[Mesa-dev] [PATCH] i965/gen8+: Skip depth stalls on state change

[Mark Janes]

Ben Widawsky <ben at bwidawsk.net> writes:

> On Fri, Sep 04, 2015 at 02:58:30PM -0700, Ben Widawsky wrote:
>> On Wed, Sep 02, 2015 at 01:03:43PM +0100, Chris Wilson wrote:
>> > On Wed, Aug 26, 2015 at 10:52:58AM -0700, Ben Widawsky wrote:
>> > > Docs suggest this is no longer required starting with Gen8.
>> > > 
>> > > Perf (no regressions in n=20)
>> > > OglMultithread       0.67%
>> > > OglTerrainPanInst    0.12%
>> > > trex                 0.45%
>> > > warsow               0.64%
>> > > 
>> > > I have a couple of spurious failures in piglit on BSW, and SKL. I have no
>> > > evidence that they are related to this patch.
>> > > 
>> > > Cc: Chris Wilson <chris at chris-wilson.co.uk>
>> > > Signed-off-by: Ben Widawsky <ben at bwidawsk.net>
>> > 
>> > I could not find any contra-indications that would suggest the workaround
>> > needed to be preserved for bdw+.
>> > 
>> > Reviewed-by: Chris Wilson <chris at chris-wilson.co.uk>
>> > -Chris
>> 
>> Apparently rebase on master and with a kernel update there are actually 3
>> regressions. Any bright ideas?
>> 
>> piglit.spec.arb_texture_cube_map_array.fbo-generatemipmap-cubemap array s3tc_dxt1.sklm64
>> piglit.spec.arb_get_texture_sub_image.arb_get_texture_sub_image-getcompressed.sklm64
>> piglit.spec.ext_texture_compression_s3tc.getteximage-targets cube_array s3tc.sklm64

I see all of these failures on mesa master.  They appear to have been
triggered by updating the kernel for the SKL machines:

38090edf0488bc28313f75f177072f1b4f233713
drm-intel-nightly: 2015y-09m-02d-15h-24m-23s UTC integration manifest

> FWIW, I don't think the mesa is the cause here. It seems to be the kernel change
> which somehow caused this regression.
>
> Mark, can you double check that you're not seeing master hitting these 3
> failures?
>> 
>> > 
>> > -- 
>> > Chris Wilson, Intel Open Source Technology Centre