS += compat/hstrerror.o
+	COMPAT_OBJS += $(OUTPUT)compat/hstrerror.o
 endif
 ifdef NO_MEMMEM
 	COMPAT_CFLAGS += -DNO_MEMMEM
-	COMPAT_OBJS += compat/memmem.o
+	COMPAT_OBJS += $(OUTPUT)compat/memmem.o
 endif
 ifdef INTERNAL_QSORT
 	COMPAT_CFLAGS += -DINTERNAL_QSORT
-	COMPAT_OBJS += compat/qsort.o
+	COMPAT_OBJS += $(OUTPUT)compat/qsort.o
 endif
 ifdef RUNTIME_PREFIX
 	COMPAT_CFLAGS += -DRUNTIME_PREFIX
@@ -803,7 +827,7 @@ export TAR INSTALL DESTDIR SHELL_PATH
 
 SHELL = $(SHELL_PATH)
 
-all:: .perf.dev.null shell_compatibility_test $(ALL_PROGRAMS) $(BUILT_INS) $(OTHER_PROGRAMS) PERF-BUILD-OPTIONS
+all:: .perf.dev.null shell_compatibility_test $(ALL_PROGRAMS) $(BUILT_INS) $(OTHER_PROGRAMS) $(OUTPUT)PERF-BUILD-OPTIONS
 ifneq (,$X)
 	$(foreach p,$(patsubst %$X,%,$(filter %$X,$(ALL_PROGRAMS) $(BUILT_INS) perf$X)), test '$p' -ef '$p$X' || $(RM) '$p';)
 endif
@@ -815,39 +839,39 @@ please_set_SHELL_PATH_to_a_more_modern_shell:
 
 shell_compatibility_test: please_set_SHELL_PATH_to_a_more_modern_shell
 
-strip: $(PROGRAMS) perf$X
-	$(STRIP) $(STRIP_OPTS) $(PROGRAMS) perf$X
+strip: $(PROGRAMS) $(OUTPUT)perf$X
+	$(STRIP) $(STRIP_OPTS) $(PROGRAMS) $(OUTPUT)perf$X
 
-perf.o: perf.c common-cmds.h PERF-CFLAGS
+$(OUTPUT)perf.o: perf.c $(OUTPUT)common-cmds.h $(OUTPUT)PERF-CFLAGS
 	$(QUIET_CC)$(CC) -DPERF_VERSION='"$(PERF_VERSION)"' \
 		'-DPERF_HTML_PATH="$(htmldir_SQ)"' \
-		$(ALL_CFLAGS) -c $(filter %.c,$^)
+		$(ALL_CFLAGS) -c $(filter %.c,$^) -o $@
 
-perf$X: perf.o $(BUILTIN_OBJS) $(PERFLIBS)
-	$(QUIET_LINK)$(CC) $(ALL_CFLAGS) -o $@ perf.o \
+$(OUTPUT)perf$X: $(OUTPUT)perf.o $(BUILTIN_OBJS) $(PERFLIBS)
+	$(QUIET_LINK)$(CC) $(ALL_CFLAGS) -o $@ $(OUTPUT)perf.o \
 		$(BUILTIN_OBJS) $(ALL_LDFLAGS) $(LIBS)
 
-builtin-help.o: builtin-help.c common-cmds.h PERF-CFLAGS
-	$(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) \
+$(OUTPUT)builtin-help.o: builtin-help.c $(OUTPUT)common-cmds.h $(OUTPUT)PERF-CFLAGS
+	$(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) \
 		'-DPERF_HTML_PATH="$(htmldir_SQ)"' \
 		'-DPERF_MAN_PATH="$(mandir_SQ)"' \
 		'-DPERF_INFO_PATH="$(infodir_SQ)"' $<
 
-builtin-timechart.o: builtin-timechart.c common-cmds.h PERF-CFLAGS
-	$(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) \
+$(OUTPUT)builtin-timechart.o: builtin-timechart.c $(OUTPUT)common-cmds.h $(OUTPUT)PERF-CFLAGS
+	$(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) \
 		'-DPERF_HTML_PATH="$(htmldir_SQ)"' \
 		'-DPERF_MAN_PATH="$(mandir_SQ)"' \
 		'-DPERF_INFO_PATH="$(infodir_SQ)"' $<
 
-$(BUILT_INS): perf$X
+$(BUILT_INS): $(OUTPUT)perf$X
 	$(QUIET_BUILT_IN)$(RM) $@ && \
 	ln perf$X $@ 2>/dev/null || \
 	ln -s perf$X $@ 2>/dev/null || \
 	cp perf$X $@
 
-common-cmds.h: util/generate-cmdlist.sh command-list.txt
+$(OUTPUT)common-cmds.h: util/generate-cmdlist.sh command-list.txt
 
-common-cmds.h: $(wildcard Documentation/perf-*.txt)
+$(OUTPUT)common-cmds.h: $(wildcard Documentation/perf-*.txt)
 	$(QUIET_GEN). util/generate-cmdlist.sh > $@+ && mv $@+ $@
 
 $(patsubst %.sh,%,$(SCRIPT_SH)) : % : %.sh
@@ -859,7 +883,7 @@ $(patsubst %.sh,%,$(SCRIPT_SH)) : % : %.sh
 	    -e 's/@@NO_CURL@@/$(NO_CURL)/g' \
 	    $@.sh >$@+ && \
 	chmod +x $@+ && \
-	mv $@+ $@
+	mv $@+ $(OUTPUT)$@
 
 configure: configure.ac
 	$(QUIET_GEN)$(RM) $@ $<+ && \
@@ -869,60 +893,60 @@ configure: configure.ac
 	$(RM) $<+
 
 # These can record PERF_VERSION
-perf.o perf.spec \
+$(OUTPUT)perf.o perf.spec \
 	$(patsubst %.sh,%,$(SCRIPT_SH)) \
 	$(patsubst %.perl,%,$(SCRIPT_PERL)) \
-	: PERF-VERSION-FILE
+	: $(OUTPUT)PERF-VERSION-FILE
 
-%.o: %.c PERF-CFLAGS
-	$(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) $<
-%.s: %.c PERF-CFLAGS
+$(OUTPUT)%.o: %.c $(OUTPUT)PERF-CFLAGS
+	$(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $<
+$(OUTPUT)%.s: %.c $(OUTPUT)PERF-CFLAGS
 	$(QUIET_CC)$(CC) -S $(ALL_CFLAGS) $<
-%.o: %.S
-	$(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) $<
+$(OUTPUT)%.o: %.S
+	$(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $<
 
-util/exec_cmd.o: util/exec_cmd.c PERF-CFLAGS
-	$(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) \
+$(OUTPUT)util/exec_cmd.o: util/exec_cmd.c $(OUTPUT)PERF-CFLAGS
+	$(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) \
 		'-DPERF_EXEC_PATH="$(perfexecdir_SQ)"' \
 		'-DBINDIR="$(bindir_relative_SQ)"' \
 		'-DPREFIX="$(prefix_SQ)"' \
 		$<
 
-builtin-init-db.o: builtin-init-db.c PERF-CFLAGS
-	$(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) -DDEFAULT_PERF_TEMPLATE_DIR='"$(template_dir_SQ)"' $<
+$(OUTPUT)builtin-init-db.o: builtin-init-db.c $(OUTPUT)PERF-CFLAGS
+	$(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DDEFAULT_PERF_TEMPLATE_DIR='"$(template_dir_SQ)"' $<
 
-util/config.o: util/config.c PERF-CFLAGS
-	$(QUIET_CC)$(CC) -o $*.o -c $(ALL_CFLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $<
+$(OUTPUT)util/config.o: util/config.c $(OUTPUT)PERF-CFLAGS
+	$(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $<
 
-util/rbtree.o: ../../lib/rbtree.c PERF-CFLAGS
-	$(QUIET_CC)$(CC) -o util/rbtree.o -c $(ALL_CFLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $<
+$(OUTPUT)util/rbtree.o: ../../lib/rbtree.c $(OUTPUT)PERF-CFLAGS
+	$(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $<
 
 # some perf warning policies can't fit to lib/bitmap.c, eg: it warns about variable shadowing
 # from <string.h> that comes from kernel headers wrapping.
 KBITMAP_FLAGS=`echo $(ALL_CFLAGS) | sed s/-Wshadow// | sed s/-Wswitch-default// | sed s/-Wextra//`
 
-util/bitmap.o: ../../lib/bitmap.c PERF-CFLAGS
-	$(QUIET_CC)$(CC) -o util/bitmap.o -c $(KBITMAP_FLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $<
+$(OUTPUT)util/bitmap.o: ../../lib/bitmap.c $(OUTPUT)PERF-CFLAGS
+	$(QUIET_CC)$(CC) -o $@ -c $(KBITMAP_FLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $<
 
-util/hweight.o: ../../lib/hweight.c PERF-CFLAGS
-	$(QUIET_CC)$(CC) -o util/hweight.o -c $(ALL_CFLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $<
+$(OUTPUT)util/hweight.o: ../../lib/hweight.c $(OUTPUT)PERF-CFLAGS
+	$(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $<
 
-util/find_next_bit.o: ../../lib/find_next_bit.c PERF-CFLAGS
-	$(QUIET_CC)$(CC) -o util/find_next_bit.o -c $(ALL_CFLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $<
+$(OUTPUT)util/find_next_bit.o: ../../lib/find_next_bit.c $(OUTPUT)PERF-CFLAGS
+	$(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) -DETC_PERFCONFIG='"$(ETC_PERFCONFIG_SQ)"' $<
 
-util/scripting-engines/trace-event-perl.o: util/scripting-engines/trace-event-perl.c PERF-CFLAGS
-	$(QUIET_CC)$(CC) -o util/scripting-engines/trace-event-perl.o -c $(ALL_CFLAGS) $(PERL_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-shadow $<
+$(OUTPUT)util/scripting-engines/trace-event-perl.o: util/scripting-engines/trace-event-perl.c $(OUTPUT)PERF-CFLAGS
+	$(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $(PERL_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-shadow $<
 
-scripts/perl/Perf-Trace-Util/Context.o: scripts/perl/Perf-Trace-Util/Context.c PERF-CFLAGS
-	$(QUIET_CC)$(CC) -o scripts/perl/Perf-Trace-Util/Context.o -c $(ALL_CFLAGS) $(PERL_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-nested-externs $<
+$(OUTPUT)scripts/perl/Perf-Trace-Util/Context.o: scripts/perl/Perf-Trace-Util/Context.c $(OUTPUT)PERF-CFLAGS
+	$(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $(PERL_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-nested-externs $<
 
-util/scripting-engines/trace-event-python.o: util/scripting-engines/trace-event-python.c PERF-CFLAGS
-	$(QUIET_CC)$(CC) -o util/scripting-engines/trace-event-python.o -c $(ALL_CFLAGS) $(PYTHON_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-shadow $<
+$(OUTPUT)util/scripting-engines/trace-event-python.o: util/scripting-engines/trace-event-python.c $(OUTPUT)PERF-CFLAGS
+	$(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $(PYTHON_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-shadow $<
 
-scripts/python/Perf-Trace-Util/Context.o: scripts/python/Perf-Trace-Util/Context.c PERF-CFLAGS
-	$(QUIET_CC)$(CC) -o scripts/python/Perf-Trace-Util/Context.o -c $(ALL_CFLAGS) $(PYTHON_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-nested-externs $<
+$(OUTPUT)scripts/python/Perf-Trace-Util/Context.o: scripts/python/Perf-Trace-Util/Context.c $(OUTPUT)PERF-CFLAGS
+	$(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) $(PYTHON_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-nested-externs $<
 
-perf-%$X: %.o $(PERFLIBS)
+$(OUTPUT)perf-%$X: %.o $(PERFLIBS)
 	$(QUIET_LINK)$(CC) $(ALL_CFLAGS) -o $@ $(ALL_LDFLAGS) $(filter %.o,$^) $(LIBS)
 
 $(LIB_OBJS) $(BUILTIN_OBJS): $(LIB_H)
@@ -963,17 +987,17 @@ cscope:
 TRACK_CFLAGS = $(subst ','\'',$(ALL_CFLAGS)):\
              $(bindir_SQ):$(perfexecdir_SQ):$(template_dir_SQ):$(prefix_SQ)
 
-PERF-CFLAGS: .FORCE-PERF-CFLAGS
+$(OUTPUT)PERF-CFLAGS: .FORCE-PERF-CFLAGS
 	@FLAGS='$(TRACK_CFLAGS)'; \
-	    if test x"$$FLAGS" != x"`cat PERF-CFLAGS 2>/dev/null`" ; then \
+	    if test x"$$FLAGS" != x"`cat $(OUTPUT)PERF-CFLAGS 2>/dev/null`" ; then \
 		echo 1>&2 "    * new build flags or prefix"; \
-		echo "$$FLAGS" >PERF-CFLAGS; \
+		echo "$$FLAGS" >$(OUTPUT)PERF-CFLAGS; \
             fi
 
 # We need to apply sq twice, once to protect from the shell
-# that runs PERF-BUILD-OPTIONS, and then again to protect it
+# that runs $(OUTPUT)PERF-BUILD-OPTIONS, and then again to protect it
 # and the first level quoting from the shell that runs "echo".
-PERF-BUILD-OPTIONS: .FORCE-PERF-BUILD-OPTIONS
+$(OUTPUT)PERF-BUILD-OPTIONS: .FORCE-PERF-BUILD-OPTIONS
 	@echo SHELL_PATH=\''$(subst ','\'',$(SHELL_PATH_SQ))'\' >$@
 	@echo TAR=\''$(subst ','\'',$(subst ','\'',$(TAR)))'\' >>$@
 	@echo NO_CURL=\''$(subst ','\'',$(subst ','\'',$(NO_CURL)))'\' >>$@
@@ -994,7 +1018,7 @@ all:: $(TEST_PROGRAMS)
 
 export NO_SVN_TESTS
 
-check: common-cmds.h
+check: $(OUTPUT)common-cmds.h
 	if sparse; \
 	then \
 		for i in *.c */*.c; \
@@ -1028,10 +1052,10 @@ export perfexec_instdir
 
 install: all
 	$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(bindir_SQ)'
-	$(INSTALL) perf$X '$(DESTDIR_SQ)$(bindir_SQ)'
+	$(INSTALL) $(OUTPUT)perf$X '$(DESTDIR_SQ)$(bindir_SQ)'
 	$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/Perf-Trace-Util/lib/Perf/Trace'
 	$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/bin'
-	$(INSTALL) perf-archive -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)'
+	$(INSTALL) $(OUTPUT)perf-archive -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)'
 	$(INSTALL) scripts/perl/Perf-Trace-Util/lib/Perf/Trace/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/Perf-Trace-Util/lib/Perf/Trace'
 	$(INSTALL) scripts/perl/*.pl -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl'
 	$(INSTALL) scripts/perl/bin/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/bin'
@@ -1045,7 +1069,7 @@ ifdef BUILT_INS
 	$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)'
 	$(INSTALL) $(BUILT_INS) '$(DESTDIR_SQ)$(perfexec_instdir_SQ)'
 ifneq (,$X)
-	$(foreach p,$(patsubst %$X,%,$(filter %$X,$(ALL_PROGRAMS) $(BUILT_INS) perf$X)), $(RM) '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/$p';)
+	$(foreach p,$(patsubst %$X,%,$(filter %$X,$(ALL_PROGRAMS) $(BUILT_INS) $(OUTPUT)perf$X)), $(RM) '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/$p';)
 endif
 endif
 
@@ -1129,14 +1153,14 @@ clean:
 	$(RM) *.o */*.o */*/*.o */*/*/*.o $(LIB_FILE)
 	$(RM) $(ALL_PROGRAMS) $(BUILT_INS) perf$X
 	$(RM) $(TEST_PROGRAMS)
-	$(RM) *.spec *.pyc *.pyo */*.pyc */*.pyo common-cmds.h TAGS tags cscope*
+	$(RM) *.spec *.pyc *.pyo */*.pyc */*.pyo $(OUTPUT)common-cmds.h TAGS tags cscope*
 	$(RM) -r autom4te.cache
 	$(RM) config.log config.mak.autogen config.mak.append config.status config.cache
 	$(RM) -r $(PERF_TARNAME) .doc-tmp-dir
 	$(RM) $(PERF_TARNAME).tar.gz perf-core_$(PERF_VERSION)-*.tar.gz
 	$(RM) $(htmldocs).tar.gz $(manpages).tar.gz
 	$(MAKE) -C Documentation/ clean
-	$(RM) PERF-VERSION-FILE PERF-CFLAGS PERF-BUILD-OPTIONS
+	$(RM) $(OUTPUT)PERF-VERSION-FILE $(OUTPUT)PERF-CFLAGS $(OUTPUT)PERF-BUILD-OPTIONS
 
