linux/kernel/trace/trace_output.c

1257 lines
27 KiB
C
Raw Normal View History

/*
* trace_output.c
*
* Copyright (C) 2008 Red Hat Inc, Steven Rostedt <srostedt@redhat.com>
*
*/
#include <linux/module.h>
#include <linux/mutex.h>
#include <linux/ftrace.h>
#include "trace_output.h"
/* must be a power of 2 */
#define EVENT_HASHSIZE 128
DECLARE_RWSEM(trace_event_sem);
static struct hlist_head event_hash[EVENT_HASHSIZE] __read_mostly;
static int next_event_type = __TRACE_LAST_TYPE + 1;
enum print_line_t trace_print_bputs_msg_only(struct trace_iterator *iter)
{
struct trace_seq *s = &iter->seq;
struct trace_entry *entry = iter->ent;
struct bputs_entry *field;
trace_assign_type(field, entry);
trace_seq_puts(s, field->str);
return trace_handle_return(s);
}
enum print_line_t trace_print_bprintk_msg_only(struct trace_iterator *iter)
{
struct trace_seq *s = &iter->seq;
struct trace_entry *entry = iter->ent;
struct bprint_entry *field;
trace_assign_type(field, entry);
trace_seq_bprintf(s, field->fmt, field->buf);
return trace_handle_return(s);
}
enum print_line_t trace_print_printk_msg_only(struct trace_iterator *iter)
{
struct trace_seq *s = &iter->seq;
struct trace_entry *entry = iter->ent;
struct print_entry *field;
trace_assign_type(field, entry);
trace_seq_puts(s, field->buf);
return trace_handle_return(s);
}
const char *
trace_print_flags_seq(struct trace_seq *p, const char *delim,
unsigned long flags,
const struct trace_print_flags *flag_array)
{
unsigned long mask;
const char *str;
const char *ret = trace_seq_buffer_ptr(p);
int i, first = 1;
for (i = 0; flag_array[i].name && flags; i++) {
mask = flag_array[i].mask;
if ((flags & mask) != mask)
continue;
str = flag_array[i].name;
flags &= ~mask;
if (!first && delim)
trace_seq_puts(p, delim);
else
first = 0;
trace_seq_puts(p, str);
}
/* check for left over flags */
if (flags) {
if (!first && delim)
trace_seq_puts(p, delim);
trace_seq_printf(p, "0x%lx", flags);
}
trace_seq_putc(p, 0);
return ret;
}
EXPORT_SYMBOL(trace_print_flags_seq);
const char *
trace_print_symbols_seq(struct trace_seq *p, unsigned long val,
const struct trace_print_flags *symbol_array)
{
int i;
const char *ret = trace_seq_buffer_ptr(p);
for (i = 0; symbol_array[i].name; i++) {
if (val != symbol_array[i].mask)
continue;
trace_seq_puts(p, symbol_array[i].name);
break;
}
if (ret == (const char *)(trace_seq_buffer_ptr(p)))
trace_seq_printf(p, "0x%lx", val);
tracing: Add additional marks to signal very large time deltas Currently, function graph tracer prints "!" or "+" just before function execution time to signal a function overhead, depending on the time. And some tracers tracing latency also print "!" or "+" just after time to signal overhead, depending on the interval between events. Even it is usually enough to do that, we sometimes need to signal for bigger execution time than 100 micro seconds. For example, I used function graph tracer to detect if there is any case that exit_mm() takes too much time. I did following steps in /sys/kernel/debug/tracing. It was easier to detect very large excution time with patched kernel than with original kernel. $ echo exit_mm > set_graph_function $ echo function_graph > current_tracer $ echo > trace $ cat trace_pipe > $LOGFILE ... (do something and terminate logging) $ grep "\\$" $LOGFILE 3) $ 22082032 us | } /* kernel_map_pages */ 3) $ 22082040 us | } /* free_pages_prepare */ 3) $ 22082113 us | } /* free_hot_cold_page */ 3) $ 22083455 us | } /* free_hot_cold_page_list */ 3) $ 22083895 us | } /* release_pages */ 3) $ 22177873 us | } /* free_pages_and_swap_cache */ 3) $ 22178929 us | } /* unmap_single_vma */ 3) $ 22198885 us | } /* unmap_vmas */ 3) $ 22206949 us | } /* exit_mmap */ 3) $ 22207659 us | } /* mmput */ 3) $ 22207793 us | } /* exit_mm */ And then, it was easy to find out that a schedule-out occured by sub_preempt_count() within kernel_map_pages(). To detect very large function exection time caused by either problematic function implementation or scheduling issues, this patch can be useful. Link: http://lkml.kernel.org/r/1416789259-24038-1-git-send-email-byungchul.park@lge.com Signed-off-by: Byungchul Park <byungchul.park@lge.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2014-11-24 00:34:19 +00:00
trace_seq_putc(p, 0);
return ret;
}
EXPORT_SYMBOL(trace_print_symbols_seq);
#if BITS_PER_LONG == 32
const char *
trace_print_symbols_seq_u64(struct trace_seq *p, unsigned long long val,
const struct trace_print_flags_u64 *symbol_array)
{
int i;
const char *ret = trace_seq_buffer_ptr(p);
for (i = 0; symbol_array[i].name; i++) {
if (val != symbol_array[i].mask)
continue;
trace_seq_puts(p, symbol_array[i].name);
break;
}
if (ret == (const char *)(trace_seq_buffer_ptr(p)))
trace_seq_printf(p, "0x%llx", val);
trace_seq_putc(p, 0);
return ret;
}
EXPORT_SYMBOL(trace_print_symbols_seq_u64);
#endif
tracing: Add __bitmask() macro to trace events to cpumasks and other bitmasks Being able to show a cpumask of events can be useful as some events may affect only some CPUs. There is no standard way to record the cpumask and converting it to a string is rather expensive during the trace as traces happen in hotpaths. It would be better to record the raw event mask and be able to parse it at print time. The following macros were added for use with the TRACE_EVENT() macro: __bitmask() __assign_bitmask() __get_bitmask() To test this, I added this to the sched_migrate_task event, which looked like this: TRACE_EVENT(sched_migrate_task, TP_PROTO(struct task_struct *p, int dest_cpu, const struct cpumask *cpus), TP_ARGS(p, dest_cpu, cpus), TP_STRUCT__entry( __array( char, comm, TASK_COMM_LEN ) __field( pid_t, pid ) __field( int, prio ) __field( int, orig_cpu ) __field( int, dest_cpu ) __bitmask( cpumask, num_possible_cpus() ) ), TP_fast_assign( memcpy(__entry->comm, p->comm, TASK_COMM_LEN); __entry->pid = p->pid; __entry->prio = p->prio; __entry->orig_cpu = task_cpu(p); __entry->dest_cpu = dest_cpu; __assign_bitmask(cpumask, cpumask_bits(cpus), num_possible_cpus()); ), TP_printk("comm=%s pid=%d prio=%d orig_cpu=%d dest_cpu=%d cpumask=%s", __entry->comm, __entry->pid, __entry->prio, __entry->orig_cpu, __entry->dest_cpu, __get_bitmask(cpumask)) ); With the output of: ksmtuned-3613 [003] d..2 485.220508: sched_migrate_task: comm=ksmtuned pid=3615 prio=120 orig_cpu=3 dest_cpu=2 cpumask=00000000,0000000f migration/1-13 [001] d..5 485.221202: sched_migrate_task: comm=ksmtuned pid=3614 prio=120 orig_cpu=1 dest_cpu=0 cpumask=00000000,0000000f awk-3615 [002] d.H5 485.221747: sched_migrate_task: comm=rcu_preempt pid=7 prio=120 orig_cpu=0 dest_cpu=1 cpumask=00000000,000000ff migration/2-18 [002] d..5 485.222062: sched_migrate_task: comm=ksmtuned pid=3615 prio=120 orig_cpu=2 dest_cpu=3 cpumask=00000000,0000000f Link: http://lkml.kernel.org/r/1399377998-14870-6-git-send-email-javi.merino@arm.com Link: http://lkml.kernel.org/r/20140506132238.22e136d1@gandalf.local.home Suggested-by: Javi Merino <javi.merino@arm.com> Tested-by: Javi Merino <javi.merino@arm.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2014-05-06 17:10:24 +00:00
const char *
trace_print_bitmask_seq(struct trace_seq *p, void *bitmask_ptr,
unsigned int bitmask_size)
tracing: Add __bitmask() macro to trace events to cpumasks and other bitmasks Being able to show a cpumask of events can be useful as some events may affect only some CPUs. There is no standard way to record the cpumask and converting it to a string is rather expensive during the trace as traces happen in hotpaths. It would be better to record the raw event mask and be able to parse it at print time. The following macros were added for use with the TRACE_EVENT() macro: __bitmask() __assign_bitmask() __get_bitmask() To test this, I added this to the sched_migrate_task event, which looked like this: TRACE_EVENT(sched_migrate_task, TP_PROTO(struct task_struct *p, int dest_cpu, const struct cpumask *cpus), TP_ARGS(p, dest_cpu, cpus), TP_STRUCT__entry( __array( char, comm, TASK_COMM_LEN ) __field( pid_t, pid ) __field( int, prio ) __field( int, orig_cpu ) __field( int, dest_cpu ) __bitmask( cpumask, num_possible_cpus() ) ), TP_fast_assign( memcpy(__entry->comm, p->comm, TASK_COMM_LEN); __entry->pid = p->pid; __entry->prio = p->prio; __entry->orig_cpu = task_cpu(p); __entry->dest_cpu = dest_cpu; __assign_bitmask(cpumask, cpumask_bits(cpus), num_possible_cpus()); ), TP_printk("comm=%s pid=%d prio=%d orig_cpu=%d dest_cpu=%d cpumask=%s", __entry->comm, __entry->pid, __entry->prio, __entry->orig_cpu, __entry->dest_cpu, __get_bitmask(cpumask)) ); With the output of: ksmtuned-3613 [003] d..2 485.220508: sched_migrate_task: comm=ksmtuned pid=3615 prio=120 orig_cpu=3 dest_cpu=2 cpumask=00000000,0000000f migration/1-13 [001] d..5 485.221202: sched_migrate_task: comm=ksmtuned pid=3614 prio=120 orig_cpu=1 dest_cpu=0 cpumask=00000000,0000000f awk-3615 [002] d.H5 485.221747: sched_migrate_task: comm=rcu_preempt pid=7 prio=120 orig_cpu=0 dest_cpu=1 cpumask=00000000,000000ff migration/2-18 [002] d..5 485.222062: sched_migrate_task: comm=ksmtuned pid=3615 prio=120 orig_cpu=2 dest_cpu=3 cpumask=00000000,0000000f Link: http://lkml.kernel.org/r/1399377998-14870-6-git-send-email-javi.merino@arm.com Link: http://lkml.kernel.org/r/20140506132238.22e136d1@gandalf.local.home Suggested-by: Javi Merino <javi.merino@arm.com> Tested-by: Javi Merino <javi.merino@arm.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2014-05-06 17:10:24 +00:00
{
const char *ret = trace_seq_buffer_ptr(p);
tracing: Add __bitmask() macro to trace events to cpumasks and other bitmasks Being able to show a cpumask of events can be useful as some events may affect only some CPUs. There is no standard way to record the cpumask and converting it to a string is rather expensive during the trace as traces happen in hotpaths. It would be better to record the raw event mask and be able to parse it at print time. The following macros were added for use with the TRACE_EVENT() macro: __bitmask() __assign_bitmask() __get_bitmask() To test this, I added this to the sched_migrate_task event, which looked like this: TRACE_EVENT(sched_migrate_task, TP_PROTO(struct task_struct *p, int dest_cpu, const struct cpumask *cpus), TP_ARGS(p, dest_cpu, cpus), TP_STRUCT__entry( __array( char, comm, TASK_COMM_LEN ) __field( pid_t, pid ) __field( int, prio ) __field( int, orig_cpu ) __field( int, dest_cpu ) __bitmask( cpumask, num_possible_cpus() ) ), TP_fast_assign( memcpy(__entry->comm, p->comm, TASK_COMM_LEN); __entry->pid = p->pid; __entry->prio = p->prio; __entry->orig_cpu = task_cpu(p); __entry->dest_cpu = dest_cpu; __assign_bitmask(cpumask, cpumask_bits(cpus), num_possible_cpus()); ), TP_printk("comm=%s pid=%d prio=%d orig_cpu=%d dest_cpu=%d cpumask=%s", __entry->comm, __entry->pid, __entry->prio, __entry->orig_cpu, __entry->dest_cpu, __get_bitmask(cpumask)) ); With the output of: ksmtuned-3613 [003] d..2 485.220508: sched_migrate_task: comm=ksmtuned pid=3615 prio=120 orig_cpu=3 dest_cpu=2 cpumask=00000000,0000000f migration/1-13 [001] d..5 485.221202: sched_migrate_task: comm=ksmtuned pid=3614 prio=120 orig_cpu=1 dest_cpu=0 cpumask=00000000,0000000f awk-3615 [002] d.H5 485.221747: sched_migrate_task: comm=rcu_preempt pid=7 prio=120 orig_cpu=0 dest_cpu=1 cpumask=00000000,000000ff migration/2-18 [002] d..5 485.222062: sched_migrate_task: comm=ksmtuned pid=3615 prio=120 orig_cpu=2 dest_cpu=3 cpumask=00000000,0000000f Link: http://lkml.kernel.org/r/1399377998-14870-6-git-send-email-javi.merino@arm.com Link: http://lkml.kernel.org/r/20140506132238.22e136d1@gandalf.local.home Suggested-by: Javi Merino <javi.merino@arm.com> Tested-by: Javi Merino <javi.merino@arm.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2014-05-06 17:10:24 +00:00
trace_seq_bitmask(p, bitmask_ptr, bitmask_size * 8);
trace_seq_putc(p, 0);
return ret;
}
EXPORT_SYMBOL_GPL(trace_print_bitmask_seq);
tracing: Add __bitmask() macro to trace events to cpumasks and other bitmasks Being able to show a cpumask of events can be useful as some events may affect only some CPUs. There is no standard way to record the cpumask and converting it to a string is rather expensive during the trace as traces happen in hotpaths. It would be better to record the raw event mask and be able to parse it at print time. The following macros were added for use with the TRACE_EVENT() macro: __bitmask() __assign_bitmask() __get_bitmask() To test this, I added this to the sched_migrate_task event, which looked like this: TRACE_EVENT(sched_migrate_task, TP_PROTO(struct task_struct *p, int dest_cpu, const struct cpumask *cpus), TP_ARGS(p, dest_cpu, cpus), TP_STRUCT__entry( __array( char, comm, TASK_COMM_LEN ) __field( pid_t, pid ) __field( int, prio ) __field( int, orig_cpu ) __field( int, dest_cpu ) __bitmask( cpumask, num_possible_cpus() ) ), TP_fast_assign( memcpy(__entry->comm, p->comm, TASK_COMM_LEN); __entry->pid = p->pid; __entry->prio = p->prio; __entry->orig_cpu = task_cpu(p); __entry->dest_cpu = dest_cpu; __assign_bitmask(cpumask, cpumask_bits(cpus), num_possible_cpus()); ), TP_printk("comm=%s pid=%d prio=%d orig_cpu=%d dest_cpu=%d cpumask=%s", __entry->comm, __entry->pid, __entry->prio, __entry->orig_cpu, __entry->dest_cpu, __get_bitmask(cpumask)) ); With the output of: ksmtuned-3613 [003] d..2 485.220508: sched_migrate_task: comm=ksmtuned pid=3615 prio=120 orig_cpu=3 dest_cpu=2 cpumask=00000000,0000000f migration/1-13 [001] d..5 485.221202: sched_migrate_task: comm=ksmtuned pid=3614 prio=120 orig_cpu=1 dest_cpu=0 cpumask=00000000,0000000f awk-3615 [002] d.H5 485.221747: sched_migrate_task: comm=rcu_preempt pid=7 prio=120 orig_cpu=0 dest_cpu=1 cpumask=00000000,000000ff migration/2-18 [002] d..5 485.222062: sched_migrate_task: comm=ksmtuned pid=3615 prio=120 orig_cpu=2 dest_cpu=3 cpumask=00000000,0000000f Link: http://lkml.kernel.org/r/1399377998-14870-6-git-send-email-javi.merino@arm.com Link: http://lkml.kernel.org/r/20140506132238.22e136d1@gandalf.local.home Suggested-by: Javi Merino <javi.merino@arm.com> Tested-by: Javi Merino <javi.merino@arm.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2014-05-06 17:10:24 +00:00
const char *
trace_print_hex_seq(struct trace_seq *p, const unsigned char *buf, int buf_len)
{
int i;
const char *ret = trace_seq_buffer_ptr(p);
for (i = 0; i < buf_len; i++)
trace_seq_printf(p, "%s%2.2x", i == 0 ? "" : " ", buf[i]);
trace_seq_putc(p, 0);
return ret;
}
EXPORT_SYMBOL(trace_print_hex_seq);
const char *
trace_print_array_seq(struct trace_seq *p, const void *buf, int count,
size_t el_size)
{
const char *ret = trace_seq_buffer_ptr(p);
const char *prefix = "";
void *ptr = (void *)buf;
size_t buf_len = count * el_size;
trace_seq_putc(p, '{');
while (ptr < buf + buf_len) {
switch (el_size) {
case 1:
trace_seq_printf(p, "%s0x%x", prefix,
*(u8 *)ptr);
break;
case 2:
trace_seq_printf(p, "%s0x%x", prefix,
*(u16 *)ptr);
break;
case 4:
trace_seq_printf(p, "%s0x%x", prefix,
*(u32 *)ptr);
break;
case 8:
trace_seq_printf(p, "%s0x%llx", prefix,
*(u64 *)ptr);
break;
default:
trace_seq_printf(p, "BAD SIZE:%zu 0x%x", el_size,
*(u8 *)ptr);
el_size = 1;
}
prefix = ",";
ptr += el_size;
}
trace_seq_putc(p, '}');
trace_seq_putc(p, 0);
return ret;
}
EXPORT_SYMBOL(trace_print_array_seq);
int ftrace_raw_output_prep(struct trace_iterator *iter,
struct trace_event *trace_event)
{
struct trace_event_call *event;
struct trace_seq *s = &iter->seq;
struct trace_seq *p = &iter->tmp_seq;
struct trace_entry *entry;
event = container_of(trace_event, struct trace_event_call, event);
entry = iter->ent;
if (entry->type != event->event.type) {
WARN_ON_ONCE(1);
return TRACE_TYPE_UNHANDLED;
}
trace_seq_init(p);
trace_seq_printf(s, "%s: ", ftrace_event_name(event));
return trace_handle_return(s);
}
EXPORT_SYMBOL(ftrace_raw_output_prep);
static int ftrace_output_raw(struct trace_iterator *iter, char *name,
char *fmt, va_list ap)
{
struct trace_seq *s = &iter->seq;
trace_seq_printf(s, "%s: ", name);
trace_seq_vprintf(s, fmt, ap);
return trace_handle_return(s);
}
int ftrace_output_call(struct trace_iterator *iter, char *name, char *fmt, ...)
{
va_list ap;
int ret;
va_start(ap, fmt);
ret = ftrace_output_raw(iter, name, fmt, ap);
va_end(ap);
return ret;
}
EXPORT_SYMBOL_GPL(ftrace_output_call);
#ifdef CONFIG_KRETPROBES
static inline const char *kretprobed(const char *name)
{
static const char tramp_name[] = "kretprobe_trampoline";
int size = sizeof(tramp_name);
if (strncmp(tramp_name, name, size) == 0)
return "[unknown/kretprobe'd]";
return name;
}
#else
static inline const char *kretprobed(const char *name)
{
return name;
}
#endif /* CONFIG_KRETPROBES */
static void
seq_print_sym_short(struct trace_seq *s, const char *fmt, unsigned long address)
{
#ifdef CONFIG_KALLSYMS
char str[KSYM_SYMBOL_LEN];
const char *name;
kallsyms_lookup(address, NULL, NULL, NULL, str);
name = kretprobed(str);
trace_seq_printf(s, fmt, name);
#endif
}
static void
seq_print_sym_offset(struct trace_seq *s, const char *fmt,
unsigned long address)
{
#ifdef CONFIG_KALLSYMS
char str[KSYM_SYMBOL_LEN];
const char *name;
sprint_symbol(str, address);
name = kretprobed(str);
trace_seq_printf(s, fmt, name);
#endif
}
#ifndef CONFIG_64BIT
# define IP_FMT "%08lx"
#else
# define IP_FMT "%016lx"
#endif
int seq_print_user_ip(struct trace_seq *s, struct mm_struct *mm,
unsigned long ip, unsigned long sym_flags)
{
struct file *file = NULL;
unsigned long vmstart = 0;
int ret = 1;
if (s->full)
return 0;
if (mm) {
const struct vm_area_struct *vma;
down_read(&mm->mmap_sem);
vma = find_vma(mm, ip);
if (vma) {
file = vma->vm_file;
vmstart = vma->vm_start;
}
if (file) {
ret = trace_seq_path(s, &file->f_path);
if (ret)
trace_seq_printf(s, "[+0x%lx]",
ip - vmstart);
}
up_read(&mm->mmap_sem);
}
if (ret && ((sym_flags & TRACE_ITER_SYM_ADDR) || !file))
trace_seq_printf(s, " <" IP_FMT ">", ip);
return !trace_seq_has_overflowed(s);
}
int
seq_print_userip_objs(const struct userstack_entry *entry, struct trace_seq *s,
unsigned long sym_flags)
{
struct mm_struct *mm = NULL;
unsigned int i;
if (trace_flags & TRACE_ITER_SYM_USEROBJ) {
struct task_struct *task;
/*
* we do the lookup on the thread group leader,
* since individual threads might have already quit!
*/
rcu_read_lock();
task = find_task_by_vpid(entry->tgid);
if (task)
mm = get_task_mm(task);
rcu_read_unlock();
}
for (i = 0; i < FTRACE_STACK_ENTRIES; i++) {
unsigned long ip = entry->caller[i];
if (ip == ULONG_MAX || trace_seq_has_overflowed(s))
break;
trace_seq_puts(s, " => ");
if (!ip) {
trace_seq_puts(s, "??");
trace_seq_putc(s, '\n');
continue;
}
seq_print_user_ip(s, mm, ip, sym_flags);
trace_seq_putc(s, '\n');
}
if (mm)
mmput(mm);
return !trace_seq_has_overflowed(s);
}
int
seq_print_ip_sym(struct trace_seq *s, unsigned long ip, unsigned long sym_flags)
{
if (!ip) {
trace_seq_putc(s, '0');
goto out;
}
if (sym_flags & TRACE_ITER_SYM_OFFSET)
seq_print_sym_offset(s, "%s", ip);
else
seq_print_sym_short(s, "%s", ip);
if (sym_flags & TRACE_ITER_SYM_ADDR)
trace_seq_printf(s, " <" IP_FMT ">", ip);
out:
return !trace_seq_has_overflowed(s);
}
/**
* trace_print_lat_fmt - print the irq, preempt and lockdep fields
* @s: trace seq struct to write to
* @entry: The trace entry field from the ring buffer
*
* Prints the generic fields of irqs off, in hard or softirq, preempt
* count.
