From: Jeremy Fitzhardinge <jeremy@goop.org>
To: Ingo Molnar <mingo@elte.hu>
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>,
Steven Rostedt <rostedt@goodmis.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Subject: Re: [PATCH 1/4] tracing: move __DO_TRACE out of line
Date: Fri, 17 Apr 2009 12:58:48 -0700 [thread overview]
Message-ID: <49E8DF78.2050203@goop.org> (raw)
In-Reply-To: <20090417194613.GA30544@elte.hu>
Ingo Molnar wrote:
> * Jeremy Fitzhardinge <jeremy@goop.org> wrote:
>
>
>> Taking __do_trace_sched_switch out of lines inserts this into the
>> hot path (6 instructions, 31 bytes):
>>
>> cmpl $0, __tracepoint_sched_switch+8(%rip) #, __tracepoint_sched_switch.state
>> je .L1748 #,
>> movq -136(%rbp), %rdx # next,
>> movq -144(%rbp), %rsi # prev,
>> movq %rbx, %rdi # rq,
>> call __do_trace_sched_switch #
>> .L1748:
>>
>
> Hm, why isnt this off-line in the function? It's marked unlikely(),
> isnt it?
>
Yes, its unlikely(). I don't know why it doesn't move it; I've never
seen unlikely() do anything useful.
> also, did you investigate the effect on the _instrumented_ function
> itself? (i.e. the non-tracing related bits) A function call clobbers
> various registers and creates pressure on gcc to shuffle registers
> around.
>
Well, there's a function call in either case, so I don't think it makes
much difference.
J
next prev parent reply other threads:[~2009-04-17 19:59 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-04-17 6:35 [PATCH] tracing WIP patches Jeremy Fitzhardinge
2009-04-17 6:35 ` [PATCH 1/4] tracing: move __DO_TRACE out of line Jeremy Fitzhardinge
2009-04-17 15:46 ` Ingo Molnar
2009-04-17 16:10 ` Mathieu Desnoyers
2009-04-17 16:23 ` Ingo Molnar
2009-04-17 16:47 ` Jeremy Fitzhardinge
2009-04-17 19:31 ` Jeremy Fitzhardinge
2009-04-17 19:46 ` Ingo Molnar
2009-04-17 19:57 ` Steven Rostedt
2009-04-17 19:58 ` Jeremy Fitzhardinge [this message]
2009-04-17 20:06 ` Steven Rostedt
2009-04-18 6:53 ` Mathieu Desnoyers
2009-04-18 14:16 ` Steven Rostedt
2009-04-19 3:59 ` Mathieu Desnoyers
2009-04-19 23:38 ` Jeremy Fitzhardinge
2009-04-20 21:39 ` Mathieu Desnoyers
2009-04-19 23:40 ` Jeremy Fitzhardinge
2009-04-20 21:47 ` Mathieu Desnoyers
2009-04-17 6:35 ` [PATCH 2/4] x86/pvops: target CREATE_TRACE_POINTS to particular subsystems Jeremy Fitzhardinge
2009-04-17 15:55 ` Steven Rostedt
2009-04-17 16:14 ` Jeremy Fitzhardinge
2009-04-17 16:32 ` Steven Rostedt
2009-04-17 16:48 ` Jeremy Fitzhardinge
2009-04-17 16:57 ` Steven Rostedt
2009-04-17 17:14 ` Jeremy Fitzhardinge
2009-04-17 17:33 ` Steven Rostedt
2009-04-17 18:11 ` Jeremy Fitzhardinge
2009-04-17 6:35 ` [PATCH 3/4] tracing: pass proto and args to DEFINE_TRACE Jeremy Fitzhardinge
2009-04-17 6:48 ` Christoph Hellwig
2009-04-17 6:58 ` Jeremy Fitzhardinge
2009-04-17 7:05 ` Christoph Hellwig
2009-04-17 12:53 ` Ingo Molnar
2009-04-17 15:21 ` Mathieu Desnoyers
2009-04-17 6:35 ` [PATCH 4/4] tracing: avoid warnings from zero-arg tracepoints Jeremy Fitzhardinge
2009-04-17 15:53 ` Steven Rostedt
2009-04-17 15:53 ` Ingo Molnar
2009-04-17 16:10 ` [tip:tracing/core] " tip-bot for Jeremy Fitzhardinge
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=49E8DF78.2050203@goop.org \
--to=jeremy@goop.org \
--cc=jeremy.fitzhardinge@citrix.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@polymtl.ca \
--cc=mingo@elte.hu \
--cc=rostedt@goodmis.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox