Merge branch 'oprofile/ring_buffer' into oprofile/oprofile-for-tip (d2852b93) · Commits · EulixOS / Software / Kernel

arch/Kconfig

+2 −0

Original line number	Diff line number	Diff line
		@@ -6,6 +6,8 @@ config OPROFILE
		tristate "OProfile system profiling (EXPERIMENTAL)"
		depends on PROFILING
		depends on HAVE_OPROFILE
		select TRACING
		select RING_BUFFER
		help
		OProfile is a profiling system capable of profiling the
		whole system, include the kernel, kernel modules, libraries,

arch/x86/oprofile/op_model_amd.c

+91 −133

Original line number	Diff line number	Diff line
		@@ -2,7 +2,7 @@
		* @file op_model_amd.c
		* athlon / K7 / K8 / Family 10h model-specific MSR operations
		*
		* @remark Copyright 2002-2008 OProfile authors
		* @remark Copyright 2002-2009 OProfile authors
		* @remark Read the file COPYING
		*
		* @author John Levon
		@@ -60,56 +60,10 @@ static unsigned long reset_value[NUM_COUNTERS];
		#define IBS_OP_LOW_VALID_BIT (1ULL<<18) /* bit 18 */
		#define IBS_OP_LOW_ENABLE (1ULL<<17) /* bit 17 */

		/* Codes used in cpu_buffer.c */
		/* This produces duplicate code, need to be fixed */
		#define IBS_FETCH_BEGIN 3
		#define IBS_OP_BEGIN 4

		/* The function interface needs to be fixed, something like add
		data. Should then be added to linux/oprofile.h. */
		extern void
		oprofile_add_ibs_sample(struct pt_regs *const regs,
		unsigned int *const ibs_sample, int ibs_code);

		struct ibs_fetch_sample {
		/* MSRC001_1031 IBS Fetch Linear Address Register */
		unsigned int ibs_fetch_lin_addr_low;
		unsigned int ibs_fetch_lin_addr_high;
		/* MSRC001_1030 IBS Fetch Control Register */
		unsigned int ibs_fetch_ctl_low;
		unsigned int ibs_fetch_ctl_high;
		/* MSRC001_1032 IBS Fetch Physical Address Register */
		unsigned int ibs_fetch_phys_addr_low;
		unsigned int ibs_fetch_phys_addr_high;
		};
		#define IBS_FETCH_SIZE 6
		#define IBS_OP_SIZE 12

		struct ibs_op_sample {
		/* MSRC001_1034 IBS Op Logical Address Register (IbsRIP) */
		unsigned int ibs_op_rip_low;
		unsigned int ibs_op_rip_high;
		/* MSRC001_1035 IBS Op Data Register */
		unsigned int ibs_op_data1_low;
		unsigned int ibs_op_data1_high;
		/* MSRC001_1036 IBS Op Data 2 Register */
		unsigned int ibs_op_data2_low;
		unsigned int ibs_op_data2_high;
		/* MSRC001_1037 IBS Op Data 3 Register */
		unsigned int ibs_op_data3_low;
		unsigned int ibs_op_data3_high;
		/* MSRC001_1038 IBS DC Linear Address Register (IbsDcLinAd) */
		unsigned int ibs_dc_linear_low;
		unsigned int ibs_dc_linear_high;
		/* MSRC001_1039 IBS DC Physical Address Register (IbsDcPhysAd) */
		unsigned int ibs_dc_phys_low;
		unsigned int ibs_dc_phys_high;
		};

		/*
		* unitialize the APIC for the IBS interrupts if needed on AMD Family10h+
		*/
		static void clear_ibs_nmi(void);

		static int ibs_allowed; /* AMD Family10h and later */
		static int has_ibs; /* AMD Family10h and later */

		struct op_ibs_config {
		unsigned long op_enabled;
		@@ -200,31 +154,29 @@ static inline int
		op_amd_handle_ibs(struct pt_regs * const regs,
		struct op_msrs const * const msrs)
		{
		unsigned int low, high;
		struct ibs_fetch_sample ibs_fetch;
		struct ibs_op_sample ibs_op;
		u32 low, high;
		u64 msr;
		struct op_entry entry;

		if (!ibs_allowed)
		if (!has_ibs)
		return 1;