*/
int trace_print_lat_fmt(struct trace_seq *s, struct trace_entry *entry)
trace: better manage the context info for events Impact: make trace_event more convenient for tracers All tracers (for the moment) that use the struct trace_event want to have the context info printed before their own output: the pid/cmdline, cpu, and timestamp. But some other tracers that want to implement their trace_event callbacks will not necessary need these information or they may want to format them as they want. This patch adds a new default-enabled trace option: TRACE_ITER_CONTEXT_INFO When disabled through: echo nocontext-info > /debugfs/tracing/trace_options The pid, cpu and timestamps headers will not be printed. IE with the sched_switch tracer with context-info (default): bash-2935 [001] 100.356561: 2935:120:S ==> [001] 0:140:R <idle> <idle>-0 [000] 100.412804: 0:140:R + [000] 11:115:S events/0 <idle>-0 [000] 100.412816: 0:140:R ==> [000] 11:115:R events/0 events/0-11 [000] 100.412829: 11:115:S ==> [000] 0:140:R <idle> Without context-info: 2935:120:S ==> [001] 0:140:R <idle> 0:140:R + [000] 11:115:S events/0 0:140:R ==> [000] 11:115:R events/0 11:115:S ==> [000] 0:140:R <idle> A tracer can disable it at runtime by clearing the bit TRACE_ITER_CONTEXT_INFO in trace_flags. The print routines were renamed to trace_print_context and trace_print_lat_context, so that they can be used by tracers if they want to use them for one of the trace_event callbacks. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-02-02 22:29:21 +00:00
{
char hardsoft_irq;
char need_resched;
char irqs_off;
int hardirq;
int softirq;
trace: better manage the context info for events Impact: make trace_event more convenient for tracers All tracers (for the moment) that use the struct trace_event want to have the context info printed before their own output: the pid/cmdline, cpu, and timestamp. But some other tracers that want to implement their trace_event callbacks will not necessary need these information or they may want to format them as they want. This patch adds a new default-enabled trace option: TRACE_ITER_CONTEXT_INFO When disabled through: echo nocontext-info > /debugfs/tracing/trace_options The pid, cpu and timestamps headers will not be printed. IE with the sched_switch tracer with context-info (default): bash-2935 [001] 100.356561: 2935:120:S ==> [001] 0:140:R <idle> <idle>-0 [000] 100.412804: 0:140:R + [000] 11:115:S events/0 <idle>-0 [000] 100.412816: 0:140:R ==> [000] 11:115:R events/0 events/0-11 [000] 100.412829: 11:115:S ==> [000] 0:140:R <idle> Without context-info: 2935:120:S ==> [001] 0:140:R <idle> 0:140:R + [000] 11:115:S events/0 0:140:R ==> [000] 11:115:R events/0 11:115:S ==> [000] 0:140:R <idle> A tracer can disable it at runtime by clearing the bit TRACE_ITER_CONTEXT_INFO in trace_flags. The print routines were renamed to trace_print_context and trace_print_lat_context, so that they can be used by tracers if they want to use them for one of the trace_event callbacks. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-02-02 22:29:21 +00:00
hardirq = entry->flags & TRACE_FLAG_HARDIRQ;
softirq = entry->flags & TRACE_FLAG_SOFTIRQ;
irqs_off =
(entry->flags & TRACE_FLAG_IRQS_OFF) ? 'd' :
(entry->flags & TRACE_FLAG_IRQS_NOSUPPORT) ? 'X' :
'.';
switch (entry->flags & (TRACE_FLAG_NEED_RESCHED |
TRACE_FLAG_PREEMPT_RESCHED)) {
case TRACE_FLAG_NEED_RESCHED | TRACE_FLAG_PREEMPT_RESCHED:
need_resched = 'N';
break;
case TRACE_FLAG_NEED_RESCHED:
need_resched = 'n';
break;
case TRACE_FLAG_PREEMPT_RESCHED:
need_resched = 'p';
break;
default:
need_resched = '.';
break;
}
hardsoft_irq =
(hardirq && softirq) ? 'H' :
hardirq ? 'h' :
softirq ? 's' :
'.';
trace_seq_printf(s, "%c%c%c",
irqs_off, need_resched, hardsoft_irq);
trace: better manage the context info for events Impact: make trace_event more convenient for tracers All tracers (for the moment) that use the struct trace_event want to have the context info printed before their own output: the pid/cmdline, cpu, and timestamp. But some other tracers that want to implement their trace_event callbacks will not necessary need these information or they may want to format them as they want. This patch adds a new default-enabled trace option: TRACE_ITER_CONTEXT_INFO When disabled through: echo nocontext-info > /debugfs/tracing/trace_options The pid, cpu and timestamps headers will not be printed. IE with the sched_switch tracer with context-info (default): bash-2935 [001] 100.356561: 2935:120:S ==> [001] 0:140:R <idle> <idle>-0 [000] 100.412804: 0:140:R + [000] 11:115:S events/0 <idle>-0 [000] 100.412816: 0:140:R ==> [000] 11:115:R events/0 events/0-11 [000] 100.412829: 11:115:S ==> [000] 0:140:R <idle> Without context-info: 2935:120:S ==> [001] 0:140:R <idle> 0:140:R + [000] 11:115:S events/0 0:140:R ==> [000] 11:115:R events/0 11:115:S ==> [000] 0:140:R <idle> A tracer can disable it at runtime by clearing the bit TRACE_ITER_CONTEXT_INFO in trace_flags. The print routines were renamed to trace_print_context and trace_print_lat_context, so that they can be used by tracers if they want to use them for one of the trace_event callbacks. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-02-02 22:29:21 +00:00
if (entry->preempt_count)
trace_seq_printf(s, "%x", entry->preempt_count);
else
trace_seq_putc(s, '.');
return !trace_seq_has_overflowed(s);
trace: better manage the context info for events Impact: make trace_event more convenient for tracers All tracers (for the moment) that use the struct trace_event want to have the context info printed before their own output: the pid/cmdline, cpu, and timestamp. But some other tracers that want to implement their trace_event callbacks will not necessary need these information or they may want to format them as they want. This patch adds a new default-enabled trace option: TRACE_ITER_CONTEXT_INFO When disabled through: echo nocontext-info > /debugfs/tracing/trace_options The pid, cpu and timestamps headers will not be printed. IE with the sched_switch tracer with context-info (default): bash-2935 [001] 100.356561: 2935:120:S ==> [001] 0:140:R <idle> <idle>-0 [000] 100.412804: 0:140:R + [000] 11:115:S events/0 <idle>-0 [000] 100.412816: 0:140:R ==> [000] 11:115:R events/0 events/0-11 [000] 100.412829: 11:115:S ==> [000] 0:140:R <idle> Without context-info: 2935:120:S ==> [001] 0:140:R <idle> 0:140:R + [000] 11:115:S events/0 0:140:R ==> [000] 11:115:R events/0 11:115:S ==> [000] 0:140:R <idle> A tracer can disable it at runtime by clearing the bit TRACE_ITER_CONTEXT_INFO in trace_flags. The print routines were renamed to trace_print_context and trace_print_lat_context, so that they can be used by tracers if they want to use them for one of the trace_event callbacks. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-02-02 22:29:21 +00:00
}
static int
lat_print_generic(struct trace_seq *s, struct trace_entry *entry, int cpu)
{
char comm[TASK_COMM_LEN];
trace_find_cmdline(entry->pid, comm);
trace_seq_printf(s, "%8.8s-%-5d %3d",
comm, entry->pid, cpu);
return trace_print_lat_fmt(s, entry);
}
tracing: Add additional marks to signal very large time deltas Currently, function graph tracer prints "!" or "+" just before function execution time to signal a function overhead, depending on the time. And some tracers tracing latency also print "!" or "+" just after time to signal overhead, depending on the interval between events. Even it is usually enough to do that, we sometimes need to signal for bigger execution time than 100 micro seconds. For example, I used function graph tracer to detect if there is any case that exit_mm() takes too much time. I did following steps in /sys/kernel/debug/tracing. It was easier to detect very large excution time with patched kernel than with original kernel. $ echo exit_mm > set_graph_function $ echo function_graph > current_tracer $ echo > trace $ cat trace_pipe > $LOGFILE ... (do something and terminate logging) $ grep "\\$" $LOGFILE 3) $ 22082032 us | } /* kernel_map_pages */ 3) $ 22082040 us | } /* free_pages_prepare */ 3) $ 22082113 us | } /* free_hot_cold_page */ 3) $ 22083455 us | } /* free_hot_cold_page_list */ 3) $ 22083895 us | } /* release_pages */ 3) $ 22177873 us | } /* free_pages_and_swap_cache */ 3) $ 22178929 us | } /* unmap_single_vma */ 3) $ 22198885 us | } /* unmap_vmas */ 3) $ 22206949 us | } /* exit_mmap */ 3) $ 22207659 us | } /* mmput */ 3) $ 22207793 us | } /* exit_mm */ And then, it was easy to find out that a schedule-out occured by sub_preempt_count() within kernel_map_pages(). To detect very large function exection time caused by either problematic function implementation or scheduling issues, this patch can be useful. Link: http://lkml.kernel.org/r/1416789259-24038-1-git-send-email-byungchul.park@lge.com Signed-off-by: Byungchul Park <byungchul.park@lge.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2014-11-24 00:34:19 +00:00
#undef MARK
#define MARK(v, s) {.val = v, .sym = s}
/* trace overhead mark */
static const struct trace_mark {
unsigned long long val; /* unit: nsec */
char sym;
} mark[] = {
MARK(1000000000ULL , '$'), /* 1 sec */
MARK(1000000ULL , '#'), /* 1000 usecs */
MARK(100000ULL , '!'), /* 100 usecs */
MARK(10000ULL , '+'), /* 10 usecs */
};
#undef MARK
char trace_find_mark(unsigned long long d)
{
int i;
int size = ARRAY_SIZE(mark);
for (i = 0; i < size; i++) {
if (d >= mark[i].val)
break;
}
return (i == size) ? ' ' : mark[i].sym;
}
trace: better manage the context info for events Impact: make trace_event more convenient for tracers All tracers (for the moment) that use the struct trace_event want to have the context info printed before their own output: the pid/cmdline, cpu, and timestamp. But some other tracers that want to implement their trace_event callbacks will not necessary need these information or they may want to format them as they want. This patch adds a new default-enabled trace option: TRACE_ITER_CONTEXT_INFO When disabled through: echo nocontext-info > /debugfs/tracing/trace_options The pid, cpu and timestamps headers will not be printed. IE with the sched_switch tracer with context-info (default): bash-2935 [001] 100.356561: 2935:120:S ==> [001] 0:140:R <idle> <idle>-0 [000] 100.412804: 0:140:R + [000] 11:115:S events/0 <idle>-0 [000] 100.412816: 0:140:R ==> [000] 11:115:R events/0 events/0-11 [000] 100.412829: 11:115:S ==> [000] 0:140:R <idle> Without context-info: 2935:120:S ==> [001] 0:140:R <idle> 0:140:R + [000] 11:115:S events/0 0:140:R ==> [000] 11:115:R events/0 11:115:S ==> [000] 0:140:R <idle> A tracer can disable it at runtime by clearing the bit TRACE_ITER_CONTEXT_INFO in trace_flags. The print routines were renamed to trace_print_context and trace_print_lat_context, so that they can be used by tracers if they want to use them for one of the trace_event callbacks. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-02-02 22:29:21 +00:00
static int
tracing: Format non-nanosec times from tsc clock without a decimal point. With the addition of the "tsc" clock, formatting timestamps to look like fractional seconds is misleading. Mark clocks as either in nanoseconds or not, and format non-nanosecond timestamps as decimal integers. Tested: $ cd /sys/kernel/debug/tracing/ $ cat trace_clock [local] global tsc $ echo sched_switch > set_event $ echo 1 > tracing_on ; sleep 0.0005 ; echo 0 > tracing_on $ cat trace <idle>-0 [000] 6330.555552: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=29964 next_prio=120 sleep-29964 [000] 6330.555628: sched_switch: prev_comm=bash prev_pid=29964 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... $ echo 1 > options/latency-format $ cat trace <idle>-0 0 4104553247us+: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=29964 next_prio=120 sleep-29964 0 4104553322us+: sched_switch: prev_comm=bash prev_pid=29964 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... $ echo tsc > trace_clock $ cat trace $ echo 1 > tracing_on ; sleep 0.0005 ; echo 0 > tracing_on $ echo 0 > options/latency-format $ cat trace <idle>-0 [000] 16490053398357: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=31128 next_prio=120 sleep-31128 [000] 16490053588518: sched_switch: prev_comm=bash prev_pid=31128 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... echo 1 > options/latency-format $ cat trace <idle>-0 0 91557653238+: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=31128 next_prio=120 sleep-31128 0 91557843399+: sched_switch: prev_comm=bash prev_pid=31128 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... v2: Move arch-specific bits out of generic code. v4: Fix x86_32 build due to 64-bit division. Google-Bug-Id: 6980623 Link: http://lkml.kernel.org/r/1352837903-32191-2-git-send-email-dhsharp@google.com Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Signed-off-by: David Sharp <dhsharp@google.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2012-11-13 20:18:22 +00:00
lat_print_timestamp(struct trace_iterator *iter, u64 next_ts)
trace: better manage the context info for events Impact: make trace_event more convenient for tracers All tracers (for the moment) that use the struct trace_event want to have the context info printed before their own output: the pid/cmdline, cpu, and timestamp. But some other tracers that want to implement their trace_event callbacks will not necessary need these information or they may want to format them as they want. This patch adds a new default-enabled trace option: TRACE_ITER_CONTEXT_INFO When disabled through: echo nocontext-info > /debugfs/tracing/trace_options The pid, cpu and timestamps headers will not be printed. IE with the sched_switch tracer with context-info (default): bash-2935 [001] 100.356561: 2935:120:S ==> [001] 0:140:R <idle> <idle>-0 [000] 100.412804: 0:140:R + [000] 11:115:S events/0 <idle>-0 [000] 100.412816: 0:140:R ==> [000] 11:115:R events/0 events/0-11 [000] 100.412829: 11:115:S ==> [000] 0:140:R <idle> Without context-info: 2935:120:S ==> [001] 0:140:R <idle> 0:140:R + [000] 11:115:S events/0 0:140:R ==> [000] 11:115:R events/0 11:115:S ==> [000] 0:140:R <idle> A tracer can disable it at runtime by clearing the bit TRACE_ITER_CONTEXT_INFO in trace_flags. The print routines were renamed to trace_print_context and trace_print_lat_context, so that they can be used by tracers if they want to use them for one of the trace_event callbacks. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-02-02 22:29:21 +00:00
{
