From: Peter Zijlstra <peterz@infradead.org>
To: Steven Rostedt <rostedt@goodmis.org>
Cc: linux-kernel@vger.kernel.org, Ingo Molnar <mingo@elte.hu>,
Thomas Gleixner <tglx@linutronix.de>,
Linus Torvalds <torvalds@linux-foundation.org>,
Andrew Morton <akpm@linux-foundation.org>
Subject: [PATCH] ftrace: sched_switch: show the wakee's cpu
Date: Thu, 04 Sep 2008 10:24:16 +0200 [thread overview]
Message-ID: <1220516656.8609.201.camel@twins> (raw)
In-Reply-To: <20080903214249.661170580@goodmis.org>
Subject: ftrace: sched_switch: show the wakee's cpu
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Tue Sep 02 21:12:13 CEST 2008
While profiling the smp behaviour of the scheduler it was needed to know to
which cpu a task got woken.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
kernel/trace/trace.c | 41 +++++++++++++++++++++++++++--------------
kernel/trace/trace.h | 2 ++
2 files changed, 29 insertions(+), 14 deletions(-)
Index: linux-2.6/kernel/trace/trace.c
===================================================================
--- linux-2.6.orig/kernel/trace/trace.c
+++ linux-2.6/kernel/trace/trace.c
@@ -976,6 +977,7 @@ tracing_sched_switch_trace(struct trace_
entry->field.ctx.next_pid = next->pid;
entry->field.ctx.next_prio = next->prio;
entry->field.ctx.next_state = next->state;
+ entry->field.ctx.next_cpu = task_cpu(next);
__trace_stack(tr, data, flags, 5);
__raw_spin_unlock(&data->lock);
raw_local_irq_restore(irq_flags);
@@ -1002,6 +1004,7 @@ tracing_sched_wakeup_trace(struct trace_
entry->field.ctx.next_pid = wakee->pid;
entry->field.ctx.next_prio = wakee->prio;
entry->field.ctx.next_state = wakee->state;
+ entry->field.ctx.next_cpu = task_cpu(wakee);
__trace_stack(tr, data, flags, 6);
__raw_spin_unlock(&data->lock);
raw_local_irq_restore(irq_flags);
@@ -1630,10 +1638,11 @@ print_lat_fmt(struct trace_iterator *ite
__ffs(field->ctx.prev_state) + 1 : 0;
S = state < sizeof(state_to_char) - 1 ? state_to_char[state] : 'X';
comm = trace_find_cmdline(field->ctx.next_pid);
- trace_seq_printf(s, " %5d:%3d:%c %s %5d:%3d:%c %s\n",
+ trace_seq_printf(s, " %5d:%3d:%c %s [%03d] %5d:%3d:%c %s\n",
field->ctx.prev_pid,
field->ctx.prev_prio,
S, entry->type == TRACE_CTX ? "==>" : " +",
+ field->ctx.next_cpu,
field->ctx.next_pid,
field->ctx.next_prio,
T, comm);
@@ -1730,11 +1739,12 @@ static int print_trace_fmt(struct trace_
state_to_char[field->ctx.prev_state] : 'X';
T = field->ctx.next_state < sizeof(state_to_char) ?
state_to_char[field->ctx.next_state] : 'X';
- ret = trace_seq_printf(s, " %5d:%3d:%c %s %5d:%3d:%c\n",
+ ret = trace_seq_printf(s, " %5d:%3d:%c %s [%03d] %5d:%3d:%c\n",
field->ctx.prev_pid,
field->ctx.prev_prio,
S,
entry->type == TRACE_CTX ? "==>" : " +",
+ field->ctx.next_cpu,
field->ctx.next_pid,
field->ctx.next_prio,
T);
@@ -1811,10 +1821,11 @@ static int print_raw_fmt(struct trace_it
state_to_char[field->ctx.next_state] : 'X';
if (entry->type == TRACE_WAKE)
S = '+';
- ret = trace_seq_printf(s, "%d %d %c %d %d %c\n",
+ ret = trace_seq_printf(s, "%d %d %c %d %d %d %c\n",
field->ctx.prev_pid,
field->ctx.prev_prio,
S,
+ field->ctx.next_cpu,
field->ctx.next_pid,
field->ctx.next_prio,
T);
@@ -1887,6 +1898,7 @@ static int print_hex_fmt(struct trace_it
SEQ_PUT_HEX_FIELD_RET(s, field->ctx.prev_pid);
SEQ_PUT_HEX_FIELD_RET(s, field->ctx.prev_prio);
SEQ_PUT_HEX_FIELD_RET(s, S);
+ SEQ_PUT_HEX_FIELD_RET(s, field->ctx.next_cpu);
SEQ_PUT_HEX_FIELD_RET(s, field->ctx.next_pid);
SEQ_PUT_HEX_FIELD_RET(s, field->ctx.next_prio);
SEQ_PUT_HEX_FIELD_RET(s, T);
Index: linux-2.6/kernel/trace/trace.h
===================================================================
--- linux-2.6.orig/kernel/trace/trace.h
+++ linux-2.6/kernel/trace/trace.h
@@ -41,6 +41,7 @@ struct ctx_switch_entry {
unsigned int next_pid;
unsigned char next_prio;
unsigned char next_state;
+ unsigned int next_cpu;
};
/*
next prev parent reply other threads:[~2008-09-04 8:27 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-09-03 21:42 [PATCH 0/2] ftrace: ftrace_printk fixes Steven Rostedt
2008-09-03 21:42 ` [PATCH 1/2] ftrace: binary and not logical for continue test Steven Rostedt
2008-09-03 21:42 ` [PATCH 2/2] ftrace: print continue index fix Steven Rostedt
2008-09-04 7:41 ` [PATCH 0/2] ftrace: ftrace_printk fixes Peter Zijlstra
2008-09-04 8:24 ` [PATCH] ftrace: make ftrace_printk usable with the other tracers Peter Zijlstra
2008-09-04 8:24 ` Peter Zijlstra [this message]
2008-09-04 10:16 ` [PATCH] ftrace: sched_switch: show the wakee's cpu Ingo Molnar
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1220516656.8609.201.camel@twins \
--to=peterz@infradead.org \
--cc=akpm@linux-foundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=rostedt@goodmis.org \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.