Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1475193
| From | Tejun Heo <tj@kernel.org> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | Re: [Question] about patch: don't use [delayed_]work_pending() |
| Date | 2016-09-02 16:30 +0200 |
| Message-ID | <scXDH-B6-7@gated-at.bofh.it> (permalink) |
| References | <scwtQ-81r-25@gated-at.bofh.it> <scHSj-7dg-69@gated-at.bofh.it> <scLjb-144-1@gated-at.bofh.it> <scXaF-9t-5@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
On Fri, Sep 02, 2016 at 09:50:07AM -0400, Tejun Heo wrote: > Hello, > > On Fri, Sep 02, 2016 at 09:17:04AM +0800, qiaozhou wrote: > > > > I don't know whether it's meaningful to still check pending work here, or > > > > it's not suggested to use pm_qos_update_request in this early boot up phase. > > > > Could you help to share some opinions? (I can fix this issue by adding the > > > > current qos value directly instead of default value, though.) > > > Hmmm... but I suppose this is super-early in the boot. Would it make > > > sense to have a static variable (e.g. bool clk_fully_initailized) to > > > gate the cancel_delayed_sync() call? > > > > You're right that it's indeed super-early stage. But currently we can't > > control the gate of can_delayed_work_sync, since it's inside > > pm_qos_update_request. Out of our control. We can choose to not call > > pm_qos_update_request to avoid this issue, and use pm_qos_add_request > > alternatively. Good to have it. > > Ah sorry, didn't understand that the offending cancel_sync call is in > the generic part. Hmm... but yeah, we should still be able to take > the same approach. I'll see what's the right thing to gate the > operation there. Does the following patch work? Subject: power: avoid calling cancel_delayed_work_sync() during early boot of_clk_init() ends up calling into pm_qos_update_request() very early during boot where irq is expected to stay disabled. pm_qos_update_request() uses cancel_delayed_work_sync() which correctly assumes that irq is enabled on invocation and unconditionally disables and re-enables it. Gate cancel_delayed_work_sync() invocation with kevented_up() to avoid enabling irq unexpectedly during early boot. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Qiao Zhou <qiaozhou@asrmicro.com> Link: http://lkml.kernel.org/r/d2501c4c-8e7b-bea3-1b01-000b36b5dfe9@asrmicro.com --- kernel/power/qos.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/kernel/power/qos.c b/kernel/power/qos.c index 97b0df7..168ff44 100644 --- a/kernel/power/qos.c +++ b/kernel/power/qos.c @@ -482,7 +482,16 @@ void pm_qos_update_request(struct pm_qos_request *req, return; } - cancel_delayed_work_sync(&req->work); + /* + * This function may be called very early during boot, for example, + * from of_clk_init(), where irq needs to stay disabled. + * cancel_delayed_work_sync() assumes that irq is enabled on + * invocation and re-enables it on return. Avoid calling it until + * workqueue is initialized. + */ + if (keventd_up()) + cancel_delayed_work_sync(&req->work); + __pm_qos_update_request(req, new_value); } EXPORT_SYMBOL_GPL(pm_qos_update_request);
Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
[Question] about patch: don't use [delayed_]work_pending() qiaozhou <qiaozhou@asrmicro.com> - 2016-09-01 11:30 +0200
Re: [Question] about patch: don't use [delayed_]work_pending() Tejun Heo <tj@kernel.org> - 2016-09-01 23:40 +0200
Re: [Question] about patch: don't use [delayed_]work_pending() qiaozhou <qiaozhou@asrmicro.com> - 2016-09-02 03:20 +0200
Re: [Question] about patch: don't use [delayed_]work_pending() Tejun Heo <tj@kernel.org> - 2016-09-02 16:00 +0200
Re: [Question] about patch: don't use [delayed_]work_pending() Tejun Heo <tj@kernel.org> - 2016-09-02 16:30 +0200
Re: [Question] about patch: don't use [delayed_]work_pending() qiaozhou <qiaozhou@asrmicro.com> - 2016-09-03 17:30 +0200
Re: [Question] about patch: don't use [delayed_]work_pending() qiaozhou <qiaozhou@asrmicro.com> - 2016-09-05 07:40 +0200
[PATCH] power: avoid calling cancel_delayed_work_sync() during early boot Tejun Heo <tj@kernel.org> - 2016-09-05 14:40 +0200
Re: [PATCH] power: avoid calling cancel_delayed_work_sync() during early boot "Rafael J. Wysocki" <rafael@kernel.org> - 2016-09-05 15:00 +0200
csiph-web