tracing: Format non-nanosec times from tsc clock without a decimal point. With the addition of the "tsc" clock, formatting timestamps to look like fractional seconds is misleading. Mark clocks as either in nanoseconds or not, and format non-nanosecond timestamps as decimal integers. Tested: $ cd /sys/kernel/debug/tracing/ $ cat trace_clock [local] global tsc $ echo sched_switch > set_event $ echo 1 > tracing_on ; sleep 0.0005 ; echo 0 > tracing_on $ cat trace <idle>-0 [000] 6330.555552: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=29964 next_prio=120 sleep-29964 [000] 6330.555628: sched_switch: prev_comm=bash prev_pid=29964 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... $ echo 1 > options/latency-format $ cat trace <idle>-0 0 4104553247us+: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=29964 next_prio=120 sleep-29964 0 4104553322us+: sched_switch: prev_comm=bash prev_pid=29964 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... $ echo tsc > trace_clock $ cat trace $ echo 1 > tracing_on ; sleep 0.0005 ; echo 0 > tracing_on $ echo 0 > options/latency-format $ cat trace <idle>-0 [000] 16490053398357: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=31128 next_prio=120 sleep-31128 [000] 16490053588518: sched_switch: prev_comm=bash prev_pid=31128 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... echo 1 > options/latency-format $ cat trace <idle>-0 0 91557653238+: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=31128 next_prio=120 sleep-31128 0 91557843399+: sched_switch: prev_comm=bash prev_pid=31128 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... v2: Move arch-specific bits out of generic code. v4: Fix x86_32 build due to 64-bit division. Google-Bug-Id: 6980623 Link: http://lkml.kernel.org/r/1352837903-32191-2-git-send-email-dhsharp@google.com Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Signed-off-by: David Sharp <dhsharp@google.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2012-11-13 20:18:22 +00:00
unsigned long verbose = trace_flags & TRACE_ITER_VERBOSE;
unsigned long in_ns = iter->iter_flags & TRACE_FILE_TIME_IN_NS;
unsigned long long abs_ts = iter->ts - iter->trace_buffer->time_start;
tracing: Format non-nanosec times from tsc clock without a decimal point. With the addition of the "tsc" clock, formatting timestamps to look like fractional seconds is misleading. Mark clocks as either in nanoseconds or not, and format non-nanosecond timestamps as decimal integers. Tested: $ cd /sys/kernel/debug/tracing/ $ cat trace_clock [local] global tsc $ echo sched_switch > set_event $ echo 1 > tracing_on ; sleep 0.0005 ; echo 0 > tracing_on $ cat trace <idle>-0 [000] 6330.555552: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=29964 next_prio=120 sleep-29964 [000] 6330.555628: sched_switch: prev_comm=bash prev_pid=29964 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... $ echo 1 > options/latency-format $ cat trace <idle>-0 0 4104553247us+: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=29964 next_prio=120 sleep-29964 0 4104553322us+: sched_switch: prev_comm=bash prev_pid=29964 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... $ echo tsc > trace_clock $ cat trace $ echo 1 > tracing_on ; sleep 0.0005 ; echo 0 > tracing_on $ echo 0 > options/latency-format $ cat trace <idle>-0 [000] 16490053398357: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=31128 next_prio=120 sleep-31128 [000] 16490053588518: sched_switch: prev_comm=bash prev_pid=31128 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... echo 1 > options/latency-format $ cat trace <idle>-0 0 91557653238+: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=31128 next_prio=120 sleep-31128 0 91557843399+: sched_switch: prev_comm=bash prev_pid=31128 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... v2: Move arch-specific bits out of generic code. v4: Fix x86_32 build due to 64-bit division. Google-Bug-Id: 6980623 Link: http://lkml.kernel.org/r/1352837903-32191-2-git-send-email-dhsharp@google.com Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Signed-off-by: David Sharp <dhsharp@google.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2012-11-13 20:18:22 +00:00
unsigned long long rel_ts = next_ts - iter->ts;
struct trace_seq *s = &iter->seq;
if (in_ns) {
abs_ts = ns2usecs(abs_ts);
rel_ts = ns2usecs(rel_ts);
}
if (verbose && in_ns) {
unsigned long abs_usec = do_div(abs_ts, USEC_PER_MSEC);
unsigned long abs_msec = (unsigned long)abs_ts;
unsigned long rel_usec = do_div(rel_ts, USEC_PER_MSEC);
unsigned long rel_msec = (unsigned long)rel_ts;
trace_seq_printf(
s, "[%08llx] %ld.%03ldms (+%ld.%03ldms): ",
ns2usecs(iter->ts),
abs_msec, abs_usec,
rel_msec, rel_usec);
tracing: Format non-nanosec times from tsc clock without a decimal point. With the addition of the "tsc" clock, formatting timestamps to look like fractional seconds is misleading. Mark clocks as either in nanoseconds or not, and format non-nanosecond timestamps as decimal integers. Tested: $ cd /sys/kernel/debug/tracing/ $ cat trace_clock [local] global tsc $ echo sched_switch > set_event $ echo 1 > tracing_on ; sleep 0.0005 ; echo 0 > tracing_on $ cat trace <idle>-0 [000] 6330.555552: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=29964 next_prio=120 sleep-29964 [000] 6330.555628: sched_switch: prev_comm=bash prev_pid=29964 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... $ echo 1 > options/latency-format $ cat trace <idle>-0 0 4104553247us+: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=29964 next_prio=120 sleep-29964 0 4104553322us+: sched_switch: prev_comm=bash prev_pid=29964 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... $ echo tsc > trace_clock $ cat trace $ echo 1 > tracing_on ; sleep 0.0005 ; echo 0 > tracing_on $ echo 0 > options/latency-format $ cat trace <idle>-0 [000] 16490053398357: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=31128 next_prio=120 sleep-31128 [000] 16490053588518: sched_switch: prev_comm=bash prev_pid=31128 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... echo 1 > options/latency-format $ cat trace <idle>-0 0 91557653238+: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=31128 next_prio=120 sleep-31128 0 91557843399+: sched_switch: prev_comm=bash prev_pid=31128 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... v2: Move arch-specific bits out of generic code. v4: Fix x86_32 build due to 64-bit division. Google-Bug-Id: 6980623 Link: http://lkml.kernel.org/r/1352837903-32191-2-git-send-email-dhsharp@google.com Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Signed-off-by: David Sharp <dhsharp@google.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2012-11-13 20:18:22 +00:00
} else if (verbose && !in_ns) {
trace_seq_printf(
s, "[%016llx] %lld (+%lld): ",
iter->ts, abs_ts, rel_ts);
tracing: Format non-nanosec times from tsc clock without a decimal point. With the addition of the "tsc" clock, formatting timestamps to look like fractional seconds is misleading. Mark clocks as either in nanoseconds or not, and format non-nanosecond timestamps as decimal integers. Tested: $ cd /sys/kernel/debug/tracing/ $ cat trace_clock [local] global tsc $ echo sched_switch > set_event $ echo 1 > tracing_on ; sleep 0.0005 ; echo 0 > tracing_on $ cat trace <idle>-0 [000] 6330.555552: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=29964 next_prio=120 sleep-29964 [000] 6330.555628: sched_switch: prev_comm=bash prev_pid=29964 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... $ echo 1 > options/latency-format $ cat trace <idle>-0 0 4104553247us+: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=29964 next_prio=120 sleep-29964 0 4104553322us+: sched_switch: prev_comm=bash prev_pid=29964 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... $ echo tsc > trace_clock $ cat trace $ echo 1 > tracing_on ; sleep 0.0005 ; echo 0 > tracing_on $ echo 0 > options/latency-format $ cat trace <idle>-0 [000] 16490053398357: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=31128 next_prio=120 sleep-31128 [000] 16490053588518: sched_switch: prev_comm=bash prev_pid=31128 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... echo 1 > options/latency-format $ cat trace <idle>-0 0 91557653238+: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=31128 next_prio=120 sleep-31128 0 91557843399+: sched_switch: prev_comm=bash prev_pid=31128 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... v2: Move arch-specific bits out of generic code. v4: Fix x86_32 build due to 64-bit division. Google-Bug-Id: 6980623 Link: http://lkml.kernel.org/r/1352837903-32191-2-git-send-email-dhsharp@google.com Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Signed-off-by: David Sharp <dhsharp@google.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2012-11-13 20:18:22 +00:00
} else if (!verbose && in_ns) {
trace_seq_printf(
s, " %4lldus%c: ",
abs_ts,
tracing: Add additional marks to signal very large time deltas Currently, function graph tracer prints "!" or "+" just before function execution time to signal a function overhead, depending on the time. And some tracers tracing latency also print "!" or "+" just after time to signal overhead, depending on the interval between events. Even it is usually enough to do that, we sometimes need to signal for bigger execution time than 100 micro seconds. For example, I used function graph tracer to detect if there is any case that exit_mm() takes too much time. I did following steps in /sys/kernel/debug/tracing. It was easier to detect very large excution time with patched kernel than with original kernel. $ echo exit_mm > set_graph_function $ echo function_graph > current_tracer $ echo > trace $ cat trace_pipe > $LOGFILE ... (do something and terminate logging) $ grep "\\$" $LOGFILE 3) $ 22082032 us | } /* kernel_map_pages */ 3) $ 22082040 us | } /* free_pages_prepare */ 3) $ 22082113 us | } /* free_hot_cold_page */ 3) $ 22083455 us | } /* free_hot_cold_page_list */ 3) $ 22083895 us | } /* release_pages */ 3) $ 22177873 us | } /* free_pages_and_swap_cache */ 3) $ 22178929 us | } /* unmap_single_vma */ 3) $ 22198885 us | } /* unmap_vmas */ 3) $ 22206949 us | } /* exit_mmap */ 3) $ 22207659 us | } /* mmput */ 3) $ 22207793 us | } /* exit_mm */ And then, it was easy to find out that a schedule-out occured by sub_preempt_count() within kernel_map_pages(). To detect very large function exection time caused by either problematic function implementation or scheduling issues, this patch can be useful. Link: http://lkml.kernel.org/r/1416789259-24038-1-git-send-email-byungchul.park@lge.com Signed-off-by: Byungchul Park <byungchul.park@lge.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2014-11-24 00:34:19 +00:00
trace_find_mark(rel_ts * NSEC_PER_USEC));
tracing: Format non-nanosec times from tsc clock without a decimal point. With the addition of the "tsc" clock, formatting timestamps to look like fractional seconds is misleading. Mark clocks as either in nanoseconds or not, and format non-nanosecond timestamps as decimal integers. Tested: $ cd /sys/kernel/debug/tracing/ $ cat trace_clock [local] global tsc $ echo sched_switch > set_event $ echo 1 > tracing_on ; sleep 0.0005 ; echo 0 > tracing_on $ cat trace <idle>-0 [000] 6330.555552: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=29964 next_prio=120 sleep-29964 [000] 6330.555628: sched_switch: prev_comm=bash prev_pid=29964 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... $ echo 1 > options/latency-format $ cat trace <idle>-0 0 4104553247us+: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=29964 next_prio=120 sleep-29964 0 4104553322us+: sched_switch: prev_comm=bash prev_pid=29964 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... $ echo tsc > trace_clock $ cat trace $ echo 1 > tracing_on ; sleep 0.0005 ; echo 0 > tracing_on $ echo 0 > options/latency-format $ cat trace <idle>-0 [000] 16490053398357: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=31128 next_prio=120 sleep-31128 [000] 16490053588518: sched_switch: prev_comm=bash prev_pid=31128 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... echo 1 > options/latency-format $ cat trace <idle>-0 0 91557653238+: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=31128 next_prio=120 sleep-31128 0 91557843399+: sched_switch: prev_comm=bash prev_pid=31128 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... v2: Move arch-specific bits out of generic code. v4: Fix x86_32 build due to 64-bit division. Google-Bug-Id: 6980623 Link: http://lkml.kernel.org/r/1352837903-32191-2-git-send-email-dhsharp@google.com Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Signed-off-by: David Sharp <dhsharp@google.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2012-11-13 20:18:22 +00:00
} else { /* !verbose && !in_ns */
trace_seq_printf(s, " %4lld: ", abs_ts);
tracing: Format non-nanosec times from tsc clock without a decimal point. With the addition of the "tsc" clock, formatting timestamps to look like fractional seconds is misleading. Mark clocks as either in nanoseconds or not, and format non-nanosecond timestamps as decimal integers. Tested: $ cd /sys/kernel/debug/tracing/ $ cat trace_clock [local] global tsc $ echo sched_switch > set_event $ echo 1 > tracing_on ; sleep 0.0005 ; echo 0 > tracing_on $ cat trace <idle>-0 [000] 6330.555552: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=29964 next_prio=120 sleep-29964 [000] 6330.555628: sched_switch: prev_comm=bash prev_pid=29964 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... $ echo 1 > options/latency-format $ cat trace <idle>-0 0 4104553247us+: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=29964 next_prio=120 sleep-29964 0 4104553322us+: sched_switch: prev_comm=bash prev_pid=29964 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... $ echo tsc > trace_clock $ cat trace $ echo 1 > tracing_on ; sleep 0.0005 ; echo 0 > tracing_on $ echo 0 > options/latency-format $ cat trace <idle>-0 [000] 16490053398357: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=31128 next_prio=120 sleep-31128 [000] 16490053588518: sched_switch: prev_comm=bash prev_pid=31128 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... echo 1 > options/latency-format $ cat trace <idle>-0 0 91557653238+: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=31128 next_prio=120 sleep-31128 0 91557843399+: sched_switch: prev_comm=bash prev_pid=31128 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... v2: Move arch-specific bits out of generic code. v4: Fix x86_32 build due to 64-bit division. Google-Bug-Id: 6980623 Link: http://lkml.kernel.org/r/1352837903-32191-2-git-send-email-dhsharp@google.com Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Signed-off-by: David Sharp <dhsharp@google.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2012-11-13 20:18:22 +00:00
