* What is the performance when using frame pointers in the kernel?
@ 2009-12-03 17:36 Justin Piszcz
2009-12-03 19:28 ` Peter Zijlstra
0 siblings, 1 reply; 6+ messages in thread
From: Justin Piszcz @ 2009-12-03 17:36 UTC (permalink / raw)
To: linux-kernel
Hello,
What is the performance hit when the kernel is compiled with frame
pointers?
Justin.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: What is the performance when using frame pointers in the kernel?
2009-12-03 17:36 What is the performance when using frame pointers in the kernel? Justin Piszcz
@ 2009-12-03 19:28 ` Peter Zijlstra
2009-12-04 13:23 ` Jiri Kosina
0 siblings, 1 reply; 6+ messages in thread
From: Peter Zijlstra @ 2009-12-03 19:28 UTC (permalink / raw)
To: Justin Piszcz; +Cc: linux-kernel
On Thu, 2009-12-03 at 12:36 -0500, Justin Piszcz wrote:
> What is the performance hit when the kernel is compiled with frame
> pointers?
build both kernels and run your favourite workload to find out.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: What is the performance when using frame pointers in the kernel?
2009-12-03 19:28 ` Peter Zijlstra
@ 2009-12-04 13:23 ` Jiri Kosina
2009-12-04 15:25 ` Justin Piszcz
2009-12-06 22:31 ` Andi Kleen
0 siblings, 2 replies; 6+ messages in thread
From: Jiri Kosina @ 2009-12-04 13:23 UTC (permalink / raw)
To: Peter Zijlstra; +Cc: Justin Piszcz, linux-kernel
On Thu, 3 Dec 2009, Peter Zijlstra wrote:
> > What is the performance hit when the kernel is compiled with frame
> > pointers?
>
> build both kernels and run your favourite workload to find out.
But generally speaking, frame pointers impose quite some performance
penalty indeed. lmbench syscall microbenchmark can give you some hint. I
expect you'll see approx. 10% performance increase for various syscalls if
you disable them, as that's what we have measured lately.
--
Jiri Kosina
SUSE Labs, Novell Inc.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: What is the performance when using frame pointers in the kernel?
2009-12-04 13:23 ` Jiri Kosina
@ 2009-12-04 15:25 ` Justin Piszcz
2009-12-05 7:02 ` Ingo Molnar
2009-12-06 22:31 ` Andi Kleen
1 sibling, 1 reply; 6+ messages in thread
From: Justin Piszcz @ 2009-12-04 15:25 UTC (permalink / raw)
To: Jiri Kosina; +Cc: Peter Zijlstra, linux-kernel
On Fri, 4 Dec 2009, Jiri Kosina wrote:
> On Thu, 3 Dec 2009, Peter Zijlstra wrote:
>
>>> What is the performance hit when the kernel is compiled with frame
>>> pointers?
>>
>> build both kernels and run your favourite workload to find out.
>
> But generally speaking, frame pointers impose quite some performance
> penalty indeed. lmbench syscall microbenchmark can give you some hint. I
> expect you'll see approx. 10% performance increase for various syscalls if
> you disable them, as that's what we have measured lately.
>
> --
> Jiri Kosina
> SUSE Labs, Novell Inc.
>
Thanks for the response.. Good to know.
Justin.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: What is the performance when using frame pointers in the kernel?
2009-12-04 15:25 ` Justin Piszcz
@ 2009-12-05 7:02 ` Ingo Molnar
0 siblings, 0 replies; 6+ messages in thread
From: Ingo Molnar @ 2009-12-05 7:02 UTC (permalink / raw)
To: Justin Piszcz; +Cc: Jiri Kosina, Peter Zijlstra, linux-kernel
* Justin Piszcz <jpiszcz@lucidpixels.com> wrote:
> On Fri, 4 Dec 2009, Jiri Kosina wrote:
>
> >On Thu, 3 Dec 2009, Peter Zijlstra wrote:
> >
> >>>What is the performance hit when the kernel is compiled with frame
> >>>pointers?
> >>
> >>build both kernels and run your favourite workload to find out.
> >
> >But generally speaking, frame pointers impose quite some performance
> >penalty indeed. lmbench syscall microbenchmark can give you some hint. I
> >expect you'll see approx. 10% performance increase for various syscalls if
> >you disable them, as that's what we have measured lately.
> >
> >--
> >Jiri Kosina
> >SUSE Labs, Novell Inc.
> >
>
> Thanks for the response.. Good to know.
I dont buy the 10% without seeing precise measurement results. 1-2%
maybe, in some cases.
Ingo
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: What is the performance when using frame pointers in the kernel?
2009-12-04 13:23 ` Jiri Kosina
2009-12-04 15:25 ` Justin Piszcz
@ 2009-12-06 22:31 ` Andi Kleen
1 sibling, 0 replies; 6+ messages in thread
From: Andi Kleen @ 2009-12-06 22:31 UTC (permalink / raw)
To: Jiri Kosina; +Cc: Peter Zijlstra, Justin Piszcz, linux-kernel
Jiri Kosina <jkosina@suse.cz> writes:
> On Thu, 3 Dec 2009, Peter Zijlstra wrote:
>
>> > What is the performance hit when the kernel is compiled with frame
>> > pointers?
>>
>> build both kernels and run your favourite workload to find out.
>
> But generally speaking, frame pointers impose quite some performance
> penalty indeed. lmbench syscall microbenchmark can give you some hint. I
> expect you'll see approx. 10% performance increase for various syscalls if
> you disable them, as that's what we have measured lately.
It'll depend on the CPU. On many common cores frame pointer cause
a a few cycles stall on each function entry, but not on all.
-Andi
--
ak@linux.intel.com -- Speaking for myself only.
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2009-12-06 22:31 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2009-12-03 17:36 What is the performance when using frame pointers in the kernel? Justin Piszcz
2009-12-03 19:28 ` Peter Zijlstra
2009-12-04 13:23 ` Jiri Kosina
2009-12-04 15:25 ` Justin Piszcz
2009-12-05 7:02 ` Ingo Molnar
2009-12-06 22:31 ` Andi Kleen
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox