Re: [Intel-gfx] [PATCH v2 2/2] drm/i915/gtt: stop caching the scratch page

[Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxxxxxxxx>]

On 28/10/2021 10:26, Matthew Auld wrote:
> Normal users shouldn't be hitting this, likely this would indicate a
> userspace bug. So don't bother caching, which should be safe now that we
> manually flush the page.

Could this have a performance impact if "things" overfetch often enough 
in normal operation?

Regards,

Tvrtko

> Suggested-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
> Signed-off-by: Matthew Auld <matthew.auld@xxxxxxxxx>
> Cc: Thomas Hellström <thomas.hellstrom@xxxxxxxxxxxxxxx>
> Cc: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
> Cc: Ramalingam C <ramalingam.c@xxxxxxxxx>
> Reviewed-by: Thomas Hellström <thomas.hellstrom@xxxxxxxxxxxxxxx>
> ---
>   drivers/gpu/drm/i915/gt/gen8_ppgtt.c | 4 ++--
>   1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
> index 8af2f709571c..9966e9dc5218 100644
> --- a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
> +++ b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c
> @@ -651,7 +651,7 @@ static int gen8_init_scratch(struct i915_address_space *vm)
>   
>   	vm->scratch[0]->encode =
>   		gen8_pte_encode(px_dma(vm->scratch[0]),
> -				I915_CACHE_LLC, pte_flags);
> +				I915_CACHE_NONE, pte_flags);
>   
>   	for (i = 1; i <= vm->top; i++) {
>   		struct drm_i915_gem_object *obj;
> @@ -667,7 +667,7 @@ static int gen8_init_scratch(struct i915_address_space *vm)
>   		}
>   
>   		fill_px(obj, vm->scratch[i - 1]->encode);
> -		obj->encode = gen8_pde_encode(px_dma(obj), I915_CACHE_LLC);
> +		obj->encode = gen8_pde_encode(px_dma(obj), I915_CACHE_NONE);
>   
>   		vm->scratch[i] = obj;
>   	}