}
return !trace_seq_has_overflowed(s);
trace: better manage the context info for events Impact: make trace_event more convenient for tracers All tracers (for the moment) that use the struct trace_event want to have the context info printed before their own output: the pid/cmdline, cpu, and timestamp. But some other tracers that want to implement their trace_event callbacks will not necessary need these information or they may want to format them as they want. This patch adds a new default-enabled trace option: TRACE_ITER_CONTEXT_INFO When disabled through: echo nocontext-info > /debugfs/tracing/trace_options The pid, cpu and timestamps headers will not be printed. IE with the sched_switch tracer with context-info (default): bash-2935 [001] 100.356561: 2935:120:S ==> [001] 0:140:R <idle> <idle>-0 [000] 100.412804: 0:140:R + [000] 11:115:S events/0 <idle>-0 [000] 100.412816: 0:140:R ==> [000] 11:115:R events/0 events/0-11 [000] 100.412829: 11:115:S ==> [000] 0:140:R <idle> Without context-info: 2935:120:S ==> [001] 0:140:R <idle> 0:140:R + [000] 11:115:S events/0 0:140:R ==> [000] 11:115:R events/0 11:115:S ==> [000] 0:140:R <idle> A tracer can disable it at runtime by clearing the bit TRACE_ITER_CONTEXT_INFO in trace_flags. The print routines were renamed to trace_print_context and trace_print_lat_context, so that they can be used by tracers if they want to use them for one of the trace_event callbacks. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-02-02 22:29:21 +00:00
}
int trace_print_context(struct trace_iterator *iter)
{
struct trace_seq *s = &iter->seq;
struct trace_entry *entry = iter->ent;
tracing: Format non-nanosec times from tsc clock without a decimal point. With the addition of the "tsc" clock, formatting timestamps to look like fractional seconds is misleading. Mark clocks as either in nanoseconds or not, and format non-nanosecond timestamps as decimal integers. Tested: $ cd /sys/kernel/debug/tracing/ $ cat trace_clock [local] global tsc $ echo sched_switch > set_event $ echo 1 > tracing_on ; sleep 0.0005 ; echo 0 > tracing_on $ cat trace <idle>-0 [000] 6330.555552: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=29964 next_prio=120 sleep-29964 [000] 6330.555628: sched_switch: prev_comm=bash prev_pid=29964 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... $ echo 1 > options/latency-format $ cat trace <idle>-0 0 4104553247us+: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=29964 next_prio=120 sleep-29964 0 4104553322us+: sched_switch: prev_comm=bash prev_pid=29964 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... $ echo tsc > trace_clock $ cat trace $ echo 1 > tracing_on ; sleep 0.0005 ; echo 0 > tracing_on $ echo 0 > options/latency-format $ cat trace <idle>-0 [000] 16490053398357: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=31128 next_prio=120 sleep-31128 [000] 16490053588518: sched_switch: prev_comm=bash prev_pid=31128 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... echo 1 > options/latency-format $ cat trace <idle>-0 0 91557653238+: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=31128 next_prio=120 sleep-31128 0 91557843399+: sched_switch: prev_comm=bash prev_pid=31128 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... v2: Move arch-specific bits out of generic code. v4: Fix x86_32 build due to 64-bit division. Google-Bug-Id: 6980623 Link: http://lkml.kernel.org/r/1352837903-32191-2-git-send-email-dhsharp@google.com Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Signed-off-by: David Sharp <dhsharp@google.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2012-11-13 20:18:22 +00:00
unsigned long long t;
unsigned long secs, usec_rem;
char comm[TASK_COMM_LEN];
trace_find_cmdline(entry->pid, comm);
trace: better manage the context info for events Impact: make trace_event more convenient for tracers All tracers (for the moment) that use the struct trace_event want to have the context info printed before their own output: the pid/cmdline, cpu, and timestamp. But some other tracers that want to implement their trace_event callbacks will not necessary need these information or they may want to format them as they want. This patch adds a new default-enabled trace option: TRACE_ITER_CONTEXT_INFO When disabled through: echo nocontext-info > /debugfs/tracing/trace_options The pid, cpu and timestamps headers will not be printed. IE with the sched_switch tracer with context-info (default): bash-2935 [001] 100.356561: 2935:120:S ==> [001] 0:140:R <idle> <idle>-0 [000] 100.412804: 0:140:R + [000] 11:115:S events/0 <idle>-0 [000] 100.412816: 0:140:R ==> [000] 11:115:R events/0 events/0-11 [000] 100.412829: 11:115:S ==> [000] 0:140:R <idle> Without context-info: 2935:120:S ==> [001] 0:140:R <idle> 0:140:R + [000] 11:115:S events/0 0:140:R ==> [000] 11:115:R events/0 11:115:S ==> [000] 0:140:R <idle> A tracer can disable it at runtime by clearing the bit TRACE_ITER_CONTEXT_INFO in trace_flags. The print routines were renamed to trace_print_context and trace_print_lat_context, so that they can be used by tracers if they want to use them for one of the trace_event callbacks. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-02-02 22:29:21 +00:00
trace_seq_printf(s, "%16s-%-5d [%03d] ",
tracing: Add irq, preempt-count and need resched info to default trace output People keep asking how to get the preempt count, irq, and need resched info and we keep telling them to enable the latency format. Some developers think that traces without this info is completely useless, and for a lot of tasks it is useless. The first option was to enable the latency trace as the default format, but the header for the latency format is pretty useless for most tracers and it also does the timestamp in straight microseconds from the time the trace started. This is sometimes more difficult to read as the default trace is seconds from the start of boot up. Latency format: # tracer: nop # # nop latency trace v1.1.5 on 3.2.0-rc1-test+ # -------------------------------------------------------------------- # latency: 0 us, #159771/64234230, CPU#1 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: -0 (uid:0 nice:0 policy:0 rt_prio:0) # ----------------- # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / migratio-6 0...2 41778231us+: rcu_note_context_switch <-__schedule migratio-6 0...2 41778233us : trace_rcu_utilization <-rcu_note_context_switch migratio-6 0...2 41778235us+: rcu_sched_qs <-rcu_note_context_switch migratio-6 0d..2 41778236us+: rcu_preempt_qs <-rcu_note_context_switch migratio-6 0...2 41778238us : trace_rcu_utilization <-rcu_note_context_switch migratio-6 0...2 41778239us+: debug_lockdep_rcu_enabled <-__schedule default format: # tracer: nop # # TASK-PID CPU# TIMESTAMP FUNCTION # | | | | | migration/0-6 [000] 50.025810: rcu_note_context_switch <-__schedule migration/0-6 [000] 50.025812: trace_rcu_utilization <-rcu_note_context_switch migration/0-6 [000] 50.025813: rcu_sched_qs <-rcu_note_context_switch migration/0-6 [000] 50.025815: rcu_preempt_qs <-rcu_note_context_switch migration/0-6 [000] 50.025817: trace_rcu_utilization <-rcu_note_context_switch migration/0-6 [000] 50.025818: debug_lockdep_rcu_enabled <-__schedule migration/0-6 [000] 50.025820: debug_lockdep_rcu_enabled <-__schedule The latency format header has latency information that is pretty meaningless for most tracers. Although some of the header is useful, and we can add that later to the default format as well. What is really useful with the latency format is the irqs-off, need-resched hard/softirq context and the preempt count. This commit adds the option irq-info which is on by default that adds this information: # tracer: nop # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | <idle>-0 [000] d..2 49.309305: cpuidle_get_driver <-cpuidle_idle_call <idle>-0 [000] d..2 49.309307: mwait_idle <-cpu_idle <idle>-0 [000] d..2 49.309309: need_resched <-mwait_idle <idle>-0 [000] d..2 49.309310: test_ti_thread_flag <-need_resched <idle>-0 [000] d..2 49.309312: trace_power_start.constprop.13 <-mwait_idle <idle>-0 [000] d..2 49.309313: trace_cpu_idle <-mwait_idle <idle>-0 [000] d..2 49.309315: need_resched <-mwait_idle If a user wants the old format, they can disable the 'irq-info' option: # tracer: nop # # TASK-PID CPU# TIMESTAMP FUNCTION # | | | | | <idle>-0 [000] 49.309305: cpuidle_get_driver <-cpuidle_idle_call <idle>-0 [000] 49.309307: mwait_idle <-cpu_idle <idle>-0 [000] 49.309309: need_resched <-mwait_idle <idle>-0 [000] 49.309310: test_ti_thread_flag <-need_resched <idle>-0 [000] 49.309312: trace_power_start.constprop.13 <-mwait_idle <idle>-0 [000] 49.309313: trace_cpu_idle <-mwait_idle <idle>-0 [000] 49.309315: need_resched <-mwait_idle Requested-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2011-11-17 14:34:33 +00:00
comm, entry->pid, iter->cpu);
if (trace_flags & TRACE_ITER_IRQ_INFO)
trace_print_lat_fmt(s, entry);
tracing: Add irq, preempt-count and need resched info to default trace output People keep asking how to get the preempt count, irq, and need resched info and we keep telling them to enable the latency format. Some developers think that traces without this info is completely useless, and for a lot of tasks it is useless. The first option was to enable the latency trace as the default format, but the header for the latency format is pretty useless for most tracers and it also does the timestamp in straight microseconds from the time the trace started. This is sometimes more difficult to read as the default trace is seconds from the start of boot up. Latency format: # tracer: nop # # nop latency trace v1.1.5 on 3.2.0-rc1-test+ # -------------------------------------------------------------------- # latency: 0 us, #159771/64234230, CPU#1 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4) # ----------------- # | task: -0 (uid:0 nice:0 policy:0 rt_prio:0) # ----------------- # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / migratio-6 0...2 41778231us+: rcu_note_context_switch <-__schedule migratio-6 0...2 41778233us : trace_rcu_utilization <-rcu_note_context_switch migratio-6 0...2 41778235us+: rcu_sched_qs <-rcu_note_context_switch migratio-6 0d..2 41778236us+: rcu_preempt_qs <-rcu_note_context_switch migratio-6 0...2 41778238us : trace_rcu_utilization <-rcu_note_context_switch migratio-6 0...2 41778239us+: debug_lockdep_rcu_enabled <-__schedule default format: # tracer: nop # # TASK-PID CPU# TIMESTAMP FUNCTION # | | | | | migration/0-6 [000] 50.025810: rcu_note_context_switch <-__schedule migration/0-6 [000] 50.025812: trace_rcu_utilization <-rcu_note_context_switch migration/0-6 [000] 50.025813: rcu_sched_qs <-rcu_note_context_switch migration/0-6 [000] 50.025815: rcu_preempt_qs <-rcu_note_context_switch migration/0-6 [000] 50.025817: trace_rcu_utilization <-rcu_note_context_switch migration/0-6 [000] 50.025818: debug_lockdep_rcu_enabled <-__schedule migration/0-6 [000] 50.025820: debug_lockdep_rcu_enabled <-__schedule The latency format header has latency information that is pretty meaningless for most tracers. Although some of the header is useful, and we can add that later to the default format as well. What is really useful with the latency format is the irqs-off, need-resched hard/softirq context and the preempt count. This commit adds the option irq-info which is on by default that adds this information: # tracer: nop # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | <idle>-0 [000] d..2 49.309305: cpuidle_get_driver <-cpuidle_idle_call <idle>-0 [000] d..2 49.309307: mwait_idle <-cpu_idle <idle>-0 [000] d..2 49.309309: need_resched <-mwait_idle <idle>-0 [000] d..2 49.309310: test_ti_thread_flag <-need_resched <idle>-0 [000] d..2 49.309312: trace_power_start.constprop.13 <-mwait_idle <idle>-0 [000] d..2 49.309313: trace_cpu_idle <-mwait_idle <idle>-0 [000] d..2 49.309315: need_resched <-mwait_idle If a user wants the old format, they can disable the 'irq-info' option: # tracer: nop # # TASK-PID CPU# TIMESTAMP FUNCTION # | | | | | <idle>-0 [000] 49.309305: cpuidle_get_driver <-cpuidle_idle_call <idle>-0 [000] 49.309307: mwait_idle <-cpu_idle <idle>-0 [000] 49.309309: need_resched <-mwait_idle <idle>-0 [000] 49.309310: test_ti_thread_flag <-need_resched <idle>-0 [000] 49.309312: trace_power_start.constprop.13 <-mwait_idle <idle>-0 [000] 49.309313: trace_cpu_idle <-mwait_idle <idle>-0 [000] 49.309315: need_resched <-mwait_idle Requested-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2011-11-17 14:34:33 +00:00
tracing: Format non-nanosec times from tsc clock without a decimal point. With the addition of the "tsc" clock, formatting timestamps to look like fractional seconds is misleading. Mark clocks as either in nanoseconds or not, and format non-nanosecond timestamps as decimal integers. Tested: $ cd /sys/kernel/debug/tracing/ $ cat trace_clock [local] global tsc $ echo sched_switch > set_event $ echo 1 > tracing_on ; sleep 0.0005 ; echo 0 > tracing_on $ cat trace <idle>-0 [000] 6330.555552: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=29964 next_prio=120 sleep-29964 [000] 6330.555628: sched_switch: prev_comm=bash prev_pid=29964 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... $ echo 1 > options/latency-format $ cat trace <idle>-0 0 4104553247us+: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=29964 next_prio=120 sleep-29964 0 4104553322us+: sched_switch: prev_comm=bash prev_pid=29964 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... $ echo tsc > trace_clock $ cat trace $ echo 1 > tracing_on ; sleep 0.0005 ; echo 0 > tracing_on $ echo 0 > options/latency-format $ cat trace <idle>-0 [000] 16490053398357: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=31128 next_prio=120 sleep-31128 [000] 16490053588518: sched_switch: prev_comm=bash prev_pid=31128 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... echo 1 > options/latency-format $ cat trace <idle>-0 0 91557653238+: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=31128 next_prio=120 sleep-31128 0 91557843399+: sched_switch: prev_comm=bash prev_pid=31128 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... v2: Move arch-specific bits out of generic code. v4: Fix x86_32 build due to 64-bit division. Google-Bug-Id: 6980623 Link: http://lkml.kernel.org/r/1352837903-32191-2-git-send-email-dhsharp@google.com Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Signed-off-by: David Sharp <dhsharp@google.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2012-11-13 20:18:22 +00:00
