[Mesa-dev] [PATCH] nvc0: re-validate images after launching a grid on Fermi

[Ilia Mirkin]

On Sun, Jun 5, 2016 at 12:13 PM, Samuel Pitoiset
<samuel.pitoiset at gmail.com> wrote:
> Images invalidation is a bit weird on Fermi and there is already a hack
> which forces invalidating all images when launching a computer shader
> to help in fixing 3D<->CP interaction.
>
> However, we need to re-validate images for compute because
> nvc0_compute_invalidate_surfaces() will destroy the previous binding.
> This is not really good for performance purposes but this might be
> improved later.
>
> This fixes the following piglits:
> - spec/arb_compute_shader/execution/basic-uniform-access
> - spec/arb_compute_shader/execution/mutiple-texture-reading
> - spec/arb_compute_shader/execution/multiple-workgroups
> - spec/glsl-4.30/execution/built-in-functions/cs-* (207 tests)
>
> Signed-off-by: Samuel Pitoiset <samuel.pitoiset at gmail.com>
> Cc: "12.0" <mesa-stable at lists.freedesktop.org>
> ---
>  src/gallium/drivers/nouveau/nvc0/nvc0_compute.c | 2 ++
>  1 file changed, 2 insertions(+)
>
> diff --git a/src/gallium/drivers/nouveau/nvc0/nvc0_compute.c b/src/gallium/drivers/nouveau/nvc0/nvc0_compute.c
> index 8c88225..baeeca5 100644
> --- a/src/gallium/drivers/nouveau/nvc0/nvc0_compute.c
> +++ b/src/gallium/drivers/nouveau/nvc0/nvc0_compute.c
> @@ -458,4 +458,6 @@ nvc0_launch_grid(struct pipe_context *pipe, const struct pipe_grid_info *info)
>
>     /* TODO: Not sure if this is really necessary. */
>     nvc0_compute_invalidate_surfaces(nvc0, 5);
> +   nvc0->dirty_cp |= NVC0_NEW_CP_SURFACES;
> +   nvc0->images_dirty[5] |= nvc0->images_valid[5];

Don't forget to clear out the bufctx. Also (in another patch), do a
pass over all your invalidation hacks - any time you do nvc0->dirty |=
foo, you should have an associated nouveau_bufctx_reset(). Otherwise
things can start piling up in there. (Since validation only adds to
the buckets.)

  -ilia