Re: [PATCH v2] drm/i915: Set the Stencil Cache eviction policy to non-LRA mode.

Re: [PATCH v2] drm/i915: Set the Stencil Cache eviction policy to non-LRA mode.

[Daniel Vetter]

On Fri, Apr 27, 2012 at 12:55:12PM -0700, Kenneth Graunke wrote:
> Clearing bit 5 of CACHE_MODE_0 is necessary to prevent GPU hangs in
> OpenGL programs such as Google MapsGL, Google Earth, and gzdoom when
> using separate stencil buffers.  Without it, the GPU tries to use the
> LRA eviction policy, which isn't supported.  This was supposed to be off
> by default, but seems to be on for many machines.
> 
> This cannot be done in gen6_init_clock_gating with most of the other
> workaround bits; the render ring needs to exist.  Otherwise, the
> register write gets dropped on the floor (one printk will show it
> changed, but a second printk immediately following shows the value
> reverts to the old one).
> 
> v2: Don't write the Gen6 registers on Gen7+.
> 
> Cc: stable@kernel.org
> Cc: Rob Castle <futuredub@gmail.com>
> Cc: Eric Appleman <erappleman@gmail.com>
> Cc: aaron667@gmx.net
> Cc: Keith Packard <keithp@keithp.com>
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=47535
> Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>

Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>

Dave, would be great if you can include this patch into drm-fixes right
away. It fixes gpu hangs on various apps when using HiZ on reported by a
lot of ppl. And it would help if this lands earlier in sorting out some
conflicts with drm-intel-next.

Thanks, Daniel

> ---
>  drivers/gpu/drm/i915/i915_reg.h         |    1 +
>  drivers/gpu/drm/i915/intel_ringbuffer.c |   10 ++++++++++
>  2 files changed, 11 insertions(+)
> 
> diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
> index b4bb1ef..9d24d65 100644
> --- a/drivers/gpu/drm/i915/i915_reg.h
> +++ b/drivers/gpu/drm/i915/i915_reg.h
> @@ -568,6 +568,7 @@
>  #define   CM0_MASK_SHIFT          16
>  #define   CM0_IZ_OPT_DISABLE      (1<<6)
>  #define   CM0_ZR_OPT_DISABLE      (1<<5)
> +#define	  CM0_STC_EVICT_DISABLE_LRA_SNB	(1<<5)
>  #define   CM0_DEPTH_EVICT_DISABLE (1<<4)
>  #define   CM0_COLOR_EVICT_DISABLE (1<<3)
>  #define   CM0_DEPTH_WRITE_DISABLE (1<<1)
> diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
> index f75806e..af76723 100644
> --- a/drivers/gpu/drm/i915/intel_ringbuffer.c
> +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
> @@ -398,6 +398,16 @@ static int init_render_ring(struct intel_ring_buffer *ring)
>  			return ret;
>  	}
>  
> +	if (IS_GEN6(dev)) {
> +		/* From the Sandybridge PRM, volume 1 part 3, page 24:
> +		 * "If this bit is set, STCunit will have LRA as replacement
> +		 *  policy. [...] This bit must be reset.  LRA replacement
> +		 *  policy is not supported."
> +		 */
> +		I915_WRITE(CACHE_MODE_0,
> +			   CM0_STC_EVICT_DISABLE_LRA_SNB << CM0_MASK_SHIFT);
> +	}
> +
>  	if (INTEL_INFO(dev)->gen >= 6) {
>  		I915_WRITE(INSTPM,
>  			   INSTPM_FORCE_ORDERING << 16 | INSTPM_FORCE_ORDERING);
> -- 
> 1.7.10
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Mail: daniel@ffwll.ch
Mobile: +41 (0)79 365 57 48