 .PHONY: all install clean strip
 .PHONY: shell_compatibility_test please_set_SHELL_PATH_to_a_more_modern_shell

tools/perf/bench/mem-memcpy.c

@@ -10,7 +10,6 @@
 #include "../perf.h"
 #include "../util/util.h"
 #include "../util/parse-options.h"
-#include "../util/string.h"
 #include "../util/header.h"
 #include "bench.h"
 
@@ -24,7 +23,7 @@
 
 static const char	*length_str	= "1MB";
 static const char	*routine	= "default";
-static int		use_clock	= 0;
+static bool		use_clock	= false;
 static int		clock_fd;
 
 static const struct option options[] = {

tools/perf/bench/sched-messaging.c

@@ -31,9 +31,9 @@
 
 #define DATASIZE 100
 
-static int use_pipes = 0;
+static bool use_pipes = false;
 static unsigned int loops = 100;
-static unsigned int thread_mode = 0;
+static bool thread_mode = false;
 static unsigned int num_groups = 10;
 
 struct sender_context {

tools/perf/bench/sched-pipe.c

@@ -93,7 +93,7 @@ int bench_sched_pipe(int argc, const char **argv,
 
 	switch (bench_format) {
 	case BENCH_FORMAT_DEFAULT:
-		printf("# Extecuted %d pipe operations between two tasks\n\n",
+		printf("# Executed %d pipe operations between two tasks\n\n",
 			loops);
 
 		result_usec = diff.tv_sec * 1000000;

tools/perf/builtin-annotate.c

@@ -14,7 +14,6 @@
 #include "util/cache.h"
 #include <linux/rbtree.h>
 #include "util/symbol.h"
-#include "util/string.h"
 
 #include "perf.h"
 #include "util/debug.h"
@@ -29,11 +28,11 @@
 
 static char		const *input_name = "perf.data";
 
-static int		force;
+static bool		force;
 
-static int		full_paths;
+static bool		full_paths;
 
-static int		print_line;
+static bool		print_line;
 
 struct sym_hist {
 	u64		sum;
@@ -69,13 +68,13 @@ static int sym__alloc_hist(struct symbol *self)
 static int annotate__hist_hit(struct hist_entry *he, u64 ip)
 {
 	unsigned int sym_size, offset;
-	struct symbol *sym = he->sym;
+	struct symbol *sym = he->ms.sym;
 	struct sym_priv *priv;
 	struct sym_hist *h;
 
 	he->count++;
 
-	if (!sym || !he->map)
+	if (!sym || !he->ms.map)
 		return 0;
 
 	priv = symbol__priv(sym);
@@ -85,7 +84,7 @@ static int annotate__hist_hit(struct hist_entry *he, u64 ip)
 	sym_size = sym->end - sym->start;
 	offset = ip - sym->start;
 
-	pr_debug3("%s: ip=%#Lx\n", __func__, he->map->unmap_ip(he->map, ip));
+	pr_debug3("%s: ip=%#Lx\n", __func__, he->ms.map->unmap_ip(he->ms.map, ip));
 
 	if (offset >= sym_size)
 		return 0;
@@ -94,8 +93,8 @@ static int annotate__hist_hit(struct hist_entry *he, u64 ip)
 	h->sum++;
 	h->ip[offset]++;
 
-	pr_debug3("%#Lx %s: count++ [ip: %#Lx, %#Lx] => %Ld\n", he->sym->start,
-		  he->sym->name, ip, ip - he->sym->start, h->ip[offset]);
+	pr_debug3("%#Lx %s: count++ [ip: %#Lx, %#Lx] => %Ld\n", he->ms.sym->start,
+		  he->ms.sym->name, ip, ip - he->ms.sym->start, h->ip[offset]);
 	return 0;
 }
 
@@ -187,7 +186,7 @@ static struct objdump_line *objdump__get_next_ip_line(struct list_head *head,
 static int parse_line(FILE *file, struct hist_entry *he,
 		      struct list_head *head)
 {
-	struct symbol *sym = he->sym;
+	struct symbol *sym = he->ms.sym;
 	struct objdump_line *objdump_line;
 	char *line = NULL, *tmp, *tmp2;
 	size_t line_len;
@@ -226,7 +225,7 @@ static int parse_line(FILE *file, struct hist_entry *he,
 	}
 
 	if (line_ip != -1) {
-		u64 start = map__rip_2objdump(he->map, sym->start);
+		u64 start = map__rip_2objdump(he->ms.map, sym->start);
 		offset = line_ip - start;
 	}
 
@@ -244,7 +243,7 @@ static int objdump_line__print(struct objdump_line *self,
 			       struct list_head *head,
 			       struct hist_entry *he, u64 len)
 {
-	struct symbol *sym = he->sym;
+	struct symbol *sym = he->ms.sym;
 	static const char *prev_line;
 	static const char *prev_color;
 
@@ -327,7 +326,7 @@ static void insert_source_line(struct sym_ext *sym_ext)
 
 static void free_source_line(struct hist_entry *he, int len)
 {
-	struct sym_priv *priv = symbol__priv(he->sym);
+	struct sym_priv *priv = symbol__priv(he->ms.sym);
 	struct sym_ext *sym_ext = priv->ext;
 	int i;
 
@@ -346,7 +345,7 @@ static void free_source_line(struct hist_entry *he, int len)
 static void
 get_source_line(struct hist_entry *he, int len, const char *filename)
 {
-	struct symbol *sym = he->sym;
+	struct symbol *sym = he->ms.sym;
 	u64 start;
 	int i;
 	char cmd[PATH_MAX * 2];
@@ -361,7 +360,7 @@ get_source_line(struct hist_entry *he, int len, const char *filename)
 	if (!priv->ext)
 		return;
 
-	start = he->map->unmap_ip(he->map, sym->start);
+	start = he->ms.map->unmap_ip(he->ms.map, sym->start);
 
 	for (i = 0; i < len; i++) {
 		char *path = NULL;
@@ -425,7 +424,7 @@ static void print_summary(const char *filename)
 
 static void hist_entry__print_hits(struct hist_entry *self)
 {
-	struct symbol *sym = self->sym;
+	struct symbol *sym = self->ms.sym;
 	struct sym_priv *priv = symbol__priv(sym);
 	struct sym_hist *h = priv->hist;
 	u64 len = sym->end - sym->start, offset;
@@ -439,9 +438,9 @@ static void hist_entry__print_hits(struct hist_entry *self)
 
 static void annotate_sym(struct hist_entry *he)
 {
-	struct map *map = he->map;
+	struct map *map = he->ms.map;
 	struct dso *dso = map->dso;
-	struct symbol *sym = he->sym;
+	struct symbol *sym = he->ms.sym;
 	const char *filename = dso->long_name, *d_filename;
 	u64 len;
 	char command[PATH_MAX*2];
@@ -452,6 +451,16 @@ static void annotate_sym(struct hist_entry *he)
 	if (!filename)
 		return;
 
+	if (dso->origin == DSO__ORIG_KERNEL) {
+		if (dso->annotate_warned)
+			return;
+		dso->annotate_warned = 1;
+		pr_err("Can't annotate %s: No vmlinux file was found in the "
+		       "path:\n", sym->name);
+		vmlinux_path__fprintf(stderr);
+		return;
+	}
+
 	pr_debug("%s: filename=%s, sym=%s, start=%#Lx, end=%#Lx\n", __func__,
 		 filename, sym->name, map->unmap_ip(map, sym->start),
 		 map->unmap_ip(map, sym->end));
@@ -516,17 +525,17 @@ static void perf_session__find_annotations(struct perf_session *self)
 		struct hist_entry *he = rb_entry(nd, struct hist_entry, rb_node);
 		struct sym_priv *priv;
 
-		if (he->sym == NULL)
+		if (he->ms.sym == NULL)
 			continue;
 
-		priv = symbol__priv(he->sym);
+		priv = symbol__priv(he->ms.sym);
 		if (priv->hist == NULL)
 			continue;
 
 		annotate_sym(he);
 		/*
 		 * Since we have a hist_entry per IP for the same symbol, free
-		 * he->sym->hist to signal we already processed this symbol.
+		 * he->ms.sym->hist to signal we already processed this symbol.
 		 */
 		free(priv->hist);
 		priv->hist = NULL;
@@ -562,7 +571,7 @@ static int __cmd_annotate(void)
 		perf_session__fprintf(session, stdout);
 
 	if (verbose > 2)
-		dsos__fprintf(stdout);
+		dsos__fprintf(&session->kerninfo_root, stdout);
 
 	perf_session__collapse_resort(&session->hists);
 	perf_session__output_resort(&session->hists, session->event_total[0]);
@@ -581,10 +590,12 @@ static const char * const annotate_usage[] = {
 static const struct option options[] = {
 	OPT_STRING('i', "input", &input_name, "file",
 		    "input file name"),
+	OPT_STRING('d', "dsos", &symbol_conf.dso_list_str, "dso[,dso...]",
+		   "only consider symbols in these dsos"),
 	OPT_STRING('s', "symbol", &sym_hist_filter, "symbol",
 		    "symbol to annotate"),
 	OPT_BOOLEAN('f', "force", &force, "don't complain, do it"),
-	OPT_BOOLEAN('v', "verbose", &verbose,
+	OPT_INCR('v', "verbose", &verbose,
 		    "be more verbose (show symbol address, etc)"),
 	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
 		    "dump raw trace in ASCII"),

tools/perf/builtin-buildid-cache.c

@@ -27,7 +27,7 @@ static const struct option buildid_cache_options[] = {
 		   "file list", "file(s) to add"),
 	OPT_STRING('r', "remove", &remove_name_list_str, "file list",
 		    "file(s) to remove"),
-	OPT_BOOLEAN('v', "verbose", &verbose, "be more verbose"),
+	OPT_INCR('v', "verbose", &verbose, "be more verbose"),
 	OPT_END()
 };
 

tools/perf/builtin-buildid-list.c

@@ -16,7 +16,7 @@
 #include "util/symbol.h"
 
 static char const *input_name = "perf.data";
-static int force;
+static bool force;
 static bool with_hits;
 
 static const char * const buildid_list_usage[] = {
@@ -29,7 +29,7 @@ static const struct option options[] = {
 	OPT_STRING('i', "input", &input_name, "file",
 		    "input file name"),
 	OPT_BOOLEAN('f', "force", &force, "don't complain, do it"),
-	OPT_BOOLEAN('v', "verbose", &verbose,
+	OPT_INCR('v', "verbose", &verbose,
 		    "be more verbose"),
 	OPT_END()
 };
@@ -46,7 +46,7 @@ static int __cmd_buildid_list(void)
 	if (with_hits)
 		perf_session__process_events(session, &build_id__mark_dso_hit_ops);
 
-	dsos__fprintf_buildid(stdout, with_hits);
+	dsos__fprintf_buildid(&session->kerninfo_root, stdout, with_hits);
 
 	perf_session__delete(session);
 	return err;

tools/perf/builtin-diff.c

@@ -19,7 +19,7 @@
 static char const *input_old = "perf.data.old",
 		  *input_new = "perf.data";
 static char	  diff__default_sort_order[] = "dso,symbol";
-static int  force;
+static bool  force;
 static bool show_displacement;
 
 static int perf_session__add_hist_entry(struct perf_session *self,
@@ -33,7 +33,7 @@ static int perf_session__add_hist_entry(struct perf_session *self,
 		return -ENOMEM;
 
 	if (hit)
-		he->count += count;
+		__perf_session__add_count(he, al, count);
 
 	return 0;
 }
@@ -188,7 +188,7 @@ static const char * const diff_usage[] = {
 };
 
 static const struct option options[] = {
-	OPT_BOOLEAN('v', "verbose", &verbose,
+	OPT_INCR('v', "verbose", &verbose,
 		    "be more verbose (show symbol address, etc)"),
 	OPT_BOOLEAN('m', "displacement", &show_displacement,
 		    "Show position displacement relative to baseline"),
@@ -225,6 +225,10 @@ int cmd_diff(int argc, const char **argv, const char *prefix __used)
 			input_new = argv[1];
 		} else
 			input_new = argv[0];
+	} else if (symbol_conf.default_guest_vmlinux_name ||
+		   symbol_conf.default_guest_kallsyms) {
+		input_old = "perf.data.host";
+		input_new = "perf.data.guest";
 	}
 
 	symbol_conf.exclude_other = false;

tools/perf/builtin-help.c

@@ -29,7 +29,7 @@ enum help_format {
 	HELP_FORMAT_WEB,
 };
 
-static int show_all = 0;
+static bool show_all = false;
 static enum help_format help_format = HELP_FORMAT_MAN;
 static struct option builtin_help_options[] = {
 	OPT_BOOLEAN('a', "all", &show_all, "print all available commands"),

tools/perf/builtin-kmem.c

@@ -351,6 +351,7 @@ static void __print_result(struct rb_root *root, struct perf_session *session,
 			   int n_lines, int is_caller)
 {
 	struct rb_node *next;
+	struct kernel_info *kerninfo;
 
 	printf("%.102s\n", graph_dotted_line);
 	printf(" %-34s |",  is_caller ? "Callsite": "Alloc Ptr");
@@ -359,23 +360,31 @@ static void __print_result(struct rb_root *root, struct perf_session *session,
 
 	next = rb_first(root);
 
+	kerninfo = kerninfo__findhost(&session->kerninfo_root);
+	if (!kerninfo) {
+		pr_err("__print_result: couldn't find kernel information\n");
+		return;
+	}
 	while (next && n_lines--) {
 		struct alloc_stat *data = rb_entry(next, struct alloc_stat,
 						   node);
 		struct symbol *sym = NULL;
+		struct map_groups *kmaps = &kerninfo->kmaps;
+		struct map *map;
 		char buf[BUFSIZ];
 		u64 addr;
 
 		if (is_caller) {
 			addr = data->call_site;
 			if (!raw_ip)
-				sym = map_groups__find_function(&session->kmaps, addr, NULL);
+				sym = map_groups__find_function(kmaps, addr,
+								&map, NULL);
 		} else
 			addr = data->ptr;
 
 		if (sym != NULL)
 			snprintf(buf, sizeof(buf), "%s+%Lx", sym->name,
-				 addr - sym->start);
+				 addr - map->unmap_ip(map, sym->start));
 		else
 			snprintf(buf, sizeof(buf), "%#Lx", addr);
 		printf(" %-34s |", buf);
@@ -488,6 +497,9 @@ static int __cmd_kmem(void)
 	if (session == NULL)
 		return -ENOMEM;
 
+	if (perf_session__create_kernel_maps(session) < 0)
+		goto out_delete;
+
 	if (!perf_session__has_traces(session, "kmem record"))
 		goto out_delete;
 

tools/perf/builtin-kvm.c

@@ -0,0 +1,144 @@
+#include "builtin.h"
+#include "perf.h"
+
+#include "util/util.h"
+#include "util/cache.h"
+#include "util/symbol.h"
+#include "util/thread.h"
+#include "util/header.h"
+#include "util/session.h"
+
+#include "util/parse-options.h"
+#include "util/trace-event.h"
+
+#include "util/debug.h"
+
+#include <sys/prctl.h>
+
+#include <semaphore.h>
+#include <pthread.h>
+#include <math.h>
+
+static char			*file_name;
+static char			name_buffer[256];
+
+int				perf_host = 1;
+int				perf_guest;
+
+static const char * const kvm_usage[] = {
+	"perf kvm [<options>] {top|record|report|diff|buildid-list}",
+	NULL
+};
+
+static const struct option kvm_options[] = {
+	OPT_STRING('i', "input", &file_name, "file",
+		   "Input file name"),
+	OPT_STRING('o', "output", &file_name, "file",
+		   "Output file name"),
+	OPT_BOOLEAN(0, "guest", &perf_guest,
+		    "Collect guest os data"),
+	OPT_BOOLEAN(0, "host", &perf_host,
+		    "Collect guest os data"),
+	OPT_STRING(0, "guestmount", &symbol_conf.guestmount, "directory",
+		   "guest mount directory under which every guest os"
+		   " instance has a subdir"),
+	OPT_STRING(0, "guestvmlinux", &symbol_conf.default_guest_vmlinux_name,
+		   "file", "file saving guest os vmlinux"),
+	OPT_STRING(0, "guestkallsyms", &symbol_conf.default_guest_kallsyms,
+		   "file", "file saving guest os /proc/kallsyms"),
+	OPT_STRING(0, "guestmodules", &symbol_conf.default_guest_modules,
+		   "file", "file saving guest os /proc/modules"),
+	OPT_END()
+};
+
+static int __cmd_record(int argc, const char **argv)
+{
+	int rec_argc, i = 0, j;
+	const char **rec_argv;
+
+	rec_argc = argc + 2;
+	rec_argv = calloc(rec_argc + 1, sizeof(char *));
+	rec_argv[i++] = strdup("record");
+	rec_argv[i++] = strdup("-o");
+	rec_argv[i++] = strdup(file_name);
+	for (j = 1; j < argc; j++, i++)
+		rec_argv[i] = argv[j];
+
+	BUG_ON(i != rec_argc);
+
+	return cmd_record(i, rec_argv, NULL);
+}
+
+static int __cmd_report(int argc, const char **argv)
+{
+	int rec_argc, i = 0, j;
+	const char **rec_argv;
+
+	rec_argc = argc + 2;
+	rec_argv = calloc(rec_argc + 1, sizeof(char *));
+	rec_argv[i++] = strdup("report");
+	rec_argv[i++] = strdup("-i");
+	rec_argv[i++] = strdup(file_name);
+	for (j = 1; j < argc; j++, i++)
+		rec_argv[i] = argv[j];
+
+	BUG_ON(i != rec_argc);
+
+	return cmd_report(i, rec_argv, NULL);
+}
+
+static int __cmd_buildid_list(int argc, const char **argv)
+{
+	int rec_argc, i = 0, j;
+	const char **rec_argv;
+
+	rec_argc = argc + 2;
+	rec_argv = calloc(rec_argc + 1, sizeof(char *));
+	rec_argv[i++] = strdup("buildid-list");
+	rec_argv[i++] = strdup("-i");
+	rec_argv[i++] = strdup(file_name);
+	for (j = 1; j < argc; j++, i++)
+		rec_argv[i] = argv[j];
+
+	BUG_ON(i != rec_argc);
+
+	return cmd_buildid_list(i, rec_argv, NULL);
+}
+
+int cmd_kvm(int argc, const char **argv, const char *prefix __used)
+{
+	perf_host = perf_guest = 0;
+
+	argc = parse_options(argc, argv, kvm_options, kvm_usage,
+			PARSE_OPT_STOP_AT_NON_OPTION);
+	if (!argc)
+		usage_with_options(kvm_usage, kvm_options);
+
+	if (!perf_host)
+		perf_guest = 1;
+
+	if (!file_name) {
+		if (perf_host && !perf_guest)
+			sprintf(name_buffer, "perf.data.host");
+		else if (!perf_host && perf_guest)
+			sprintf(name_buffer, "perf.data.guest");
+		else
+			sprintf(name_buffer, "perf.data.kvm");
+		file_name = name_buffer;
+	}
+
+	if (!strncmp(argv[0], "rec", 3))
+		return __cmd_record(argc, argv);
+	else if (!strncmp(argv[0], "rep", 3))
+		return __cmd_report(argc, argv);
+	else if (!strncmp(argv[0], "diff", 4))
+		return cmd_diff(argc, argv, NULL);
+	else if (!strncmp(argv[0], "top", 3))
+		return cmd_top(argc, argv, NULL);
+	else if (!strncmp(argv[0], "buildid-list", 12))
+		return __cmd_buildid_list(argc, argv);
+	else
+		usage_with_options(kvm_usage, kvm_options);
+
+	return 0;
+}

tools/perf/builtin-lock.c

@@ -744,7 +744,7 @@ static const char * const lock_usage[] = {
 
 static const struct option lock_options[] = {
 	OPT_STRING('i', "input", &input_name, "file", "input file name"),
-	OPT_BOOLEAN('v', "verbose", &verbose, "be more verbose (show symbol address, etc)"),
+	OPT_INCR('v', "verbose", &verbose, "be more verbose (show symbol address, etc)"),
 	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace, "dump raw trace in ASCII"),
 	OPT_END()
 };

tools/perf/builtin-probe.c

@@ -36,13 +36,10 @@
 #include "builtin.h"
 #include "util/util.h"
 #include "util/strlist.h"
-#include "util/event.h"
+#include "util/symbol.h"
 #include "util/debug.h"
 #include "util/debugfs.h"
-#include "util/symbol.h"
-#include "util/thread.h"
 #include "util/parse-options.h"
-#include "util/parse-events.h"	/* For debugfs_path */
 #include "util/probe-finder.h"
 #include "util/probe-event.h"
 
@@ -50,103 +47,83 @@
 
 /* Session management structure */
 static struct {
-	bool need_dwarf;
 	bool list_events;
 	bool force_add;
 	bool show_lines;
-	int nr_probe;
-	struct probe_point probes[MAX_PROBES];
+	int nevents;
+	struct perf_probe_event events[MAX_PROBES];
 	struct strlist *dellist;
-	struct map_groups kmap_groups;
-	struct map *kmaps[MAP__NR_TYPES];
 	struct line_range line_range;
-} session;
+} params;
 
 
 /* Parse an event definition. Note that any error must die. */
-static void parse_probe_event(const char *str)
+static int parse_probe_event(const char *str)
 {
-	struct probe_point *pp = &session.probes[session.nr_probe];
+	struct perf_probe_event *pev = &params.events[params.nevents];
+	int ret;
 
-	pr_debug("probe-definition(%d): %s\n", session.nr_probe, str);
-	if (++session.nr_probe == MAX_PROBES)
+	pr_debug("probe-definition(%d): %s\n", params.nevents, str);
+	if (++params.nevents == MAX_PROBES)
 		die("Too many probes (> %d) are specified.", MAX_PROBES);
 
-	/* Parse perf-probe event into probe_point */
-	parse_perf_probe_event(str, pp, &session.need_dwarf);
+	/* Parse a perf-probe command into event */
+	ret = parse_perf_probe_command(str, pev);
+	pr_debug("%d arguments\n", pev->nargs);
 
-	pr_debug("%d arguments\n", pp->nr_args);
+	return ret;
 }
 
-static void parse_probe_event_argv(int argc, const char **argv)
+static int parse_probe_event_argv(int argc, const char **argv)
 {
-	int i, len;
+	int i, len, ret;
 	char *buf;
 
 	/* Bind up rest arguments */
 	len = 0;
 	for (i = 0; i < argc; i++)
 		len += strlen(argv[i]) + 1;
-	buf = zalloc(len + 1);
-	if (!buf)
-		die("Failed to allocate memory for binding arguments.");
+	buf = xzalloc(len + 1);
 	len = 0;
 	for (i = 0; i < argc; i++)
 		len += sprintf(&buf[len], "%s ", argv[i]);
-	parse_probe_event(buf);
+	ret = parse_probe_event(buf);
 	free(buf);
+	return ret;
 }
 
 static int opt_add_probe_event(const struct option *opt __used,
 			      const char *str, int unset __used)
 {
 	if (str)
-		parse_probe_event(str);
-	return 0;
+		return parse_probe_event(str);
+	else
+		return 0;
 }
 
 static int opt_del_probe_event(const struct option *opt __used,
 			       const char *str, int unset __used)
 {
 	if (str) {
-		if (!session.dellist)
-			session.dellist = strlist__new(true, NULL);
-		strlist__add(session.dellist, str);
+		if (!params.dellist)
+			params.dellist = strlist__new(true, NULL);
+		strlist__add(params.dellist, str);
 	}
 	return 0;
 }
 
-/* Currently just checking function name from symbol map */
-static void evaluate_probe_point(struct probe_point *pp)
-{
-	struct symbol *sym;
-	sym = map__find_symbol_by_name(session.kmaps[MAP__FUNCTION],
-				       pp->function, NULL);
-	if (!sym)
-		die("Kernel symbol \'%s\' not found - probe not added.",
-		    pp->function);
-}
-
-#ifndef NO_DWARF_SUPPORT
-static int open_vmlinux(void)
-{
-	if (map__load(session.kmaps[MAP__FUNCTION], NULL) < 0) {
-		pr_debug("Failed to load kernel map.\n");
-		return -EINVAL;
-	}
-	pr_debug("Try to open %s\n",
-		 session.kmaps[MAP__FUNCTION]->dso->long_name);
-	return open(session.kmaps[MAP__FUNCTION]->dso->long_name, O_RDONLY);
-}
-
+#ifdef DWARF_SUPPORT
 static int opt_show_lines(const struct option *opt __used,
 			  const char *str, int unset __used)
 {
+	int ret = 0;
+
 	if (str)
-		parse_line_range_desc(str, &session.line_range);
-	INIT_LIST_HEAD(&session.line_range.line_list);
-	session.show_lines = true;
-	return 0;
+		ret = parse_line_range_desc(str, &params.line_range);
+	INIT_LIST_HEAD(&params.line_range.line_list);
+	params.show_lines = true;
+
+	return ret;
 }
 #endif
 
@@ -155,29 +132,25 @@ static const char * const probe_usage[] = {
 	"perf probe [<options>] --add 'PROBEDEF' [--add 'PROBEDEF' ...]",
 	"perf probe [<options>] --del '[GROUP:]EVENT' ...",
 	"perf probe --list",
-#ifndef NO_DWARF_SUPPORT
+#ifdef DWARF_SUPPORT
 	"perf probe --line 'LINEDESC'",
 #endif
 	NULL
 };
 
 static const struct option options[] = {
-	OPT_BOOLEAN('v', "verbose", &verbose,
+	OPT_INCR('v', "verbose", &verbose,
 		    "be more verbose (show parsed arguments, etc)"),
-#ifndef NO_DWARF_SUPPORT
-	OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name,
-		   "file", "vmlinux pathname"),
-#endif
-	OPT_BOOLEAN('l', "list", &session.list_events,
+	OPT_BOOLEAN('l', "list", &params.list_events,
 		    "list up current probe events"),
 	OPT_CALLBACK('d', "del", NULL, "[GROUP:]EVENT", "delete a probe event.",
 		opt_del_probe_event),
 	OPT_CALLBACK('a', "add", NULL,
-#ifdef NO_DWARF_SUPPORT
-		"[EVENT=]FUNC[+OFF|%return] [ARG ...]",
-#else
+#ifdef DWARF_SUPPORT
 		"[EVENT=]FUNC[@SRC][+OFF|%return|:RL|;PT]|SRC:AL|SRC;PT"
-		" [ARG ...]",
+		" [[NAME=]ARG ...]",
+#else
+		"[EVENT=]FUNC[+OFF|%return] [[NAME=]ARG ...]",
 #endif
 		"probe point definition, where\n"
 		"\t\tGROUP:\tGroup name (optional)\n"
@@ -185,51 +158,33 @@ static const struct option options[] = {
 		"\t\tFUNC:\tFunction name\n"
 		"\t\tOFF:\tOffset from function entry (in byte)\n"
 		"\t\t%return:\tPut the probe at function return\n"
-#ifdef NO_DWARF_SUPPORT
-		"\t\tARG:\tProbe argument (only \n"
-#else
+#ifdef DWARF_SUPPORT
 		"\t\tSRC:\tSource code path\n"
 		"\t\tRL:\tRelative line number from function entry.\n"
 		"\t\tAL:\tAbsolute line number in file.\n"
 		"\t\tPT:\tLazy expression of line code.\n"
 		"\t\tARG:\tProbe argument (local variable name or\n"
-#endif
 		"\t\t\tkprobe-tracer argument format.)\n",
+#else
+		"\t\tARG:\tProbe argument (kprobe-tracer argument format.)\n",
+#endif
 		opt_add_probe_event),
-	OPT_BOOLEAN('f', "force", &session.force_add, "forcibly add events"
+	OPT_BOOLEAN('f', "force", &params.force_add, "forcibly add events"
 		    " with existing name"),
-#ifndef NO_DWARF_SUPPORT
+#ifdef DWARF_SUPPORT
 	OPT_CALLBACK('L', "line", NULL,
-		     "FUNC[:RLN[+NUM|:RLN2]]|SRC:ALN[+NUM|:ALN2]",
+		     "FUNC[:RLN[+NUM|-RLN2]]|SRC:ALN[+NUM|-ALN2]",
 		     "Show source code lines.", opt_show_lines),
+	OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name,
+		   "file", "vmlinux pathname"),
 #endif
+	OPT__DRY_RUN(&probe_event_dry_run),
 	OPT_END()
 };
 
-/* Initialize symbol maps for vmlinux */
-static void init_vmlinux(void)
-{
-	symbol_conf.sort_by_name = true;
-	if (symbol_conf.vmlinux_name == NULL)
-		symbol_conf.try_vmlinux_path = true;
-	else
-		pr_debug("Use vmlinux: %s\n", symbol_conf.vmlinux_name);
-	if (symbol__init() < 0)
-		die("Failed to init symbol map.");
-
-	map_groups__init(&session.kmap_groups);
-	if (map_groups__create_kernel_maps(&session.kmap_groups,
-					   session.kmaps) < 0)
-		die("Failed to create kernel maps.");
-}
-
 int cmd_probe(int argc, const char **argv, const char *prefix __used)
 {
-	int i, ret;
-#ifndef NO_DWARF_SUPPORT
-	int fd;
-#endif
-	struct probe_point *pp;
+	int ret;
 
 	argc = parse_options(argc, argv, options, probe_usage,
 			     PARSE_OPT_STOP_AT_NON_OPTION);
@@ -238,123 +193,65 @@ int cmd_probe(int argc, const char **argv, const char *prefix __used)
 			pr_warning("  Error: '-' is not supported.\n");
 			usage_with_options(probe_usage, options);
 		}
-		parse_probe_event_argv(argc, argv);
+		ret = parse_probe_event_argv(argc, argv);
+		if (ret < 0) {
+			pr_err("  Error: Parse Error.  (%d)\n", ret);
+			return ret;
+		}
 	}
 
-	if ((!session.nr_probe && !session.dellist && !session.list_events &&
-	     !session.show_lines))
+	if ((!params.nevents && !params.dellist && !params.list_events &&
+	     !params.show_lines))
 		usage_with_options(probe_usage, options);
 
-	if (debugfs_valid_mountpoint(debugfs_path) < 0)
-		die("Failed to find debugfs path.");
-
-	if (session.list_events) {
-		if (session.nr_probe != 0 || session.dellist) {
-			pr_warning("  Error: Don't use --list with"
-				   " --add/--del.\n");
+	if (params.list_events) {
+		if (params.nevents != 0 || params.dellist) {
+			pr_err("  Error: Don't use --list with --add/--del.\n");
 			usage_with_options(probe_usage, options);
 		}
-		if (session.show_lines) {
-			pr_warning("  Error: Don't use --list with --line.\n");
+		if (params.show_lines) {
+			pr_err("  Error: Don't use --list with --line.\n");
 			usage_with_options(probe_usage, options);
 		}
-		show_perf_probe_events();
-		return 0;
+		ret = show_perf_probe_events();
+		if (ret < 0)
+			pr_err("  Error: Failed to show event list. (%d)\n",
+			       ret);
+		return ret;
 	}
 
-#ifndef NO_DWARF_SUPPORT
-	if (session.show_lines) {
-		if (session.nr_probe != 0 || session.dellist) {
+#ifdef DWARF_SUPPORT
+	if (params.show_lines) {
+		if (params.nevents != 0 || params.dellist) {
 			pr_warning("  Error: Don't use --line with"
 				   " --add/--del.\n");
 			usage_with_options(probe_usage, options);
 		}
-		init_vmlinux();
-		fd = open_vmlinux();
-		if (fd < 0)
-			die("Could not open debuginfo file.");
-		ret = find_line_range(fd, &session.line_range);
-		if (ret <= 0)
-			die("Source line is not found.\n");
-		close(fd);
-		show_line_range(&session.line_range);
-		return 0;
-	}
-#endif
 
-	if (session.dellist) {
-		del_trace_kprobe_events(session.dellist);
-		strlist__delete(session.dellist);
-		if (session.nr_probe == 0)
-			return 0;
+		ret = show_line_range(&params.line_range);
+		if (ret < 0)
+			pr_err("  Error: Failed to show lines. (%d)\n", ret);
+		return ret;
 	}
+#endif
 
-	/* Add probes */
-	init_vmlinux();
-
-	if (session.need_dwarf)
-#ifdef NO_DWARF_SUPPORT
-		die("Debuginfo-analysis is not supported");
-#else	/* !NO_DWARF_SUPPORT */
-		pr_debug("Some probes require debuginfo.\n");
-
-	fd = open_vmlinux();
-	if (fd < 0) {
-		if (session.need_dwarf)
-			die("Could not open debuginfo file.");
-
-		pr_debug("Could not open vmlinux/module file."
-			 " Try to use symbols.\n");
-		goto end_dwarf;
-	}
-
-	/* Searching probe points */
-	for (i = 0; i < session.nr_probe; i++) {
-		pp = &session.probes[i];
-		if (pp->found)
-			continue;
-
-		lseek(fd, SEEK_SET, 0);
-		ret = find_probe_point(fd, pp);
-		if (ret > 0)
-			continue;
-		if (ret == 0) {	/* No error but failed to find probe point. */
-			synthesize_perf_probe_point(pp);
-			die("Probe point '%s' not found. - probe not added.",
-			    pp->probes[0]);
-		}
-		/* Error path */
-		if (session.need_dwarf) {
-			if (ret == -ENOENT)
-				pr_warning("No dwarf info found in the vmlinux - please rebuild with CONFIG_DEBUG_INFO=y.\n");
-			die("Could not analyze debuginfo.");
+	if (params.dellist) {
+		ret = del_perf_probe_events(params.dellist);
+		strlist__delete(params.dellist);
+		if (ret < 0) {
+			pr_err("  Error: Failed to delete events. (%d)\n", ret);
+			return ret;
 		}
-		pr_debug("An error occurred in debuginfo analysis."
-			 " Try to use symbols.\n");
-		break;
 	}
-	close(fd);
-
-end_dwarf:
-#endif /* !NO_DWARF_SUPPORT */
 
