Merge branch 'perfcounters-core-for-linus' of...

Merge branch 'perfcounters-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip

* 'perfcounters-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (60 commits)
  perf tools: Avoid unnecessary work in directory lookups
  perf stat: Clean up statistics calculations a bit more
  perf stat: More advanced variance computation
  perf stat: Use stddev_mean in stead of stddev
  perf stat: Remove the limit on repeat
  perf stat: Change noise calculation to use stddev
  x86, perf_counter, bts: Do not allow kernel BTS tracing for now
  x86, perf_counter, bts: Correct pointer-to-u64 casts
  x86, perf_counter, bts: Fail if BTS is not available
  perf_counter: Fix output-sharing error path
  perf trace: Fix read_string()
  perf trace: Print out in nanoseconds
  perf tools: Seek to the end of the header area
  perf trace: Fix parsing of perf.data
  perf trace: Sample timestamps as well
  perf_counter: Introduce new (non-)paranoia level to allow raw tracepoint access
  perf trace: Sample the CPU too
  perf tools: Work around strict aliasing related warnings
  perf tools: Clean up warnings list in the Makefile
  perf tools: Complete support for dynamic strings
  ...

arch/powerpc/include/asm/pgtable.h

@@ -104,8 +104,8 @@ static inline void __set_pte_at(struct mm_struct *mm, unsigned long addr,
 	else
 		pte_update(ptep, ~_PAGE_HASHPTE, pte_val(pte));
 
-#elif defined(CONFIG_PPC32) && defined(CONFIG_PTE_64BIT) && defined(CONFIG_SMP)
-	/* Second case is 32-bit with 64-bit PTE in SMP mode. In this case, we
+#elif defined(CONFIG_PPC32) && defined(CONFIG_PTE_64BIT)
+	/* Second case is 32-bit with 64-bit PTE.  In this case, we
 	 * can just store as long as we do the two halves in the right order
 	 * with a barrier in between. This is possible because we take care,
 	 * in the hash code, to pre-invalidate if the PTE was already hashed,
@@ -140,7 +140,7 @@ static inline void __set_pte_at(struct mm_struct *mm, unsigned long addr,
 
 #else
 	/* Anything else just stores the PTE normally. That covers all 64-bit
-	 * cases, and 32-bit non-hash with 64-bit PTEs in UP mode
+	 * cases, and 32-bit non-hash with 32-bit PTEs.
 	 */
 	*ptep = pte;
 #endif

arch/powerpc/kernel/Makefile

@@ -97,7 +97,7 @@ obj64-$(CONFIG_AUDIT)		+= compat_audit.o
 
 obj-$(CONFIG_DYNAMIC_FTRACE)	+= ftrace.o
 obj-$(CONFIG_FUNCTION_GRAPH_TRACER)	+= ftrace.o
-obj-$(CONFIG_PPC_PERF_CTRS)	+= perf_counter.o
+obj-$(CONFIG_PPC_PERF_CTRS)	+= perf_counter.o perf_callchain.o
 obj64-$(CONFIG_PPC_PERF_CTRS)	+= power4-pmu.o ppc970-pmu.o power5-pmu.o \
 				   power5+-pmu.o power6-pmu.o power7-pmu.o
 obj32-$(CONFIG_PPC_PERF_CTRS)	+= mpc7450-pmu.o

arch/powerpc/kernel/asm-offsets.c

@@ -67,6 +67,8 @@ int main(void)
 	DEFINE(MMCONTEXTID, offsetof(struct mm_struct, context.id));
 #ifdef CONFIG_PPC64
 	DEFINE(AUDITCONTEXT, offsetof(struct task_struct, audit_context));
+	DEFINE(SIGSEGV, SIGSEGV);
+	DEFINE(NMI_MASK, NMI_MASK);
 #else
 	DEFINE(THREAD_INFO, offsetof(struct task_struct, stack));
 #endif /* CONFIG_PPC64 */

arch/powerpc/kernel/exceptions-64s.S

@@ -729,6 +729,11 @@ BEGIN_FTR_SECTION
 	bne-	do_ste_alloc		/* If so handle it */
 END_FTR_SECTION_IFCLR(CPU_FTR_SLB)
 
+	clrrdi	r11,r1,THREAD_SHIFT
+	lwz	r0,TI_PREEMPT(r11)	/* If we're in an "NMI" */
+	andis.	r0,r0,NMI_MASK@h	/* (i.e. an irq when soft-disabled) */
+	bne	77f			/* then don't call hash_page now */
+
 	/*
 	 * On iSeries, we soft-disable interrupts here, then
 	 * hard-enable interrupts so that the hash_page code can spin on
@@ -833,6 +838,20 @@ handle_page_fault:
 	bl	.low_hash_fault
 	b	.ret_from_except
 
+/*
+ * We come here as a result of a DSI at a point where we don't want
+ * to call hash_page, such as when we are accessing memory (possibly
+ * user memory) inside a PMU interrupt that occurred while interrupts
+ * were soft-disabled.  We want to invoke the exception handler for
+ * the access, or panic if there isn't a handler.
+ */
+77:	bl	.save_nvgprs
+	mr	r4,r3
+	addi	r3,r1,STACK_FRAME_OVERHEAD
+	li	r5,SIGSEGV
+	bl	.bad_page_fault
+	b	.ret_from_except
+
 	/* here we have a segment miss */
 do_ste_alloc:
 	bl	.ste_allocate		/* try to insert stab entry */

arch/powerpc/mm/slb.c

@@ -92,15 +92,13 @@ static inline void create_shadowed_slbe(unsigned long ea, int ssize,
 		     : "memory" );
 }
 
-void slb_flush_and_rebolt(void)
+static void __slb_flush_and_rebolt(void)
 {
 	/* If you change this make sure you change SLB_NUM_BOLTED
 	 * appropriately too. */
 	unsigned long linear_llp, vmalloc_llp, lflags, vflags;
 	unsigned long ksp_esid_data, ksp_vsid_data;
 
-	WARN_ON(!irqs_disabled());
-
 	linear_llp = mmu_psize_defs[mmu_linear_psize].sllp;
 	vmalloc_llp = mmu_psize_defs[mmu_vmalloc_psize].sllp;
 	lflags = SLB_VSID_KERNEL | linear_llp;
@@ -117,12 +115,6 @@ void slb_flush_and_rebolt(void)
 		ksp_vsid_data = get_slb_shadow()->save_area[2].vsid;
 	}
 
-	/*
-	 * We can't take a PMU exception in the following code, so hard
-	 * disable interrupts.
-	 */
-	hard_irq_disable();
-
 	/* We need to do this all in asm, so we're sure we don't touch
 	 * the stack between the slbia and rebolting it. */
 	asm volatile("isync\n"
@@ -139,6 +131,21 @@ void slb_flush_and_rebolt(void)
 		     : "memory");
 }
 
+void slb_flush_and_rebolt(void)
+{
+
+	WARN_ON(!irqs_disabled());
+
+	/*
+	 * We can't take a PMU exception in the following code, so hard
+	 * disable interrupts.
+	 */
+	hard_irq_disable();
+
+	__slb_flush_and_rebolt();
+	get_paca()->slb_cache_ptr = 0;
+}
+
 void slb_vmalloc_update(void)
 {
 	unsigned long vflags;
@@ -180,12 +187,20 @@ static inline int esids_match(unsigned long addr1, unsigned long addr2)
 /* Flush all user entries from the segment table of the current processor. */
 void switch_slb(struct task_struct *tsk, struct mm_struct *mm)
 {
-	unsigned long offset = get_paca()->slb_cache_ptr;
+	unsigned long offset;
 	unsigned long slbie_data = 0;
 	unsigned long pc = KSTK_EIP(tsk);
 	unsigned long stack = KSTK_ESP(tsk);
 	unsigned long unmapped_base;
 
+	/*
+	 * We need interrupts hard-disabled here, not just soft-disabled,
+	 * so that a PMU interrupt can't occur, which might try to access
+	 * user memory (to get a stack trace) and possible cause an SLB miss
+	 * which would update the slb_cache/slb_cache_ptr fields in the PACA.
+	 */
+	hard_irq_disable();
+	offset = get_paca()->slb_cache_ptr;
 	if (!cpu_has_feature(CPU_FTR_NO_SLBIE_B) &&
 	    offset <= SLB_CACHE_ENTRIES) {
 		int i;
@@ -200,7 +215,7 @@ void switch_slb(struct task_struct *tsk, struct mm_struct *mm)
 		}
 		asm volatile("isync" : : : "memory");
 	} else {
-		slb_flush_and_rebolt();
+		__slb_flush_and_rebolt();
 	}
 
 	/* Workaround POWER5 < DD2.1 issue */

arch/powerpc/mm/stab.c

@@ -164,7 +164,7 @@ void switch_stab(struct task_struct *tsk, struct mm_struct *mm)
 {
 	struct stab_entry *stab = (struct stab_entry *) get_paca()->stab_addr;
 	struct stab_entry *ste;
-	unsigned long offset = __get_cpu_var(stab_cache_ptr);
+	unsigned long offset;
 	unsigned long pc = KSTK_EIP(tsk);
 	unsigned long stack = KSTK_ESP(tsk);
 	unsigned long unmapped_base;
@@ -172,6 +172,15 @@ void switch_stab(struct task_struct *tsk, struct mm_struct *mm)
 	/* Force previous translations to complete. DRENG */
 	asm volatile("isync" : : : "memory");
 
+	/*
+	 * We need interrupts hard-disabled here, not just soft-disabled,
+	 * so that a PMU interrupt can't occur, which might try to access
+	 * user memory (to get a stack trace) and possible cause an STAB miss
+	 * which would update the stab_cache/stab_cache_ptr per-cpu variables.
+	 */
+	hard_irq_disable();
+
+	offset = __get_cpu_var(stab_cache_ptr);
 	if (offset <= NR_STAB_CACHE_ENTRIES) {
 		int i;
 

arch/x86/include/asm/perf_counter.h

@@ -84,6 +84,16 @@ union cpuid10_edx {
 #define MSR_ARCH_PERFMON_FIXED_CTR2			0x30b
 #define X86_PMC_IDX_FIXED_BUS_CYCLES			(X86_PMC_IDX_FIXED + 2)
 
