All of lore.kernel.org
 help / color / mirror / Atom feed
* Loop block overhead
@ 2014-04-30 14:26 lhmaster2 .
  2014-04-30 20:10 ` Andi Kleen
  0 siblings, 1 reply; 4+ messages in thread
From: lhmaster2 . @ 2014-04-30 14:26 UTC (permalink / raw)
  To: linux-perf-users

Dear all,

Is there any way to make perf report retrieve the total overhead due
to loop blocks, like what is done with total function overhead?

Lucas

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Loop block overhead
  2014-04-30 14:26 Loop block overhead lhmaster2 .
@ 2014-04-30 20:10 ` Andi Kleen
  2014-05-07 17:55   ` lhmaster2 .
  0 siblings, 1 reply; 4+ messages in thread
From: Andi Kleen @ 2014-04-30 20:10 UTC (permalink / raw)
  To: lhmaster2 .; +Cc: linux-perf-users

"lhmaster2 ." <lhmaster2@gmail.com> writes:

> Dear all,
>
> Is there any way to make perf report retrieve the total overhead due
> to loop blocks, like what is done with total function overhead?

perf has no idea what a loop is unfortunately.

You can use srcline as a sort key to account per line.

-Andi

-- 
ak@linux.intel.com -- Speaking for myself only

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Loop block overhead
  2014-04-30 20:10 ` Andi Kleen
@ 2014-05-07 17:55   ` lhmaster2 .
  2014-05-14  0:37     ` Namhyung Kim
  0 siblings, 1 reply; 4+ messages in thread
From: lhmaster2 . @ 2014-05-07 17:55 UTC (permalink / raw)
  To: Andi Kleen; +Cc: linux-perf-users

Thank you for the info.

I researched a bit and found that srcline uses addr2line capabilities
to map memory addressed into source lines. Nevertheless, as I invoke

$ perf report -s srcline, sys

only the original mem addresses are printed. Should I pass some
special parameter for perf record to make it work?

Best regards,
Lucas


On 30 April 2014 17:10, Andi Kleen <andi@firstfloor.org> wrote:
>
> "lhmaster2 ." <lhmaster2@gmail.com> writes:
>
> > Dear all,
> >
> > Is there any way to make perf report retrieve the total overhead due
> > to loop blocks, like what is done with total function overhead?
>
> perf has no idea what a loop is unfortunately.
>
> You can use srcline as a sort key to account per line.
>
> -Andi
>
> --
> ak@linux.intel.com -- Speaking for myself only

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Loop block overhead
  2014-05-07 17:55   ` lhmaster2 .
@ 2014-05-14  0:37     ` Namhyung Kim
  0 siblings, 0 replies; 4+ messages in thread
From: Namhyung Kim @ 2014-05-14  0:37 UTC (permalink / raw)
  To: lhmaster2 .; +Cc: Andi Kleen, linux-perf-users

Hi lhmaster,

On Wed, 7 May 2014 14:55:05 -0300, lhmaster wrote:
> Thank you for the info.
>
> I researched a bit and found that srcline uses addr2line capabilities
> to map memory addressed into source lines. Nevertheless, as I invoke
>
> $ perf report -s srcline, sys
>
> only the original mem addresses are printed. Should I pass some
> special parameter for perf record to make it work?

Nope.  But you need to make sure that your binary was built with
debuginfo and the binutils is available on your system.

Thanks,
Namhyung

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2014-05-14  0:37 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2014-04-30 14:26 Loop block overhead lhmaster2 .
2014-04-30 20:10 ` Andi Kleen
2014-05-07 17:55   ` lhmaster2 .
2014-05-14  0:37     ` Namhyung Kim

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.