[Intel-gfx] [PATCH] drm/i915/bdw: Cleanup pre prod workarounds

[Daniel Vetter]

On Tue, Sep 23, 2014 at 02:48:47PM +0300, Jani Nikula wrote:
> On Tue, 23 Sep 2014, Daniel Vetter <daniel at ffwll.ch> wrote:
> > On Fri, Sep 19, 2014 at 08:49:06PM +0300, Ville Syrjälä wrote:
> >> On Fri, Sep 19, 2014 at 08:05:26PM +0300, Mika Kuoppala wrote:
> >> > as these have been fixed in production hw and hurt performance
> >> > if applied.
> >> > 
> >> > v2: adjust requested ring space (Ville)
> >> > 
> >> > Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=83482
> >> > Tested-by: zhoujian <jianx.zhou at intel.com>
> >> > Signed-off-by: Mika Kuoppala <mika.kuoppala at intel.com>
> >> 
> >> Documentation agrees that these can go.
> >> Reviewed-by: Ville Syrjälä <ville.syrjala at linux.intel.com>
> >
> > Cc: stable at vger.kernel.org imo. Jani?
> 
> This depends on
> 
> commit 86d7f23842f1bce3ab5e8c8d0c676112bbc4c99b
> Author: Arun Siluvery <arun.siluvery at linux.intel.com>
> Date:   Tue Aug 26 14:44:50 2014 +0100
> 
>     drm/i915/bdw: Apply workarounds in render ring init function
> 
> which brought the problem to daylight to begin with. That's not in
> 3.17-rc6. I'm not sure if that one has additional dependencies, but
> perhaps a combined backport without an intermediate broken step is
> plausible.
> 
> In any case I don't think this is something we want to rush for 3.17
> now, since a release is imminent, so it'll be 3.18. Adding cc: stable is
> up to you, but do note the size is beyond stable rules.

Hm right, so a patch for drm-intel-next-fixes (first one even!). I'll
leave out the cc: stable since backporting all the register wa stuff is a
pile of patches ...
-Daniel

> 
> 
> BR,
> Jani.
> 
> 
> 
> > -Daniel
> >
> >> 
> >> > ---
> >> >  drivers/gpu/drm/i915/intel_ringbuffer.c | 15 ++-------------
> >> >  1 file changed, 2 insertions(+), 13 deletions(-)
> >> > 
> >> > diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
> >> > index 681ea86..679a3c7 100644
> >> > --- a/drivers/gpu/drm/i915/intel_ringbuffer.c
> >> > +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
> >> > @@ -707,7 +707,7 @@ static int bdw_init_workarounds(struct intel_engine_cs *ring)
> >> >  	 * update the number of dwords required based on the
> >> >  	 * actual number of workarounds applied
> >> >  	 */
> >> > -	ret = intel_ring_begin(ring, 24);
> >> > +	ret = intel_ring_begin(ring, 18);
> >> >  	if (ret)
> >> >  		return ret;
> >> >  
> >> > @@ -722,19 +722,8 @@ static int bdw_init_workarounds(struct intel_engine_cs *ring)
> >> >  	intel_ring_emit_wa(ring, GEN7_ROW_CHICKEN2,
> >> >  			   _MASKED_BIT_ENABLE(DOP_CLOCK_GATING_DISABLE));
> >> >  
> >> > -	/*
> >> > -	 * This GEN8_CENTROID_PIXEL_OPT_DIS W/A is only needed for
> >> > -	 * pre-production hardware
> >> > -	 */
> >> >  	intel_ring_emit_wa(ring, HALF_SLICE_CHICKEN3,
> >> > -			   _MASKED_BIT_ENABLE(GEN8_CENTROID_PIXEL_OPT_DIS
> >> > -					      | GEN8_SAMPLER_POWER_BYPASS_DIS));
> >> > -
> >> > -	intel_ring_emit_wa(ring, GEN7_HALF_SLICE_CHICKEN1,
> >> > -			   _MASKED_BIT_ENABLE(GEN7_SINGLE_SUBSCAN_DISPATCH_ENABLE));
> >> > -
> >> > -	intel_ring_emit_wa(ring, COMMON_SLICE_CHICKEN2,
> >> > -			   _MASKED_BIT_ENABLE(GEN8_CSC2_SBE_VUE_CACHE_CONSERVATIVE));
> >> > +			   _MASKED_BIT_ENABLE(GEN8_SAMPLER_POWER_BYPASS_DIS));
> >> >  
> >> >  	/* Use Force Non-Coherent whenever executing a 3D context. This is a
> >> >  	 * workaround for for a possible hang in the unlikely event a TLB
> >> > -- 
> >> > 1.9.1
> >> 
> >> -- 
> >> Ville Syrjälä
> >> Intel OTC
> >> _______________________________________________
> >> Intel-gfx mailing list
> >> Intel-gfx at lists.freedesktop.org
> >> http://lists.freedesktop.org/mailman/listinfo/intel-gfx
> >
> > -- 
> > Daniel Vetter
> > Software Engineer, Intel Corporation
> > +41 (0) 79 365 57 48 - http://blog.ffwll.ch
> 
> -- 
> Jani Nikula, Intel Open Source Technology Center

-- 
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch