Thanks Peter. We'll give that patch a try as part of our refactoring.
Looking at finer-grained locking and we'll try going back to rt_mutex
plus this patch.
On Wed, Sep 14, 2016 at 9:55 AM, Peter Zijlstra wrote:
> On Wed, Sep 14, 2016 at 06:13:40PM +0200, Peter Zijlstra wrote:
>> On Wed, Sep 14,
On Wed, Sep 14, 2016 at 06:13:40PM +0200, Peter Zijlstra wrote:
> On Wed, Sep 14, 2016 at 06:11:03PM +0200, Peter Zijlstra wrote:
> > On Tue, Sep 13, 2016 at 12:53:27PM -0700, Arve Hjønnevåg wrote:
> > > On Tue, Sep 13, 2016 at 12:32 AM, Peter Zijlstra
> > > wrote:
> > > > cgroups should be irrel
On Wed, Sep 14, 2016 at 06:11:03PM +0200, Peter Zijlstra wrote:
> On Tue, Sep 13, 2016 at 12:53:27PM -0700, Arve Hjønnevåg wrote:
> > On Tue, Sep 13, 2016 at 12:32 AM, Peter Zijlstra
> > wrote:
> > > cgroups should be irrelevant, PI is unaware of them.
> >
> > I don't think cgroups are irrelevan
On Tue, Sep 13, 2016 at 12:53:27PM -0700, Arve Hjønnevåg wrote:
> On Tue, Sep 13, 2016 at 12:32 AM, Peter Zijlstra wrote:
> > cgroups should be irrelevant, PI is unaware of them.
>
> I don't think cgroups are irrelevant. PI being unaware of them
> explains the problem I described. If the task tha
On Wed, Sep 14, 2016 at 09:10:01AM +0200, Peter Zijlstra wrote:
> On Tue, Sep 13, 2016 at 12:53:27PM -0700, Arve Hjønnevåg wrote:
> > Most of the tasks here are not RR/FIFO/DL tasks. I don't see anything
> > in the rtmutex code or documentation that indicates that they don't
> > work for normal ta
On Wed, Sep 14, 2016 at 09:10:01AM +0200, Peter Zijlstra wrote:
> We could define a meaningful something for CFS and implement that, but
> it isn't currently done.
So the generalization of the Priority Inheritance Protocol is Proxy
Execution Protocol, which basically lets the boosted task run _as_
On Tue, Sep 13, 2016 at 12:53:27PM -0700, Arve Hjønnevåg wrote:
> On Tue, Sep 13, 2016 at 12:32 AM, Peter Zijlstra wrote:
> > On Mon, Sep 12, 2016 at 08:44:09PM -0700, Arve Hjønnevåg wrote:
> >
> >> A previous attempt to fix this problem, changed the lock to use
> >> rt_mutex instead of mutex, but
On Tue, Sep 13, 2016 at 12:32 AM, Peter Zijlstra wrote:
> On Mon, Sep 12, 2016 at 08:44:09PM -0700, Arve Hjønnevåg wrote:
>
>> A previous attempt to fix this problem, changed the lock to use
>> rt_mutex instead of mutex, but this apparently did not work as well as
>> this patch. I believe the adde
On Mon, Sep 12, 2016 at 11:42 PM, Greg Kroah-Hartman
wrote:
> On Mon, Sep 12, 2016 at 08:44:09PM -0700, Arve Hjønnevåg wrote:
>> On Sat, Sep 10, 2016 at 10:28 AM, Greg Kroah-Hartman
>> wrote:
>> > On Sat, Sep 10, 2016 at 06:37:29PM +0200, Thomas Gleixner wrote:
>> >> On Sat, 10 Sep 2016, Peter Zi
On Mon, Sep 12, 2016 at 08:44:09PM -0700, Arve Hjønnevåg wrote:
> A previous attempt to fix this problem, changed the lock to use
> rt_mutex instead of mutex, but this apparently did not work as well as
> this patch. I believe the added overhead was noticeable, and it did
> not work when the preem
On Mon, Sep 12, 2016 at 08:44:09PM -0700, Arve Hjønnevåg wrote:
> On Sat, Sep 10, 2016 at 10:28 AM, Greg Kroah-Hartman
> wrote:
> > On Sat, Sep 10, 2016 at 06:37:29PM +0200, Thomas Gleixner wrote:
> >> On Sat, 10 Sep 2016, Peter Zijlstra wrote:
> >>
> >> > On Sat, Sep 10, 2016 at 09:16:59AM -0700,
On Sat, Sep 10, 2016 at 10:28 AM, Greg Kroah-Hartman
wrote:
> On Sat, Sep 10, 2016 at 06:37:29PM +0200, Thomas Gleixner wrote:
>> On Sat, 10 Sep 2016, Peter Zijlstra wrote:
>>
>> > On Sat, Sep 10, 2016 at 09:16:59AM -0700, Christoph Hellwig wrote:
>> > > On Thu, Sep 08, 2016 at 09:12:50AM -0700, T
Thanks for the reviews. We'll come up with a different solution.
On Sat, Sep 10, 2016 at 10:28 AM, Greg Kroah-Hartman
wrote:
> On Sat, Sep 10, 2016 at 06:37:29PM +0200, Thomas Gleixner wrote:
>> On Sat, 10 Sep 2016, Peter Zijlstra wrote:
>>
>> > On Sat, Sep 10, 2016 at 09:16:59AM -0700, Christoph
On Sat, Sep 10, 2016 at 06:37:29PM +0200, Thomas Gleixner wrote:
> On Sat, 10 Sep 2016, Peter Zijlstra wrote:
>
> > On Sat, Sep 10, 2016 at 09:16:59AM -0700, Christoph Hellwig wrote:
> > > On Thu, Sep 08, 2016 at 09:12:50AM -0700, Todd Kjos wrote:
> > > > In Android systems, the display pipeline r
On Sat, Sep 10, 2016 at 09:16:59AM -0700, Christoph Hellwig wrote:
> On Thu, Sep 08, 2016 at 09:12:50AM -0700, Todd Kjos wrote:
> > In Android systems, the display pipeline relies on low
> > latency binder transactions and is therefore sensitive to
> > delays caused by contention for the global bin
On Thu, Sep 08, 2016 at 09:12:50AM -0700, Todd Kjos wrote:
> In Android systems, the display pipeline relies on low
> latency binder transactions and is therefore sensitive to
> delays caused by contention for the global binder lock.
> Jank is siginificantly reduced by disabling preemption
> while
On Sat, 10 Sep 2016, Peter Zijlstra wrote:
> On Sat, Sep 10, 2016 at 09:16:59AM -0700, Christoph Hellwig wrote:
> > On Thu, Sep 08, 2016 at 09:12:50AM -0700, Todd Kjos wrote:
> > > In Android systems, the display pipeline relies on low
> > > latency binder transactions and is therefore sensitive t
On Sat, Sep 10, 2016 at 01:18:47PM +0200, Greg KH wrote:
> On Fri, Sep 09, 2016 at 10:39:32AM -0700, Todd Kjos wrote:
> > On Fri, Sep 9, 2016 at 8:44 AM, Greg KH wrote:
> > > On Fri, Sep 09, 2016 at 08:17:44AM -0700, Todd Kjos wrote:
> > >> From: Todd Kjos
> > >>
> > >> In Android systems, the di
On Fri, Sep 09, 2016 at 10:39:32AM -0700, Todd Kjos wrote:
> On Fri, Sep 9, 2016 at 8:44 AM, Greg KH wrote:
> > On Fri, Sep 09, 2016 at 08:17:44AM -0700, Todd Kjos wrote:
> >> From: Todd Kjos
> >>
> >> In Android systems, the display pipeline relies on low
> >> latency binder transactions and is
On Fri, Sep 9, 2016 at 8:44 AM, Greg KH wrote:
> On Fri, Sep 09, 2016 at 08:17:44AM -0700, Todd Kjos wrote:
>> From: Todd Kjos
>>
>> In Android systems, the display pipeline relies on low
>> latency binder transactions and is therefore sensitive to
>> delays caused by contention for the global bi
Hi Todd,
[auto build test WARNING on staging/staging-testing]
[also build test WARNING on v4.8-rc5 next-20160909]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system]
[Suggest to use git(>=2.9.0) format-patch --base= (or --base=auto for
convenience)
On Fri, Sep 09, 2016 at 08:17:44AM -0700, Todd Kjos wrote:
> From: Todd Kjos
>
> In Android systems, the display pipeline relies on low
> latency binder transactions and is therefore sensitive to
> delays caused by contention for the global binder lock.
> Jank is significantly reduced by disablin
From: Todd Kjos
In Android systems, the display pipeline relies on low
latency binder transactions and is therefore sensitive to
delays caused by contention for the global binder lock.
Jank is significantly reduced by disabling preemption
while the global binder lock is held.
This patch was orig
On Thu, Sep 08, 2016 at 09:12:50AM -0700, Todd Kjos wrote:
> In Android systems, the display pipeline relies on low
> latency binder transactions and is therefore sensitive to
> delays caused by contention for the global binder lock.
> Jank is siginificantly reduced by disabling preemption
> while
In Android systems, the display pipeline relies on low
latency binder transactions and is therefore sensitive to
delays caused by contention for the global binder lock.
Jank is siginificantly reduced by disabling preemption
while the global binder lock is held.
Originally-from: Riley Andrews
Sign
25 matches
Mail list logo