		if (ibs_config.fetch_enabled) {
		rdmsr(MSR_AMD64_IBSFETCHCTL, low, high);
		if (high & IBS_FETCH_HIGH_VALID_BIT) {
		ibs_fetch.ibs_fetch_ctl_high = high;
		ibs_fetch.ibs_fetch_ctl_low = low;
		rdmsr(MSR_AMD64_IBSFETCHLINAD, low, high);
		ibs_fetch.ibs_fetch_lin_addr_high = high;
		ibs_fetch.ibs_fetch_lin_addr_low = low;
		rdmsr(MSR_AMD64_IBSFETCHPHYSAD, low, high);
		ibs_fetch.ibs_fetch_phys_addr_high = high;
		ibs_fetch.ibs_fetch_phys_addr_low = low;

		oprofile_add_ibs_sample(regs,
		(unsigned int *)&ibs_fetch,
		IBS_FETCH_BEGIN);
		rdmsrl(MSR_AMD64_IBSFETCHLINAD, msr);
		oprofile_write_reserve(&entry, regs, msr,
		IBS_FETCH_CODE, IBS_FETCH_SIZE);
		oprofile_add_data(&entry, (u32)msr);
		oprofile_add_data(&entry, (u32)(msr >> 32));
		oprofile_add_data(&entry, low);
		oprofile_add_data(&entry, high);
		rdmsrl(MSR_AMD64_IBSFETCHPHYSAD, msr);
		oprofile_add_data(&entry, (u32)msr);
		oprofile_add_data(&entry, (u32)(msr >> 32));
		oprofile_write_commit(&entry);

		/* reenable the IRQ */
		rdmsr(MSR_AMD64_IBSFETCHCTL, low, high);
		high &= ~IBS_FETCH_HIGH_VALID_BIT;
		high \|= IBS_FETCH_HIGH_ENABLE;
		low &= IBS_FETCH_LOW_MAX_CNT_MASK;
		@@ -235,30 +187,29 @@ op_amd_handle_ibs(struct pt_regs * const regs,
		if (ibs_config.op_enabled) {
		rdmsr(MSR_AMD64_IBSOPCTL, low, high);
		if (low & IBS_OP_LOW_VALID_BIT) {
		rdmsr(MSR_AMD64_IBSOPRIP, low, high);
		ibs_op.ibs_op_rip_low = low;
		ibs_op.ibs_op_rip_high = high;
		rdmsr(MSR_AMD64_IBSOPDATA, low, high);
		ibs_op.ibs_op_data1_low = low;
		ibs_op.ibs_op_data1_high = high;
		rdmsr(MSR_AMD64_IBSOPDATA2, low, high);
		ibs_op.ibs_op_data2_low = low;
		ibs_op.ibs_op_data2_high = high;
		rdmsr(MSR_AMD64_IBSOPDATA3, low, high);
		ibs_op.ibs_op_data3_low = low;
		ibs_op.ibs_op_data3_high = high;
		rdmsr(MSR_AMD64_IBSDCLINAD, low, high);
		ibs_op.ibs_dc_linear_low = low;
		ibs_op.ibs_dc_linear_high = high;
		rdmsr(MSR_AMD64_IBSDCPHYSAD, low, high);
		ibs_op.ibs_dc_phys_low = low;
		ibs_op.ibs_dc_phys_high = high;
		rdmsrl(MSR_AMD64_IBSOPRIP, msr);
		oprofile_write_reserve(&entry, regs, msr,
		IBS_OP_CODE, IBS_OP_SIZE);
		oprofile_add_data(&entry, (u32)msr);
		oprofile_add_data(&entry, (u32)(msr >> 32));
		rdmsrl(MSR_AMD64_IBSOPDATA, msr);
		oprofile_add_data(&entry, (u32)msr);
		oprofile_add_data(&entry, (u32)(msr >> 32));
		rdmsrl(MSR_AMD64_IBSOPDATA2, msr);
		oprofile_add_data(&entry, (u32)msr);
		oprofile_add_data(&entry, (u32)(msr >> 32));
		rdmsrl(MSR_AMD64_IBSOPDATA3, msr);
		oprofile_add_data(&entry, (u32)msr);
		oprofile_add_data(&entry, (u32)(msr >> 32));
		rdmsrl(MSR_AMD64_IBSDCLINAD, msr);
		oprofile_add_data(&entry, (u32)msr);
		oprofile_add_data(&entry, (u32)(msr >> 32));
		rdmsrl(MSR_AMD64_IBSDCPHYSAD, msr);
		oprofile_add_data(&entry, (u32)msr);
		oprofile_add_data(&entry, (u32)(msr >> 32));
		oprofile_write_commit(&entry);

