Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1415714 > unrolled thread

Re: [PATCH v1] PM / sleep: make pm notifier called symmetrically

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: [PATCH v1] PM / sleep: make pm notifier called symmetrically Lianwei Wang <lianwei.wang@gmail.com> - 2016-06-07 07:50 +0200
    Re: [PATCH v1] PM / sleep: make pm notifier called symmetrically "Rafael J. Wysocki" <rjw@rjwysocki.net> - 2016-06-08 02:40 +0200

#1415714 — Re: [PATCH v1] PM / sleep: make pm notifier called symmetrically

On Wed, May 18, 2016 at 11:06 PM, Lianwei Wang <lianwei.wang@gmail.com> wrote:
> On Wed, May 18, 2016 at 1:34 PM, Rafael J. Wysocki <rafael@kernel.org> wrote:
>> On Wed, May 18, 2016 at 8:07 AM, Lianwei Wang <lianwei.wang@gmail.com> wrote:
>>> This makes pm notifier PREPARE/POST symmetrical: if PREPARE
>>> fails, we will only undo what ever happened on PREPARE.
>>>
>>> It fixes the unbalanced cpu hotplug enable in cpu pm notifier.
>>
>> Can you please describe the problem you're trying to address in a bit
>> more detail?
>
> It is described in another patch, and it is a common issue. I fixed
> the similar issue several times, all because that the CPU_POST_SUSPEND
> notifier callback is called without call CPU_SUSPEND_PREPARE notifier
> callback before. Something is DO in PM_XXX_PREPARE and UNDO in
> POST_XXX. If we don't DO it in XXX_PREPARE then we don't need to UNDO
> it in  PM_POST_XXX.
>
> Here is the link for the unbalanced cpu hotplug enable case:
> https://lkml.org/lkml/2016/5/6/61 .
>
> Thanks,
> Lianwei

Did anyone from the maintainer review the patch yet?

[toc] | [next] | [standalone]

#1416785

On Monday, June 06, 2016 10:41:25 PM Lianwei Wang wrote:
> On Wed, May 18, 2016 at 11:06 PM, Lianwei Wang <lianwei.wang@gmail.com> wrote:
> > On Wed, May 18, 2016 at 1:34 PM, Rafael J. Wysocki <rafael@kernel.org> wrote:
> >> On Wed, May 18, 2016 at 8:07 AM, Lianwei Wang <lianwei.wang@gmail.com> wrote:
> >>> This makes pm notifier PREPARE/POST symmetrical: if PREPARE
> >>> fails, we will only undo what ever happened on PREPARE.
> >>>
> >>> It fixes the unbalanced cpu hotplug enable in cpu pm notifier.
> >>
> >> Can you please describe the problem you're trying to address in a bit
> >> more detail?
> >
> > It is described in another patch, and it is a common issue. I fixed
> > the similar issue several times, all because that the CPU_POST_SUSPEND
> > notifier callback is called without call CPU_SUSPEND_PREPARE notifier
> > callback before. Something is DO in PM_XXX_PREPARE and UNDO in
> > POST_XXX. If we don't DO it in XXX_PREPARE then we don't need to UNDO
> > it in  PM_POST_XXX.
> >
> > Here is the link for the unbalanced cpu hotplug enable case:
> > https://lkml.org/lkml/2016/5/6/61 .
> >
> > Thanks,
> > Lianwei
> 
> Did anyone from the maintainer review the patch yet?

Yes, I have reviewed it, but not yet decided whether or not I will accept it.

Thanks,
Rafael

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web