-	/* Synthesize probes without dwarf */
-	for (i = 0; i < session.nr_probe; i++) {
-		pp = &session.probes[i];
-		if (pp->found)	/* This probe is already found. */
-			continue;
-
-		evaluate_probe_point(pp);
-		ret = synthesize_trace_kprobe_event(pp);
-		if (ret == -E2BIG)
-			die("probe point definition becomes too long.");
-		else if (ret < 0)
-			die("Failed to synthesize a probe point.");
+	if (params.nevents) {
+		ret = add_perf_probe_events(params.events, params.nevents,
+					    params.force_add);
+		if (ret < 0) {
+			pr_err("  Error: Failed to add events. (%d)\n", ret);
+			return ret;
+		}
 	}
-
-	/* Settng up probe points */
-	add_trace_kprobe_events(session.probes, session.nr_probe,
-				session.force_add);
 	return 0;
 }
 

tools/perf/builtin-record.c

@@ -15,7 +15,6 @@
 #include "util/util.h"
 #include "util/parse-options.h"
 #include "util/parse-events.h"
-#include "util/string.h"
 
 #include "util/header.h"
 #include "util/event.h"
@@ -27,31 +26,41 @@
 #include <unistd.h>
 #include <sched.h>
 
-static int			fd[MAX_NR_CPUS][MAX_COUNTERS];
+enum write_mode_t {
+	WRITE_FORCE,
+	WRITE_APPEND
+};
+
+static int			*fd[MAX_NR_CPUS][MAX_COUNTERS];
 
+static unsigned int		user_interval 			= UINT_MAX;
 static long			default_interval		=      0;
 
 static int			nr_cpus				=      0;
 static unsigned int		page_size;
 static unsigned int		mmap_pages			=    128;
+static unsigned int		user_freq 			= UINT_MAX;
 static int			freq				=   1000;
 static int			output;
+static int			pipe_output			=      0;
 static const char		*output_name			= "perf.data";
 static int			group				=      0;
 static unsigned int		realtime_prio			=      0;
-static int			raw_samples			=      0;
-static int			system_wide			=      0;
+static bool			raw_samples			=  false;
+static bool			system_wide			=  false;
 static int			profile_cpu			=     -1;
 static pid_t			target_pid			=     -1;
+static pid_t			target_tid			=     -1;
+static pid_t			*all_tids			=      NULL;
+static int			thread_num			=      0;
 static pid_t			child_pid			=     -1;
-static int			inherit				=      1;
-static int			force				=      0;
-static int			append_file			=      0;
-static int			call_graph			=      0;
-static int			inherit_stat			=      0;
-static int			no_samples			=      0;
-static int			sample_address			=      0;
-static int			multiplex			=      0;
+static bool			inherit				=   true;
+static enum write_mode_t	write_mode			= WRITE_FORCE;
+static bool			call_graph			=  false;
+static bool			inherit_stat			=  false;
+static bool			no_samples			=  false;
+static bool			sample_address			=  false;
+static bool			multiplex			=  false;
 static int			multiplex_fd			=     -1;
 
 static long			samples				=      0;
@@ -60,7 +69,7 @@ static struct timeval		this_read;
 
 static u64			bytes_written			=      0;
 
-static struct pollfd		event_array[MAX_NR_CPUS * MAX_COUNTERS];
+static struct pollfd		*event_array;
 
 static int			nr_poll				=      0;
 static int			nr_cpu				=      0;
@@ -77,7 +86,7 @@ struct mmap_data {
 	unsigned int		prev;
 };
 
-static struct mmap_data		mmap_array[MAX_NR_CPUS][MAX_COUNTERS];
+static struct mmap_data		*mmap_array[MAX_NR_CPUS][MAX_COUNTERS];
 
 static unsigned long mmap_read_head(struct mmap_data *md)
 {
@@ -101,6 +110,11 @@ static void mmap_write_tail(struct mmap_data *md, unsigned long tail)
 	pc->data_tail = tail;
 }
 
+static void advance_output(size_t size)
+{
+	bytes_written += size;
+}
+
 static void write_output(void *buf, size_t size)
 {
 	while (size) {
@@ -225,12 +239,13 @@ static struct perf_header_attr *get_header_attr(struct perf_event_attr *a, int n
 	return h_attr;
 }
 
-static void create_counter(int counter, int cpu, pid_t pid)
+static void create_counter(int counter, int cpu)
 {
 	char *filter = filters[counter];
 	struct perf_event_attr *attr = attrs + counter;
 	struct perf_header_attr *h_attr;
 	int track = !counter; /* only the first counter needs these */
+	int thread_index;
 	int ret;
 	struct {
 		u64 count;
@@ -248,10 +263,19 @@ static void create_counter(int counter, int cpu, pid_t pid)
 	if (nr_counters > 1)
 		attr->sample_type |= PERF_SAMPLE_ID;
 
-	if (freq) {
-		attr->sample_type	|= PERF_SAMPLE_PERIOD;
-		attr->freq		= 1;
-		attr->sample_freq	= freq;
+	/*
+	 * We default some events to a 1 default interval. But keep
+	 * it a weak assumption overridable by the user.
+	 */
+	if (!attr->sample_period || (user_freq != UINT_MAX &&
+				     user_interval != UINT_MAX)) {
+		if (freq) {
+			attr->sample_type	|= PERF_SAMPLE_PERIOD;
+			attr->freq		= 1;
+			attr->sample_freq	= freq;
+		} else {
+			attr->sample_period = default_interval;
+		}
 	}
 
 	if (no_samples)
@@ -275,118 +299,129 @@ static void create_counter(int counter, int cpu, pid_t pid)
 	attr->mmap		= track;
 	attr->comm		= track;
 	attr->inherit		= inherit;
-	attr->disabled		= 1;
+	if (target_pid == -1 && !system_wide) {
+		attr->disabled = 1;
+		attr->enable_on_exec = 1;
+	}
 
+	for (thread_index = 0; thread_index < thread_num; thread_index++) {
 try_again:
-	fd[nr_cpu][counter] = sys_perf_event_open(attr, pid, cpu, group_fd, 0);
-
-	if (fd[nr_cpu][counter] < 0) {
-		int err = errno;
-
-		if (err == EPERM || err == EACCES)
-			die("Permission error - are you root?\n");
-		else if (err ==  ENODEV && profile_cpu != -1)
-			die("No such device - did you specify an out-of-range profile CPU?\n");
+		fd[nr_cpu][counter][thread_index] = sys_perf_event_open(attr,
+				all_tids[thread_index], cpu, group_fd, 0);
+
+		if (fd[nr_cpu][counter][thread_index] < 0) {
+			int err = errno;
+
+			if (err == EPERM || err == EACCES)
+				die("Permission error - are you root?\n"
+					"\t Consider tweaking"
+					" /proc/sys/kernel/perf_event_paranoid.\n");
+			else if (err ==  ENODEV && profile_cpu != -1) {
+				die("No such device - did you specify"
+					" an out-of-range profile CPU?\n");
+			}
 
-		/*
-		 * If it's cycles then fall back to hrtimer
-		 * based cpu-clock-tick sw counter, which
-		 * is always available even if no PMU support:
-		 */
-		if (attr->type == PERF_TYPE_HARDWARE
-			&& attr->config == PERF_COUNT_HW_CPU_CYCLES) {
-
-			if (verbose)
-				warning(" ... trying to fall back to cpu-clock-ticks\n");
-			attr->type = PERF_TYPE_SOFTWARE;
-			attr->config = PERF_COUNT_SW_CPU_CLOCK;
-			goto try_again;
-		}
-		printf("\n");
-		error("perfcounter syscall returned with %d (%s)\n",
-			fd[nr_cpu][counter], strerror(err));
+			/*
+			 * If it's cycles then fall back to hrtimer
+			 * based cpu-clock-tick sw counter, which
+			 * is always available even if no PMU support:
+			 */
+			if (attr->type == PERF_TYPE_HARDWARE
+					&& attr->config == PERF_COUNT_HW_CPU_CYCLES) {
+
+				if (verbose)
+					warning(" ... trying to fall back to cpu-clock-ticks\n");
+				attr->type = PERF_TYPE_SOFTWARE;
+				attr->config = PERF_COUNT_SW_CPU_CLOCK;
+				goto try_again;
+			}
+			printf("\n");
+			error("perfcounter syscall returned with %d (%s)\n",
+					fd[nr_cpu][counter][thread_index], strerror(err));
 
 #if defined(__i386__) || defined(__x86_64__)
-		if (attr->type == PERF_TYPE_HARDWARE && err == EOPNOTSUPP)
-			die("No hardware sampling interrupt available. No APIC? If so then you can boot the kernel with the \"lapic\" boot parameter to force-enable it.\n");
+			if (attr->type == PERF_TYPE_HARDWARE && err == EOPNOTSUPP)
+				die("No hardware sampling interrupt available."
+				    " No APIC? If so then you can boot the kernel"
+				    " with the \"lapic\" boot parameter to"
+				    " force-enable it.\n");
 #endif
 
-		die("No CONFIG_PERF_EVENTS=y kernel support configured?\n");
-		exit(-1);
-	}
+			die("No CONFIG_PERF_EVENTS=y kernel support configured?\n");
+			exit(-1);
+		}
 
-	h_attr = get_header_attr(attr, counter);
-	if (h_attr == NULL)
-		die("nomem\n");
+		h_attr = get_header_attr(attr, counter);
+		if (h_attr == NULL)
+			die("nomem\n");
 
-	if (!file_new) {
-		if (memcmp(&h_attr->attr, attr, sizeof(*attr))) {
-			fprintf(stderr, "incompatible append\n");
-			exit(-1);
+		if (!file_new) {
+			if (memcmp(&h_attr->attr, attr, sizeof(*attr))) {
+				fprintf(stderr, "incompatible append\n");
+				exit(-1);
+			}
 		}
-	}
 
-	if (read(fd[nr_cpu][counter], &read_data, sizeof(read_data)) == -1) {
-		perror("Unable to read perf file descriptor\n");
-		exit(-1);
-	}
+		if (read(fd[nr_cpu][counter][thread_index], &read_data, sizeof(read_data)) == -1) {
+			perror("Unable to read perf file descriptor\n");
+			exit(-1);
+		}
 
-	if (perf_header_attr__add_id(h_attr, read_data.id) < 0) {
-		pr_warning("Not enough memory to add id\n");
-		exit(-1);
-	}
+		if (perf_header_attr__add_id(h_attr, read_data.id) < 0) {
+			pr_warning("Not enough memory to add id\n");
+			exit(-1);
+		}
 
-	assert(fd[nr_cpu][counter] >= 0);
-	fcntl(fd[nr_cpu][counter], F_SETFL, O_NONBLOCK);
+		assert(fd[nr_cpu][counter][thread_index] >= 0);
+		fcntl(fd[nr_cpu][counter][thread_index], F_SETFL, O_NONBLOCK);
 
-	/*
-	 * First counter acts as the group leader:
-	 */
-	if (group && group_fd == -1)
-		group_fd = fd[nr_cpu][counter];
-	if (multiplex && multiplex_fd == -1)
-		multiplex_fd = fd[nr_cpu][counter];
+		/*
+		 * First counter acts as the group leader:
+		 */
+		if (group && group_fd == -1)
+			group_fd = fd[nr_cpu][counter][thread_index];
+		if (multiplex && multiplex_fd == -1)
+			multiplex_fd = fd[nr_cpu][counter][thread_index];
 
-	if (multiplex && fd[nr_cpu][counter] != multiplex_fd) {
+		if (multiplex && fd[nr_cpu][counter][thread_index] != multiplex_fd) {
 
-		ret = ioctl(fd[nr_cpu][counter], PERF_EVENT_IOC_SET_OUTPUT, multiplex_fd);
-		assert(ret != -1);
-	} else {
-		event_array[nr_poll].fd = fd[nr_cpu][counter];
-		event_array[nr_poll].events = POLLIN;
-		nr_poll++;
-
-		mmap_array[nr_cpu][counter].counter = counter;
-		mmap_array[nr_cpu][counter].prev = 0;
-		mmap_array[nr_cpu][counter].mask = mmap_pages*page_size - 1;
-		mmap_array[nr_cpu][counter].base = mmap(NULL, (mmap_pages+1)*page_size,
-				PROT_READ|PROT_WRITE, MAP_SHARED, fd[nr_cpu][counter], 0);
-		if (mmap_array[nr_cpu][counter].base == MAP_FAILED) {
-			error("failed to mmap with %d (%s)\n", errno, strerror(errno));
-			exit(-1);
+			ret = ioctl(fd[nr_cpu][counter][thread_index], PERF_EVENT_IOC_SET_OUTPUT, multiplex_fd);
+			assert(ret != -1);
+		} else {
+			event_array[nr_poll].fd = fd[nr_cpu][counter][thread_index];
+			event_array[nr_poll].events = POLLIN;
+			nr_poll++;
+
+			mmap_array[nr_cpu][counter][thread_index].counter = counter;
+			mmap_array[nr_cpu][counter][thread_index].prev = 0;
+			mmap_array[nr_cpu][counter][thread_index].mask = mmap_pages*page_size - 1;
+			mmap_array[nr_cpu][counter][thread_index].base = mmap(NULL, (mmap_pages+1)*page_size,
+				PROT_READ|PROT_WRITE, MAP_SHARED, fd[nr_cpu][counter][thread_index], 0);
+			if (mmap_array[nr_cpu][counter][thread_index].base == MAP_FAILED) {
+				error("failed to mmap with %d (%s)\n", errno, strerror(errno));
+				exit(-1);
+			}
 		}
-	}
 
-	if (filter != NULL) {
-		ret = ioctl(fd[nr_cpu][counter],
-			    PERF_EVENT_IOC_SET_FILTER, filter);
-		if (ret) {
-			error("failed to set filter with %d (%s)\n", errno,
-			      strerror(errno));
-			exit(-1);
+		if (filter != NULL) {
+			ret = ioctl(fd[nr_cpu][counter][thread_index],
+					PERF_EVENT_IOC_SET_FILTER, filter);
+			if (ret) {
+				error("failed to set filter with %d (%s)\n", errno,
+						strerror(errno));
+				exit(-1);
+			}
 		}
 	}
-
-	ioctl(fd[nr_cpu][counter], PERF_EVENT_IOC_ENABLE);
 }
 
-static void open_counters(int cpu, pid_t pid)
+static void open_counters(int cpu)
 {
 	int counter;
 
 	group_fd = -1;
 	for (counter = 0; counter < nr_counters; counter++)
-		create_counter(counter, cpu, pid);
+		create_counter(counter, cpu);
 
 	nr_cpu++;
 }
@@ -406,10 +441,65 @@ static int process_buildids(void)
 
 static void atexit_header(void)
 {
-	session->header.data_size += bytes_written;
+	if (!pipe_output) {
+		session->header.data_size += bytes_written;
+
+		process_buildids();
+		perf_header__write(&session->header, output, true);
+	} else {
+		int err;
+
+		err = event__synthesize_build_ids(process_synthesized_event,
+						  session);
+		if (err < 0)
+			pr_err("Couldn't synthesize build ids.\n");
+	}
+}
+
+static void event__synthesize_guest_os(struct kernel_info *kerninfo,
+		void *data __attribute__((unused)))
+{
+	int err;
+	char *guest_kallsyms;
+	char path[PATH_MAX];
+
+	if (is_host_kernel(kerninfo))
+		return;
+
+	/*
+	 *As for guest kernel when processing subcommand record&report,
+	 *we arrange module mmap prior to guest kernel mmap and trigger
+	 *a preload dso because default guest module symbols are loaded
+	 *from guest kallsyms instead of /lib/modules/XXX/XXX. This
+	 *method is used to avoid symbol missing when the first addr is
+	 *in module instead of in guest kernel.
+	 */
+	err = event__synthesize_modules(process_synthesized_event,
+			session,
+			kerninfo);
+	if (err < 0)
+		pr_err("Couldn't record guest kernel [%d]'s reference"
+			" relocation symbol.\n", kerninfo->pid);
+
+	if (is_default_guest(kerninfo))
+		guest_kallsyms = (char *) symbol_conf.default_guest_kallsyms;
+	else {
+		sprintf(path, "%s/proc/kallsyms", kerninfo->root_dir);
+		guest_kallsyms = path;
+	}
 
-	process_buildids();
-	perf_header__write(&session->header, output, true);
+	/*
+	 * We use _stext for guest kernel because guest kernel's /proc/kallsyms
+	 * have no _text sometimes.
+	 */
+	err = event__synthesize_kernel_mmap(process_synthesized_event,
+			session, kerninfo, "_text");
+	if (err < 0)
+		err = event__synthesize_kernel_mmap(process_synthesized_event,
+				session, kerninfo, "_stext");
+	if (err < 0)
+		pr_err("Couldn't record guest kernel [%d]'s reference"
+			" relocation symbol.\n", kerninfo->pid);
 }
 
 static int __cmd_record(int argc, const char **argv)
@@ -421,8 +511,9 @@ static int __cmd_record(int argc, const char **argv)
 	int err;
 	unsigned long waking = 0;
 	int child_ready_pipe[2], go_pipe[2];
-	const bool forks = target_pid == -1 && argc > 0;
+	const bool forks = argc > 0;
 	char buf;
+	struct kernel_info *kerninfo;
 
 	page_size = sysconf(_SC_PAGE_SIZE);
 
@@ -435,45 +526,44 @@ static int __cmd_record(int argc, const char **argv)
 		exit(-1);
 	}
 
-	if (!stat(output_name, &st) && st.st_size) {
-		if (!force) {
-			if (!append_file) {
-				pr_err("Error, output file %s exists, use -A "
-				       "to append or -f to overwrite.\n",
-				       output_name);
-				exit(-1);
-			}
-		} else {
+	if (!strcmp(output_name, "-"))
+		pipe_output = 1;
+	else if (!stat(output_name, &st) && st.st_size) {
+		if (write_mode == WRITE_FORCE) {
 			char oldname[PATH_MAX];
 			snprintf(oldname, sizeof(oldname), "%s.old",
 				 output_name);
 			unlink(oldname);
 			rename(output_name, oldname);
 		}
-	} else {
-		append_file = 0;
+	} else if (write_mode == WRITE_APPEND) {
+		write_mode = WRITE_FORCE;
 	}
 
 	flags = O_CREAT|O_RDWR;
-	if (append_file)
+	if (write_mode == WRITE_APPEND)
 		file_new = 0;
 	else
 		flags |= O_TRUNC;
 
-	output = open(output_name, flags, S_IRUSR|S_IWUSR);
+	if (pipe_output)
+		output = STDOUT_FILENO;
+	else
+		output = open(output_name, flags, S_IRUSR | S_IWUSR);
 	if (output < 0) {
 		perror("failed to create output file");
 		exit(-1);
 	}
 
-	session = perf_session__new(output_name, O_WRONLY, force);
+	session = perf_session__new(output_name, O_WRONLY,
+				    write_mode == WRITE_FORCE);
 	if (session == NULL) {
 		pr_err("Not enough memory for reading perf file header\n");
 		return -1;
 	}
 
 	if (!file_new) {
-		err = perf_header__read(&session->header, output);
+		err = perf_header__read(session, output);
 		if (err < 0)
 			return err;
 	}
@@ -492,13 +582,15 @@ static int __cmd_record(int argc, const char **argv)
 	atexit(atexit_header);
 
 	if (forks) {
-		pid = fork();
+		child_pid = fork();
 		if (pid < 0) {
 			perror("failed to fork");
 			exit(-1);
 		}
 
-		if (!pid) {
+		if (!child_pid) {
+			if (pipe_output)
+				dup2(2, 1);
 			close(child_ready_pipe[0]);
 			close(go_pipe[1]);
 			fcntl(go_pipe[0], F_SETFD, FD_CLOEXEC);
@@ -527,10 +619,8 @@ static int __cmd_record(int argc, const char **argv)
 			exit(-1);
 		}
 
-		child_pid = pid;
-
-		if (!system_wide)
-			target_pid = pid;
+		if (!system_wide && target_tid == -1 && target_pid == -1)
+			all_tids[0] = child_pid;
 
 		close(child_ready_pipe[1]);
 		close(go_pipe[0]);
@@ -544,16 +634,19 @@ static int __cmd_record(int argc, const char **argv)
 		close(child_ready_pipe[0]);
 	}
 
-
 	if ((!system_wide && !inherit) || profile_cpu != -1) {
-		open_counters(profile_cpu, target_pid);
+		open_counters(profile_cpu);
 	} else {
 		nr_cpus = read_cpu_map();
 		for (i = 0; i < nr_cpus; i++)
-			open_counters(cpumap[i], target_pid);
+			open_counters(cpumap[i]);
 	}
 
-	if (file_new) {
+	if (pipe_output) {
+		err = perf_header__write_pipe(output);
+		if (err < 0)
+			return err;
+	} else if (file_new) {
 		err = perf_header__write(&session->header, output, false);
 		if (err < 0)
 			return err;
@@ -561,21 +654,62 @@ static int __cmd_record(int argc, const char **argv)
 
 	post_processing_offset = lseek(output, 0, SEEK_CUR);
 
+	if (pipe_output) {
+		err = event__synthesize_attrs(&session->header,
+					      process_synthesized_event,
+					      session);
+		if (err < 0) {
+			pr_err("Couldn't synthesize attrs.\n");
+			return err;
+		}
+
+		err = event__synthesize_event_types(process_synthesized_event,
+						    session);
+		if (err < 0) {
+			pr_err("Couldn't synthesize event_types.\n");
+			return err;
+		}
+
+		err = event__synthesize_tracing_data(output, attrs,
+						     nr_counters,
+						     process_synthesized_event,
+						     session);
+		if (err <= 0) {
+			pr_err("Couldn't record tracing data.\n");
+			return err;
+		}
+
+		advance_output(err);
+	}
+
+	kerninfo = kerninfo__findhost(&session->kerninfo_root);
+	if (!kerninfo) {
+		pr_err("Couldn't find native kernel information.\n");
+		return -1;
+	}
+
 	err = event__synthesize_kernel_mmap(process_synthesized_event,
-					    session, "_text");
+			session, kerninfo, "_text");
+	if (err < 0)
+		err = event__synthesize_kernel_mmap(process_synthesized_event,
+				session, kerninfo, "_stext");
 	if (err < 0) {
 		pr_err("Couldn't record kernel reference relocation symbol.\n");
 		return err;
 	}
 
-	err = event__synthesize_modules(process_synthesized_event, session);
+	err = event__synthesize_modules(process_synthesized_event,
+				session, kerninfo);
 	if (err < 0) {
 		pr_err("Couldn't record kernel reference relocation symbol.\n");
 		return err;
 	}
+	if (perf_guest)
+		kerninfo__process_allkernels(&session->kerninfo_root,
+			event__synthesize_guest_os, session);
 
 	if (!system_wide && profile_cpu == -1)
-		event__synthesize_thread(target_pid, process_synthesized_event,
+		event__synthesize_thread(target_tid, process_synthesized_event,
 					 session);
 	else
 		event__synthesize_threads(process_synthesized_event, session);
@@ -598,11 +732,16 @@ static int __cmd_record(int argc, const char **argv)
 
 	for (;;) {
 		int hits = samples;
+		int thread;
 
 		for (i = 0; i < nr_cpu; i++) {
 			for (counter = 0; counter < nr_counters; counter++) {
-				if (mmap_array[i][counter].base)
-					mmap_read(&mmap_array[i][counter]);
+				for (thread = 0;
+					thread < thread_num; thread++) {
+					if (mmap_array[i][counter][thread].base)
+						mmap_read(&mmap_array[i][counter][thread]);
+				}
+
 			}
 		}
 
@@ -615,8 +754,15 @@ static int __cmd_record(int argc, const char **argv)
 
 		if (done) {
 			for (i = 0; i < nr_cpu; i++) {
-				for (counter = 0; counter < nr_counters; counter++)
-					ioctl(fd[i][counter], PERF_EVENT_IOC_DISABLE);
+				for (counter = 0;
+					counter < nr_counters;
+					counter++) {
+					for (thread = 0;
+						thread < thread_num;
+						thread++)
+						ioctl(fd[i][counter][thread],
+							PERF_EVENT_IOC_DISABLE);
+				}
 			}
 		}
 	}
@@ -641,6 +787,8 @@ static const char * const record_usage[] = {
 	NULL
 };
 
+static bool force, append_file;
+
 static const struct option options[] = {
 	OPT_CALLBACK('e', "event", NULL, "event",
 		     "event selector. use 'perf list' to list available events",
@@ -648,7 +796,9 @@ static const struct option options[] = {
 	OPT_CALLBACK(0, "filter", NULL, "filter",
 		     "event filter", parse_filter),
 	OPT_INTEGER('p', "pid", &target_pid,
-		    "record events on existing pid"),
+		    "record events on existing process id"),
+	OPT_INTEGER('t', "tid", &target_tid,
+		    "record events on existing thread id"),
 	OPT_INTEGER('r', "realtime", &realtime_prio,
 		    "collect data with this RT SCHED_FIFO priority"),
 	OPT_BOOLEAN('R', "raw-samples", &raw_samples,
@@ -660,20 +810,20 @@ static const struct option options[] = {
 	OPT_INTEGER('C', "profile_cpu", &profile_cpu,
 			    "CPU to profile on"),
 	OPT_BOOLEAN('f', "force", &force,
-			"overwrite existing data file"),
-	OPT_LONG('c', "count", &default_interval,
+			"overwrite existing data file (deprecated)"),
+	OPT_LONG('c', "count", &user_interval,
 		    "event period to sample"),
 	OPT_STRING('o', "output", &output_name, "file",
 		    "output file name"),
 	OPT_BOOLEAN('i', "inherit", &inherit,
 		    "child tasks inherit counters"),
-	OPT_INTEGER('F', "freq", &freq,
+	OPT_INTEGER('F', "freq", &user_freq,
 		    "profile at this frequency"),
 	OPT_INTEGER('m', "mmap-pages", &mmap_pages,
 		    "number of mmap data pages"),
 	OPT_BOOLEAN('g', "call-graph", &call_graph,
 		    "do call-graph (stack chain/backtrace) recording"),
-	OPT_BOOLEAN('v', "verbose", &verbose,
+	OPT_INCR('v', "verbose", &verbose,
 		    "be more verbose (show counter open errors, etc)"),
 	OPT_BOOLEAN('s', "stat", &inherit_stat,
 		    "per thread counts"),
@@ -688,13 +838,24 @@ static const struct option options[] = {
 
 int cmd_record(int argc, const char **argv, const char *prefix __used)
 {
-	int counter;
+	int i,j;
 
 	argc = parse_options(argc, argv, options, record_usage,
 			    PARSE_OPT_STOP_AT_NON_OPTION);
-	if (!argc && target_pid == -1 && !system_wide && profile_cpu == -1)
+	if (!argc && target_pid == -1 && target_tid == -1 &&
+		!system_wide && profile_cpu == -1)
 		usage_with_options(record_usage, options);
 
+	if (force && append_file) {
+		fprintf(stderr, "Can't overwrite and append at the same time."
+				" You need to choose between -f and -A");
+		usage_with_options(record_usage, options);
+	} else if (append_file) {
+		write_mode = WRITE_APPEND;
+	} else {
+		write_mode = WRITE_FORCE;
+	}
+
 	symbol__init();
 
 	if (!nr_counters) {
@@ -703,6 +864,42 @@ int cmd_record(int argc, const char **argv, const char *prefix __used)
 		attrs[0].config = PERF_COUNT_HW_CPU_CYCLES;
 	}
 
+	if (target_pid != -1) {
+		target_tid = target_pid;
+		thread_num = find_all_tid(target_pid, &all_tids);
+		if (thread_num <= 0) {
+			fprintf(stderr, "Can't find all threads of pid %d\n",
+					target_pid);
+			usage_with_options(record_usage, options);
+		}
+	} else {
+		all_tids=malloc(sizeof(pid_t));
+		if (!all_tids)
+			return -ENOMEM;
+
+		all_tids[0] = target_tid;
+		thread_num = 1;
+	}
+
+	for (i = 0; i < MAX_NR_CPUS; i++) {
+		for (j = 0; j < MAX_COUNTERS; j++) {
+			fd[i][j] = malloc(sizeof(int)*thread_num);
+			mmap_array[i][j] = zalloc(
+				sizeof(struct mmap_data)*thread_num);
+			if (!fd[i][j] || !mmap_array[i][j])
+				return -ENOMEM;
+		}
+	}
+	event_array = malloc(
+		sizeof(struct pollfd)*MAX_NR_CPUS*MAX_COUNTERS*thread_num);
+	if (!event_array)
+		return -ENOMEM;
+
+	if (user_interval != UINT_MAX)
+		default_interval = user_interval;
+	if (user_freq != UINT_MAX)
+		freq = user_freq;
+
 	/*
 	 * User specified count overrides default frequency.
 	 */
@@ -715,12 +912,5 @@ int cmd_record(int argc, const char **argv, const char *prefix __used)
 		exit(EXIT_FAILURE);
 	}
 
-	for (counter = 0; counter < nr_counters; counter++) {
-		if (attrs[counter].sample_period)
-			continue;
-
-		attrs[counter].sample_period = default_interval;
-	}
-
 	return __cmd_record(argc, argv);
 }

tools/perf/builtin-report.c

@@ -14,7 +14,6 @@
 #include "util/cache.h"
 #include <linux/rbtree.h>
 #include "util/symbol.h"
-#include "util/string.h"
 #include "util/callchain.h"
 #include "util/strlist.h"
 #include "util/values.h"
@@ -33,11 +32,11 @@
 
 static char		const *input_name = "perf.data";
 
-static int		force;
+static bool		force;
 static bool		hide_unresolved;
 static bool		dont_use_callchains;
 
-static int		show_threads;
+static bool		show_threads;
 static struct perf_read_values	show_threads_values;
 
 static char		default_pretty_printing_style[] = "normal";
@@ -81,15 +80,20 @@ static int perf_session__add_hist_entry(struct perf_session *self,
 					struct addr_location *al,
 					struct sample_data *data)
 {
-	struct symbol **syms = NULL, *parent = NULL;
+	struct map_symbol *syms = NULL;
+	struct symbol *parent = NULL;
 	bool hit;
+	int err;
 	struct hist_entry *he;
 	struct event_stat_id *stats;
 	struct perf_event_attr *attr;
 
-	if ((sort__has_parent || symbol_conf.use_callchain) && data->callchain)
+	if ((sort__has_parent || symbol_conf.use_callchain) && data->callchain) {
 		syms = perf_session__resolve_callchain(self, al->thread,
 						       data->callchain, &parent);
+		if (syms == NULL)
+			return -ENOMEM;
+	}
 
 	attr = perf_header__find_attr(data->id, &self->header);
 	if (attr)
@@ -104,13 +108,16 @@ static int perf_session__add_hist_entry(struct perf_session *self,
 		return -ENOMEM;
 
 	if (hit)
-		he->count += data->period;
+		__perf_session__add_count(he, al,  data->period);
 
 	if (symbol_conf.use_callchain) {
 		if (!hit)
-			callchain_init(&he->callchain);
-		append_chain(&he->callchain, data->callchain, syms);
+			callchain_init(he->callchain);
+		err = append_chain(he->callchain, data->callchain, syms);
 		free(syms);
+
+		if (err)
+			return err;
 	}
 
 	return 0;
@@ -260,13 +267,27 @@ static struct perf_event_ops event_ops = {
 	.fork	= event__process_task,
 	.lost	= event__process_lost,
 	.read	= process_read_event,
+	.attr	= event__process_attr,
+	.event_type = event__process_event_type,
+	.tracing_data = event__process_tracing_data,
+	.build_id = event__process_build_id,
 };
 