if (iter->iter_flags & TRACE_FILE_TIME_IN_NS) {
t = ns2usecs(iter->ts);
usec_rem = do_div(t, USEC_PER_SEC);
secs = (unsigned long)t;
trace_seq_printf(s, " %5lu.%06lu: ", secs, usec_rem);
tracing: Format non-nanosec times from tsc clock without a decimal point. With the addition of the "tsc" clock, formatting timestamps to look like fractional seconds is misleading. Mark clocks as either in nanoseconds or not, and format non-nanosecond timestamps as decimal integers. Tested: $ cd /sys/kernel/debug/tracing/ $ cat trace_clock [local] global tsc $ echo sched_switch > set_event $ echo 1 > tracing_on ; sleep 0.0005 ; echo 0 > tracing_on $ cat trace <idle>-0 [000] 6330.555552: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=29964 next_prio=120 sleep-29964 [000] 6330.555628: sched_switch: prev_comm=bash prev_pid=29964 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... $ echo 1 > options/latency-format $ cat trace <idle>-0 0 4104553247us+: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=29964 next_prio=120 sleep-29964 0 4104553322us+: sched_switch: prev_comm=bash prev_pid=29964 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... $ echo tsc > trace_clock $ cat trace $ echo 1 > tracing_on ; sleep 0.0005 ; echo 0 > tracing_on $ echo 0 > options/latency-format $ cat trace <idle>-0 [000] 16490053398357: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=31128 next_prio=120 sleep-31128 [000] 16490053588518: sched_switch: prev_comm=bash prev_pid=31128 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... echo 1 > options/latency-format $ cat trace <idle>-0 0 91557653238+: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=31128 next_prio=120 sleep-31128 0 91557843399+: sched_switch: prev_comm=bash prev_pid=31128 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... v2: Move arch-specific bits out of generic code. v4: Fix x86_32 build due to 64-bit division. Google-Bug-Id: 6980623 Link: http://lkml.kernel.org/r/1352837903-32191-2-git-send-email-dhsharp@google.com Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Signed-off-by: David Sharp <dhsharp@google.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2012-11-13 20:18:22 +00:00
} else
trace_seq_printf(s, " %12llu: ", iter->ts);
return !trace_seq_has_overflowed(s);
trace: better manage the context info for events Impact: make trace_event more convenient for tracers All tracers (for the moment) that use the struct trace_event want to have the context info printed before their own output: the pid/cmdline, cpu, and timestamp. But some other tracers that want to implement their trace_event callbacks will not necessary need these information or they may want to format them as they want. This patch adds a new default-enabled trace option: TRACE_ITER_CONTEXT_INFO When disabled through: echo nocontext-info > /debugfs/tracing/trace_options The pid, cpu and timestamps headers will not be printed. IE with the sched_switch tracer with context-info (default): bash-2935 [001] 100.356561: 2935:120:S ==> [001] 0:140:R <idle> <idle>-0 [000] 100.412804: 0:140:R + [000] 11:115:S events/0 <idle>-0 [000] 100.412816: 0:140:R ==> [000] 11:115:R events/0 events/0-11 [000] 100.412829: 11:115:S ==> [000] 0:140:R <idle> Without context-info: 2935:120:S ==> [001] 0:140:R <idle> 0:140:R + [000] 11:115:S events/0 0:140:R ==> [000] 11:115:R events/0 11:115:S ==> [000] 0:140:R <idle> A tracer can disable it at runtime by clearing the bit TRACE_ITER_CONTEXT_INFO in trace_flags. The print routines were renamed to trace_print_context and trace_print_lat_context, so that they can be used by tracers if they want to use them for one of the trace_event callbacks. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-02-02 22:29:21 +00:00
}
int trace_print_lat_context(struct trace_iterator *iter)
{
u64 next_ts;
/* trace_find_next_entry will reset ent_size */
int ent_size = iter->ent_size;
trace: better manage the context info for events Impact: make trace_event more convenient for tracers All tracers (for the moment) that use the struct trace_event want to have the context info printed before their own output: the pid/cmdline, cpu, and timestamp. But some other tracers that want to implement their trace_event callbacks will not necessary need these information or they may want to format them as they want. This patch adds a new default-enabled trace option: TRACE_ITER_CONTEXT_INFO When disabled through: echo nocontext-info > /debugfs/tracing/trace_options The pid, cpu and timestamps headers will not be printed. IE with the sched_switch tracer with context-info (default): bash-2935 [001] 100.356561: 2935:120:S ==> [001] 0:140:R <idle> <idle>-0 [000] 100.412804: 0:140:R + [000] 11:115:S events/0 <idle>-0 [000] 100.412816: 0:140:R ==> [000] 11:115:R events/0 events/0-11 [000] 100.412829: 11:115:S ==> [000] 0:140:R <idle> Without context-info: 2935:120:S ==> [001] 0:140:R <idle> 0:140:R + [000] 11:115:S events/0 0:140:R ==> [000] 11:115:R events/0 11:115:S ==> [000] 0:140:R <idle> A tracer can disable it at runtime by clearing the bit TRACE_ITER_CONTEXT_INFO in trace_flags. The print routines were renamed to trace_print_context and trace_print_lat_context, so that they can be used by tracers if they want to use them for one of the trace_event callbacks. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-02-02 22:29:21 +00:00
struct trace_seq *s = &iter->seq;
struct trace_entry *entry = iter->ent,
*next_entry = trace_find_next_entry(iter, NULL,
&next_ts);
unsigned long verbose = (trace_flags & TRACE_ITER_VERBOSE);
/* Restore the original ent_size */
iter->ent_size = ent_size;
trace: better manage the context info for events Impact: make trace_event more convenient for tracers All tracers (for the moment) that use the struct trace_event want to have the context info printed before their own output: the pid/cmdline, cpu, and timestamp. But some other tracers that want to implement their trace_event callbacks will not necessary need these information or they may want to format them as they want. This patch adds a new default-enabled trace option: TRACE_ITER_CONTEXT_INFO When disabled through: echo nocontext-info > /debugfs/tracing/trace_options The pid, cpu and timestamps headers will not be printed. IE with the sched_switch tracer with context-info (default): bash-2935 [001] 100.356561: 2935:120:S ==> [001] 0:140:R <idle> <idle>-0 [000] 100.412804: 0:140:R + [000] 11:115:S events/0 <idle>-0 [000] 100.412816: 0:140:R ==> [000] 11:115:R events/0 events/0-11 [000] 100.412829: 11:115:S ==> [000] 0:140:R <idle> Without context-info: 2935:120:S ==> [001] 0:140:R <idle> 0:140:R + [000] 11:115:S events/0 0:140:R ==> [000] 11:115:R events/0 11:115:S ==> [000] 0:140:R <idle> A tracer can disable it at runtime by clearing the bit TRACE_ITER_CONTEXT_INFO in trace_flags. The print routines were renamed to trace_print_context and trace_print_lat_context, so that they can be used by tracers if they want to use them for one of the trace_event callbacks. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-02-02 22:29:21 +00:00
if (!next_entry)
next_ts = iter->ts;
if (verbose) {
char comm[TASK_COMM_LEN];
trace_find_cmdline(entry->pid, comm);
trace_seq_printf(
s, "%16s %5d %3d %d %08x %08lx ",
comm, entry->pid, iter->cpu, entry->flags,
entry->preempt_count, iter->idx);
trace: better manage the context info for events Impact: make trace_event more convenient for tracers All tracers (for the moment) that use the struct trace_event want to have the context info printed before their own output: the pid/cmdline, cpu, and timestamp. But some other tracers that want to implement their trace_event callbacks will not necessary need these information or they may want to format them as they want. This patch adds a new default-enabled trace option: TRACE_ITER_CONTEXT_INFO When disabled through: echo nocontext-info > /debugfs/tracing/trace_options The pid, cpu and timestamps headers will not be printed. IE with the sched_switch tracer with context-info (default): bash-2935 [001] 100.356561: 2935:120:S ==> [001] 0:140:R <idle> <idle>-0 [000] 100.412804: 0:140:R + [000] 11:115:S events/0 <idle>-0 [000] 100.412816: 0:140:R ==> [000] 11:115:R events/0 events/0-11 [000] 100.412829: 11:115:S ==> [000] 0:140:R <idle> Without context-info: 2935:120:S ==> [001] 0:140:R <idle> 0:140:R + [000] 11:115:S events/0 0:140:R ==> [000] 11:115:R events/0 11:115:S ==> [000] 0:140:R <idle> A tracer can disable it at runtime by clearing the bit TRACE_ITER_CONTEXT_INFO in trace_flags. The print routines were renamed to trace_print_context and trace_print_lat_context, so that they can be used by tracers if they want to use them for one of the trace_event callbacks. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-02-02 22:29:21 +00:00
} else {
lat_print_generic(s, entry, iter->cpu);
trace: better manage the context info for events Impact: make trace_event more convenient for tracers All tracers (for the moment) that use the struct trace_event want to have the context info printed before their own output: the pid/cmdline, cpu, and timestamp. But some other tracers that want to implement their trace_event callbacks will not necessary need these information or they may want to format them as they want. This patch adds a new default-enabled trace option: TRACE_ITER_CONTEXT_INFO When disabled through: echo nocontext-info > /debugfs/tracing/trace_options The pid, cpu and timestamps headers will not be printed. IE with the sched_switch tracer with context-info (default): bash-2935 [001] 100.356561: 2935:120:S ==> [001] 0:140:R <idle> <idle>-0 [000] 100.412804: 0:140:R + [000] 11:115:S events/0 <idle>-0 [000] 100.412816: 0:140:R ==> [000] 11:115:R events/0 events/0-11 [000] 100.412829: 11:115:S ==> [000] 0:140:R <idle> Without context-info: 2935:120:S ==> [001] 0:140:R <idle> 0:140:R + [000] 11:115:S events/0 0:140:R ==> [000] 11:115:R events/0 11:115:S ==> [000] 0:140:R <idle> A tracer can disable it at runtime by clearing the bit TRACE_ITER_CONTEXT_INFO in trace_flags. The print routines were renamed to trace_print_context and trace_print_lat_context, so that they can be used by tracers if they want to use them for one of the trace_event callbacks. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-02-02 22:29:21 +00:00
}
lat_print_timestamp(iter, next_ts);
tracing: Format non-nanosec times from tsc clock without a decimal point. With the addition of the "tsc" clock, formatting timestamps to look like fractional seconds is misleading. Mark clocks as either in nanoseconds or not, and format non-nanosecond timestamps as decimal integers. Tested: $ cd /sys/kernel/debug/tracing/ $ cat trace_clock [local] global tsc $ echo sched_switch > set_event $ echo 1 > tracing_on ; sleep 0.0005 ; echo 0 > tracing_on $ cat trace <idle>-0 [000] 6330.555552: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=29964 next_prio=120 sleep-29964 [000] 6330.555628: sched_switch: prev_comm=bash prev_pid=29964 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... $ echo 1 > options/latency-format $ cat trace <idle>-0 0 4104553247us+: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=29964 next_prio=120 sleep-29964 0 4104553322us+: sched_switch: prev_comm=bash prev_pid=29964 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... $ echo tsc > trace_clock $ cat trace $ echo 1 > tracing_on ; sleep 0.0005 ; echo 0 > tracing_on $ echo 0 > options/latency-format $ cat trace <idle>-0 [000] 16490053398357: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=31128 next_prio=120 sleep-31128 [000] 16490053588518: sched_switch: prev_comm=bash prev_pid=31128 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... echo 1 > options/latency-format $ cat trace <idle>-0 0 91557653238+: sched_switch: prev_comm=swapper prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=bash next_pid=31128 next_prio=120 sleep-31128 0 91557843399+: sched_switch: prev_comm=bash prev_pid=31128 prev_prio=120 prev_state=S ==> next_comm=swapper next_pid=0 next_prio=120 ... v2: Move arch-specific bits out of generic code. v4: Fix x86_32 build due to 64-bit division. Google-Bug-Id: 6980623 Link: http://lkml.kernel.org/r/1352837903-32191-2-git-send-email-dhsharp@google.com Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Signed-off-by: David Sharp <dhsharp@google.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2012-11-13 20:18:22 +00:00
return !trace_seq_has_overflowed(s);
trace: better manage the context info for events Impact: make trace_event more convenient for tracers All tracers (for the moment) that use the struct trace_event want to have the context info printed before their own output: the pid/cmdline, cpu, and timestamp. But some other tracers that want to implement their trace_event callbacks will not necessary need these information or they may want to format them as they want. This patch adds a new default-enabled trace option: TRACE_ITER_CONTEXT_INFO When disabled through: echo nocontext-info > /debugfs/tracing/trace_options The pid, cpu and timestamps headers will not be printed. IE with the sched_switch tracer with context-info (default): bash-2935 [001] 100.356561: 2935:120:S ==> [001] 0:140:R <idle> <idle>-0 [000] 100.412804: 0:140:R + [000] 11:115:S events/0 <idle>-0 [000] 100.412816: 0:140:R ==> [000] 11:115:R events/0 events/0-11 [000] 100.412829: 11:115:S ==> [000] 0:140:R <idle> Without context-info: 2935:120:S ==> [001] 0:140:R <idle> 0:140:R + [000] 11:115:S events/0 0:140:R ==> [000] 11:115:R events/0 11:115:S ==> [000] 0:140:R <idle> A tracer can disable it at runtime by clearing the bit TRACE_ITER_CONTEXT_INFO in trace_flags. The print routines were renamed to trace_print_context and trace_print_lat_context, so that they can be used by tracers if they want to use them for one of the trace_event callbacks. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-02-02 22:29:21 +00:00
}
static const char state_to_char[] = TASK_STATE_TO_CHAR_STR;
static int task_state_char(unsigned long state)
{
int bit = state ? __ffs(state) + 1 : 0;
return bit < sizeof(state_to_char) - 1 ? state_to_char[bit] : '?';
}
/**
* ftrace_find_event - find a registered event
* @type: the type of event to look for
*
* Returns an event of type @type otherwise NULL
* Called with trace_event_read_lock() held.
*/
struct trace_event *ftrace_find_event(int type)
{
struct trace_event *event;
unsigned key;
key = type & (EVENT_HASHSIZE - 1);
hlist: drop the node parameter from iterators I'm not sure why, but the hlist for each entry iterators were conceived list_for_each_entry(pos, head, member) The hlist ones were greedy and wanted an extra parameter: hlist_for_each_entry(tpos, pos, head, member) Why did they need an extra pos parameter? I'm not quite sure. Not only they don't really need it, it also prevents the iterator from looking exactly like the list iterator, which is unfortunate. Besides the semantic patch, there was some manual work required: - Fix up the actual hlist iterators in linux/list.h - Fix up the declaration of other iterators based on the hlist ones. - A very small amount of places were using the 'node' parameter, this was modified to use 'obj->member' instead. - Coccinelle didn't handle the hlist_for_each_entry_safe iterator properly, so those had to be fixed up manually. The semantic patch which is mostly the work of Peter Senna Tschudin is here: @@ iterator name hlist_for_each_entry, hlist_for_each_entry_continue, hlist_for_each_entry_from, hlist_for_each_entry_rcu, hlist_for_each_entry_rcu_bh, hlist_for_each_entry_continue_rcu_bh, for_each_busy_worker, ax25_uid_for_each, ax25_for_each, inet_bind_bucket_for_each, sctp_for_each_hentry, sk_for_each, sk_for_each_rcu, sk_for_each_from, sk_for_each_safe, sk_for_each_bound, hlist_for_each_entry_safe, hlist_for_each_entry_continue_rcu, nr_neigh_for_each, nr_neigh_for_each_safe, nr_node_for_each, nr_node_for_each_safe, for_each_gfn_indirect_valid_sp, for_each_gfn_sp, for_each_host; type T; expression a,c,d,e; identifier b; statement S; @@ -T b; <+... when != b ( hlist_for_each_entry(a, - b, c, d) S | hlist_for_each_entry_continue(a, - b, c) S | hlist_for_each_entry_from(a, - b, c) S | hlist_for_each_entry_rcu(a, - b, c, d) S | hlist_for_each_entry_rcu_bh(a, - b, c, d) S | hlist_for_each_entry_continue_rcu_bh(a, - b, c) S | for_each_busy_worker(a, c, - b, d) S | ax25_uid_for_each(a, - b, c) S | ax25_for_each(a, - b, c) S | inet_bind_bucket_for_each(a, - b, c) S | sctp_for_each_hentry(a, - b, c) S | sk_for_each(a, - b, c) S | sk_for_each_rcu(a, - b, c) S | sk_for_each_from -(a, b) +(a) S + sk_for_each_from(a) S | sk_for_each_safe(a, - b, c, d) S | sk_for_each_bound(a, - b, c) S | hlist_for_each_entry_safe(a, - b, c, d, e) S | hlist_for_each_entry_continue_rcu(a, - b, c) S | nr_neigh_for_each(a, - b, c) S | nr_neigh_for_each_safe(a, - b, c, d) S | nr_node_for_each(a, - b, c) S | nr_node_for_each_safe(a, - b, c, d) S | - for_each_gfn_sp(a, c, d, b) S + for_each_gfn_sp(a, c, d) S | - for_each_gfn_indirect_valid_sp(a, c, d, b) S + for_each_gfn_indirect_valid_sp(a, c, d) S | for_each_host(a, - b, c) S | for_each_host_safe(a, - b, c, d) S | for_each_mesh_entry(a, - b, c, d) S ) ...+> [akpm@linux-foundation.org: drop bogus change from net/ipv4/raw.c] [akpm@linux-foundation.org: drop bogus hunk from net/ipv6/raw.c] [akpm@linux-foundation.org: checkpatch fixes] [akpm@linux-foundation.org: fix warnings] [akpm@linux-foudnation.org: redo intrusive kvm changes] Tested-by: Peter Senna Tschudin <peter.senna@gmail.com> Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Signed-off-by: Sasha Levin <sasha.levin@oracle.com> Cc: Wu Fengguang <fengguang.wu@intel.com> Cc: Marcelo Tosatti <mtosatti@redhat.com> Cc: Gleb Natapov <gleb@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-28 01:06:00 +00:00
hlist_for_each_entry(event, &event_hash[key], node) {
if (event->type == type)
return event;
}
return NULL;
}
static LIST_HEAD(ftrace_event_list);
static int trace_search_list(struct list_head **list)
{
struct trace_event *e;
int last = __TRACE_LAST_TYPE;
if (list_empty(&ftrace_event_list)) {
*list = &ftrace_event_list;
return last + 1;
}
/*
* We used up all possible max events,
* lets see if somebody freed one.
*/
list_for_each_entry(e, &ftrace_event_list, list) {
if (e->type != last + 1)
break;
last++;
}
/* Did we used up all 65 thousand events??? */
if ((last + 1) > FTRACE_MAX_EVENT)
return 0;
*list = &e->list;
return last + 1;
}
void trace_event_read_lock(void)
{
down_read(&trace_event_sem);
}
void trace_event_read_unlock(void)
{
up_read(&trace_event_sem);
}
/**
* register_trace_event - register output for an event type
* @event: the event type to register
*
* Event types are stored in a hash and this hash is used to
* find a way to print an event. If the @event->type is set
* then it will use that type, otherwise it will assign a
* type to use.
*
* If you assign your own type, please make sure it is added
* to the trace_type enum in trace.h, to avoid collisions
* with the dynamic types.
*
* Returns the event type number or zero on error.
*/
int register_trace_event(struct trace_event *event)
{
unsigned key;
int ret = 0;
down_write(&trace_event_sem);
if (WARN_ON(!event))
goto out;
if (WARN_ON(!event->funcs))
goto out;
INIT_LIST_HEAD(&event->list);
if (!event->type) {
struct list_head *list = NULL;
if (next_event_type > FTRACE_MAX_EVENT) {
event->type = trace_search_list(&list);
if (!event->type)
goto out;
} else {
tracing: Add additional marks to signal very large time deltas Currently, function graph tracer prints "!" or "+" just before function execution time to signal a function overhead, depending on the time. And some tracers tracing latency also print "!" or "+" just after time to signal overhead, depending on the interval between events. Even it is usually enough to do that, we sometimes need to signal for bigger execution time than 100 micro seconds. For example, I used function graph tracer to detect if there is any case that exit_mm() takes too much time. I did following steps in /sys/kernel/debug/tracing. It was easier to detect very large excution time with patched kernel than with original kernel. $ echo exit_mm > set_graph_function $ echo function_graph > current_tracer $ echo > trace $ cat trace_pipe > $LOGFILE ... (do something and terminate logging) $ grep "\\$" $LOGFILE 3) $ 22082032 us | } /* kernel_map_pages */ 3) $ 22082040 us | } /* free_pages_prepare */ 3) $ 22082113 us | } /* free_hot_cold_page */ 3) $ 22083455 us | } /* free_hot_cold_page_list */ 3) $ 22083895 us | } /* release_pages */ 3) $ 22177873 us | } /* free_pages_and_swap_cache */ 3) $ 22178929 us | } /* unmap_single_vma */ 3) $ 22198885 us | } /* unmap_vmas */ 3) $ 22206949 us | } /* exit_mmap */ 3) $ 22207659 us | } /* mmput */ 3) $ 22207793 us | } /* exit_mm */ And then, it was easy to find out that a schedule-out occured by sub_preempt_count() within kernel_map_pages(). To detect very large function exection time caused by either problematic function implementation or scheduling issues, this patch can be useful. Link: http://lkml.kernel.org/r/1416789259-24038-1-git-send-email-byungchul.park@lge.com Signed-off-by: Byungchul Park <byungchul.park@lge.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2014-11-24 00:34:19 +00:00
event->type = next_event_type++;
list = &ftrace_event_list;
}
if (WARN_ON(ftrace_find_event(event->type)))
goto out;
list_add_tail(&event->list, list);
} else if (event->type > __TRACE_LAST_TYPE) {
printk(KERN_WARNING "Need to add type to trace.h\n");
WARN_ON(1);
goto out;
} else {
/* Is this event already used */
if (ftrace_find_event(event->type))
goto out;
}
if (event->funcs->trace == NULL)
event->funcs->trace = trace_nop_print;
if (event->funcs->raw == NULL)
event->funcs->raw = trace_nop_print;
if (event->funcs->hex == NULL)
event->funcs->hex = trace_nop_print;
if (event->funcs->binary == NULL)
event->funcs->binary = trace_nop_print;
key = event->type & (EVENT_HASHSIZE - 1);
hlist_add_head(&event->node, &event_hash[key]);
ret = event->type;
out:
up_write(&trace_event_sem);
return ret;
}
EXPORT_SYMBOL_GPL(register_trace_event);
/*
* Used by module code with the trace_event_sem held for write.
*/
int __unregister_trace_event(struct trace_event *event)
{
hlist_del(&event->node);
list_del(&event->list);
return 0;
}
/**
* unregister_trace_event - remove a no longer used event
* @event: the event to remove
*/
int unregister_trace_event(struct trace_event *event)
{
down_write(&trace_event_sem);
__unregister_trace_event(event);
up_write(&trace_event_sem);
return 0;
}
EXPORT_SYMBOL_GPL(unregister_trace_event);
/*
* Standard events
*/
enum print_line_t trace_nop_print(struct trace_iterator *iter, int flags,
struct trace_event *event)
{
trace_seq_printf(&iter->seq, "type: %d\n", iter->ent->type);
return trace_handle_return(&iter->seq);
}
/* TRACE_FN */
static enum print_line_t trace_fn_trace(struct trace_iterator *iter, int flags,
struct trace_event *event)
{
struct ftrace_entry *field;
struct trace_seq *s = &iter->seq;
trace_assign_type(field, iter->ent);
seq_print_ip_sym(s, field->ip, flags);
if ((flags & TRACE_ITER_PRINT_PARENT) && field->parent_ip) {
trace_seq_puts(s, " <-");
seq_print_ip_sym(s, field->parent_ip, flags);
}
trace_seq_putc(s, '\n');
return trace_handle_return(s);
}
static enum print_line_t trace_fn_raw(struct trace_iterator *iter, int flags,
struct trace_event *event)
{
struct ftrace_entry *field;
trace_assign_type(field, iter->ent);
trace_seq_printf(&iter->seq, "%lx %lx\n",
field->ip,
field->parent_ip);
return trace_handle_return(&iter->seq);
}
static enum print_line_t trace_fn_hex(struct trace_iterator *iter, int flags,
struct trace_event *event)
{
struct ftrace_entry *field;
struct trace_seq *s = &iter->seq;
trace_assign_type(field, iter->ent);
SEQ_PUT_HEX_FIELD(s, field->ip);
SEQ_PUT_HEX_FIELD(s, field->parent_ip);
return trace_handle_return(s);
}
static enum print_line_t trace_fn_bin(struct trace_iterator *iter, int flags,
struct trace_event *event)
{
struct ftrace_entry *field;
struct trace_seq *s = &iter->seq;
trace_assign_type(field, iter->ent);
SEQ_PUT_FIELD(s, field->ip);
SEQ_PUT_FIELD(s, field->parent_ip);
return trace_handle_return(s);
}
static struct trace_event_functions trace_fn_funcs = {
.trace = trace_fn_trace,
.raw = trace_fn_raw,
.hex = trace_fn_hex,
.binary = trace_fn_bin,
};
static struct trace_event trace_fn_event = {
.type = TRACE_FN,
.funcs = &trace_fn_funcs,
};
/* TRACE_CTX an TRACE_WAKE */
static enum print_line_t trace_ctxwake_print(struct trace_iterator *iter,
char *delim)
{
struct ctx_switch_entry *field;
char comm[TASK_COMM_LEN];
int S, T;
trace_assign_type(field, iter->ent);
T = task_state_char(field->next_state);
S = task_state_char(field->prev_state);
trace_find_cmdline(field->next_pid, comm);
trace_seq_printf(&iter->seq,
" %5d:%3d:%c %s [%03d] %5d:%3d:%c %s\n",
field->prev_pid,
field->prev_prio,
S, delim,
field->next_cpu,
field->next_pid,
field->next_prio,
T, comm);
return trace_handle_return(&iter->seq);
}
static enum print_line_t trace_ctx_print(struct trace_iterator *iter, int flags,
struct trace_event *event)
{
return trace_ctxwake_print(iter, "==>");
}
static enum print_line_t trace_wake_print(struct trace_iterator *iter,
int flags, struct trace_event *event)
{
return trace_ctxwake_print(iter, " +");
}
static int trace_ctxwake_raw(struct trace_iterator *iter, char S)
{
struct ctx_switch_entry *field;
int T;
trace_assign_type(field, iter->ent);
if (!S)
S = task_state_char(field->prev_state);
T = task_state_char(field->next_state);
trace_seq_printf(&iter->seq, "%d %d %c %d %d %d %c\n",
field->prev_pid,
field->prev_prio,
S,
field->next_cpu,
field->next_pid,
field->next_prio,
T);
return trace_handle_return(&iter->seq);
}
static enum print_line_t trace_ctx_raw(struct trace_iterator *iter, int flags,
struct trace_event *event)
{
return trace_ctxwake_raw(iter, 0);
}
static enum print_line_t trace_wake_raw(struct trace_iterator *iter, int flags,
struct trace_event *event)
{
return trace_ctxwake_raw(iter, '+');
}
static int trace_ctxwake_hex(struct trace_iterator *iter, char S)
{
struct ctx_switch_entry *field;
struct trace_seq *s = &iter->seq;
int T;
trace_assign_type(field, iter->ent);
if (!S)
S = task_state_char(field->prev_state);
T = task_state_char(field->next_state);
SEQ_PUT_HEX_FIELD(s, field->prev_pid);
SEQ_PUT_HEX_FIELD(s, field->prev_prio);
SEQ_PUT_HEX_FIELD(s, S);
SEQ_PUT_HEX_FIELD(s, field->next_cpu);
SEQ_PUT_HEX_FIELD(s, field->next_pid);
