Mailing List Archive: [PATCH 0 of 1] sched: rework locking for dump debugkey.

[PATCH 0 of 1] sched: rework locking for dump debugkey.

Jan 18, 2012, 8:13 AM

Post #1 of 3 (93 views)

Hi George, everyone,

I noticed this issue while going through the scheduling code. Basically,
for the runq debugkey, locking happens in schedule.c (which we usually
don't want) and doesn't seem take credit2's runqs<-->CPUs mapping into
account.

I think this patch correctly deals with locking in this specific case,
but of course I'm happy to hear what you think! :-)
I tested the patch with all schedulers, but as you can imagine it's
quite hard to produce an actual race and see if it is being handled
properly...

Thanks and Regards,
Dario

--
<<This happens because I choose it to happen!>> (Raistlin Majere)
-------------------------------------------------------------------
Dario Faggioli, http://retis.sssup.it/people/faggioli
Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)
PhD Candidate, ReTiS Lab, Scuola Superiore Sant'Anna, Pisa (Italy)

Re: [PATCH 0 of 1] sched: rework locking for dump debugkey. [ In reply to ]

JBeulich at suse

Jan 18, 2012, 8:22 AM

Post #2 of 3 (91 views)

Permalink

>>> On 18.01.12 at 17:13, Dario Faggioli <raistlin@linux.it> wrote:
> Hi George, everyone,
>
> I noticed this issue while going through the scheduling code. Basically,
> for the runq debugkey, locking happens in schedule.c (which we usually
> don't want) and doesn't seem take credit2's runqs<-->CPUs mapping into
> account.
>
>
> I think this patch correctly deals with locking in this specific case,
> but of course I'm happy to hear what you think! :-)

Forgot to include/attach the patch?

> I tested the patch with all schedulers, but as you can imagine it's
> quite hard to produce an actual race and see if it is being handled
> properly...
>
> Thanks and Regards,
> Dario
>
> --
> <<This happens because I choose it to happen!>> (Raistlin Majere)
> -------------------------------------------------------------------
> Dario Faggioli, http://retis.sssup.it/people/faggioli
> Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)
> PhD Candidate, ReTiS Lab, Scuola Superiore Sant'Anna, Pisa (Italy)

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Re: [PATCH 0 of 1] sched: rework locking for dump debugkey. [ In reply to ]

raistlin at linux

Jan 18, 2012, 8:28 AM

Post #3 of 3 (88 views)

Permalink

On Wed, 2012-01-18 at 16:22 +0000, Jan Beulich wrote:
> > I think this patch correctly deals with locking in this specific case,
> > but of course I'm happy to hear what you think! :-)
>
> Forgot to include/attach the patch?
>
No, it's coming but you're right, my "submission machinery" definitely
need some refinements! :-)

Will do that ASAP, as I expect to start submitting more complex series
in the coming period, and this method I'm using now does not scale.

Thanks and Regards,
Dario

--
<<This happens because I choose it to happen!>> (Raistlin Majere)
-------------------------------------------------------------------
Dario Faggioli, http://retis.sssup.it/people/faggioli
Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)
PhD Candidate, ReTiS Lab, Scuola Superiore Sant'Anna, Pisa (Italy)

Mailing List Archive

Attached Files:

Attached Files: