[Mesa-dev] [PATCH] Revert "radv: do not allocate CMASK or DCC for small surfaces"

[Bas Nieuwenhuizen]

Reviewed-by: Bas Nieuwenhuizen <bas at basnieuwenhuizen.nl>

On Wed, Dec 20, 2017 at 9:07 PM, Samuel Pitoiset
<samuel.pitoiset at gmail.com> wrote:
> This reverts commit ff0f17da1446e7aa965e06c04a6ad5a55d95463d.
>
> See the TODO.
>
> Signed-off-by: Samuel Pitoiset <samuel.pitoiset at gmail.com>
> ---
>  src/amd/vulkan/radv_image.c | 17 +++++++----------
>  1 file changed, 7 insertions(+), 10 deletions(-)
>
> diff --git a/src/amd/vulkan/radv_image.c b/src/amd/vulkan/radv_image.c
> index efd17e4889..d2acfea86a 100644
> --- a/src/amd/vulkan/radv_image.c
> +++ b/src/amd/vulkan/radv_image.c
> @@ -812,16 +812,13 @@ radv_image_alloc_htile(struct radv_image *image)
>  static inline bool
>  radv_image_can_enable_dcc_or_cmask(struct radv_image *image)
>  {
> -       if (image->info.samples <= 1 &&
> -           image->info.width <= 512 && image->info.height <= 512) {
> -               /* Do not enable CMASK or DCC for small surfaces where the cost
> -                * of the eliminate pass can be higher than the benefit of fast
> -                * clear. RadeonSI does this, but the image threshold is
> -                * different.
> -                */
> -               return false;
> -       }
> -
> +       /* TODO: We should not enable CMASK or DCC for small surfaces because
> +        * the cost of the eliminate pass can be higher than the benefit of
> +        * fast clear. Note that radv does something bad when it's disabled,
> +        * for example dEQP-VK.pipeline.multisample_interpolation.sample_interpolate_at_single_sample.128_128_1.samples_1
> +        * reports VM faults and hangs on Vega. This has to be fixed at some
> +        * point.
> +        */
>         return image->usage & VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT &&
>                (image->exclusive || image->queue_family_mask == 1);
>  }
> --
> 2.15.1
>
> _______________________________________________
> mesa-dev mailing list
> mesa-dev at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/mesa-dev