+extern volatile int session_done;
+
+static void sig_handler(int sig __attribute__((__unused__)))
+{
+	session_done = 1;
+}
+
 static int __cmd_report(void)
 {
 	int ret = -EINVAL;
 	struct perf_session *session;
 	struct rb_node *next;
+	const char *help = "For a higher level overview, try: perf report --sort comm,dso";
+
+	signal(SIGINT, sig_handler);
 
 	session = perf_session__new(input_name, O_RDONLY, force);
 	if (session == NULL)
@@ -292,39 +313,49 @@ static int __cmd_report(void)
 		perf_session__fprintf(session, stdout);
 
 	if (verbose > 2)
-		dsos__fprintf(stdout);
+		dsos__fprintf(&session->kerninfo_root, stdout);
 
 	next = rb_first(&session->stats_by_id);
 	while (next) {
 		struct event_stat_id *stats;
+		u64 nr_hists;
 
 		stats = rb_entry(next, struct event_stat_id, rb_node);
 		perf_session__collapse_resort(&stats->hists);
-		perf_session__output_resort(&stats->hists, stats->stats.total);
-		if (rb_first(&session->stats_by_id) ==
-		    rb_last(&session->stats_by_id))
-			fprintf(stdout, "# Samples: %Ld\n#\n",
-				stats->stats.total);
-		else
-			fprintf(stdout, "# Samples: %Ld %s\n#\n",
-				stats->stats.total,
-				__event_name(stats->type, stats->config));
-
-		perf_session__fprintf_hists(&stats->hists, NULL, false, stdout,
+		nr_hists = perf_session__output_resort(&stats->hists,
+						       stats->stats.total);
+		if (use_browser)
+			perf_session__browse_hists(&stats->hists, nr_hists,
+						   stats->stats.total, help,
+						   input_name);
+		else {
+			if (rb_first(&session->stats_by_id) ==
+			    rb_last(&session->stats_by_id))
+				fprintf(stdout, "# Samples: %Ld\n#\n",
+					stats->stats.total);
+			else
+				fprintf(stdout, "# Samples: %Ld %s\n#\n",
+					stats->stats.total,
+					__event_name(stats->type, stats->config));
+
+			perf_session__fprintf_hists(&stats->hists, NULL, false, stdout,
 					    stats->stats.total);
-		fprintf(stdout, "\n\n");
+			fprintf(stdout, "\n\n");
+		}
+
 		next = rb_next(&stats->rb_node);
 	}
 
-	if (sort_order == default_sort_order &&
-	    parent_pattern == default_parent_pattern)
-		fprintf(stdout, "#\n# (For a higher level overview, try: perf report --sort comm,dso)\n#\n");
+	if (!use_browser && sort_order == default_sort_order &&
+	    parent_pattern == default_parent_pattern) {
+		fprintf(stdout, "#\n# (%s)\n#\n", help);
 
-	if (show_threads) {
-		bool raw_printing_style = !strcmp(pretty_printing_style, "raw");
-		perf_read_values_display(stdout, &show_threads_values,
-					 raw_printing_style);
-		perf_read_values_destroy(&show_threads_values);
+		if (show_threads) {
+			bool style = !strcmp(pretty_printing_style, "raw");
+			perf_read_values_display(stdout, &show_threads_values,
+						 style);
+			perf_read_values_destroy(&show_threads_values);
+		}
 	}
 out_delete:
 	perf_session__delete(session);
@@ -400,7 +431,7 @@ static const char * const report_usage[] = {
 static const struct option options[] = {
 	OPT_STRING('i', "input", &input_name, "file",
 		    "input file name"),
-	OPT_BOOLEAN('v', "verbose", &verbose,
+	OPT_INCR('v', "verbose", &verbose,
 		    "be more verbose (show symbol address, etc)"),
 	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
 		    "dump raw trace in ASCII"),
@@ -419,6 +450,8 @@ static const struct option options[] = {
 		   "sort by key(s): pid, comm, dso, symbol, parent"),
 	OPT_BOOLEAN('P', "full-paths", &symbol_conf.full_paths,
 		    "Don't shorten the pathnames taking into account the cwd"),
+	OPT_BOOLEAN(0, "showcpuutilization", &symbol_conf.show_cpu_utilization,
+		    "Show sample percentage for different cpu modes"),
 	OPT_STRING('p', "parent", &parent_pattern, "regex",
 		   "regex filter to identify parent, see: '--sort parent'"),
 	OPT_BOOLEAN('x', "exclude-other", &symbol_conf.exclude_other,
@@ -447,7 +480,8 @@ int cmd_report(int argc, const char **argv, const char *prefix __used)
 {
 	argc = parse_options(argc, argv, options, report_usage, 0);
 
-	setup_pager();
+	if (strcmp(input_name, "-") != 0)
+		setup_browser();
 
 	if (symbol__init() < 0)
 		return -1;
@@ -455,7 +489,8 @@ int cmd_report(int argc, const char **argv, const char *prefix __used)
 	setup_sorting(report_usage, options);
 
 	if (parent_pattern != default_parent_pattern) {
-		sort_dimension__add("parent");
+		if (sort_dimension__add("parent") < 0)
+			return -1;
 		sort_parent.elide = 1;
 	} else
 		symbol_conf.exclude_other = false;

tools/perf/builtin-sched.c

@@ -68,10 +68,10 @@ enum sched_event_type {
 
 struct sched_atom {
 	enum sched_event_type	type;
+	int			specific_wait;
 	u64			timestamp;
 	u64			duration;
 	unsigned long		nr;
-	int			specific_wait;
 	sem_t			*wait_sem;
 	struct task_desc	*wakee;
 };
@@ -1790,7 +1790,7 @@ static const char * const sched_usage[] = {
 static const struct option sched_options[] = {
 	OPT_STRING('i', "input", &input_name, "file",
 		    "input file name"),
-	OPT_BOOLEAN('v', "verbose", &verbose,
+	OPT_INCR('v', "verbose", &verbose,
 		    "be more verbose (show symbol address, etc)"),
 	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
 		    "dump raw trace in ASCII"),
@@ -1805,7 +1805,7 @@ static const char * const latency_usage[] = {
 static const struct option latency_options[] = {
 	OPT_STRING('s', "sort", &sort_order, "key[,key2...]",
 		   "sort by key(s): runtime, switch, avg, max"),
-	OPT_BOOLEAN('v', "verbose", &verbose,
+	OPT_INCR('v', "verbose", &verbose,
 		    "be more verbose (show symbol address, etc)"),
 	OPT_INTEGER('C', "CPU", &profile_cpu,
 		    "CPU to profile on"),
@@ -1822,7 +1822,7 @@ static const char * const replay_usage[] = {
 static const struct option replay_options[] = {
 	OPT_INTEGER('r', "repeat", &replay_repeat,
 		    "repeat the workload replay N times (-1: infinite)"),
-	OPT_BOOLEAN('v', "verbose", &verbose,
+	OPT_INCR('v', "verbose", &verbose,
 		    "be more verbose (show symbol address, etc)"),
 	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
 		    "dump raw trace in ASCII"),

tools/perf/builtin-stat.c

@@ -46,6 +46,7 @@
 #include "util/debug.h"
 #include "util/header.h"
 #include "util/cpumap.h"
+#include "util/thread.h"
 
 #include <sys/prctl.h>
 #include <math.h>
@@ -66,18 +67,21 @@ static struct perf_event_attr default_attrs[] = {
 
 };
 
-static int			system_wide			=  0;
+static bool			system_wide			=  false;
 static unsigned int		nr_cpus				=  0;
 static int			run_idx				=  0;
 
 static int			run_count			=  1;
-static int			inherit				=  1;
-static int			scale				=  1;
+static bool			inherit				=  true;
+static bool			scale				=  true;
 static pid_t			target_pid			= -1;
+static pid_t			target_tid			= -1;
+static pid_t			*all_tids			=  NULL;
+static int			thread_num			=  0;
 static pid_t			child_pid			= -1;
-static int			null_run			=  0;
+static bool			null_run			=  false;
 
-static int			fd[MAX_NR_CPUS][MAX_COUNTERS];
+static int			*fd[MAX_NR_CPUS][MAX_COUNTERS];
 
 static int			event_scaled[MAX_COUNTERS];
 
@@ -140,9 +144,11 @@ struct stats			runtime_branches_stats;
 #define ERR_PERF_OPEN \
 "Error: counter %d, sys_perf_event_open() syscall returned with %d (%s)\n"
 
-static void create_perf_stat_counter(int counter, int pid)
+static int create_perf_stat_counter(int counter)
 {
 	struct perf_event_attr *attr = attrs + counter;
+	int thread;
+	int ncreated = 0;
 
 	if (scale)
 		attr->read_format = PERF_FORMAT_TOTAL_TIME_ENABLED |
@@ -152,21 +158,33 @@ static void create_perf_stat_counter(int counter, int pid)
 		unsigned int cpu;
 
 		for (cpu = 0; cpu < nr_cpus; cpu++) {
-			fd[cpu][counter] = sys_perf_event_open(attr, -1, cpumap[cpu], -1, 0);
-			if (fd[cpu][counter] < 0 && verbose)
-				fprintf(stderr, ERR_PERF_OPEN, counter,
-					fd[cpu][counter], strerror(errno));
+			fd[cpu][counter][0] = sys_perf_event_open(attr,
+					-1, cpumap[cpu], -1, 0);
+			if (fd[cpu][counter][0] < 0)
+				pr_debug(ERR_PERF_OPEN, counter,
+					 fd[cpu][counter][0], strerror(errno));
+			else
+				++ncreated;
 		}
 	} else {
 		attr->inherit	     = inherit;
-		attr->disabled	     = 1;
-		attr->enable_on_exec = 1;
-
-		fd[0][counter] = sys_perf_event_open(attr, pid, -1, -1, 0);
-		if (fd[0][counter] < 0 && verbose)
-			fprintf(stderr, ERR_PERF_OPEN, counter,
-				fd[0][counter], strerror(errno));
+		if (target_pid == -1) {
+			attr->disabled = 1;
+			attr->enable_on_exec = 1;
+		}
+		for (thread = 0; thread < thread_num; thread++) {
+			fd[0][counter][thread] = sys_perf_event_open(attr,
+				all_tids[thread], -1, -1, 0);
+			if (fd[0][counter][thread] < 0)
+				pr_debug(ERR_PERF_OPEN, counter,
+					 fd[0][counter][thread],
+					 strerror(errno));
+			else
+				++ncreated;
+		}
 	}
+
+	return ncreated;
 }
 
 /*
@@ -190,25 +208,28 @@ static void read_counter(int counter)
 	unsigned int cpu;
 	size_t res, nv;
 	int scaled;
-	int i;
+	int i, thread;
 
 	count[0] = count[1] = count[2] = 0;
 
 	nv = scale ? 3 : 1;
 	for (cpu = 0; cpu < nr_cpus; cpu++) {
-		if (fd[cpu][counter] < 0)
-			continue;
-
-		res = read(fd[cpu][counter], single_count, nv * sizeof(u64));
-		assert(res == nv * sizeof(u64));
-
-		close(fd[cpu][counter]);
-		fd[cpu][counter] = -1;
-
-		count[0] += single_count[0];
-		if (scale) {
-			count[1] += single_count[1];
-			count[2] += single_count[2];
+		for (thread = 0; thread < thread_num; thread++) {
+			if (fd[cpu][counter][thread] < 0)
+				continue;
+
+			res = read(fd[cpu][counter][thread],
+					single_count, nv * sizeof(u64));
+			assert(res == nv * sizeof(u64));
+
+			close(fd[cpu][counter][thread]);
+			fd[cpu][counter][thread] = -1;
+
+			count[0] += single_count[0];
+			if (scale) {
+				count[1] += single_count[1];
+				count[2] += single_count[2];
+			}
 		}
 	}
 
@@ -250,10 +271,9 @@ static int run_perf_stat(int argc __used, const char **argv)
 {
 	unsigned long long t0, t1;
 	int status = 0;
-	int counter;
-	int pid = target_pid;
+	int counter, ncreated = 0;
 	int child_ready_pipe[2], go_pipe[2];
-	const bool forks = (target_pid == -1 && argc > 0);
+	const bool forks = (argc > 0);
 	char buf;
 
 	if (!system_wide)
@@ -265,10 +285,10 @@ static int run_perf_stat(int argc __used, const char **argv)
 	}
 
 	if (forks) {
-		if ((pid = fork()) < 0)
+		if ((child_pid = fork()) < 0)
 			perror("failed to fork");
 
-		if (!pid) {
+		if (!child_pid) {
 			close(child_ready_pipe[0]);
 			close(go_pipe[1]);
 			fcntl(go_pipe[0], F_SETFD, FD_CLOEXEC);
@@ -297,7 +317,8 @@ static int run_perf_stat(int argc __used, const char **argv)
 			exit(-1);
 		}
 
-		child_pid = pid;
+		if (target_tid == -1 && target_pid == -1 && !system_wide)
+			all_tids[0] = child_pid;
 
 		/*
 		 * Wait for the child to be ready to exec.
@@ -310,7 +331,16 @@ static int run_perf_stat(int argc __used, const char **argv)
 	}
 
 	for (counter = 0; counter < nr_counters; counter++)
-		create_perf_stat_counter(counter, pid);
+		ncreated += create_perf_stat_counter(counter);
+
+	if (ncreated == 0) {
+		pr_err("No permission to collect %sstats.\n"
+		       "Consider tweaking /proc/sys/kernel/perf_event_paranoid.\n",
+		       system_wide ? "system-wide " : "");
+		if (child_pid != -1)
+			kill(child_pid, SIGTERM);
+		return -1;
+	}
 
 	/*
 	 * Enable counters and exec the command:
@@ -321,7 +351,7 @@ static int run_perf_stat(int argc __used, const char **argv)
 		close(go_pipe[1]);
 		wait(&status);
 	} else {
-		while(!done);
+		while(!done) sleep(1);
 	}
 
 	t1 = rdclock();
@@ -429,12 +459,14 @@ static void print_stat(int argc, const char **argv)
 
 	fprintf(stderr, "\n");
 	fprintf(stderr, " Performance counter stats for ");
-	if(target_pid == -1) {
+	if(target_pid == -1 && target_tid == -1) {
 		fprintf(stderr, "\'%s", argv[0]);
 		for (i = 1; i < argc; i++)
 			fprintf(stderr, " %s", argv[i]);
-	}else
-		fprintf(stderr, "task pid \'%d", target_pid);
+	} else if (target_pid != -1)
+		fprintf(stderr, "process id \'%d", target_pid);
+	else
+		fprintf(stderr, "thread id \'%d", target_tid);
 
 	fprintf(stderr, "\'");
 	if (run_count > 1)
@@ -459,7 +491,7 @@ static volatile int signr = -1;
 
 static void skip_signal(int signo)
 {
-	if(target_pid != -1)
+	if(child_pid == -1)
 		done = 1;
 
 	signr = signo;
@@ -489,12 +521,14 @@ static const struct option options[] = {
 	OPT_BOOLEAN('i', "inherit", &inherit,
 		    "child tasks inherit counters"),
 	OPT_INTEGER('p', "pid", &target_pid,
-		    "stat events on existing pid"),
+		    "stat events on existing process id"),
+	OPT_INTEGER('t', "tid", &target_tid,
+		    "stat events on existing thread id"),
 	OPT_BOOLEAN('a', "all-cpus", &system_wide,
 		    "system-wide collection from all CPUs"),
 	OPT_BOOLEAN('c', "scale", &scale,
 		    "scale/normalize counters"),
-	OPT_BOOLEAN('v', "verbose", &verbose,
+	OPT_INCR('v', "verbose", &verbose,
 		    "be more verbose (show counter open errors, etc)"),
 	OPT_INTEGER('r', "repeat", &run_count,
 		    "repeat command and print average + stddev (max: 100)"),
@@ -506,10 +540,11 @@ static const struct option options[] = {
 int cmd_stat(int argc, const char **argv, const char *prefix __used)
 {
 	int status;
+	int i,j;
 
 	argc = parse_options(argc, argv, options, stat_usage,
 		PARSE_OPT_STOP_AT_NON_OPTION);
-	if (!argc && target_pid == -1)
+	if (!argc && target_pid == -1 && target_tid == -1)
 		usage_with_options(stat_usage, options);
 	if (run_count <= 0)
 		usage_with_options(stat_usage, options);
@@ -525,6 +560,31 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used)
 	else
 		nr_cpus = 1;
 
+	if (target_pid != -1) {
+		target_tid = target_pid;
+		thread_num = find_all_tid(target_pid, &all_tids);
+		if (thread_num <= 0) {
+			fprintf(stderr, "Can't find all threads of pid %d\n",
+					target_pid);
+			usage_with_options(stat_usage, options);
+		}
+	} else {
+		all_tids=malloc(sizeof(pid_t));
+		if (!all_tids)
+			return -ENOMEM;
+
+		all_tids[0] = target_tid;
+		thread_num = 1;
+	}
+
+	for (i = 0; i < MAX_NR_CPUS; i++) {
+		for (j = 0; j < MAX_COUNTERS; j++) {
+			fd[i][j] = malloc(sizeof(int)*thread_num);
+			if (!fd[i][j])
+				return -ENOMEM;
+		}
+	}
+
 	/*
 	 * We dont want to block the signals - that would cause
 	 * child tasks to inherit that and Ctrl-C would not work.
@@ -543,7 +603,8 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used)
 		status = run_perf_stat(argc, argv);
 	}
 
