[Intel-gfx] [PATCH] drm/i915/execlists: Flush the post-sync breadcrumb write harder

Mika Kuoppala mika.kuoppala at linux.intel.com
Wed Aug 28 12:49:55 UTC 2019


Chris Wilson <chris at chris-wilson.co.uk> writes:

> Quite rarely we see that the CS completion event fires before the
> breadcrumb is coherent, which presumably is a result of the CS_STALL not
> waiting for the post-sync operation. Try following in a DC_FLUSH into
> the following pipecontrol to see if that makes any difference.
>
> Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
> Cc: Mika Kuoppala <mika.kuoppala at linux.intel.com>

Acked-by: Mika Kuoppala <mika.kuoppala at linux.intel.com>

> ---
>  drivers/gpu/drm/i915/gt/intel_lrc.c | 1 +
>  1 file changed, 1 insertion(+)
>
> diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c
> index 80a3f1dbb456..48046d445733 100644
> --- a/drivers/gpu/drm/i915/gt/intel_lrc.c
> +++ b/drivers/gpu/drm/i915/gt/intel_lrc.c
> @@ -2971,6 +2971,7 @@ static u32 *gen8_emit_fini_breadcrumb_rcs(struct i915_request *request, u32 *cs)
>  	/* XXX flush+write+CS_STALL all in one upsets gem_concurrent_blt:kbl */
>  	cs = gen8_emit_pipe_control(cs,
>  				    PIPE_CONTROL_FLUSH_ENABLE |
> +				    PIPE_CONTROL_DC_FLUSH_ENABLE |
>  				    PIPE_CONTROL_CS_STALL,
>  				    0);
>  
> -- 
> 2.23.0


More information about the Intel-gfx mailing list