		/* reenable the IRQ */
		oprofile_add_ibs_sample(regs,
		(unsigned int *)&ibs_op,
		IBS_OP_BEGIN);
		rdmsr(MSR_AMD64_IBSOPCTL, low, high);
		high = 0;
		low &= ~IBS_OP_LOW_VALID_BIT;
		low \|= IBS_OP_LOW_ENABLE;
		@@ -308,14 +259,14 @@ static void op_amd_start(struct op_msrs const * const msrs)
		}

		#ifdef CONFIG_OPROFILE_IBS
		if (ibs_allowed && ibs_config.fetch_enabled) {
		if (has_ibs && ibs_config.fetch_enabled) {
		low = (ibs_config.max_cnt_fetch >> 4) & 0xFFFF;
		high = ((ibs_config.rand_en & 0x1) << 25) /* bit 57 */
		+ IBS_FETCH_HIGH_ENABLE;
		wrmsr(MSR_AMD64_IBSFETCHCTL, low, high);
		}

		if (ibs_allowed && ibs_config.op_enabled) {
		if (has_ibs && ibs_config.op_enabled) {
		low = ((ibs_config.max_cnt_op >> 4) & 0xFFFF)
		+ ((ibs_config.dispatched_ops & 0x1) << 19) /* bit 19 */
		+ IBS_OP_LOW_ENABLE;
		@@ -331,8 +282,10 @@ static void op_amd_stop(struct op_msrs const * const msrs)
		unsigned int low, high;
		int i;

		/* Subtle: stop on all counters to avoid race with
		* setting our pm callback */
		/*
		* Subtle: stop on all counters to avoid race with setting our
		* pm callback
		*/
		for (i = 0 ; i < NUM_COUNTERS ; ++i) {
		if (!reset_value[i])
		continue;
		@@ -342,14 +295,16 @@ static void op_amd_stop(struct op_msrs const * const msrs)
		}

		#ifdef CONFIG_OPROFILE_IBS
		if (ibs_allowed && ibs_config.fetch_enabled) {
		low = 0; /* clear max count and enable */
		if (has_ibs && ibs_config.fetch_enabled) {
		/* clear max count and enable */
		low = 0;
		high = 0;
		wrmsr(MSR_AMD64_IBSFETCHCTL, low, high);
		}

		if (ibs_allowed && ibs_config.op_enabled) {
		low = 0; /* clear max count and enable */
		if (has_ibs && ibs_config.op_enabled) {
		/* clear max count and enable */
		low = 0;
		high = 0;
		wrmsr(MSR_AMD64_IBSOPCTL, low, high);
		}
		@@ -370,18 +325,7 @@ static void op_amd_shutdown(struct op_msrs const * const msrs)
		}
		}

		#ifndef CONFIG_OPROFILE_IBS

		/* no IBS support */

		static int op_amd_init(struct oprofile_operations *ops)
		{
		return 0;
		}

		static void op_amd_exit(void) {}

		#else
		#ifdef CONFIG_OPROFILE_IBS

		static u8 ibs_eilvt_off;

		@@ -395,7 +339,7 @@ static inline void apic_clear_ibs_nmi_per_cpu(void *arg)
		setup_APIC_eilvt_ibs(0, APIC_EILVT_MSG_FIX, 1);
		}

		static int pfm_amd64_setup_eilvt(void)
		static int init_ibs_nmi(void)
		{
		#define IBSCTL_LVTOFFSETVAL (1 << 8)
		#define IBSCTL 0x1cc
		@@ -419,6 +363,7 @@ static int pfm_amd64_setup_eilvt(void)
		\| IBSCTL_LVTOFFSETVAL);
		pci_read_config_dword(cpu_cfg, IBSCTL, &value);
		if (value != (ibs_eilvt_off \| IBSCTL_LVTOFFSETVAL)) {
		pci_dev_put(cpu_cfg);
		printk(KERN_DEBUG "Failed to setup IBS LVT offset, "
		"IBSCTL = 0x%08x", value);
		return 1;
		@@ -443,33 +388,35 @@ static int pfm_amd64_setup_eilvt(void)
		return 0;
		}

		/*
		* initialize the APIC for the IBS interrupts
		* if available (AMD Family10h rev B0 and later)
		*/
		static void setup_ibs(void)
		/* uninitialize the APIC for the IBS interrupts if needed */
		static void clear_ibs_nmi(void)
		{
		ibs_allowed = boot_cpu_has(X86_FEATURE_IBS);
		if (has_ibs)
		on_each_cpu(apic_clear_ibs_nmi_per_cpu, NULL, 1);
		}

		if (!ibs_allowed)
		/* initialize the APIC for the IBS interrupts if available */
		static void ibs_init(void)
		{
		has_ibs = boot_cpu_has(X86_FEATURE_IBS);

		if (!has_ibs)
		return;

		if (pfm_amd64_setup_eilvt()) {
		ibs_allowed = 0;
		if (init_ibs_nmi()) {
		has_ibs = 0;
		return;
		}

		printk(KERN_INFO "oprofile: AMD IBS detected\n");
		}


		/*
		* unitialize the APIC for the IBS interrupts if needed on AMD Family10h
		* rev B0 and later */
		static void clear_ibs_nmi(void)
		static void ibs_exit(void)
		{
		if (ibs_allowed)
		on_each_cpu(apic_clear_ibs_nmi_per_cpu, NULL, 1);
		if (!has_ibs)
		return;

		clear_ibs_nmi();
		}

		static int (create_arch_files)(struct super_block sb, struct dentry *root);
		@@ -486,7 +433,7 @@ static int setup_ibs_files(struct super_block sb, struct dentry root)
		if (ret)
		return ret;

		if (!ibs_allowed)
		if (!has_ibs)
		return ret;

		/* model specific files */
		@@ -519,7 +466,7 @@ static int setup_ibs_files(struct super_block sb, struct dentry root)

		static int op_amd_init(struct oprofile_operations *ops)
		{
		setup_ibs();
		ibs_init();
		create_arch_files = ops->create_files;
		ops->create_files = setup_ibs_files;
		return 0;
		@@ -527,10 +474,21 @@ static int op_amd_init(struct oprofile_operations *ops)

		static void op_amd_exit(void)
		{
		clear_ibs_nmi();
		ibs_exit();
		}

		#endif
		#else

		/* no IBS support */

		static int op_amd_init(struct oprofile_operations *ops)
		{
		return 0;
		}

		static void op_amd_exit(void) {}

		#endif /* CONFIG_OPROFILE_IBS */

		struct op_x86_model_spec const op_amd_spec = {
		.init = op_amd_init,

drivers/oprofile/buffer_sync.c

+81 −148

Original line number	Diff line number	Diff line
		/**
		* @file buffer_sync.c
		*
		* @remark Copyright 2002 OProfile authors
		* @remark Copyright 2002-2009 OProfile authors
		* @remark Read the file COPYING
		*
		* @author John Levon <levon@movementarian.org>
		* @author Barry Kasindorf
		* @author Robert Richter <robert.richter@amd.com>
		*
		* This is the core of the buffer management. Each
		* CPU buffer is processed and entered into the
		@@ -268,18 +269,6 @@ lookup_dcookie(struct mm_struct mm, unsigned long addr, off_t offset)
		return cookie;
		}

		static void increment_tail(struct oprofile_cpu_buffer *b)
		{
		unsigned long new_tail = b->tail_pos + 1;

		rmb(); /* be sure fifo pointers are synchromized */

		if (new_tail < b->buffer_size)
		b->tail_pos = new_tail;
		else
		b->tail_pos = 0;
		}

		static unsigned long last_cookie = INVALID_COOKIE;

		static void add_cpu_switch(int i)
		@@ -327,84 +316,73 @@ static void add_trace_begin(void)
		add_event_entry(TRACE_BEGIN_CODE);
		}

		#ifdef CONFIG_OPROFILE_IBS

		#define IBS_FETCH_CODE_SIZE 2
		#define IBS_OP_CODE_SIZE 5
		#define IBS_EIP(offset) \
		(((struct op_sample *)&cpu_buf->buffer[(offset)])->eip)
		#define IBS_EVENT(offset) \
		(((struct op_sample *)&cpu_buf->buffer[(offset)])->event)

		/*
		* Add IBS fetch and op entries to event buffer
		*/
		static void add_ibs_begin(struct oprofile_cpu_buffer *cpu_buf, int code,
		struct mm_struct *mm)
		static void add_data(struct op_entry entry, struct mm_struct mm)
		{
		unsigned long rip;
		int i, count;
		unsigned long ibs_cookie = 0;
		unsigned long code, pc, val;
		unsigned long cookie;
		off_t offset;

		increment_tail(cpu_buf); /* move to RIP entry */

		rip = IBS_EIP(cpu_buf->tail_pos);

		#ifdef __LP64__
		rip += IBS_EVENT(cpu_buf->tail_pos) << 32;
		#endif
		if (!op_cpu_buffer_get_data(entry, &code))
		return;
		if (!op_cpu_buffer_get_data(entry, &pc))
		return;
		if (!op_cpu_buffer_get_size(entry))
		return;

		if (mm) {
		ibs_cookie = lookup_dcookie(mm, rip, &offset);
		cookie = lookup_dcookie(mm, pc, &offset);

		if (ibs_cookie == NO_COOKIE)
		offset = rip;
		if (ibs_cookie == INVALID_COOKIE) {
		if (cookie == NO_COOKIE)
		offset = pc;
		if (cookie == INVALID_COOKIE) {
		atomic_inc(&oprofile_stats.sample_lost_no_mapping);
		offset = rip;
		offset = pc;
		}
		if (ibs_cookie != last_cookie) {
		add_cookie_switch(ibs_cookie);
		last_cookie = ibs_cookie;
		if (cookie != last_cookie) {
		add_cookie_switch(cookie);
		last_cookie = cookie;
		}
		} else
		offset = rip;
		offset = pc;

		add_event_entry(ESCAPE_CODE);
		add_event_entry(code);
		add_event_entry(offset); /* Offset from Dcookie */

		/* we send the Dcookie offset, but send the raw Linear Add also*/
		add_event_entry(IBS_EIP(cpu_buf->tail_pos));
		add_event_entry(IBS_EVENT(cpu_buf->tail_pos));

		if (code == IBS_FETCH_CODE)
		count = IBS_FETCH_CODE_SIZE; /IBS FETCH is 2 int64s/
		else
		count = IBS_OP_CODE_SIZE; /IBS OP is 5 int64s/

		for (i = 0; i < count; i++) {
		increment_tail(cpu_buf);
		add_event_entry(IBS_EIP(cpu_buf->tail_pos));
		add_event_entry(IBS_EVENT(cpu_buf->tail_pos));
		}
		while (op_cpu_buffer_get_data(entry, &val))
		add_event_entry(val);
		}

		#endif

		static void add_sample_entry(unsigned long offset, unsigned long event)
		static inline void add_sample_entry(unsigned long offset, unsigned long event)
		{
		add_event_entry(offset);
		add_event_entry(event);
		}


		static int add_us_sample(struct mm_struct mm, struct op_sample s)
		/*
		* Add a sample to the global event buffer. If possible the
		* sample is converted into a persistent dentry/offset pair
		* for later lookup from userspace. Return 0 on failure.
		*/
		static int
		add_sample(struct mm_struct mm, struct op_sample s, int in_kernel)
		{
		unsigned long cookie;
		off_t offset;

		if (in_kernel) {
		add_sample_entry(s->eip, s->event);
		return 1;
		}

		/* add userspace sample */

		if (!mm) {
		atomic_inc(&oprofile_stats.sample_lost_no_mm);
		return 0;
		}

		cookie = lookup_dcookie(mm, s->eip, &offset);

		if (cookie == INVALID_COOKIE) {
		@@ -423,25 +401,6 @@ static int add_us_sample(struct mm_struct mm, struct op_sample s)
		}


		/* Add a sample to the global event buffer. If possible the
		* sample is converted into a persistent dentry/offset pair
		* for later lookup from userspace.
		*/
		static int
		add_sample(struct mm_struct mm, struct op_sample s, int in_kernel)
		{
		if (in_kernel) {
		add_sample_entry(s->eip, s->event);
		return 1;
		} else if (mm) {
		return add_us_sample(mm, s);
		} else {
		atomic_inc(&oprofile_stats.sample_lost_no_mm);
		}
		return 0;
		}


		static void release_mm(struct mm_struct *mm)
		{
		if (!mm)
		@@ -466,33 +425,6 @@ static inline int is_code(unsigned long val)
		}


		/* "acquire" as many cpu buffer slots as we can */
		static unsigned long get_slots(struct oprofile_cpu_buffer *b)
		{
		unsigned long head = b->head_pos;
		unsigned long tail = b->tail_pos;

		/*
		* Subtle. This resets the persistent last_task
		* and in_kernel values used for switching notes.
		* BUT, there is a small window between reading
		* head_pos, and this call, that means samples
		* can appear at the new head position, but not
		* be prefixed with the notes for switching
		* kernel mode or a task switch. This small hole
		* can lead to mis-attribution or samples where
		* we don't know if it's in the kernel or not,
		* at the start of an event buffer.
		*/
		cpu_buffer_reset(b);

		if (head >= tail)
		return head - tail;

		return head + (b->buffer_size - tail);
		}


		/* Move tasks along towards death. Any tasks on dead_tasks
		* will definitely have no remaining references in any
		* CPU buffers at this point, because we use two lists,
		@@ -559,72 +491,73 @@ typedef enum {
		*/
		void sync_buffer(int cpu)
		{
		struct oprofile_cpu_buffer *cpu_buf = &per_cpu(cpu_buffer, cpu);
		struct mm_struct *mm = NULL;
		struct mm_struct *oldmm;
		unsigned long val;
		struct task_struct *new;
		unsigned long cookie = 0;
		int in_kernel = 1;
		sync_buffer_state state = sb_buffer_start;
		#ifndef CONFIG_OPROFILE_IBS
		unsigned int i;
		unsigned long available;
		#endif
		unsigned long flags;
		struct op_entry entry;
		struct op_sample *sample;

		mutex_lock(&buffer_mutex);

		add_cpu_switch(cpu);

		/* Remember, only we can modify tail_pos */

		#ifndef CONFIG_OPROFILE_IBS
		available = get_slots(cpu_buf);
		op_cpu_buffer_reset(cpu);
		available = op_cpu_buffer_entries(cpu);

		for (i = 0; i < available; ++i) {
		#else
		while (get_slots(cpu_buf)) {
		#endif
		struct op_sample *s = &cpu_buf->buffer[cpu_buf->tail_pos];
		sample = op_cpu_buffer_read_entry(&entry, cpu);
		if (!sample)
		break;

		if (is_code(s->eip)) {
		if (s->event <= CPU_IS_KERNEL) {
		if (is_code(sample->eip)) {
		flags = sample->event;
		if (flags & TRACE_BEGIN) {
		state = sb_bt_start;
		add_trace_begin();
		}
		if (flags & KERNEL_CTX_SWITCH) {
		/* kernel/userspace switch */
		in_kernel = s->event;
		in_kernel = flags & IS_KERNEL;
		if (state == sb_buffer_start)
		state = sb_sample_start;
		add_kernel_ctx_switch(s->event);
		} else if (s->event == CPU_TRACE_BEGIN) {
		state = sb_bt_start;
		add_trace_begin();
		#ifdef CONFIG_OPROFILE_IBS
		} else if (s->event == IBS_FETCH_BEGIN) {
		state = sb_bt_start;
		add_ibs_begin(cpu_buf, IBS_FETCH_CODE, mm);
		} else if (s->event == IBS_OP_BEGIN) {
		state = sb_bt_start;
		add_ibs_begin(cpu_buf, IBS_OP_CODE, mm);
		#endif
		} else {
		struct mm_struct *oldmm = mm;

		add_kernel_ctx_switch(flags & IS_KERNEL);
		}
		if (flags & USER_CTX_SWITCH
		&& op_cpu_buffer_get_data(&entry, &val)) {
		/* userspace context switch */
		new = (struct task_struct *)s->event;

		new = (struct task_struct *)val;
		oldmm = mm;
		release_mm(oldmm);
		mm = take_tasks_mm(new);
		if (mm != oldmm)
		cookie = get_exec_dcookie(mm);
		add_user_ctx_switch(new, cookie);
		}
		} else if (state >= sb_bt_start &&
		!add_sample(mm, s, in_kernel)) {
		if (op_cpu_buffer_get_size(&entry))
		add_data(&entry, mm);
		continue;
		}

		if (state < sb_bt_start)
		/* ignore sample */
		continue;

		if (add_sample(mm, sample, in_kernel))
		continue;

		/* ignore backtraces if failed to add a sample */
		if (state == sb_bt_start) {
		state = sb_bt_ignore;
		atomic_inc(&oprofile_stats.bt_lost_no_mapping);
		}
		}

		increment_tail(cpu_buf);
		}
		release_mm(mm);

		mark_done(cpu);

drivers/oprofile/cpu_buffer.c

+240 −153

File changed.