+/*
+ * We model BTS tracing as another fixed-mode PMC.
+ *
+ * We choose a value in the middle of the fixed counter range, since lower
+ * values are used by actual fixed counters and higher values are used
+ * to indicate other overflow conditions in the PERF_GLOBAL_STATUS msr.
+ */
+#define X86_PMC_IDX_FIXED_BTS				(X86_PMC_IDX_FIXED + 16)
+
+
 #ifdef CONFIG_PERF_COUNTERS
 extern void init_hw_perf_counters(void);
 extern void perf_counters_lapic_init(void);

include/linux/perf_counter.h

@@ -216,6 +216,7 @@ struct perf_counter_attr {
 #define PERF_COUNTER_IOC_REFRESH	_IO ('$', 2)
 #define PERF_COUNTER_IOC_RESET		_IO ('$', 3)
 #define PERF_COUNTER_IOC_PERIOD		_IOW('$', 4, u64)
+#define PERF_COUNTER_IOC_SET_OUTPUT	_IO ('$', 5)
 
 enum perf_counter_ioc_flags {
 	PERF_IOC_FLAG_GROUP		= 1U << 0,
@@ -415,6 +416,9 @@ enum perf_callchain_context {
 	PERF_CONTEXT_MAX		= (__u64)-4095,
 };
 
+#define PERF_FLAG_FD_NO_GROUP	(1U << 0)
+#define PERF_FLAG_FD_OUTPUT	(1U << 1)
+
 #ifdef __KERNEL__
 /*
  * Kernel-internal data types and definitions:
@@ -536,6 +540,7 @@ struct perf_counter {
 	struct list_head		sibling_list;
 	int				nr_siblings;
 	struct perf_counter		*group_leader;
+	struct perf_counter		*output;
 	const struct pmu		*pmu;
 
 	enum perf_counter_active_state	state;

kernel/perf_counter.c

@@ -46,12 +46,18 @@ static atomic_t nr_task_counters __read_mostly;
 
 /*
  * perf counter paranoia level:
- *  0 - not paranoid
- *  1 - disallow cpu counters to unpriv
- *  2 - disallow kernel profiling to unpriv
+ *  -1 - not paranoid at all
+ *   0 - disallow raw tracepoint access for unpriv
+ *   1 - disallow cpu counters for unpriv
+ *   2 - disallow kernel profiling for unpriv
  */
 int sysctl_perf_counter_paranoid __read_mostly = 1;
 
+static inline bool perf_paranoid_tracepoint_raw(void)
+{
+	return sysctl_perf_counter_paranoid > -1;
+}
+
 static inline bool perf_paranoid_cpu(void)
 {
 	return sysctl_perf_counter_paranoid > 0;
@@ -469,7 +475,8 @@ static void update_counter_times(struct perf_counter *counter)
 	struct perf_counter_context *ctx = counter->ctx;
 	u64 run_end;
 
-	if (counter->state < PERF_COUNTER_STATE_INACTIVE)
+	if (counter->state < PERF_COUNTER_STATE_INACTIVE ||
+	    counter->group_leader->state < PERF_COUNTER_STATE_INACTIVE)
 		return;
 
 	counter->total_time_enabled = ctx->time - counter->tstamp_enabled;
@@ -518,7 +525,7 @@ static void __perf_counter_disable(void *info)
 	 */
 	if (counter->state >= PERF_COUNTER_STATE_INACTIVE) {
 		update_context_time(ctx);
-		update_counter_times(counter);
+		update_group_times(counter);
 		if (counter == counter->group_leader)
 			group_sched_out(counter, cpuctx, ctx);
 		else
@@ -573,7 +580,7 @@ static void perf_counter_disable(struct perf_counter *counter)
 	 * in, so we can change the state safely.
 	 */
 	if (counter->state == PERF_COUNTER_STATE_INACTIVE) {
-		update_counter_times(counter);
+		update_group_times(counter);
 		counter->state = PERF_COUNTER_STATE_OFF;
 	}
 
@@ -850,6 +857,27 @@ perf_install_in_context(struct perf_counter_context *ctx,
 	spin_unlock_irq(&ctx->lock);
 }
 
+/*
+ * Put a counter into inactive state and update time fields.
+ * Enabling the leader of a group effectively enables all
+ * the group members that aren't explicitly disabled, so we
+ * have to update their ->tstamp_enabled also.
+ * Note: this works for group members as well as group leaders
+ * since the non-leader members' sibling_lists will be empty.
+ */
+static void __perf_counter_mark_enabled(struct perf_counter *counter,
+					struct perf_counter_context *ctx)
+{
+	struct perf_counter *sub;
+
+	counter->state = PERF_COUNTER_STATE_INACTIVE;
+	counter->tstamp_enabled = ctx->time - counter->total_time_enabled;
+	list_for_each_entry(sub, &counter->sibling_list, list_entry)
+		if (sub->state >= PERF_COUNTER_STATE_INACTIVE)
+			sub->tstamp_enabled =
+				ctx->time - sub->total_time_enabled;
+}
+
 /*
  * Cross CPU call to enable a performance counter
  */
@@ -877,8 +905,7 @@ static void __perf_counter_enable(void *info)
 
 	if (counter->state >= PERF_COUNTER_STATE_INACTIVE)
 		goto unlock;
-	counter->state = PERF_COUNTER_STATE_INACTIVE;
-	counter->tstamp_enabled = ctx->time - counter->total_time_enabled;
+	__perf_counter_mark_enabled(counter, ctx);
 
 	/*
 	 * If the counter is in a group and isn't the group leader,
@@ -971,11 +998,9 @@ static void perf_counter_enable(struct perf_counter *counter)
 	 * Since we have the lock this context can't be scheduled
 	 * in, so we can change the state safely.
 	 */
-	if (counter->state == PERF_COUNTER_STATE_OFF) {
-		counter->state = PERF_COUNTER_STATE_INACTIVE;
-		counter->tstamp_enabled =
-			ctx->time - counter->total_time_enabled;
-	}
+	if (counter->state == PERF_COUNTER_STATE_OFF)
+		__perf_counter_mark_enabled(counter, ctx);
+
  out:
 	spin_unlock_irq(&ctx->lock);
 }
@@ -1479,9 +1504,7 @@ static void perf_counter_enable_on_exec(struct task_struct *task)
 		counter->attr.enable_on_exec = 0;
 		if (counter->state >= PERF_COUNTER_STATE_INACTIVE)
 			continue;
-		counter->state = PERF_COUNTER_STATE_INACTIVE;
-		counter->tstamp_enabled =
-			ctx->time - counter->total_time_enabled;
+		__perf_counter_mark_enabled(counter, ctx);
 		enabled = 1;
 	}
 
@@ -1675,6 +1698,11 @@ static void free_counter(struct perf_counter *counter)
 			atomic_dec(&nr_task_counters);
 	}
 
+	if (counter->output) {
+		fput(counter->output->filp);
+		counter->output = NULL;
+	}
+
 	if (counter->destroy)
 		counter->destroy(counter);
 
@@ -1960,6 +1988,8 @@ static int perf_counter_period(struct perf_counter *counter, u64 __user *arg)
 	return ret;
 }
 
+int perf_counter_set_output(struct perf_counter *counter, int output_fd);
+
 static long perf_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 {
 	struct perf_counter *counter = file->private_data;
@@ -1983,6 +2013,9 @@ static long perf_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	case PERF_COUNTER_IOC_PERIOD:
 		return perf_counter_period(counter, (u64 __user *)arg);
 
+	case PERF_COUNTER_IOC_SET_OUTPUT:
+		return perf_counter_set_output(counter, arg);
+
 	default:
 		return -ENOTTY;
 	}
@@ -2253,6 +2286,11 @@ static int perf_mmap(struct file *file, struct vm_area_struct *vma)
 
 	WARN_ON_ONCE(counter->ctx->parent_ctx);
 	mutex_lock(&counter->mmap_mutex);
+	if (counter->output) {
+		ret = -EINVAL;
+		goto unlock;
+	}
+
 	if (atomic_inc_not_zero(&counter->mmap_count)) {
 		if (nr_pages != counter->data->nr_pages)
 			ret = -EINVAL;
@@ -2638,6 +2676,7 @@ static int perf_output_begin(struct perf_output_handle *handle,
 			     struct perf_counter *counter, unsigned int size,
 			     int nmi, int sample)
 {
+	struct perf_counter *output_counter;
 	struct perf_mmap_data *data;
 	unsigned int offset, head;
 	int have_lost;
@@ -2647,13 +2686,17 @@ static int perf_output_begin(struct perf_output_handle *handle,
 		u64			 lost;
 	} lost_event;
 
+	rcu_read_lock();
 	/*
 	 * For inherited counters we send all the output towards the parent.
 	 */
 	if (counter->parent)
 		counter = counter->parent;
 
-	rcu_read_lock();
+	output_counter = rcu_dereference(counter->output);
+	if (output_counter)
+		counter = output_counter;
+
 	data = rcu_dereference(counter->data);
 	if (!data)
 		goto out;
@@ -3934,6 +3977,7 @@ static const struct pmu *tp_perf_counter_init(struct perf_counter *counter)
 	 * have these.
 	 */
 	if ((counter->attr.sample_type & PERF_SAMPLE_RAW) &&
+			perf_paranoid_tracepoint_raw() &&
 			!capable(CAP_SYS_ADMIN))
 		return ERR_PTR(-EPERM);
 
@@ -4202,6 +4246,57 @@ static int perf_copy_attr(struct perf_counter_attr __user *uattr,
 	goto out;
 }
 
+int perf_counter_set_output(struct perf_counter *counter, int output_fd)
+{
+	struct perf_counter *output_counter = NULL;
+	struct file *output_file = NULL;
+	struct perf_counter *old_output;
+	int fput_needed = 0;
+	int ret = -EINVAL;
+
+	if (!output_fd)
+		goto set;
+
+	output_file = fget_light(output_fd, &fput_needed);
+	if (!output_file)
+		return -EBADF;
+
+	if (output_file->f_op != &perf_fops)
+		goto out;
+
+	output_counter = output_file->private_data;
+
+	/* Don't chain output fds */
+	if (output_counter->output)
+		goto out;
+
+	/* Don't set an output fd when we already have an output channel */
+	if (counter->data)
+		goto out;
+
+	atomic_long_inc(&output_file->f_count);
+
+set:
+	mutex_lock(&counter->mmap_mutex);
+	old_output = counter->output;
+	rcu_assign_pointer(counter->output, output_counter);
+	mutex_unlock(&counter->mmap_mutex);
+
+	if (old_output) {
+		/*
+		 * we need to make sure no existing perf_output_*()
+		 * is still referencing this counter.
+		 */
+		synchronize_rcu();
+		fput(old_output->filp);
+	}
+
+	ret = 0;
+out:
+	fput_light(output_file, fput_needed);
+	return ret;
+}
+
 /**
  * sys_perf_counter_open - open a performance counter, associate it to a task/cpu
  *
@@ -4221,15 +4316,15 @@ SYSCALL_DEFINE5(perf_counter_open,
 	struct file *group_file = NULL;
 	int fput_needed = 0;
 	int fput_needed2 = 0;
-	int ret;
+	int err;
 
 	/* for future expandability... */
-	if (flags)
+	if (flags & ~(PERF_FLAG_FD_NO_GROUP | PERF_FLAG_FD_OUTPUT))
 		return -EINVAL;
 
-	ret = perf_copy_attr(attr_uptr, &attr);
-	if (ret)
-		return ret;
+	err = perf_copy_attr(attr_uptr, &attr);
+	if (err)
+		return err;
 
 	if (!attr.exclude_kernel) {
 		if (perf_paranoid_kernel() && !capable(CAP_SYS_ADMIN))
@@ -4252,8 +4347,8 @@ SYSCALL_DEFINE5(perf_counter_open,
 	 * Look up the group leader (we will attach this counter to it):
 	 */
 	group_leader = NULL;
-	if (group_fd != -1) {
-		ret = -EINVAL;
+	if (group_fd != -1 && !(flags & PERF_FLAG_FD_NO_GROUP)) {
+		err = -EINVAL;
 		group_file = fget_light(group_fd, &fput_needed);
 		if (!group_file)
 			goto err_put_context;
@@ -4282,18 +4377,24 @@ SYSCALL_DEFINE5(perf_counter_open,
 
 	counter = perf_counter_alloc(&attr, cpu, ctx, group_leader,
 				     NULL, GFP_KERNEL);
-	ret = PTR_ERR(counter);
+	err = PTR_ERR(counter);
 	if (IS_ERR(counter))
 		goto err_put_context;
 
-	ret = anon_inode_getfd("[perf_counter]", &perf_fops, counter, 0);
-	if (ret < 0)
+	err = anon_inode_getfd("[perf_counter]", &perf_fops, counter, 0);
+	if (err < 0)
 		goto err_free_put_context;
 
-	counter_file = fget_light(ret, &fput_needed2);
+	counter_file = fget_light(err, &fput_needed2);
 	if (!counter_file)
 		goto err_free_put_context;
 
+	if (flags & PERF_FLAG_FD_OUTPUT) {
+		err = perf_counter_set_output(counter, group_fd);
+		if (err)
+			goto err_fput_free_put_context;
+	}
+
 	counter->filp = counter_file;
 	WARN_ON_ONCE(ctx->parent_ctx);
 	mutex_lock(&ctx->mutex);
@@ -4307,20 +4408,20 @@ SYSCALL_DEFINE5(perf_counter_open,
 	list_add_tail(&counter->owner_entry, &current->perf_counter_list);
 	mutex_unlock(&current->perf_counter_mutex);
 
+err_fput_free_put_context:
 	fput_light(counter_file, fput_needed2);
 
-out_fput:
-	fput_light(group_file, fput_needed);
-
-	return ret;
-
 err_free_put_context:
+	if (err < 0)
 		kfree(counter);
 
 err_put_context:
+	if (err < 0)
 		put_ctx(ctx);
 
-	goto out_fput;
+	fput_light(group_file, fput_needed);
+
+	return err;
 }
 
 /*

tools/perf/Documentation/perf-record.txt

@@ -91,6 +91,10 @@ OPTIONS
 --no-samples::
 	Don't sample.
 
+-R::
+--raw-samples::
+Collect raw sample records from all opened counters (typically for tracepoint counters).
+
 SEE ALSO
 --------
 linkperf:perf-stat[1], linkperf:perf-list[1]

tools/perf/Documentation/perf-report.txt

@@ -27,6 +27,9 @@ OPTIONS
 -n
 --show-nr-samples
 	Show the number of samples for each symbol
+-T
+--threads
+	Show per-thread event counters
 -C::
 --comms=::
 	Only consider symbols in these comms. CSV that understands
@@ -48,6 +51,16 @@ OPTIONS
 	all occurances of this separator in symbol names (and other output)
 	with a '.' character, that thus it's the only non valid separator.
 
+-g [type,min]::
+--call-graph::
+        Display callchains using type and min percent threshold.
+	type can be either:
+	- flat: single column, linear exposure of callchains.
+	- graph: use a graph tree, displaying absolute overhead rates.
+	- fractal: like graph, but displays relative rates. Each branch of
+		 the tree is considered as a new profiled object. +
+	Default: fractal,0.5.
+
 SEE ALSO
 --------
 linkperf:perf-stat[1]

tools/perf/Makefile

@@ -166,7 +166,35 @@ endif
 
 # CFLAGS and LDFLAGS are for the users to override from the command line.
 
-CFLAGS = $(M64) -ggdb3 -Wall -Wextra -Wstrict-prototypes -Wmissing-declarations -Wmissing-prototypes -std=gnu99 -Wdeclaration-after-statement -Werror -O6
+#
+# Include saner warnings here, which can catch bugs:
+#
+
+EXTRA_WARNINGS := -Wcast-align
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wformat
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wformat-security
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wformat-y2k
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wshadow
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Winit-self
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wpacked
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wredundant-decls
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wstack-protector
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wstrict-aliasing=3
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wswitch-default
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wswitch-enum
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wno-system-headers
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wundef
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wvolatile-register-var
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wwrite-strings
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wbad-function-cast
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wmissing-declarations
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wmissing-prototypes
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wnested-externs
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wold-style-definition
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wstrict-prototypes
+EXTRA_WARNINGS := $(EXTRA_WARNINGS) -Wdeclaration-after-statement
+
+CFLAGS = $(M64) -ggdb3 -Wall -Wextra -std=gnu99 -Werror -O6 -fstack-protector-all -D_FORTIFY_SOURCE=2 $(EXTRA_WARNINGS)
 LDFLAGS = -lpthread -lrt -lelf -lm
 ALL_CFLAGS = $(CFLAGS)
 ALL_LDFLAGS = $(LDFLAGS)
@@ -310,6 +338,7 @@ LIB_H += util/sigchain.h
 LIB_H += util/symbol.h
 LIB_H += util/module.h
 LIB_H += util/color.h
+LIB_H += util/values.h
 
 LIB_OBJS += util/abspath.o
 LIB_OBJS += util/alias.o
@@ -337,6 +366,13 @@ 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/thread.o
+LIB_OBJS += util/trace-event-parse.o
+LIB_OBJS += util/trace-event-read.o
+LIB_OBJS += util/trace-event-info.o
 
 BUILTIN_OBJS += builtin-annotate.o
 BUILTIN_OBJS += builtin-help.o
@@ -345,6 +381,7 @@ BUILTIN_OBJS += builtin-record.o
 BUILTIN_OBJS += builtin-report.o
 BUILTIN_OBJS += builtin-stat.o
 BUILTIN_OBJS += builtin-top.o
+BUILTIN_OBJS += builtin-trace.o
 
 PERFLIBS = $(LIB_FILE)
 

tools/perf/builtin-help.c

@@ -456,6 +456,7 @@ int cmd_help(int argc, const char **argv, const char *prefix __used)
 		break;
 	case HELP_FORMAT_WEB:
 		show_html_page(argv[0]);
+	default:
 		break;
 	}
 

tools/perf/builtin-record.c

@@ -15,6 +15,9 @@
 #include "util/string.h"
 
 #include "util/header.h"
+#include "util/event.h"
+#include "util/debug.h"
+#include "util/trace-event.h"
 
 #include <unistd.h>
 #include <sched.h>
@@ -42,7 +45,6 @@ static int			inherit				= 1;
 static int			force				= 0;
 static int			append_file			= 0;
 static int			call_graph			= 0;
-static int			verbose				= 0;
 static int			inherit_stat			= 0;
 static int			no_samples			= 0;
 static int			sample_address			= 0;
@@ -62,24 +64,6 @@ static int			file_new = 1;
 
 struct perf_header		*header;
 
-struct mmap_event {
-	struct perf_event_header	header;
-	u32				pid;
-	u32				tid;
-	u64				start;
-	u64				len;
-	u64				pgoff;
-	char				filename[PATH_MAX];
-};
-
-struct comm_event {
-	struct perf_event_header	header;
-	u32				pid;
-	u32				tid;
-	char				comm[16];
-};
-
-
 struct mmap_data {
 	int			counter;
 	void			*base;
@@ -419,8 +403,11 @@ static void create_counter(int counter, int cpu, pid_t pid)
 	if (call_graph)
 		attr->sample_type	|= PERF_SAMPLE_CALLCHAIN;
 
-	if (raw_samples)
+	if (raw_samples) {
+		attr->sample_type	|= PERF_SAMPLE_TIME;
 		attr->sample_type	|= PERF_SAMPLE_RAW;
+		attr->sample_type	|= PERF_SAMPLE_CPU;
+	}
 
 	attr->mmap		= track;
 	attr->comm		= track;
@@ -563,6 +550,17 @@ static int __cmd_record(int argc, const char **argv)
 	else
 		header = perf_header__new();
 
+
+	if (raw_samples) {
+		read_tracing_data(attrs, nr_counters);
+	} else {
+		for (i = 0; i < nr_counters; i++) {
+			if (attrs[i].sample_type & PERF_SAMPLE_RAW) {
+				read_tracing_data(attrs, nr_counters);
+				break;
+			}
+		}
+	}
 	atexit(atexit_header);
 
 	if (!system_wide) {

tools/perf/builtin-stat.c

@@ -42,6 +42,8 @@
 #include "util/util.h"
 #include "util/parse-options.h"
 #include "util/parse-events.h"
+#include "util/event.h"
+#include "util/debug.h"
 
 #include <sys/prctl.h>
 #include <math.h>
@@ -60,10 +62,7 @@ static struct perf_counter_attr default_attrs[] = {
 
 };
 
-#define MAX_RUN			100
-
 static int			system_wide			=  0;
-static int			verbose				=  0;
 static unsigned int		nr_cpus				=  0;
 static int			run_idx				=  0;
 
@@ -75,26 +74,56 @@ static int			null_run			=  0;
 
 static int			fd[MAX_NR_CPUS][MAX_COUNTERS];
 
-static u64			runtime_nsecs[MAX_RUN];
-static u64			walltime_nsecs[MAX_RUN];
-static u64			runtime_cycles[MAX_RUN];
+static int			event_scaled[MAX_COUNTERS];
 
-static u64			event_res[MAX_RUN][MAX_COUNTERS][3];
-static u64			event_scaled[MAX_RUN][MAX_COUNTERS];
+struct stats
+{
+	double n, mean, M2;
+};
 
-static u64			event_res_avg[MAX_COUNTERS][3];
-static u64			event_res_noise[MAX_COUNTERS][3];
+static void update_stats(struct stats *stats, u64 val)
+{
+	double delta;
 
-static u64			event_scaled_avg[MAX_COUNTERS];
+	stats->n++;
+	delta = val - stats->mean;
+	stats->mean += delta / stats->n;
+	stats->M2 += delta*(val - stats->mean);
+}
 
-static u64			runtime_nsecs_avg;
-static u64			runtime_nsecs_noise;
+static double avg_stats(struct stats *stats)
+{
+	return stats->mean;
+}
 
-static u64			walltime_nsecs_avg;
-static u64			walltime_nsecs_noise;
+/*
+ * http://en.wikipedia.org/wiki/Algorithms_for_calculating_variance
+ *
+ *       (\Sum n_i^2) - ((\Sum n_i)^2)/n
+ * s^2 = -------------------------------
+ *                  n - 1
+ *
+ * http://en.wikipedia.org/wiki/Stddev
+ *
+ * The std dev of the mean is related to the std dev by:
+ *
+ *             s
+ * s_mean = -------
+ *          sqrt(n)
+ *
+ */
+static double stddev_stats(struct stats *stats)
+{
+	double variance = stats->M2 / (stats->n - 1);
+	double variance_mean = variance / stats->n;
 
-static u64			runtime_cycles_avg;
-static u64			runtime_cycles_noise;
+	return sqrt(variance_mean);
+}
+
+struct stats			event_res_stats[MAX_COUNTERS][3];
+struct stats			runtime_nsecs_stats;
+struct stats			walltime_nsecs_stats;
+struct stats			runtime_cycles_stats;
 
 #define MATCH_EVENT(t, c, counter)			\
 	(attrs[counter].type == PERF_TYPE_##t &&	\
@@ -149,12 +178,11 @@ static inline int nsec_counter(int counter)
  */
 static void read_counter(int counter)
 {
-	u64 *count, single_count[3];
+	u64 count[3], single_count[3];
 	unsigned int cpu;
 	size_t res, nv;
 	int scaled;
-
-	count = event_res[run_idx][counter];
+	int i;
 
 	count[0] = count[1] = count[2] = 0;
 
@@ -179,24 +207,33 @@ static void read_counter(int counter)
 	scaled = 0;
 	if (scale) {
 		if (count[2] == 0) {
-			event_scaled[run_idx][counter] = -1;
+			event_scaled[counter] = -1;
 			count[0] = 0;
 			return;
 		}
 
 		if (count[2] < count[1]) {
-			event_scaled[run_idx][counter] = 1;
+			event_scaled[counter] = 1;
 			count[0] = (unsigned long long)
 				((double)count[0] * count[1] / count[2] + 0.5);
 		}
 	}
+
+	for (i = 0; i < 3; i++)
+		update_stats(&event_res_stats[counter][i], count[i]);
+
+	if (verbose) {
+		fprintf(stderr, "%s: %Ld %Ld %Ld\n", event_name(counter),
+				count[0], count[1], count[2]);
+	}
+
 	/*
 	 * Save the full runtime - to allow normalization during printout:
 	 */
 	if (MATCH_EVENT(SOFTWARE, SW_TASK_CLOCK, counter))
-		runtime_nsecs[run_idx] = count[0];
+		update_stats(&runtime_nsecs_stats, count[0]);
 	if (MATCH_EVENT(HARDWARE, HW_CPU_CYCLES, counter))
-		runtime_cycles[run_idx] = count[0];
+		update_stats(&runtime_cycles_stats, count[0]);
 }
 
 static int run_perf_stat(int argc __used, const char **argv)
@@ -270,7 +307,7 @@ static int run_perf_stat(int argc __used, const char **argv)
 
 	t1 = rdclock();
 
-	walltime_nsecs[run_idx] = t1 - t0;
+	update_stats(&walltime_nsecs_stats, t1 - t0);
 
 	for (counter = 0; counter < nr_counters; counter++)
 		read_counter(counter);
@@ -278,42 +315,38 @@ static int run_perf_stat(int argc __used, const char **argv)
 	return WEXITSTATUS(status);
 }
 
-static void print_noise(u64 *count, u64 *noise)
+static void print_noise(int counter, double avg)
 {
-	if (run_count > 1)
+	if (run_count == 1)
+		return;
+
 	fprintf(stderr, "   ( +- %7.3f%% )",
-			(double)noise[0]/(count[0]+1)*100.0);
+			100 * stddev_stats(&event_res_stats[counter][0]) / avg);
 }
 
-static void nsec_printout(int counter, u64 *count, u64 *noise)
+static void nsec_printout(int counter, double avg)
 {
-	double msecs = (double)count[0] / 1000000;
+	double msecs = avg / 1e6;
 
 	fprintf(stderr, " %14.6f  %-24s", msecs, event_name(counter));
 
 	if (MATCH_EVENT(SOFTWARE, SW_TASK_CLOCK, counter)) {
-		if (walltime_nsecs_avg)
 		fprintf(stderr, " # %10.3f CPUs ",
-				(double)count[0] / (double)walltime_nsecs_avg);
+				avg / avg_stats(&walltime_nsecs_stats));
 	}
-	print_noise(count, noise);
 }
 
-static void abs_printout(int counter, u64 *count, u64 *noise)
+static void abs_printout(int counter, double avg)
 {
-	fprintf(stderr, " %14Ld  %-24s", count[0], event_name(counter));
+	fprintf(stderr, " %14.0f  %-24s", avg, event_name(counter));
 
-	if (runtime_cycles_avg &&
-	    MATCH_EVENT(HARDWARE, HW_INSTRUCTIONS, counter)) {
+	if (MATCH_EVENT(HARDWARE, HW_INSTRUCTIONS, counter)) {
 		fprintf(stderr, " # %10.3f IPC  ",
-			(double)count[0] / (double)runtime_cycles_avg);
+				avg / avg_stats(&runtime_cycles_stats));
 	} else {
-		if (runtime_nsecs_avg) {
 		fprintf(stderr, " # %10.3f M/sec",
-				(double)count[0]/runtime_nsecs_avg*1000.0);
-		}
+				1000.0 * avg / avg_stats(&runtime_nsecs_stats));
 	}
-	print_noise(count, noise);
 }
 
 /*
@@ -321,12 +354,8 @@ static void abs_printout(int counter, u64 *count, u64 *noise)
  */
 static void print_counter(int counter)
 {
-	u64 *count, *noise;
-	int scaled;
-
-	count = event_res_avg[counter];
-	noise = event_res_noise[counter];
-	scaled = event_scaled_avg[counter];
+	double avg = avg_stats(&event_res_stats[counter][0]);
+	int scaled = event_scaled[counter];
 
 	if (scaled == -1) {
 		fprintf(stderr, " %14s  %-24s\n",
@@ -335,110 +364,29 @@ static void print_counter(int counter)
 	}
 
 	if (nsec_counter(counter))
-		nsec_printout(counter, count, noise);
+		nsec_printout(counter, avg);
 	else
-		abs_printout(counter, count, noise);
+		abs_printout(counter, avg);
 
-	if (scaled)
-		fprintf(stderr, "  (scaled from %.2f%%)",
-			(double) count[2] / count[1] * 100);
+	print_noise(counter, avg);
 
-	fprintf(stderr, "\n");
-}
-
-/*
- * normalize_noise noise values down to stddev:
- */
-static void normalize_noise(u64 *val)
-{
-	double res;
-
-	res = (double)*val / (run_count * sqrt((double)run_count));
-
-	*val = (u64)res;
-}
-
-static void update_avg(const char *name, int idx, u64 *avg, u64 *val)
-{
-	*avg += *val;
-
-	if (verbose > 1)
-		fprintf(stderr, "debug: %20s[%d]: %Ld\n", name, idx, *val);
-}
-/*
- * Calculate the averages and noises:
- */
-static void calc_avg(void)
-{
-	int i, j;
-
-	if (verbose > 1)
-		fprintf(stderr, "\n");
+	if (scaled) {
+		double avg_enabled, avg_running;
 
-	for (i = 0; i < run_count; i++) {
-		update_avg("runtime", 0, &runtime_nsecs_avg, runtime_nsecs + i);
-		update_avg("walltime", 0, &walltime_nsecs_avg, walltime_nsecs + i);
-		update_avg("runtime_cycles", 0, &runtime_cycles_avg, runtime_cycles + i);
-
-		for (j = 0; j < nr_counters; j++) {
-			update_avg("counter/0", j,
-				event_res_avg[j]+0, event_res[i][j]+0);
-			update_avg("counter/1", j,
-				event_res_avg[j]+1, event_res[i][j]+1);
-			update_avg("counter/2", j,
-				event_res_avg[j]+2, event_res[i][j]+2);
-			if (event_scaled[i][j] != (u64)-1)
-				update_avg("scaled", j,
-					event_scaled_avg + j, event_scaled[i]+j);
-			else
-				event_scaled_avg[j] = -1;
-		}
-	}
-	runtime_nsecs_avg /= run_count;
-	walltime_nsecs_avg /= run_count;
-	runtime_cycles_avg /= run_count;
-
-	for (j = 0; j < nr_counters; j++) {
-		event_res_avg[j][0] /= run_count;
-		event_res_avg[j][1] /= run_count;
-		event_res_avg[j][2] /= run_count;
-	}
+		avg_enabled = avg_stats(&event_res_stats[counter][1]);
+		avg_running = avg_stats(&event_res_stats[counter][2]);
 
-	for (i = 0; i < run_count; i++) {
-		runtime_nsecs_noise +=
-			abs((s64)(runtime_nsecs[i] - runtime_nsecs_avg));
-		walltime_nsecs_noise +=
-			abs((s64)(walltime_nsecs[i] - walltime_nsecs_avg));
-		runtime_cycles_noise +=
-			abs((s64)(runtime_cycles[i] - runtime_cycles_avg));
-
-		for (j = 0; j < nr_counters; j++) {
-			event_res_noise[j][0] +=
-				abs((s64)(event_res[i][j][0] - event_res_avg[j][0]));
-			event_res_noise[j][1] +=
-				abs((s64)(event_res[i][j][1] - event_res_avg[j][1]));
-			event_res_noise[j][2] +=
-				abs((s64)(event_res[i][j][2] - event_res_avg[j][2]));
-		}
+		fprintf(stderr, "  (scaled from %.2f%%)",
+				100 * avg_running / avg_enabled);
 	}
 
-	normalize_noise(&runtime_nsecs_noise);
-	normalize_noise(&walltime_nsecs_noise);
-	normalize_noise(&runtime_cycles_noise);
-
-	for (j = 0; j < nr_counters; j++) {
-		normalize_noise(&event_res_noise[j][0]);
-		normalize_noise(&event_res_noise[j][1]);
-		normalize_noise(&event_res_noise[j][2]);
-	}
+	fprintf(stderr, "\n");
 }
 
 static void print_stat(int argc, const char **argv)
 {
 	int i, counter;
 
