* Perf degradation from -rt14 onwards
@ 2005-12-01 20:42 Dinakar Guniguntala
2005-12-01 20:59 ` David Singleton
2005-12-01 21:21 ` Ingo Molnar
0 siblings, 2 replies; 3+ messages in thread
From: Dinakar Guniguntala @ 2005-12-01 20:42 UTC (permalink / raw)
To: linux-kernel; +Cc: Ingo Molnar, david singleton
I was wondering why the following change was made from -rt14
onwards.
@@ -1634,7 +1531,7 @@ asmlinkage long sys_futex(u32 __user *ua
int val3)
{
struct timespec t;
- unsigned long timeout = MAX_SCHEDULE_TIMEOUT;
+ unsigned long timeout = 0;
This was introduced in patch-2.6.14-rt13-rf3 by David.
This seems to return spurious -ETIMEDOUT errors even in the
non-robust code and results in userspace (glibc) retrying
several mutex operations before it succeeds. I was chasing
down a degradation of performance of some testcases and was
able to fix those by reverting this change back.
-Dinakar
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Perf degradation from -rt14 onwards
2005-12-01 20:42 Perf degradation from -rt14 onwards Dinakar Guniguntala
@ 2005-12-01 20:59 ` David Singleton
2005-12-01 21:21 ` Ingo Molnar
1 sibling, 0 replies; 3+ messages in thread
From: David Singleton @ 2005-12-01 20:59 UTC (permalink / raw)
To: dino; +Cc: linux-kernel, Ingo Molnar
Dinakar Guniguntala wrote:
>I was wondering why the following change was made from -rt14
>onwards.
>
>
>@@ -1634,7 +1531,7 @@ asmlinkage long sys_futex(u32 __user *ua
> int val3)
> {
> struct timespec t;
>- unsigned long timeout = MAX_SCHEDULE_TIMEOUT;
>+ unsigned long timeout = 0;
>
>This was introduced in patch-2.6.14-rt13-rf3 by David.
>
>This seems to return spurious -ETIMEDOUT errors even in the
>non-robust code and results in userspace (glibc) retrying
>several mutex operations before it succeeds. I was chasing
>down a degradation of performance of some testcases and was
>able to fix those by reverting this change back.
>
>
Yes. The default timeout should be set to MAX_SCHEDULE_TIMEOUT, not zero.
David
> -Dinakar
>
>
>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Perf degradation from -rt14 onwards
2005-12-01 20:42 Perf degradation from -rt14 onwards Dinakar Guniguntala
2005-12-01 20:59 ` David Singleton
@ 2005-12-01 21:21 ` Ingo Molnar
1 sibling, 0 replies; 3+ messages in thread
From: Ingo Molnar @ 2005-12-01 21:21 UTC (permalink / raw)
To: Dinakar Guniguntala; +Cc: linux-kernel, david singleton
* Dinakar Guniguntala <dino@in.ibm.com> wrote:
> I was wondering why the following change was made from -rt14
> onwards.
>
>
> @@ -1634,7 +1531,7 @@ asmlinkage long sys_futex(u32 __user *ua
> int val3)
> {
> struct timespec t;
> - unsigned long timeout = MAX_SCHEDULE_TIMEOUT;
> + unsigned long timeout = 0;
>
> This was introduced in patch-2.6.14-rt13-rf3 by David.
>
> This seems to return spurious -ETIMEDOUT errors even in the non-robust
> code and results in userspace (glibc) retrying several mutex
> operations before it succeeds. I was chasing down a degradation of
> performance of some testcases and was able to fix those by reverting
> this change back.
nice catch! I've undone this change in my tree.
Ingo
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2005-12-01 21:20 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2005-12-01 20:42 Perf degradation from -rt14 onwards Dinakar Guniguntala
2005-12-01 20:59 ` David Singleton
2005-12-01 21:21 ` Ingo Molnar
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox