On Wed, Jun 27, 2018 at 04:34:27PM +0200, Lucas Stach wrote:
> When the hangcheck handler was replaced by the DRM scheduler timeout
> handling we dropped the forward progress check, as this might allow
> clients to hog the GPU for a long time with a big job.
>
> It turns out that even reasonably
On Wed, Jun 27, 2018 at 04:34:27PM +0200, Lucas Stach wrote:
> When the hangcheck handler was replaced by the DRM scheduler timeout
> handling we dropped the forward progress check, as this might allow
> clients to hog the GPU for a long time with a big job.
>
> It turns out that even reasonably
Am Dienstag, den 03.07.2018, 10:44 +0100 schrieb Russell King - ARM Linux:
> On Wed, Jun 27, 2018 at 04:34:27PM +0200, Lucas Stach wrote:
> > When the hangcheck handler was replaced by the DRM scheduler timeout
> > handling we dropped the forward progress check, as this might allow
> > clients to
Hi Russell,
can you tell me if you will be able to provide some test feedback for
this patch during this week? I would like to get this into a fixes pull
towards the end of the week and would feel a bit more confident to do
so knowing that this really fixes the issue you are hitting on GC600.
Lucas Stach writes:
> Am Mittwoch, den 27.06.2018, 10:25 -0700 schrieb Eric Anholt:
>> > Lucas Stach writes:
>>
>> > When the hangcheck handler was replaced by the DRM scheduler timeout
>> > handling we dropped the forward progress check, as this might allow
>> > clients to hog the GPU for a
Am Mittwoch, den 27.06.2018, 10:25 -0700 schrieb Eric Anholt:
> > Lucas Stach writes:
>
> > When the hangcheck handler was replaced by the DRM scheduler timeout
> > handling we dropped the forward progress check, as this might allow
> > clients to hog the GPU for a long time with a big job.
> >
Lucas Stach writes:
> When the hangcheck handler was replaced by the DRM scheduler timeout
> handling we dropped the forward progress check, as this might allow
> clients to hog the GPU for a long time with a big job.
>
> It turns out that even reasonably well behaved clients like the
> Armada
Hi Lucas,
On Wed, Jun 27, 2018 at 11:34 AM, Lucas Stach wrote:
> When the hangcheck handler was replaced by the DRM scheduler timeout
> handling we dropped the forward progress check, as this might allow
> clients to hog the GPU for a long time with a big job.
>
> It turns out that even
When the hangcheck handler was replaced by the DRM scheduler timeout
handling we dropped the forward progress check, as this might allow
clients to hog the GPU for a long time with a big job.
It turns out that even reasonably well behaved clients like the
Armada Xorg driver occasionally trip over