-	calc_avg();
-
 	fflush(stdout);
 
 	fprintf(stderr, "\n");
@@ -457,10 +405,11 @@ static void print_stat(int argc, const char **argv)
 
 	fprintf(stderr, "\n");
 	fprintf(stderr, " %14.9f  seconds time elapsed",
-			(double)walltime_nsecs_avg/1e9);
+			avg_stats(&walltime_nsecs_stats)/1e9);
 	if (run_count > 1) {
 		fprintf(stderr, "   ( +- %7.3f%% )",
-			100.0*(double)walltime_nsecs_noise/(double)walltime_nsecs_avg);
+				100*stddev_stats(&walltime_nsecs_stats) /
+				avg_stats(&walltime_nsecs_stats));
 	}
 	fprintf(stderr, "\n\n");
 }
@@ -515,7 +464,7 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used)
 		PARSE_OPT_STOP_AT_NON_OPTION);
 	if (!argc)
 		usage_with_options(stat_usage, options);
-	if (run_count <= 0 || run_count > MAX_RUN)
+	if (run_count <= 0)
 		usage_with_options(stat_usage, options);
 
 	/* Set attrs and nr_counters if no event is selected and !null_run */

tools/perf/builtin-top.c

@@ -27,6 +27,8 @@
 #include "util/parse-options.h"
 #include "util/parse-events.h"
 
+#include "util/debug.h"
+
 #include <assert.h>
 #include <fcntl.h>
 
@@ -68,8 +70,6 @@ static int			group				=  0;
 static unsigned int		page_size;
 static unsigned int		mmap_pages			= 16;
 static int			freq				=  0;
-static int			verbose				=  0;
-static char			*vmlinux			=  NULL;
 
 static int			delay_secs			=  2;
 static int			zero;
@@ -122,7 +122,8 @@ static void parse_source(struct sym_entry *syme)
 	struct module *module;
 	struct section *section = NULL;
 	FILE *file;
-	char command[PATH_MAX*2], *path = vmlinux;
+	char command[PATH_MAX*2];
+	const char *path = vmlinux_name;
 	u64 start, end, len;
 
 	if (!syme)
@@ -338,8 +339,6 @@ static void show_details(struct sym_entry *syme)
 		printf("%d lines not displayed, maybe increase display entries [e]\n", more);
 }
 
-struct dso			*kernel_dso;
-
 /*
  * Symbols will be added here in record_ip and will get out
  * after decayed.
@@ -486,15 +485,22 @@ static void print_sym_table(void)
 	else
 		printf("   weight    samples    pcnt");
 
-	printf("         RIP          kernel function\n"
-	       	       "  ______     _______   _____   ________________   _______________\n\n"
-	);
+	if (verbose)
+		printf("         RIP       ");
+	printf("   kernel function\n");
+	printf("   %s    _______   _____",
+	       nr_counters == 1 ? "      " : "______");
+	if (verbose)
+		printf("   ________________");
+	printf("   _______________\n\n");
 
 	for (nd = rb_first(&tmp); nd; nd = rb_next(nd)) {
-		struct sym_entry *syme = rb_entry(nd, struct sym_entry, rb_node);
-		struct symbol *sym = (struct symbol *)(syme + 1);
+		struct symbol *sym;
 		double pcnt;
 
+		syme = rb_entry(nd, struct sym_entry, rb_node);
+		sym = (struct symbol *)(syme + 1);
+
 		if (++printed > print_entries || (int)syme->snap_count < count_filter)
 			continue;
 
@@ -507,7 +513,9 @@ static void print_sym_table(void)
 			printf("%9.1f %10ld - ", syme->weight, syme->snap_count);
 
 		percent_color_fprintf(stdout, "%4.1f%%", pcnt);
-		printf(" - %016llx : %s", sym->start, sym->name);
+		if (verbose)
+			printf(" - %016llx", sym->start);
+		printf(" : %s", sym->name);
 		if (sym->module)
 			printf("\t[%s]", sym->module->name);
 		printf("\n");
@@ -613,7 +621,7 @@ static void print_mapped_keys(void)
 
 	fprintf(stdout, "\t[f]     profile display filter (count).    \t(%d)\n", count_filter);
 
-	if (vmlinux) {
+	if (vmlinux_name) {
 		fprintf(stdout, "\t[F]     annotate display filter (percent). \t(%d%%)\n", sym_pcnt_filter);
 		fprintf(stdout, "\t[s]     annotate symbol.                   \t(%s)\n", name?: "NULL");
 		fprintf(stdout, "\t[S]     stop annotation.\n");
@@ -642,7 +650,9 @@ static int key_mapped(int c)
 		case 'F':
 		case 's':
 		case 'S':
-			return vmlinux ? 1 : 0;
+			return vmlinux_name ? 1 : 0;
+		default:
+			break;
 	}
 
 	return 0;
@@ -728,6 +738,8 @@ static void handle_keypress(int c)
 		case 'z':
 			zero = ~zero;
 			break;
+		default:
+			break;
 	}
 }
 
@@ -816,13 +828,13 @@ static int parse_symbols(void)
 {
 	struct rb_node *node;
 	struct symbol  *sym;
-	int modules = vmlinux ? 1 : 0;
+	int use_modules = vmlinux_name ? 1 : 0;
 
 	kernel_dso = dso__new("[kernel]", sizeof(struct sym_entry));
 	if (kernel_dso == NULL)
 		return -1;
 
-	if (dso__load_kernel(kernel_dso, vmlinux, symbol_filter, verbose, modules) <= 0)
+	if (dso__load_kernel(kernel_dso, vmlinux_name, symbol_filter, verbose, use_modules) <= 0)
 		goto out_delete_dso;
 
 	node = rb_first(&kernel_dso->syms);
@@ -937,26 +949,6 @@ static void mmap_read_counter(struct mmap_data *md)
 	last_read = this_read;
 
 	for (; old != head;) {
-		struct ip_event {
-			struct perf_event_header header;
-			u64 ip;
-			u32 pid, target_pid;
-		};
-		struct mmap_event {
-			struct perf_event_header header;
-			u32 pid, target_pid;
-			u64 start;
-			u64 len;
-			u64 pgoff;
-			char filename[PATH_MAX];
-		};
-
-		typedef union event_union {
-			struct perf_event_header header;
-			struct ip_event ip;
-			struct mmap_event mmap;
-		} event_t;
-
 		event_t *event = (event_t *)&data[old & md->mask];
 
 		event_t event_copy;
@@ -1138,7 +1130,7 @@ static const struct option options[] = {
 			    "system-wide collection from all CPUs"),
 	OPT_INTEGER('C', "CPU", &profile_cpu,
 		    "CPU to profile on"),
-	OPT_STRING('k', "vmlinux", &vmlinux, "file", "vmlinux pathname"),
+	OPT_STRING('k', "vmlinux", &vmlinux_name, "file", "vmlinux pathname"),
 	OPT_INTEGER('m', "mmap-pages", &mmap_pages,
 		    "number of mmap data pages"),
 	OPT_INTEGER('r', "realtime", &realtime_prio,

tools/perf/builtin-trace.c

+#include "builtin.h"
+
+#include "util/util.h"
+#include "util/cache.h"
+#include "util/symbol.h"
+#include "util/thread.h"
+#include "util/header.h"
+
+#include "util/parse-options.h"
+
+#include "perf.h"
+#include "util/debug.h"
+
+#include "util/trace-event.h"
+
+static char		const *input_name = "perf.data";
+static int		input;
+static unsigned long	page_size;
+static unsigned long	mmap_window = 32;
+
+static unsigned long	total = 0;
+static unsigned long	total_comm = 0;
+
+static struct rb_root	threads;
+static struct thread	*last_match;
+
+static struct perf_header *header;
+static u64		sample_type;
+
+
+static int
+process_comm_event(event_t *event, unsigned long offset, unsigned long head)
+{
+	struct thread *thread;
+
+	thread = threads__findnew(event->comm.pid, &threads, &last_match);
+
+	dump_printf("%p [%p]: PERF_EVENT_COMM: %s:%d\n",
+		(void *)(offset + head),
+		(void *)(long)(event->header.size),
+		event->comm.comm, event->comm.pid);
+
+	if (thread == NULL ||
+	    thread__set_comm(thread, event->comm.comm)) {
+		dump_printf("problem processing PERF_EVENT_COMM, skipping event.\n");
+		return -1;
+	}
+	total_comm++;
+
+	return 0;
+}
+
+static int
+process_sample_event(event_t *event, unsigned long offset, unsigned long head)
+{
+	char level;
+	int show = 0;
+	struct dso *dso = NULL;
+	struct thread *thread;
+	u64 ip = event->ip.ip;
+	u64 timestamp = -1;
+	u32 cpu = -1;
+	u64 period = 1;
+	void *more_data = event->ip.__more_data;
+	int cpumode;
+
+	thread = threads__findnew(event->ip.pid, &threads, &last_match);
+
+	if (sample_type & PERF_SAMPLE_TIME) {
+		timestamp = *(u64 *)more_data;
+		more_data += sizeof(u64);
+	}
+
+	if (sample_type & PERF_SAMPLE_CPU) {
+		cpu = *(u32 *)more_data;
+		more_data += sizeof(u32);
+		more_data += sizeof(u32); /* reserved */
+	}
+
+	if (sample_type & PERF_SAMPLE_PERIOD) {
+		period = *(u64 *)more_data;
+		more_data += sizeof(u64);
+	}
+
+	dump_printf("%p [%p]: PERF_EVENT_SAMPLE (IP, %d): %d/%d: %p period: %Ld\n",
+		(void *)(offset + head),
+		(void *)(long)(event->header.size),
+		event->header.misc,
+		event->ip.pid, event->ip.tid,
+		(void *)(long)ip,
+		(long long)period);
+
+	dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid);
+
+	if (thread == NULL) {
+		eprintf("problem processing %d event, skipping it.\n",
+			event->header.type);
+		return -1;
+	}
+
+	cpumode = event->header.misc & PERF_EVENT_MISC_CPUMODE_MASK;
+
+	if (cpumode == PERF_EVENT_MISC_KERNEL) {
+		show = SHOW_KERNEL;
+		level = 'k';
+
+		dso = kernel_dso;
+
+		dump_printf(" ...... dso: %s\n", dso->name);
+
+	} else if (cpumode == PERF_EVENT_MISC_USER) {
+
+		show = SHOW_USER;
+		level = '.';
+
+	} else {
+		show = SHOW_HV;
+		level = 'H';
+
+		dso = hypervisor_dso;
+
+		dump_printf(" ...... dso: [hypervisor]\n");
+	}
+
+	if (sample_type & PERF_SAMPLE_RAW) {
+		struct {
+			u32 size;
+			char data[0];
+		} *raw = more_data;
+
+		/*
+		 * FIXME: better resolve from pid from the struct trace_entry
+		 * field, although it should be the same than this perf
+		 * event pid
+		 */
+		print_event(cpu, raw->data, raw->size, timestamp, thread->comm);
+	}
+	total += period;
+
+	return 0;
+}
+
+static int
+process_event(event_t *event, unsigned long offset, unsigned long head)
+{
+	trace_event(event);
+
+	switch (event->header.type) {
+	case PERF_EVENT_MMAP ... PERF_EVENT_LOST:
+		return 0;
+
+	case PERF_EVENT_COMM:
+		return process_comm_event(event, offset, head);
+
+	case PERF_EVENT_EXIT ... PERF_EVENT_READ:
+		return 0;
+
+	case PERF_EVENT_SAMPLE:
+		return process_sample_event(event, offset, head);
+
+	case PERF_EVENT_MAX:
+	default:
+		return -1;
+	}
+
+	return 0;
+}
+
+static int __cmd_trace(void)
+{
+	int ret, rc = EXIT_FAILURE;
+	unsigned long offset = 0;
+	unsigned long head = 0;
+	struct stat perf_stat;
+	event_t *event;
+	uint32_t size;
+	char *buf;
+
+	trace_report();
+	register_idle_thread(&threads, &last_match);
+
+	input = open(input_name, O_RDONLY);
+	if (input < 0) {
+		perror("failed to open file");
+		exit(-1);
+	}
+
+	ret = fstat(input, &perf_stat);
+	if (ret < 0) {
+		perror("failed to stat file");
+		exit(-1);
+	}
+
+	if (!perf_stat.st_size) {
+		fprintf(stderr, "zero-sized file, nothing to do!\n");
+		exit(0);
+	}
+	header = perf_header__read(input);
+	head = header->data_offset;
+	sample_type = perf_header__sample_type(header);
+
+	if (!(sample_type & PERF_SAMPLE_RAW))
+		die("No trace sample to read. Did you call perf record "
+		    "without -R?");
+
+	if (load_kernel() < 0) {
+		perror("failed to load kernel symbols");
+		return EXIT_FAILURE;
+	}
+
+remap:
+	buf = (char *)mmap(NULL, page_size * mmap_window, PROT_READ,
+			   MAP_SHARED, input, offset);
+	if (buf == MAP_FAILED) {
+		perror("failed to mmap file");
+		exit(-1);
+	}
+
+more:
+	event = (event_t *)(buf + head);
+
+	size = event->header.size;
+	if (!size)
+		size = 8;
+
+	if (head + event->header.size >= page_size * mmap_window) {
+		unsigned long shift = page_size * (head / page_size);
+		int res;
+
+		res = munmap(buf, page_size * mmap_window);
+		assert(res == 0);
+
+		offset += shift;
+		head -= shift;
+		goto remap;
+	}
+
+	size = event->header.size;
+
+
+	if (!size || process_event(event, offset, head) < 0) {
+
+		/*
+		 * assume we lost track of the stream, check alignment, and
+		 * increment a single u64 in the hope to catch on again 'soon'.
+		 */
+
+		if (unlikely(head & 7))
+			head &= ~7ULL;
+
+		size = 8;
+	}
+
+	head += size;
+
+	if (offset + head < (unsigned long)perf_stat.st_size)
+		goto more;
+
+	rc = EXIT_SUCCESS;
+	close(input);
+
+	return rc;
+}
+
+static const char * const annotate_usage[] = {
+	"perf trace [<options>] <command>",
+	NULL
+};
+
+static const struct option options[] = {
+	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
+		    "dump raw trace in ASCII"),
+	OPT_BOOLEAN('v', "verbose", &verbose,
+		    "be more verbose (show symbol address, etc)"),
+	OPT_END()
+};
+
+int cmd_trace(int argc, const char **argv, const char *prefix __used)
+{
+	symbol__init();
+	page_size = getpagesize();
+
+	argc = parse_options(argc, argv, options, annotate_usage, 0);
+	if (argc) {
+		/*
+		 * Special case: if there's an argument left then assume tha
+		 * it's a symbol filter:
+		 */
+		if (argc > 1)
+			usage_with_options(annotate_usage, options);
+	}
+
+
+	setup_pager();
+
+	return __cmd_trace();
+}

tools/perf/builtin.h

@@ -22,5 +22,6 @@ extern int cmd_stat(int argc, const char **argv, const char *prefix);
 extern int cmd_top(int argc, const char **argv, const char *prefix);
 extern int cmd_version(int argc, const char **argv, const char *prefix);
 extern int cmd_list(int argc, const char **argv, const char *prefix);
+extern int cmd_trace(int argc, const char **argv, const char *prefix);
 
 #endif

tools/perf/perf.c

@@ -292,6 +292,7 @@ static void handle_internal_command(int argc, const char **argv)
 		{ "top", cmd_top, 0 },
 		{ "annotate", cmd_annotate, 0 },
 		{ "version", cmd_version, 0 },
+		{ "trace", cmd_trace, 0 },
 	};
 	unsigned int i;
 	static const char ext[] = STRIP_EXTENSION;

tools/perf/util/abspath.c

@@ -50,7 +50,8 @@ const char *make_absolute_path(const char *path)
 			die ("Could not get current working directory");
 
 		if (last_elem) {
-			int len = strlen(buf);
+			len = strlen(buf);
+
 			if (len + strlen(last_elem) + 2 > PATH_MAX)
 				die ("Too long path name: '%s/%s'",
 						buf, last_elem);

tools/perf/util/cache.h

@@ -52,7 +52,6 @@ extern const char *perf_mailmap_file;
 extern void maybe_flush_or_die(FILE *, const char *);
 extern int copy_fd(int ifd, int ofd);
 extern int copy_file(const char *dst, const char *src, int mode);
-extern ssize_t read_in_full(int fd, void *buf, size_t count);
 extern ssize_t write_in_full(int fd, const void *buf, size_t count);
 extern void write_or_die(int fd, const void *buf, size_t count);
 extern int write_or_whine(int fd, const void *buf, size_t count, const char *msg);

tools/perf/util/callchain.c

@@ -50,6 +50,7 @@ rb_insert_callchain(struct rb_root *root, struct callchain_node *chain,
 			else
 				p = &(*p)->rb_right;
 			break;
+		case CHAIN_NONE:
 		default:
 			break;
 		}
@@ -143,6 +144,7 @@ int register_callchain_param(struct callchain_param *param)
 	case CHAIN_FLAT:
 		param->sort = sort_chain_flat;
 		break;
+	case CHAIN_NONE:
 	default:
 		return -1;
 	}

tools/perf/util/callchain.h

@@ -4,6 +4,7 @@
 #include "../perf.h"
 #include <linux/list.h>
 #include <linux/rbtree.h>
+#include "util.h"
 #include "symbol.h"
 
 enum chain_mode {

tools/perf/util/color.c

@@ -166,7 +166,7 @@ int perf_color_default_config(const char *var, const char *value, void *cb)
 	return perf_default_config(var, value, cb);
 }
 
-static int color_vfprintf(FILE *fp, const char *color, const char *fmt,
+static int __color_vfprintf(FILE *fp, const char *color, const char *fmt,
 		va_list args, const char *trail)
 {
 	int r = 0;
@@ -191,6 +191,10 @@ static int color_vfprintf(FILE *fp, const char *color, const char *fmt,
 	return r;
 }
 
+int color_vfprintf(FILE *fp, const char *color, const char *fmt, va_list args)
+{
+	return __color_vfprintf(fp, color, fmt, args, NULL);
+}
 
 
 int color_fprintf(FILE *fp, const char *color, const char *fmt, ...)
@@ -199,7 +203,7 @@ int color_fprintf(FILE *fp, const char *color, const char *fmt, ...)
 	int r;
 
 	va_start(args, fmt);
-	r = color_vfprintf(fp, color, fmt, args, NULL);
+	r = color_vfprintf(fp, color, fmt, args);
 	va_end(args);
 	return r;
 }
@@ -209,7 +213,7 @@ int color_fprintf_ln(FILE *fp, const char *color, const char *fmt, ...)
 	va_list args;
 	int r;
 	va_start(args, fmt);
-	r = color_vfprintf(fp, color, fmt, args, "\n");
+	r = __color_vfprintf(fp, color, fmt, args, "\n");
 	va_end(args);
 	return r;
 }
@@ -242,9 +246,9 @@ int color_fwrite_lines(FILE *fp, const char *color,
 	return 0;
 }
 
-char *get_percent_color(double percent)
+const char *get_percent_color(double percent)
 {
-	char *color = PERF_COLOR_NORMAL;
+	const char *color = PERF_COLOR_NORMAL;
 
 	/*
 	 * We color high-overhead entries in red, mid-overhead
@@ -263,7 +267,7 @@ char *get_percent_color(double percent)
 int percent_color_fprintf(FILE *fp, const char *fmt, double percent)
 {
 	int r;
-	char *color;
+	const char *color;
 
 	color = get_percent_color(percent);
 	r = color_fprintf(fp, color, fmt, percent);

tools/perf/util/color.h

@@ -32,10 +32,11 @@ int perf_color_default_config(const char *var, const char *value, void *cb);
 int perf_config_colorbool(const char *var, const char *value, int stdout_is_tty);
 void color_parse(const char *value, const char *var, char *dst);
 void color_parse_mem(const char *value, int len, const char *var, char *dst);
+int color_vfprintf(FILE *fp, const char *color, const char *fmt, va_list args);
 int color_fprintf(FILE *fp, const char *color, const char *fmt, ...);
 int color_fprintf_ln(FILE *fp, const char *color, const char *fmt, ...);
 int color_fwrite_lines(FILE *fp, const char *color, size_t count, const char *buf);
 int percent_color_fprintf(FILE *fp, const char *fmt, double percent);
-char *get_percent_color(double percent);
+const char *get_percent_color(double percent);
 
 #endif /* COLOR_H */

tools/perf/util/config.c

@@ -160,17 +160,18 @@ static int get_extended_base_var(char *name, int baselen, int c)
 	name[baselen++] = '.';
 
 	for (;;) {
-		int c = get_next_char();
-		if (c == '\n')
+		int ch = get_next_char();
+
+		if (ch == '\n')
 			return -1;
-		if (c == '"')
+		if (ch == '"')
 			break;
-		if (c == '\\') {
-			c = get_next_char();
-			if (c == '\n')
+		if (ch == '\\') {
+			ch = get_next_char();
+			if (ch == '\n')
 				return -1;
 		}
-		name[baselen++] = c;
+		name[baselen++] = ch;
 		if (baselen > MAXNAME / 2)
 			return -1;
 	}
@@ -530,6 +531,8 @@ static int store_aux(const char* key, const char* value, void *cb __used)
 					store.offset[store.seen] = ftell(config_file);
 			}
 		}
+	default:
+		break;
 	}
 	return 0;
 }
@@ -619,6 +622,7 @@ static ssize_t find_beginning_of_line(const char* contents, size_t size,
 		switch (contents[offset]) {
 			case '=': equal_offset = offset; break;
 			case ']': bracket_offset = offset; break;
+			default: break;
 		}
 	if (offset > 0 && contents[offset-1] == '\\') {
 		offset_ = offset;
@@ -742,9 +746,9 @@ int perf_config_set_multivar(const char* key, const char* value,
 			goto write_err_out;
 	} else {
 		struct stat st;
-		char* contents;
+		char *contents;
 		ssize_t contents_sz, copy_begin, copy_end;
-		int i, new_line = 0;
+		int new_line = 0;
 
 		if (value_regex == NULL)
 			store.value_regex = NULL;

tools/perf/util/debug.c

+/* For general debugging purposes */
+
+#include "../perf.h"
+
+#include <string.h>
+#include <stdarg.h>
+#include <stdio.h>
+
+#include "color.h"
+#include "event.h"
+#include "debug.h"
+
+int verbose = 0;
+int dump_trace = 0;
+
+int eprintf(const char *fmt, ...)
+{
+	va_list args;
+	int ret = 0;
+
+	if (verbose) {
+		va_start(args, fmt);
+		ret = vfprintf(stderr, fmt, args);
+		va_end(args);
+	}
+
+	return ret;
+}
+
+int dump_printf(const char *fmt, ...)
+{
+	va_list args;
+	int ret = 0;
+
+	if (dump_trace) {
+		va_start(args, fmt);
+		ret = vprintf(fmt, args);
+		va_end(args);
+	}
+
+	return ret;
+}
+
+static int dump_printf_color(const char *fmt, const char *color, ...)
+{
+	va_list args;
+	int ret = 0;
+
+	if (dump_trace) {
+		va_start(args, color);
+		ret = color_vfprintf(stdout, color, fmt, args);
+		va_end(args);
+	}
+
+	return ret;
+}
+
+
+void trace_event(event_t *event)
+{
+	unsigned char *raw_event = (void *)event;
+	const char *color = PERF_COLOR_BLUE;
+	int i, j;
+
+	if (!dump_trace)
+		return;
+
+	dump_printf(".");
+	dump_printf_color("\n. ... raw event: size %d bytes\n", color,
+			  event->header.size);
+
+	for (i = 0; i < event->header.size; i++) {
+		if ((i & 15) == 0) {
+			dump_printf(".");
+			dump_printf_color("  %04x: ", color, i);
+		}
+
+		dump_printf_color(" %02x", color, raw_event[i]);
+
+		if (((i & 15) == 15) || i == event->header.size-1) {
+			dump_printf_color("  ", color);
+			for (j = 0; j < 15-(i & 15); j++)
+				dump_printf_color("   ", color);
+			for (j = 0; j < (i & 15); j++) {
+				if (isprint(raw_event[i-15+j]))
+					dump_printf_color("%c", color,
+							  raw_event[i-15+j]);
+				else
+					dump_printf_color(".", color);
+			}
+			dump_printf_color("\n", color);
+		}
+	}
+	dump_printf(".\n");
+}

tools/perf/util/debug.h

+/* For debugging general purposes */
+
+extern int verbose;
+extern int dump_trace;
+
+int eprintf(const char *fmt, ...) __attribute__((format(printf, 1, 2)));
+int dump_printf(const char *fmt, ...) __attribute__((format(printf, 1, 2)));
+void trace_event(event_t *event);

tools/perf/util/event.h

+#ifndef __PERF_EVENT_H
+#define __PERF_EVENT_H
+#include "../perf.h"
+#include "util.h"
+#include <linux/list.h>
+
+enum {
+	SHOW_KERNEL	= 1,
+	SHOW_USER	= 2,
+	SHOW_HV		= 4,
+};
+
+/*
+ * PERF_SAMPLE_IP | PERF_SAMPLE_TID | *
+ */
+struct ip_event {
+	struct perf_event_header header;
+	u64 ip;
+	u32 pid, tid;
+	unsigned char __more_data[];
+};
+
+struct mmap_event {
+	struct perf_event_header header;
+	u32 pid, tid;
+	u64 start;
+	u64 len;
+	u64 pgoff;
+	char filename[PATH_MAX];
+};
+
+struct comm_event {
+	struct perf_event_header header;
+	u32 pid, tid;
+	char comm[16];
+};
+
+struct fork_event {
+	struct perf_event_header header;
+	u32 pid, ppid;
+	u32 tid, ptid;
+};
+
+struct lost_event {
+	struct perf_event_header header;
+	u64 id;
+	u64 lost;
+};
+
+/*
+ * PERF_FORMAT_ENABLED | PERF_FORMAT_RUNNING | PERF_FORMAT_ID
+ */
+struct read_event {
+	struct perf_event_header header;
+	u32 pid,tid;
+	u64 value;
+	u64 time_enabled;
+	u64 time_running;
+	u64 id;
+};
+
+typedef union event_union {
+	struct perf_event_header	header;
+	struct ip_event			ip;
+	struct mmap_event		mmap;
+	struct comm_event		comm;
+	struct fork_event		fork;
+	struct lost_event		lost;
+	struct read_event		read;
+} event_t;
+
+struct map {
+	struct list_head	node;
+	u64			start;
+	u64			end;
+	u64			pgoff;
+	u64			(*map_ip)(struct map *, u64);
+	struct dso		*dso;
+};
+
+static inline u64 map__map_ip(struct map *map, u64 ip)
+{
+	return ip - map->start + map->pgoff;
+}
+
+static inline u64 vdso__map_ip(struct map *map __used, u64 ip)
+{
+	return ip;
+}
+
+struct map *map__new(struct mmap_event *event, char *cwd, int cwdlen);
+struct map *map__clone(struct map *self);
+int map__overlap(struct map *l, struct map *r);
+size_t map__fprintf(struct map *self, FILE *fp);
+
+#endif

tools/perf/util/exec_cmd.c

@@ -6,7 +6,6 @@
 
 #define MAX_ARGS	32
 
-extern char **environ;
 static const char *argv_exec_path;
 static const char *argv0_path;
 

tools/perf/util/header.c

@@ -237,9 +237,44 @@ struct perf_header *perf_header__read(int fd)
 	self->data_offset = f_header.data.offset;
 	self->data_size   = f_header.data.size;
 
-	lseek(fd, self->data_offset + self->data_size, SEEK_SET);
+	lseek(fd, self->data_offset, SEEK_SET);
 
 	self->frozen = 1;
 
 	return self;
 }
+
+u64 perf_header__sample_type(struct perf_header *header)
+{
+	u64 type = 0;
+	int i;
+
+	for (i = 0; i < header->attrs; i++) {
+		struct perf_header_attr *attr = header->attr[i];
+
+		if (!type)
+			type = attr->attr.sample_type;
+		else if (type != attr->attr.sample_type)
+			die("non matching sample_type");
+	}
+
+	return type;
+}
+
+struct perf_counter_attr *
+perf_header__find_attr(u64 id, struct perf_header *header)
+{
+	int i;
+
+	for (i = 0; i < header->attrs; i++) {
+		struct perf_header_attr *attr = header->attr[i];
+		int j;
+
+		for (j = 0; j < attr->ids; j++) {
+			if (attr->id[j] == id)
+				return &attr->attr;
+		}
+	}
+
+	return NULL;
+}

tools/perf/util/header.h

@@ -31,6 +31,10 @@ struct perf_header_attr *
 perf_header_attr__new(struct perf_counter_attr *attr);
 void perf_header_attr__add_id(struct perf_header_attr *self, u64 id);
 
+u64 perf_header__sample_type(struct perf_header *header);
+struct perf_counter_attr *
+perf_header__find_attr(u64 id, struct perf_header *header);
+
 
 struct perf_header *perf_header__new(void);
 

tools/perf/util/map.c

+#include "event.h"
+#include "symbol.h"
+#include <stdlib.h>
+#include <string.h>
+#include <stdio.h>
+
+static inline int is_anon_memory(const char *filename)
+{
+	return strcmp(filename, "//anon") == 0;
+}
+
+static int strcommon(const char *pathname, char *cwd, int cwdlen)
+{
+	int n = 0;
+
+	while (n < cwdlen && pathname[n] == cwd[n])
+		++n;
+
+	return n;
+}
+
+ struct map *map__new(struct mmap_event *event, char *cwd, int cwdlen)
+{
+	struct map *self = malloc(sizeof(*self));
+
+	if (self != NULL) {
+		const char *filename = event->filename;
+		char newfilename[PATH_MAX];
+		int anon;
+
+		if (cwd) {
+			int n = strcommon(filename, cwd, cwdlen);
+
+			if (n == cwdlen) {
+				snprintf(newfilename, sizeof(newfilename),
+					 ".%s", filename + n);
+				filename = newfilename;
+			}
+		}
+
+		anon = is_anon_memory(filename);
+
+		if (anon) {
+			snprintf(newfilename, sizeof(newfilename), "/tmp/perf-%d.map", event->pid);
+			filename = newfilename;
+		}
+
+		self->start = event->start;
+		self->end   = event->start + event->len;
+		self->pgoff = event->pgoff;
+
+		self->dso = dsos__findnew(filename);
+		if (self->dso == NULL)
+			goto out_delete;
+
+		if (self->dso == vdso || anon)
+			self->map_ip = vdso__map_ip;
+		else
+			self->map_ip = map__map_ip;
+	}
+	return self;
+out_delete:
+	free(self);
+	return NULL;
+}
+
+struct map *map__clone(struct map *self)
+{
+	struct map *map = malloc(sizeof(*self));
+
+	if (!map)
+		return NULL;
+
+	memcpy(map, self, sizeof(*self));
+
+	return map;
+}
+
+int map__overlap(struct map *l, struct map *r)
+{
+	if (l->start > r->start) {
+		struct map *t = l;
+		l = r;
+		r = t;
+	}
+
+	if (l->end > r->start)
+		return 1;
+
+	return 0;
+}
+
+size_t map__fprintf(struct map *self, FILE *fp)
+{
+	return fprintf(fp, " %Lx-%Lx %Lx %s\n",
+		       self->start, self->end, self->pgoff, self->dso->name);
+}

tools/perf/util/module.c

@@ -436,9 +436,9 @@ static int mod_dso__load_module_paths(struct mod_dso *self)
 		goto out_failure;
 
 	while (!feof(file)) {
-		char *path, *name, *tmp;
+		char *name, *tmp;
 		struct module *module;
-		int line_len, len;
+		int line_len;
 
 		line_len = getline(&line, &n, file);
 		if (line_len < 0)

tools/perf/util/parse-events.c

 
-#include "../perf.h"
 #include "util.h"
+#include "../perf.h"
 #include "parse-options.h"
 #include "parse-events.h"
 #include "exec_cmd.h"
 #include "string.h"
 #include "cache.h"
 
-extern char *strcasestr(const char *haystack, const char *needle);
-
 int					nr_counters;
 
 struct perf_counter_attr		attrs[MAX_COUNTERS];
@@ -16,8 +14,8 @@ struct perf_counter_attr		attrs[MAX_COUNTERS];
 struct event_symbol {
 	u8		type;
 	u64		config;
-	char	*symbol;
-	char	*alias;
+	const char	*symbol;
+	const char	*alias;
 };
 
 char debugfs_path[MAXPATHLEN];
@@ -51,7 +49,7 @@ static struct event_symbol event_symbols[] = {
 #define PERF_COUNTER_TYPE(config)	__PERF_COUNTER_FIELD(config, TYPE)
 #define PERF_COUNTER_ID(config)		__PERF_COUNTER_FIELD(config, EVENT)
 
-static char *hw_event_names[] = {
+static const char *hw_event_names[] = {
 	"cycles",
 	"instructions",
 	"cache-references",
@@ -61,7 +59,7 @@ static char *hw_event_names[] = {
 	"bus-cycles",
 };
 
-static char *sw_event_names[] = {
+static const char *sw_event_names[] = {
 	"cpu-clock-msecs",
 	"task-clock-msecs",
 	"page-faults",
@@ -73,7 +71,7 @@ static char *sw_event_names[] = {
 
 #define MAX_ALIASES 8
 
-static char *hw_cache[][MAX_ALIASES] = {
+static const char *hw_cache[][MAX_ALIASES] = {
  { "L1-dcache",	"l1-d",		"l1d",		"L1-data",		},
  { "L1-icache",	"l1-i",		"l1i",		"L1-instruction",	},
  { "LLC",	"L2"							},
@@ -82,13 +80,13 @@ static char *hw_cache[][MAX_ALIASES] = {
  { "branch",	"branches",	"bpu",		"btb",		"bpc",	},
 };
 
-static char *hw_cache_op[][MAX_ALIASES] = {
+static const char *hw_cache_op[][MAX_ALIASES] = {
  { "load",	"loads",	"read",					},
  { "store",	"stores",	"write",				},
  { "prefetch",	"prefetches",	"speculative-read", "speculative-load",	},
 };
 
-static char *hw_cache_result[][MAX_ALIASES] = {
+static const char *hw_cache_result[][MAX_ALIASES] = {
  { "refs",	"Reference",	"ops",		"access",		},
  { "misses",	"miss",							},
 };
@@ -113,11 +111,9 @@ static unsigned long hw_cache_stat[C(MAX)] = {
  [C(BPU)]	= (CACHE_READ),
 };
 
-#define for_each_subsystem(sys_dir, sys_dirent, sys_next, file, st)	       \
+#define for_each_subsystem(sys_dir, sys_dirent, sys_next)	       \
 	while (!readdir_r(sys_dir, &sys_dirent, &sys_next) && sys_next)	       \
-	if (snprintf(file, MAXPATHLEN, "%s/%s", debugfs_path,	       	       \
-			sys_dirent.d_name) &&		       		       \
-	   (!stat(file, &st)) && (S_ISDIR(st.st_mode)) &&		       \
+	if (sys_dirent.d_type == DT_DIR &&				       \
 	   (strcmp(sys_dirent.d_name, ".")) &&				       \
 	   (strcmp(sys_dirent.d_name, "..")))
 
@@ -136,11 +132,9 @@ static int tp_event_has_id(struct dirent *sys_dir, struct dirent *evt_dir)
 	return 0;
 }
 
-#define for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next, file, st)    \
+#define for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next)	       \
 	while (!readdir_r(evt_dir, &evt_dirent, &evt_next) && evt_next)        \
-	if (snprintf(file, MAXPATHLEN, "%s/%s/%s", debugfs_path,	       \
-		     sys_dirent.d_name, evt_dirent.d_name) &&		       \
-	   (!stat(file, &st)) && (S_ISDIR(st.st_mode)) &&		       \
+	if (evt_dirent.d_type == DT_DIR &&				       \
 	   (strcmp(evt_dirent.d_name, ".")) &&				       \
 	   (strcmp(evt_dirent.d_name, "..")) &&				       \
 	   (!tp_event_has_id(&sys_dirent, &evt_dirent)))
@@ -158,34 +152,39 @@ int valid_debugfs_mount(const char *debugfs)
 	return 0;
 }
 
-static char *tracepoint_id_to_name(u64 config)
+struct tracepoint_path *tracepoint_id_to_path(u64 config)
 {
-	static char tracepoint_name[2 * MAX_EVENT_LENGTH];
+	struct tracepoint_path *path = NULL;
 	DIR *sys_dir, *evt_dir;
 	struct dirent *sys_next, *evt_next, sys_dirent, evt_dirent;
-	struct stat st;
 	char id_buf[4];
-	int fd;
+	int sys_dir_fd, fd;
 	u64 id;
 	char evt_path[MAXPATHLEN];
 
 	if (valid_debugfs_mount(debugfs_path))
-		return "unkown";
+		return NULL;
 
 	sys_dir = opendir(debugfs_path);
 	if (!sys_dir)
 		goto cleanup;
+	sys_dir_fd = dirfd(sys_dir);
 
-	for_each_subsystem(sys_dir, sys_dirent, sys_next, evt_path, st) {
-		evt_dir = opendir(evt_path);
-		if (!evt_dir)
-			goto cleanup;
-		for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next,
-								evt_path, st) {
-			snprintf(evt_path, MAXPATHLEN, "%s/%s/%s/id",
-				 debugfs_path, sys_dirent.d_name,
+	for_each_subsystem(sys_dir, sys_dirent, sys_next) {
+		int dfd = openat(sys_dir_fd, sys_dirent.d_name,
+				 O_RDONLY|O_DIRECTORY), evt_dir_fd;
+		if (dfd == -1)
+			continue;
+		evt_dir = fdopendir(dfd);
+		if (!evt_dir) {
+			close(dfd);
+			continue;
+		}
+		evt_dir_fd = dirfd(evt_dir);
+		for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next) {
+			snprintf(evt_path, MAXPATHLEN, "%s/id",
 				 evt_dirent.d_name);
-			fd = open(evt_path, O_RDONLY);
+			fd = openat(evt_dir_fd, evt_path, O_RDONLY);
 			if (fd < 0)
 				continue;
 			if (read(fd, id_buf, sizeof(id_buf)) < 0) {
@@ -197,10 +196,23 @@ static char *tracepoint_id_to_name(u64 config)
 			if (id == config) {
 				closedir(evt_dir);
 				closedir(sys_dir);
-				snprintf(tracepoint_name, 2 * MAX_EVENT_LENGTH,
-					"%s:%s", sys_dirent.d_name,
-					evt_dirent.d_name);
-				return tracepoint_name;
+				path = calloc(1, sizeof(path));
+				path->system = malloc(MAX_EVENT_LENGTH);
+				if (!path->system) {
+					free(path);
+					return NULL;
+				}
+				path->name = malloc(MAX_EVENT_LENGTH);
+				if (!path->name) {
+					free(path->system);
+					free(path);
+					return NULL;
+				}
+				strncpy(path->system, sys_dirent.d_name,
+					MAX_EVENT_LENGTH);
+				strncpy(path->name, evt_dirent.d_name,
+					MAX_EVENT_LENGTH);
+				return path;
 			}
 		}
 		closedir(evt_dir);
@@ -208,7 +220,25 @@ static char *tracepoint_id_to_name(u64 config)
 
 cleanup:
 	closedir(sys_dir);
-	return "unkown";
+	return NULL;
+}
+
+#define TP_PATH_LEN (MAX_EVENT_LENGTH * 2 + 1)
+static const char *tracepoint_id_to_name(u64 config)
+{
+	static char buf[TP_PATH_LEN];
+	struct tracepoint_path *path;
+
+	path = tracepoint_id_to_path(config);
+	if (path) {
+		snprintf(buf, TP_PATH_LEN, "%s:%s", path->system, path->name);
+		free(path->name);
+		free(path->system);
+		free(path);
+	} else
+		snprintf(buf, TP_PATH_LEN, "%s:%s", "unknown", "unknown");
+
+	return buf;
 }
 
 static int is_cache_op_valid(u8 cache_type, u8 cache_op)
@@ -235,7 +265,7 @@ static char *event_cache_name(u8 cache_type, u8 cache_op, u8 cache_result)
 	return name;
 }
 
-char *event_name(int counter)
+const char *event_name(int counter)
 {
 	u64 config = attrs[counter].config;
 	int type = attrs[counter].type;
@@ -243,7 +273,7 @@ char *event_name(int counter)
 	return __event_name(type, config);
 }
 
-char *__event_name(int type, u64 config)
+const char *__event_name(int type, u64 config)
 {
 	static char buf[32];
 
@@ -294,7 +324,7 @@ char *__event_name(int type, u64 config)
 	return "unknown";
 }
 
-static int parse_aliases(const char **str, char *names[][MAX_ALIASES], int size)
+static int parse_aliases(const char **str, const char *names[][MAX_ALIASES], int size)
 {
 	int i, j;
 	int n, longest = -1;
@@ -598,7 +628,7 @@ static void print_tracepoint_events(void)
 {
 	DIR *sys_dir, *evt_dir;
 	struct dirent *sys_next, *evt_next, sys_dirent, evt_dirent;
-	struct stat st;
+	int sys_dir_fd;
 	char evt_path[MAXPATHLEN];
 
 	if (valid_debugfs_mount(debugfs_path))
@@ -607,13 +637,20 @@ static void print_tracepoint_events(void)
 	sys_dir = opendir(debugfs_path);
 	if (!sys_dir)
 		goto cleanup;
+	sys_dir_fd = dirfd(sys_dir);
 
-	for_each_subsystem(sys_dir, sys_dirent, sys_next, evt_path, st) {
-		evt_dir = opendir(evt_path);
-		if (!evt_dir)
-			goto cleanup;
-		for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next,
-								evt_path, st) {
+	for_each_subsystem(sys_dir, sys_dirent, sys_next) {
+		int dfd = openat(sys_dir_fd, sys_dirent.d_name,
+				 O_RDONLY|O_DIRECTORY), evt_dir_fd;
+		if (dfd == -1)
+			continue;
+		evt_dir = fdopendir(dfd);
+		if (!evt_dir) {
+			close(dfd);
+			continue;
+		}
+		evt_dir_fd = dirfd(evt_dir);
+		for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next) {
 			snprintf(evt_path, MAXPATHLEN, "%s:%s",
 				 sys_dirent.d_name, evt_dirent.d_name);
 			fprintf(stderr, "  %-40s [%s]\n", evt_path,