-	print_stat(argc, argv);
+	if (status != -1)
+		print_stat(argc, argv);
 
 	return status;
 }

tools/perf/builtin-timechart.c

@@ -21,7 +21,6 @@
 #include "util/cache.h"
 #include <linux/rbtree.h>
 #include "util/symbol.h"
-#include "util/string.h"
 #include "util/callchain.h"
 #include "util/strlist.h"
 
@@ -43,7 +42,7 @@ static u64		turbo_frequency;
 
 static u64		first_time, last_time;
 
-static int		power_only;
+static bool		power_only;
 
 
 struct per_pid;
@@ -78,8 +77,6 @@ struct per_pid {
 
 	struct per_pidcomm *all;
 	struct per_pidcomm *current;
-
-	int painted;
 };
 
 

tools/perf/builtin-top.c

@@ -55,9 +55,9 @@
 #include <linux/unistd.h>
 #include <linux/types.h>
 
-static int			fd[MAX_NR_CPUS][MAX_COUNTERS];
+static int			*fd[MAX_NR_CPUS][MAX_COUNTERS];
 
-static int			system_wide			=      0;
+static bool			system_wide			=  false;
 
 static int			default_interval		=      0;
 
@@ -65,18 +65,21 @@ static int			count_filter			=      5;
 static int			print_entries;
 
 static int			target_pid			=     -1;
-static int			inherit				=      0;
+static int			target_tid			=     -1;
+static pid_t			*all_tids			=      NULL;
+static int			thread_num			=      0;
+static bool			inherit				=  false;
 static int			profile_cpu			=     -1;
 static int			nr_cpus				=      0;
 static unsigned int		realtime_prio			=      0;
-static int			group				=      0;
+static bool			group				=  false;
 static unsigned int		page_size;
 static unsigned int		mmap_pages			=     16;
 static int			freq				=   1000; /* 1 KHz */
 
 static int			delay_secs			=      2;
-static int			zero                            =      0;
-static int			dump_symtab                     =      0;
+static bool			zero                            =  false;
+static bool			dump_symtab                     =  false;
 
 static bool			hide_kernel_symbols		=  false;
 static bool			hide_user_symbols		=  false;
@@ -133,7 +136,7 @@ static inline struct symbol *sym_entry__symbol(struct sym_entry *self)
        return ((void *)self) + symbol_conf.priv_size;
 }
 
-static void get_term_dimensions(struct winsize *ws)
+void get_term_dimensions(struct winsize *ws)
 {
 	char *s = getenv("LINES");
 
@@ -169,7 +172,7 @@ static void sig_winch_handler(int sig __used)
 	update_print_entries(&winsize);
 }
 
-static void parse_source(struct sym_entry *syme)
+static int parse_source(struct sym_entry *syme)
 {
 	struct symbol *sym;
 	struct sym_entry_source *source;
@@ -180,12 +183,21 @@ static void parse_source(struct sym_entry *syme)
 	u64 len;
 
 	if (!syme)
-		return;
+		return -1;
+
+	sym = sym_entry__symbol(syme);
+	map = syme->map;
+
+	/*
+	 * We can't annotate with just /proc/kallsyms
+	 */
+	if (map->dso->origin == DSO__ORIG_KERNEL)
+		return -1;
 
 	if (syme->src == NULL) {
 		syme->src = zalloc(sizeof(*source));
 		if (syme->src == NULL)
-			return;
+			return -1;
 		pthread_mutex_init(&syme->src->lock, NULL);
 	}
 
@@ -195,9 +207,6 @@ static void parse_source(struct sym_entry *syme)
 		pthread_mutex_lock(&source->lock);
 		goto out_assign;
 	}
-
-	sym = sym_entry__symbol(syme);
-	map = syme->map;
 	path = map->dso->long_name;
 
 	len = sym->end - sym->start;
@@ -209,7 +218,7 @@ static void parse_source(struct sym_entry *syme)
 
 	file = popen(command, "r");
 	if (!file)
-		return;
+		return -1;
 
 	pthread_mutex_lock(&source->lock);
 	source->lines_tail = &source->lines;
@@ -245,6 +254,7 @@ static void parse_source(struct sym_entry *syme)
 out_assign:
 	sym_filter_entry = syme;
 	pthread_mutex_unlock(&source->lock);
+	return 0;
 }
 
 static void __zero_source_counters(struct sym_entry *syme)
@@ -410,7 +420,9 @@ static double sym_weight(const struct sym_entry *sym)
 }
 
 static long			samples;
-static long			userspace_samples;
+static long			kernel_samples, us_samples;
+static long			exact_samples;
+static long			guest_us_samples, guest_kernel_samples;
 static const char		CONSOLE_CLEAR[] = "";
 
 static void __list_insert_active_sym(struct sym_entry *syme)
@@ -450,7 +462,11 @@ static void print_sym_table(void)
 	int printed = 0, j;
 	int counter, snap = !display_weighted ? sym_counter : 0;
 	float samples_per_sec = samples/delay_secs;
-	float ksamples_per_sec = (samples-userspace_samples)/delay_secs;
+	float ksamples_per_sec = kernel_samples/delay_secs;
+	float us_samples_per_sec = (us_samples)/delay_secs;
+	float guest_kernel_samples_per_sec = (guest_kernel_samples)/delay_secs;
+	float guest_us_samples_per_sec = (guest_us_samples)/delay_secs;
+	float esamples_percent = (100.0*exact_samples)/samples;
 	float sum_ksamples = 0.0;
 	struct sym_entry *syme, *n;
 	struct rb_root tmp = RB_ROOT;
@@ -458,7 +474,8 @@ static void print_sym_table(void)
 	int sym_width = 0, dso_width = 0, dso_short_width = 0;
 	const int win_width = winsize.ws_col - 1;
 
-	samples = userspace_samples = 0;
+	samples = us_samples = kernel_samples = exact_samples = 0;
+	guest_kernel_samples = guest_us_samples = 0;
 
 	/* Sort the active symbols */
 	pthread_mutex_lock(&active_symbols_lock);
@@ -489,9 +506,30 @@ static void print_sym_table(void)
 	puts(CONSOLE_CLEAR);
 
 	printf("%-*.*s\n", win_width, win_width, graph_dotted_line);
-	printf( "   PerfTop:%8.0f irqs/sec  kernel:%4.1f%% [",
-		samples_per_sec,
-		100.0 - (100.0*((samples_per_sec-ksamples_per_sec)/samples_per_sec)));
+	if (!perf_guest) {
+		printf("   PerfTop:%8.0f irqs/sec  kernel:%4.1f%%"
+			"  exact: %4.1f%% [",
+			samples_per_sec,
+			100.0 - (100.0 * ((samples_per_sec - ksamples_per_sec) /
+					 samples_per_sec)),
+			esamples_percent);
+	} else {
+		printf("   PerfTop:%8.0f irqs/sec  kernel:%4.1f%% us:%4.1f%%"
+			" guest kernel:%4.1f%% guest us:%4.1f%%"
+			" exact: %4.1f%% [",
+			samples_per_sec,
+			100.0 - (100.0 * ((samples_per_sec-ksamples_per_sec) /
+					  samples_per_sec)),
+			100.0 - (100.0 * ((samples_per_sec-us_samples_per_sec) /
+					  samples_per_sec)),
+			100.0 - (100.0 * ((samples_per_sec -
+						guest_kernel_samples_per_sec) /
+					  samples_per_sec)),
+			100.0 - (100.0 * ((samples_per_sec -
+					   guest_us_samples_per_sec) /
+					  samples_per_sec)),
+			esamples_percent);
+	}
 
 	if (nr_counters == 1 || !display_weighted) {
 		printf("%Ld", (u64)attrs[0].sample_period);
@@ -514,13 +552,15 @@ static void print_sym_table(void)
 
 	if (target_pid != -1)
 		printf(" (target_pid: %d", target_pid);
+	else if (target_tid != -1)
+		printf(" (target_tid: %d", target_tid);
 	else
 		printf(" (all");
 
 	if (profile_cpu != -1)
 		printf(", cpu: %d)\n", profile_cpu);
 	else {
-		if (target_pid != -1)
+		if (target_tid != -1)
 			printf(")\n");
 		else
 			printf(", %d CPUs)\n", nr_cpus);
@@ -582,7 +622,6 @@ static void print_sym_table(void)
 
 		syme = rb_entry(nd, struct sym_entry, rb_node);
 		sym = sym_entry__symbol(syme);
-
 		if (++printed > print_entries || (int)syme->snap_count < count_filter)
 			continue;
 
@@ -746,7 +785,7 @@ static int key_mapped(int c)
 	return 0;
 }
 
-static void handle_keypress(int c)
+static void handle_keypress(struct perf_session *session, int c)
 {
 	if (!key_mapped(c)) {
 		struct pollfd stdin_poll = { .fd = 0, .events = POLLIN };
@@ -815,7 +854,7 @@ static void handle_keypress(int c)
 		case 'Q':
 			printf("exiting.\n");
 			if (dump_symtab)
-				dsos__fprintf(stderr);
+				dsos__fprintf(&session->kerninfo_root, stderr);
 			exit(0);
 		case 's':
 			prompt_symbol(&sym_filter_entry, "Enter details symbol");
@@ -839,7 +878,7 @@ static void handle_keypress(int c)
 			display_weighted = ~display_weighted;
 			break;
 		case 'z':
-			zero = ~zero;
+			zero = !zero;
 			break;
 		default:
 			break;
@@ -851,6 +890,7 @@ static void *display_thread(void *arg __used)
 	struct pollfd stdin_poll = { .fd = 0, .events = POLLIN };
 	struct termios tc, save;
 	int delay_msecs, c;
+	struct perf_session *session = (struct perf_session *) arg;
 
 	tcgetattr(0, &save);
 	tc = save;
@@ -871,7 +911,7 @@ repeat:
 	c = getc(stdin);
 	tcsetattr(0, TCSAFLUSH, &save);
 
-	handle_keypress(c);
+	handle_keypress(session, c);
 	goto repeat;
 
 	return NULL;
@@ -942,24 +982,49 @@ static void event__process_sample(const event_t *self,
 	u64 ip = self->ip.ip;
 	struct sym_entry *syme;
 	struct addr_location al;
+	struct kernel_info *kerninfo;
 	u8 origin = self->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
 
 	++samples;
 
 	switch (origin) {
 	case PERF_RECORD_MISC_USER:
-		++userspace_samples;
+		++us_samples;
 		if (hide_user_symbols)
 			return;
+		kerninfo = kerninfo__findhost(&session->kerninfo_root);
 		break;
 	case PERF_RECORD_MISC_KERNEL:
+		++kernel_samples;
 		if (hide_kernel_symbols)
 			return;
+		kerninfo = kerninfo__findhost(&session->kerninfo_root);
+		break;
+	case PERF_RECORD_MISC_GUEST_KERNEL:
+		++guest_kernel_samples;
+		kerninfo = kerninfo__find(&session->kerninfo_root,
+					  self->ip.pid);
 		break;
+	case PERF_RECORD_MISC_GUEST_USER:
+		++guest_us_samples;
+		/*
+		 * TODO: we don't process guest user from host side
+		 * except simple counting.
+		 */
+		return;
 	default:
 		return;
 	}
 
+	if (!kerninfo && perf_guest) {
+		pr_err("Can't find guest [%d]'s kernel information\n",
+			self->ip.pid);
+		return;
+	}
+
+	if (self->header.misc & PERF_RECORD_MISC_EXACT)
+		exact_samples++;
+
 	if (event__preprocess_sample(self, session, &al, symbol_filter) < 0 ||
 	    al.filtered)
 		return;
@@ -976,7 +1041,7 @@ static void event__process_sample(const event_t *self,
 		 * --hide-kernel-symbols, even if the user specifies an
 		 * invalid --vmlinux ;-)
 		 */
-		if (al.map == session->vmlinux_maps[MAP__FUNCTION] &&
+		if (al.map == kerninfo->vmlinux_maps[MAP__FUNCTION] &&
 		    RB_EMPTY_ROOT(&al.map->dso->symbols[MAP__FUNCTION])) {
 			pr_err("The %s file can't be used\n",
 			       symbol_conf.vmlinux_name);
@@ -990,7 +1055,17 @@ static void event__process_sample(const event_t *self,
 	if (sym_filter_entry_sched) {
 		sym_filter_entry = sym_filter_entry_sched;
 		sym_filter_entry_sched = NULL;
-		parse_source(sym_filter_entry);
+		if (parse_source(sym_filter_entry) < 0) {
+			struct symbol *sym = sym_entry__symbol(sym_filter_entry);
+
+			pr_err("Can't annotate %s", sym->name);
+			if (sym_filter_entry->map->dso->origin == DSO__ORIG_KERNEL) {
+				pr_err(": No vmlinux file was found in the path:\n");
+				vmlinux_path__fprintf(stderr);
+			} else
+				pr_err(".\n");
+			exit(1);
+		}
 	}
 
 	syme = symbol__priv(al.sym);
@@ -1106,16 +1181,21 @@ static void perf_session__mmap_read_counter(struct perf_session *self,
 	md->prev = old;
 }
 
-static struct pollfd event_array[MAX_NR_CPUS * MAX_COUNTERS];
-static struct mmap_data mmap_array[MAX_NR_CPUS][MAX_COUNTERS];
+static struct pollfd *event_array;
+static struct mmap_data *mmap_array[MAX_NR_CPUS][MAX_COUNTERS];
 
 static void perf_session__mmap_read(struct perf_session *self)
 {
-	int i, counter;
+	int i, counter, thread_index;
 
 	for (i = 0; i < nr_cpus; i++) {
 		for (counter = 0; counter < nr_counters; counter++)
-			perf_session__mmap_read_counter(self, &mmap_array[i][counter]);
+			for (thread_index = 0;
+				thread_index < thread_num;
+				thread_index++) {
+				perf_session__mmap_read_counter(self,
+					&mmap_array[i][counter][thread_index]);
+			}
 	}
 }
 
@@ -1126,9 +1206,10 @@ static void start_counter(int i, int counter)
 {
 	struct perf_event_attr *attr;
 	int cpu;
+	int thread_index;
 
 	cpu = profile_cpu;
-	if (target_pid == -1 && profile_cpu == -1)
+	if (target_tid == -1 && profile_cpu == -1)
 		cpu = cpumap[i];
 
 	attr = attrs + counter;
@@ -1144,55 +1225,58 @@ static void start_counter(int i, int counter)
 	attr->inherit		= (cpu < 0) && inherit;
 	attr->mmap		= 1;
 
+	for (thread_index = 0; thread_index < thread_num; thread_index++) {
 try_again:
-	fd[i][counter] = sys_perf_event_open(attr, target_pid, cpu, group_fd, 0);
-
-	if (fd[i][counter] < 0) {
-		int err = errno;
+		fd[i][counter][thread_index] = sys_perf_event_open(attr,
+				all_tids[thread_index], cpu, group_fd, 0);
+
+		if (fd[i][counter][thread_index] < 0) {
+			int err = errno;
+
+			if (err == EPERM || err == EACCES)
+				die("No permission - are you root?\n");
+			/*
+			 * If it's cycles then fall back to hrtimer
+			 * based cpu-clock-tick sw counter, which
+			 * is always available even if no PMU support:
+			 */
+			if (attr->type == PERF_TYPE_HARDWARE
+					&& attr->config == PERF_COUNT_HW_CPU_CYCLES) {
+
+				if (verbose)
+					warning(" ... trying to fall back to cpu-clock-ticks\n");
+
+				attr->type = PERF_TYPE_SOFTWARE;
+				attr->config = PERF_COUNT_SW_CPU_CLOCK;
+				goto try_again;
+			}
+			printf("\n");
+			error("perfcounter syscall returned with %d (%s)\n",
+					fd[i][counter][thread_index], strerror(err));
+			die("No CONFIG_PERF_EVENTS=y kernel support configured?\n");
+			exit(-1);
+		}
+		assert(fd[i][counter][thread_index] >= 0);
+		fcntl(fd[i][counter][thread_index], F_SETFL, O_NONBLOCK);
 
-		if (err == EPERM || err == EACCES)
-			die("No permission - are you root?\n");
 		/*
-		 * If it's cycles then fall back to hrtimer
-		 * based cpu-clock-tick sw counter, which
-		 * is always available even if no PMU support:
+		 * First counter acts as the group leader:
 		 */
-		if (attr->type == PERF_TYPE_HARDWARE
-			&& attr->config == PERF_COUNT_HW_CPU_CYCLES) {
-
-			if (verbose)
-				warning(" ... trying to fall back to cpu-clock-ticks\n");
-
-			attr->type = PERF_TYPE_SOFTWARE;
-			attr->config = PERF_COUNT_SW_CPU_CLOCK;
-			goto try_again;
-		}
-		printf("\n");
-		error("perfcounter syscall returned with %d (%s)\n",
-			fd[i][counter], strerror(err));
-		die("No CONFIG_PERF_EVENTS=y kernel support configured?\n");
-		exit(-1);
+		if (group && group_fd == -1)
+			group_fd = fd[i][counter][thread_index];
+
+		event_array[nr_poll].fd = fd[i][counter][thread_index];
+		event_array[nr_poll].events = POLLIN;
+		nr_poll++;
+
+		mmap_array[i][counter][thread_index].counter = counter;
+		mmap_array[i][counter][thread_index].prev = 0;
+		mmap_array[i][counter][thread_index].mask = mmap_pages*page_size - 1;
+		mmap_array[i][counter][thread_index].base = mmap(NULL, (mmap_pages+1)*page_size,
+				PROT_READ, MAP_SHARED, fd[i][counter][thread_index], 0);
+		if (mmap_array[i][counter][thread_index].base == MAP_FAILED)
+			die("failed to mmap with %d (%s)\n", errno, strerror(errno));
 	}
-	assert(fd[i][counter] >= 0);
-	fcntl(fd[i][counter], F_SETFL, O_NONBLOCK);
-
-	/*
-	 * First counter acts as the group leader:
-	 */
-	if (group && group_fd == -1)
-		group_fd = fd[i][counter];
-
-	event_array[nr_poll].fd = fd[i][counter];
-	event_array[nr_poll].events = POLLIN;
-	nr_poll++;
-
-	mmap_array[i][counter].counter = counter;
-	mmap_array[i][counter].prev = 0;
-	mmap_array[i][counter].mask = mmap_pages*page_size - 1;
-	mmap_array[i][counter].base = mmap(NULL, (mmap_pages+1)*page_size,
-			PROT_READ, MAP_SHARED, fd[i][counter], 0);
-	if (mmap_array[i][counter].base == MAP_FAILED)
-		die("failed to mmap with %d (%s)\n", errno, strerror(errno));
 }
 
 static int __cmd_top(void)
@@ -1208,8 +1292,8 @@ static int __cmd_top(void)
 	if (session == NULL)
 		return -ENOMEM;
 
-	if (target_pid != -1)
-		event__synthesize_thread(target_pid, event__process, session);
+	if (target_tid != -1)
+		event__synthesize_thread(target_tid, event__process, session);
 	else
 		event__synthesize_threads(event__process, session);
 
@@ -1220,11 +1304,11 @@ static int __cmd_top(void)
 	}
 
 	/* Wait for a minimal set of events before starting the snapshot */
-	poll(event_array, nr_poll, 100);
+	poll(&event_array[0], nr_poll, 100);
 
 	perf_session__mmap_read(session);
 
-	if (pthread_create(&thread, NULL, display_thread, NULL)) {
+	if (pthread_create(&thread, NULL, display_thread, session)) {
 		printf("Could not create display thread.\n");
 		exit(-1);
 	}
@@ -1263,7 +1347,9 @@ static const struct option options[] = {
 	OPT_INTEGER('c', "count", &default_interval,
 		    "event period to sample"),
 	OPT_INTEGER('p', "pid", &target_pid,
-		    "profile events on existing pid"),
+		    "profile events on existing process id"),
+	OPT_INTEGER('t', "tid", &target_tid,
+		    "profile events on existing thread id"),
 	OPT_BOOLEAN('a', "all-cpus", &system_wide,
 			    "system-wide collection from all CPUs"),
 	OPT_INTEGER('C', "CPU", &profile_cpu,
@@ -1296,7 +1382,7 @@ static const struct option options[] = {
 		    "display this many functions"),
 	OPT_BOOLEAN('U', "hide_user_symbols", &hide_user_symbols,
 		    "hide user symbols"),
-	OPT_BOOLEAN('v', "verbose", &verbose,
+	OPT_INCR('v', "verbose", &verbose,
 		    "be more verbose (show counter open errors, etc)"),
 	OPT_END()
 };
@@ -1304,6 +1390,7 @@ static const struct option options[] = {
 int cmd_top(int argc, const char **argv, const char *prefix __used)
 {
 	int counter;
+	int i,j;
 
 	page_size = sysconf(_SC_PAGE_SIZE);
 
@@ -1311,8 +1398,39 @@ int cmd_top(int argc, const char **argv, const char *prefix __used)
 	if (argc)
 		usage_with_options(top_usage, options);
 
+	if (target_pid != -1) {
+		target_tid = target_pid;
+		thread_num = find_all_tid(target_pid, &all_tids);
+		if (thread_num <= 0) {
+			fprintf(stderr, "Can't find all threads of pid %d\n",
+				target_pid);
+			usage_with_options(top_usage, options);
+		}
+	} else {
+		all_tids=malloc(sizeof(pid_t));
+		if (!all_tids)
+			return -ENOMEM;
+
+		all_tids[0] = target_tid;
+		thread_num = 1;
+	}
+
+	for (i = 0; i < MAX_NR_CPUS; i++) {
+		for (j = 0; j < MAX_COUNTERS; j++) {
+			fd[i][j] = malloc(sizeof(int)*thread_num);
+			mmap_array[i][j] = zalloc(
+				sizeof(struct mmap_data)*thread_num);
+			if (!fd[i][j] || !mmap_array[i][j])
+				return -ENOMEM;
+		}
+	}
+	event_array = malloc(
+		sizeof(struct pollfd)*MAX_NR_CPUS*MAX_COUNTERS*thread_num);
+	if (!event_array)
+		return -ENOMEM;
+
 	/* CPU and PID are mutually exclusive */
-	if (target_pid != -1 && profile_cpu != -1) {
+	if (target_tid > 0 && profile_cpu != -1) {
 		printf("WARNING: PID switch overriding CPU\n");
 		sleep(1);
 		profile_cpu = -1;
@@ -1353,7 +1471,7 @@ int cmd_top(int argc, const char **argv, const char *prefix __used)
 		attrs[counter].sample_period = default_interval;
 	}
 
-	if (target_pid != -1 || profile_cpu != -1)
+	if (target_tid != -1 || profile_cpu != -1)
 		nr_cpus = 1;
 	else
 		nr_cpus = read_cpu_map();

tools/perf/builtin-trace.c

@@ -104,10 +104,23 @@ static int process_sample_event(event_t *event, struct perf_session *session)
 static struct perf_event_ops event_ops = {
 	.sample	= process_sample_event,
 	.comm	= event__process_comm,
+	.attr	= event__process_attr,
+	.event_type = event__process_event_type,
+	.tracing_data = event__process_tracing_data,
+	.build_id = event__process_build_id,
 };
 
+extern volatile int session_done;
+
+static void sig_handler(int sig __unused)
+{
+	session_done = 1;
+}
+
 static int __cmd_trace(struct perf_session *session)
 {
+	signal(SIGINT, sig_handler);
+
 	return perf_session__process_events(session, &event_ops);
 }
 
@@ -505,7 +518,7 @@ static const char * const trace_usage[] = {
 static const struct option options[] = {
 	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
 		    "dump raw trace in ASCII"),
-	OPT_BOOLEAN('v', "verbose", &verbose,
+	OPT_INCR('v', "verbose", &verbose,
 		    "be more verbose (show symbol address, etc)"),
 	OPT_BOOLEAN('L', "Latency", &latency_format,
 		    "show latency attributes (irqs/preemption disabled, etc)"),
@@ -548,6 +561,65 @@ int cmd_trace(int argc, const char **argv, const char *prefix __used)
 		suffix = REPORT_SUFFIX;
 	}
 
+	if (!suffix && argc >= 2 && strncmp(argv[1], "-", strlen("-")) != 0) {
+		char *record_script_path, *report_script_path;
+		int live_pipe[2];
+		pid_t pid;
+
+		record_script_path = get_script_path(argv[1], RECORD_SUFFIX);
+		if (!record_script_path) {
+			fprintf(stderr, "record script not found\n");
+			return -1;
+		}
+
+		report_script_path = get_script_path(argv[1], REPORT_SUFFIX);
+		if (!report_script_path) {
+			fprintf(stderr, "report script not found\n");
+			return -1;
+		}
+
+		if (pipe(live_pipe) < 0) {
+			perror("failed to create pipe");
+			exit(-1);
+		}
+
+		pid = fork();
+		if (pid < 0) {
+			perror("failed to fork");
+			exit(-1);
+		}
+
+		if (!pid) {
+			dup2(live_pipe[1], 1);
+			close(live_pipe[0]);
+
+			__argv = malloc(5 * sizeof(const char *));
+			__argv[0] = "/bin/sh";
+			__argv[1] = record_script_path;
+			__argv[2] = "-o";
+			__argv[3] = "-";
+			__argv[4] = NULL;
+
+			execvp("/bin/sh", (char **)__argv);
+			exit(-1);
+		}
+
+		dup2(live_pipe[0], 0);
+		close(live_pipe[1]);
+
+		__argv = malloc((argc + 3) * sizeof(const char *));
+		__argv[0] = "/bin/sh";
+		__argv[1] = report_script_path;
+		for (i = 2; i < argc; i++)
+			__argv[i] = argv[i];
+		__argv[i++] = "-i";
+		__argv[i++] = "-";
+		__argv[i++] = NULL;
+
+		execvp("/bin/sh", (char **)__argv);
+		exit(-1);
+	}
+
 	if (suffix) {
 		script_path = get_script_path(argv[2], suffix);
 		if (!script_path) {
@@ -580,7 +652,8 @@ int cmd_trace(int argc, const char **argv, const char *prefix __used)
 	if (session == NULL)
 		return -ENOMEM;
 
-	if (!perf_session__has_traces(session, "record -R"))
+	if (strcmp(input_name, "-") &&
+	    !perf_session__has_traces(session, "record -R"))
 		return -EINVAL;
 
 	if (generate_script_lang) {

tools/perf/builtin.h

@@ -32,5 +32,6 @@ extern int cmd_version(int argc, const char **argv, const char *prefix);
 extern int cmd_probe(int argc, const char **argv, const char *prefix);
 extern int cmd_kmem(int argc, const char **argv, const char *prefix);
 extern int cmd_lock(int argc, const char **argv, const char *prefix);
+extern int cmd_kvm(int argc, const char **argv, const char *prefix);
 
 #endif

tools/perf/command-list.txt

@@ -19,3 +19,4 @@ perf-trace			mainporcelain common
 perf-probe			mainporcelain common
 perf-kmem			mainporcelain common
 perf-lock			mainporcelain common
+perf-kvm			mainporcelain common

tools/perf/perf-archive.sh

@@ -30,4 +30,7 @@ done
 
 tar cfj $PERF_DATA.tar.bz2 -C $DEBUGDIR -T $MANIFEST
 rm -f $MANIFEST $BUILDIDS
+echo -e "Now please run:\n"
+echo -e "$ tar xvf $PERF_DATA.tar.bz2 -C ~/.debug\n"
+echo "wherever you need to run 'perf report' on."
 exit 0

tools/perf/perf.c

@@ -13,9 +13,10 @@
 #include "util/quote.h"
 #include "util/run-command.h"
 #include "util/parse-events.h"
-#include "util/string.h"
 #include "util/debugfs.h"
 
+bool use_browser;
+
 const char perf_usage_string[] =
 	"perf [--version] [--help] COMMAND [ARGS]";
 
@@ -262,6 +263,8 @@ static int run_builtin(struct cmd_struct *p, int argc, const char **argv)
 	set_debugfs_path();
 
 	status = p->fn(argc, argv, prefix);
+	exit_browser(status);
+
 	if (status)
 		return status & 0xff;
 
@@ -304,6 +307,7 @@ static void handle_internal_command(int argc, const char **argv)
 		{ "probe",	cmd_probe,	0 },
 		{ "kmem",	cmd_kmem,	0 },
 		{ "lock",	cmd_lock,	0 },
+		{ "kvm",	cmd_kvm,	0 },
 	};
 	unsigned int i;
 	static const char ext[] = STRIP_EXTENSION;

tools/perf/perf.h

@@ -1,6 +1,10 @@
 #ifndef _PERF_PERF_H
 #define _PERF_PERF_H
 
+struct winsize;
+
+void get_term_dimensions(struct winsize *ws);
+
 #if defined(__i386__)
 #include "../../arch/x86/include/asm/unistd.h"
 #define rmb()		asm volatile("lock; addl $0,0(%%esp)" ::: "memory")
@@ -102,8 +106,6 @@ static inline unsigned long long rdclock(void)
 #define __user
 #define asmlinkage
 
-#define __used		__attribute__((__unused__))
-
 #define unlikely(x)	__builtin_expect(!!(x), 0)
 #define min(x, y) ({				\
 	typeof(x) _min1 = (x);			\
@@ -129,4 +131,6 @@ struct ip_callchain {
 	u64 ips[0];
 };
 
+extern int perf_host, perf_guest;
+
 #endif

tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Util.pm

@@ -15,6 +15,7 @@ our @EXPORT_OK = ( @{ $EXPORT_TAGS{'all'} } );
 
 our @EXPORT = qw(
 avg nsecs nsecs_secs nsecs_nsecs nsecs_usecs print_nsecs
+clear_term
 );
 
 our $VERSION = '0.01';
@@ -55,6 +56,11 @@ sub nsecs_str {
     return $str;
 }
 
+sub clear_term
+{
+    print "\x1b[H\x1b[2J";
+}
+
 1;
 __END__
 =head1 NAME

tools/perf/scripts/perl/bin/failed-syscalls-record

@@ -1,2 +1,2 @@
 #!/bin/bash
-perf record -c 1 -f -a -M -R -e raw_syscalls:sys_exit
+perf record -c 1 -f -a -M -R -e raw_syscalls:sys_exit $@

tools/perf/scripts/perl/bin/failed-syscalls-report

@@ -1,4 +1,10 @@
 #!/bin/bash
 # description: system-wide failed syscalls
 # args: [comm]
-perf trace -s ~/libexec/perf-core/scripts/perl/failed-syscalls.pl $1
+if [ $# -gt 0 ] ; then
+    if ! expr match "$1" "-" ; then
+	comm=$1
+	shift
+    fi
+fi
+perf trace $@ -s ~/libexec/perf-core/scripts/perl/failed-syscalls.pl $comm

tools/perf/scripts/perl/bin/rw-by-file-record

@@ -1,2 +1,3 @@
 #!/bin/bash
-perf record -c 1 -f -a -M -R -e syscalls:sys_enter_read -e syscalls:sys_enter_write
+perf record -c 1 -f -a -M -R -e syscalls:sys_enter_read -e syscalls:sys_enter_write $@
+

tools/perf/scripts/perl/bin/rw-by-file-report

@@ -1,7 +1,13 @@
 #!/bin/bash
 # description: r/w activity for a program, by file
 # args: <comm>
-perf trace -s ~/libexec/perf-core/scripts/perl/rw-by-file.pl $1
+if [ $# -lt 1 ] ; then
+    echo "usage: rw-by-file <comm>"
+    exit
+fi
+comm=$1
+shift
+perf trace $@ -s ~/libexec/perf-core/scripts/perl/rw-by-file.pl $comm
 
 
 

tools/perf/scripts/perl/bin/rw-by-pid-record

@@ -1,2 +1,2 @@
 #!/bin/bash
-perf record -c 1 -f -a -M -R -e syscalls:sys_enter_read -e syscalls:sys_exit_read -e syscalls:sys_enter_write -e syscalls:sys_exit_write
+perf record -c 1 -f -a -M -R -e syscalls:sys_enter_read -e syscalls:sys_exit_read -e syscalls:sys_enter_write -e syscalls:sys_exit_write $@

tools/perf/scripts/perl/bin/rw-by-pid-report

@@ -1,6 +1,6 @@
 #!/bin/bash
 # description: system-wide r/w activity
-perf trace -s ~/libexec/perf-core/scripts/perl/rw-by-pid.pl
+perf trace $@ -s ~/libexec/perf-core/scripts/perl/rw-by-pid.pl
 
 
 

tools/perf/scripts/perl/bin/rwtop-record

@@ -0,0 +1,2 @@
+#!/bin/bash
+perf record -c 1 -f -a -M -R -e syscalls:sys_enter_read -e syscalls:sys_exit_read -e syscalls:sys_enter_write -e syscalls:sys_exit_write $@

tools/perf/scripts/perl/bin/rwtop-report

@@ -0,0 +1,23 @@
+#!/bin/bash
+# description: system-wide r/w top
+# args: [interval]
+n_args=0
+for i in "$@"
+do
+    if expr match "$i" "-" > /dev/null ; then
+	break
+    fi
+    n_args=$(( $n_args + 1 ))
+done
+if [ "$n_args" -gt 1 ] ; then
+    echo "usage: rwtop-report [interval]"
+    exit
+fi
+if [ "$n_args" -gt 0 ] ; then
+    interval=$1
+    shift
+fi
+perf trace $@ -s ~/libexec/perf-core/scripts/perl/rwtop.pl $interval
+
+
+