Preview size limit exceeded, changes collapsed.

drivers/oprofile/cpu_buffer.h

+62 −10

Original line number	Diff line number	Diff line
		/**
		* @file cpu_buffer.h
		*
		* @remark Copyright 2002 OProfile authors
		* @remark Copyright 2002-2009 OProfile authors
		* @remark Read the file COPYING
		*
		* @author John Levon <levon@movementarian.org>
		* @author Robert Richter <robert.richter@amd.com>
		*/

		#ifndef OPROFILE_CPU_BUFFER_H
		@@ -15,6 +16,7 @@
		#include <linux/workqueue.h>
		#include <linux/cache.h>
		#include <linux/sched.h>
		#include <linux/ring_buffer.h>

		struct task_struct;

		@@ -30,16 +32,16 @@ void end_cpu_work(void);
		struct op_sample {
		unsigned long eip;
		unsigned long event;
		unsigned long data[0];
		};

		struct op_entry;

		struct oprofile_cpu_buffer {
		volatile unsigned long head_pos;
		volatile unsigned long tail_pos;
		unsigned long buffer_size;
		struct task_struct *last_task;
		int last_is_kernel;
		int tracing;
		struct op_sample *buffer;
		unsigned long sample_received;
		unsigned long sample_lost_overflow;
		unsigned long backtrace_aborted;
		@@ -50,12 +52,62 @@ struct oprofile_cpu_buffer {

		DECLARE_PER_CPU(struct oprofile_cpu_buffer, cpu_buffer);

		void cpu_buffer_reset(struct oprofile_cpu_buffer *cpu_buf);
		/*
		* Resets the cpu buffer to a sane state.
		*
		* reset these to invalid values; the next sample collected will
		* populate the buffer with proper values to initialize the buffer
		*/
		static inline void op_cpu_buffer_reset(int cpu)
		{
		struct oprofile_cpu_buffer *cpu_buf = &per_cpu(cpu_buffer, cpu);

		cpu_buf->last_is_kernel = -1;
		cpu_buf->last_task = NULL;
		}

		struct op_sample
		op_cpu_buffer_write_reserve(struct op_entry entry, unsigned long size);
		int op_cpu_buffer_write_commit(struct op_entry *entry);
		struct op_sample op_cpu_buffer_read_entry(struct op_entry entry, int cpu);
		unsigned long op_cpu_buffer_entries(int cpu);

		/* returns the remaining free size of data in the entry */
		static inline
		int op_cpu_buffer_add_data(struct op_entry *entry, unsigned long val)
		{
		if (!entry->size)
		return 0;
		*entry->data = val;
		entry->size--;
		entry->data++;
		return entry->size;
		}

		/* returns the size of data in the entry */
		static inline
		int op_cpu_buffer_get_size(struct op_entry *entry)
		{
		return entry->size;
		}

		/* returns 0 if empty or the size of data including the current value */
		static inline
		int op_cpu_buffer_get_data(struct op_entry entry, unsigned long val)
		{
		int size = entry->size;
		if (!size)
		return 0;
		val = entry->data;
		entry->size--;
		entry->data++;
		return size;
		}

		/* transient events for the CPU buffer -> event buffer */
		#define CPU_IS_KERNEL 1
		#define CPU_TRACE_BEGIN 2
		#define IBS_FETCH_BEGIN 3
		#define IBS_OP_BEGIN 4
		/* extra data flags */
		#define KERNEL_CTX_SWITCH (1UL << 0)
		#define IS_KERNEL (1UL << 1)
		#define TRACE_BEGIN (1UL << 2)
		#define USER_CTX_SWITCH (1UL << 3)

		#endif /* OPROFILE_CPU_BUFFER_H */