SEQ_PUT_HEX_FIELD(s, field->next_prio);
SEQ_PUT_HEX_FIELD(s, T);
return trace_handle_return(s);
}
static enum print_line_t trace_ctx_hex(struct trace_iterator *iter, int flags,
struct trace_event *event)
{
return trace_ctxwake_hex(iter, 0);
}
static enum print_line_t trace_wake_hex(struct trace_iterator *iter, int flags,
struct trace_event *event)
{
return trace_ctxwake_hex(iter, '+');
}
static enum print_line_t trace_ctxwake_bin(struct trace_iterator *iter,
int flags, struct trace_event *event)
{
struct ctx_switch_entry *field;
struct trace_seq *s = &iter->seq;
trace_assign_type(field, iter->ent);
SEQ_PUT_FIELD(s, field->prev_pid);
SEQ_PUT_FIELD(s, field->prev_prio);
SEQ_PUT_FIELD(s, field->prev_state);
SEQ_PUT_FIELD(s, field->next_cpu);
SEQ_PUT_FIELD(s, field->next_pid);
SEQ_PUT_FIELD(s, field->next_prio);
SEQ_PUT_FIELD(s, field->next_state);
return trace_handle_return(s);
}
static struct trace_event_functions trace_ctx_funcs = {
.trace = trace_ctx_print,
.raw = trace_ctx_raw,
.hex = trace_ctx_hex,
.binary = trace_ctxwake_bin,
};
static struct trace_event trace_ctx_event = {
.type = TRACE_CTX,
.funcs = &trace_ctx_funcs,
};
static struct trace_event_functions trace_wake_funcs = {
.trace = trace_wake_print,
.raw = trace_wake_raw,
.hex = trace_wake_hex,
.binary = trace_ctxwake_bin,
};
static struct trace_event trace_wake_event = {
.type = TRACE_WAKE,
.funcs = &trace_wake_funcs,
};
/* TRACE_STACK */
static enum print_line_t trace_stack_print(struct trace_iterator *iter,
int flags, struct trace_event *event)
{
struct stack_entry *field;
struct trace_seq *s = &iter->seq;
unsigned long *p;
unsigned long *end;
trace_assign_type(field, iter->ent);
end = (unsigned long *)((long)iter->ent + iter->ent_size);
trace_seq_puts(s, "<stack trace>\n");
for (p = field->caller; p && *p != ULONG_MAX && p < end; p++) {
if (trace_seq_has_overflowed(s))
break;
trace_seq_puts(s, " => ");
seq_print_ip_sym(s, *p, flags);
trace_seq_putc(s, '\n');
}
return trace_handle_return(s);
}
static struct trace_event_functions trace_stack_funcs = {
.trace = trace_stack_print,
};
static struct trace_event trace_stack_event = {
.type = TRACE_STACK,
.funcs = &trace_stack_funcs,
};
/* TRACE_USER_STACK */
static enum print_line_t trace_user_stack_print(struct trace_iterator *iter,
int flags, struct trace_event *event)
{
struct userstack_entry *field;
struct trace_seq *s = &iter->seq;
trace_assign_type(field, iter->ent);
trace_seq_puts(s, "<user stack trace>\n");
seq_print_userip_objs(field, s, flags);
return trace_handle_return(s);
}
static struct trace_event_functions trace_user_stack_funcs = {
.trace = trace_user_stack_print,
};
static struct trace_event trace_user_stack_event = {
.type = TRACE_USER_STACK,
.funcs = &trace_user_stack_funcs,
};
/* TRACE_BPUTS */
static enum print_line_t
trace_bputs_print(struct trace_iterator *iter, int flags,
struct trace_event *event)
{
struct trace_entry *entry = iter->ent;
struct trace_seq *s = &iter->seq;
struct bputs_entry *field;
trace_assign_type(field, entry);
seq_print_ip_sym(s, field->ip, flags);
trace_seq_puts(s, ": ");
trace_seq_puts(s, field->str);
return trace_handle_return(s);
}
static enum print_line_t
trace_bputs_raw(struct trace_iterator *iter, int flags,
struct trace_event *event)
{
struct bputs_entry *field;
struct trace_seq *s = &iter->seq;
trace_assign_type(field, iter->ent);
trace_seq_printf(s, ": %lx : ", field->ip);
trace_seq_puts(s, field->str);
return trace_handle_return(s);
}
static struct trace_event_functions trace_bputs_funcs = {
.trace = trace_bputs_print,
.raw = trace_bputs_raw,
};
static struct trace_event trace_bputs_event = {
.type = TRACE_BPUTS,
.funcs = &trace_bputs_funcs,
};
/* TRACE_BPRINT */
static enum print_line_t
trace_bprint_print(struct trace_iterator *iter, int flags,
struct trace_event *event)
{
struct trace_entry *entry = iter->ent;
struct trace_seq *s = &iter->seq;
struct bprint_entry *field;
trace_assign_type(field, entry);
seq_print_ip_sym(s, field->ip, flags);
trace_seq_puts(s, ": ");
trace_seq_bprintf(s, field->fmt, field->buf);
return trace_handle_return(s);
}
tracing/core: drop the old trace_printk() implementation in favour of trace_bprintk() Impact: faster and lighter tracing Now that we have trace_bprintk() which is faster and consume lesser memory than trace_printk() and has the same purpose, we can now drop the old implementation in favour of the binary one from trace_bprintk(), which means we move all the implementation of trace_bprintk() to trace_printk(), so the Api doesn't change except that we must now use trace_seq_bprintk() to print the TRACE_PRINT entries. Some changes result of this: - Previously, trace_bprintk depended of a single tracer and couldn't work without. This tracer has been dropped and the whole implementation of trace_printk() (like the module formats management) is now integrated in the tracing core (comes with CONFIG_TRACING), though we keep the file trace_printk (previously trace_bprintk.c) where we can find the module management. Thus we don't overflow trace.c - changes some parts to use trace_seq_bprintk() to print TRACE_PRINT entries. - change a bit trace_printk/trace_vprintk macros to support non-builtin formats constants, and fix 'const' qualifiers warnings. But this is all transparent for developers. - etc... V2: - Rebase against last changes - Fix mispell on the changelog V3: - Rebase against last changes (moving trace_printk() to kernel.h) Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Acked-by: Steven Rostedt <rostedt@goodmis.org> LKML-Reference: <1236356510-8381-5-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-03-06 16:21:49 +00:00
static enum print_line_t
trace_bprint_raw(struct trace_iterator *iter, int flags,
struct trace_event *event)
{
struct bprint_entry *field;
struct trace_seq *s = &iter->seq;
tracing/core: drop the old trace_printk() implementation in favour of trace_bprintk() Impact: faster and lighter tracing Now that we have trace_bprintk() which is faster and consume lesser memory than trace_printk() and has the same purpose, we can now drop the old implementation in favour of the binary one from trace_bprintk(), which means we move all the implementation of trace_bprintk() to trace_printk(), so the Api doesn't change except that we must now use trace_seq_bprintk() to print the TRACE_PRINT entries. Some changes result of this: - Previously, trace_bprintk depended of a single tracer and couldn't work without. This tracer has been dropped and the whole implementation of trace_printk() (like the module formats management) is now integrated in the tracing core (comes with CONFIG_TRACING), though we keep the file trace_printk (previously trace_bprintk.c) where we can find the module management. Thus we don't overflow trace.c - changes some parts to use trace_seq_bprintk() to print TRACE_PRINT entries. - change a bit trace_printk/trace_vprintk macros to support non-builtin formats constants, and fix 'const' qualifiers warnings. But this is all transparent for developers. - etc... V2: - Rebase against last changes - Fix mispell on the changelog V3: - Rebase against last changes (moving trace_printk() to kernel.h) Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Acked-by: Steven Rostedt <rostedt@goodmis.org> LKML-Reference: <1236356510-8381-5-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-03-06 16:21:49 +00:00
trace_assign_type(field, iter->ent);
trace_seq_printf(s, ": %lx : ", field->ip);
trace_seq_bprintf(s, field->fmt, field->buf);
return trace_handle_return(s);
}
static struct trace_event_functions trace_bprint_funcs = {
.trace = trace_bprint_print,
.raw = trace_bprint_raw,
};
tracing/core: drop the old trace_printk() implementation in favour of trace_bprintk() Impact: faster and lighter tracing Now that we have trace_bprintk() which is faster and consume lesser memory than trace_printk() and has the same purpose, we can now drop the old implementation in favour of the binary one from trace_bprintk(), which means we move all the implementation of trace_bprintk() to trace_printk(), so the Api doesn't change except that we must now use trace_seq_bprintk() to print the TRACE_PRINT entries. Some changes result of this: - Previously, trace_bprintk depended of a single tracer and couldn't work without. This tracer has been dropped and the whole implementation of trace_printk() (like the module formats management) is now integrated in the tracing core (comes with CONFIG_TRACING), though we keep the file trace_printk (previously trace_bprintk.c) where we can find the module management. Thus we don't overflow trace.c - changes some parts to use trace_seq_bprintk() to print TRACE_PRINT entries. - change a bit trace_printk/trace_vprintk macros to support non-builtin formats constants, and fix 'const' qualifiers warnings. But this is all transparent for developers. - etc... V2: - Rebase against last changes - Fix mispell on the changelog V3: - Rebase against last changes (moving trace_printk() to kernel.h) Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Acked-by: Steven Rostedt <rostedt@goodmis.org> LKML-Reference: <1236356510-8381-5-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-03-06 16:21:49 +00:00
static struct trace_event trace_bprint_event = {
.type = TRACE_BPRINT,
.funcs = &trace_bprint_funcs,
};
/* TRACE_PRINT */
static enum print_line_t trace_print_print(struct trace_iterator *iter,
int flags, struct trace_event *event)
{
struct print_entry *field;
struct trace_seq *s = &iter->seq;
trace_assign_type(field, iter->ent);
seq_print_ip_sym(s, field->ip, flags);
trace_seq_printf(s, ": %s", field->buf);
return trace_handle_return(s);
}
static enum print_line_t trace_print_raw(struct trace_iterator *iter, int flags,
struct trace_event *event)
{
struct print_entry *field;
trace_assign_type(field, iter->ent);
trace_seq_printf(&iter->seq, "# %lx %s", field->ip, field->buf);
return trace_handle_return(&iter->seq);
}
static struct trace_event_functions trace_print_funcs = {
tracing/core: drop the old trace_printk() implementation in favour of trace_bprintk() Impact: faster and lighter tracing Now that we have trace_bprintk() which is faster and consume lesser memory than trace_printk() and has the same purpose, we can now drop the old implementation in favour of the binary one from trace_bprintk(), which means we move all the implementation of trace_bprintk() to trace_printk(), so the Api doesn't change except that we must now use trace_seq_bprintk() to print the TRACE_PRINT entries. Some changes result of this: - Previously, trace_bprintk depended of a single tracer and couldn't work without. This tracer has been dropped and the whole implementation of trace_printk() (like the module formats management) is now integrated in the tracing core (comes with CONFIG_TRACING), though we keep the file trace_printk (previously trace_bprintk.c) where we can find the module management. Thus we don't overflow trace.c - changes some parts to use trace_seq_bprintk() to print TRACE_PRINT entries. - change a bit trace_printk/trace_vprintk macros to support non-builtin formats constants, and fix 'const' qualifiers warnings. But this is all transparent for developers. - etc... V2: - Rebase against last changes - Fix mispell on the changelog V3: - Rebase against last changes (moving trace_printk() to kernel.h) Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Acked-by: Steven Rostedt <rostedt@goodmis.org> LKML-Reference: <1236356510-8381-5-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-03-06 16:21:49 +00:00
.trace = trace_print_print,
.raw = trace_print_raw,
};
static struct trace_event trace_print_event = {
.type = TRACE_PRINT,
.funcs = &trace_print_funcs,
};
static struct trace_event *events[] __initdata = {
&trace_fn_event,
&trace_ctx_event,
&trace_wake_event,
&trace_stack_event,
&trace_user_stack_event,
&trace_bputs_event,
&trace_bprint_event,
&trace_print_event,
NULL
};
__init static int init_events(void)
{
struct trace_event *event;
int i, ret;
for (i = 0; events[i]; i++) {
event = events[i];
ret = register_trace_event(event);
if (!ret) {
printk(KERN_WARNING "event %d failed to register\n",
event->type);
WARN_ON_ONCE(1);
}
}
return 0;
}
early_initcall(init_events);