[Mesa-dev] [PATCH 1/2] i965/state: Set CullTestEnableBitmask based on the last geometry stage

[Emil Velikov]

On Sat, 27 Oct 2018 at 22:34, Jason Ekstrand <jason at jlekstrand.net> wrote:
>
> Instead of hard-coding it to look at the VS stage, look at whatever the
> last geometry stage is.
>
> Cc: mesa-stable at lists.freedesktop.org
> ---
>  src/mesa/drivers/dri/i965/genX_state_upload.c | 14 +++++++++++++-
>  1 file changed, 13 insertions(+), 1 deletion(-)
>
> diff --git a/src/mesa/drivers/dri/i965/genX_state_upload.c b/src/mesa/drivers/dri/i965/genX_state_upload.c
> index 740cb0c4d2e..56fa5969725 100644
> --- a/src/mesa/drivers/dri/i965/genX_state_upload.c
> +++ b/src/mesa/drivers/dri/i965/genX_state_upload.c
> @@ -1458,6 +1458,17 @@ genX(upload_clip_state)(struct brw_context *brw)
>     struct brw_wm_prog_data *wm_prog_data =
>        brw_wm_prog_data(brw->wm.base.prog_data);
>
> +#if GEN_GEN < 8
> +   /* BRW_NEW_VS_PROG_DATA | BRW_NEW_TES_PROG_DATA | BRW_NEW_GS_PROG_DATA */
> +   const struct brw_vue_prog_data *vue_prog_data;
> +   if (brw->gs.base.prog_data)
> +      vue_prog_data = brw_vue_prog_data(brw->gs.base.prog_data);
> +   else if (brw->tes.base.prog_data)
> +      vue_prog_data = brw_vue_prog_data(brw->tes.base.prog_data);
> +   else
> +      vue_prog_data = brw_vue_prog_data(brw->vs.base.prog_data);
> +#endif
> +
>     brw_batch_emit(brw, GENX(3DSTATE_CLIP), clip) {
>        clip.StatisticsEnable = !brw->meta_in_progress;
>
> @@ -1493,7 +1504,7 @@ genX(upload_clip_state)(struct brw_context *brw)
>
>  #if GEN_GEN < 8
>        clip.UserClipDistanceCullTestEnableBitmask =
> -         brw_vue_prog_data(brw->vs.base.prog_data)->cull_distance_mask;
> +         vue_prog_data->cull_distance_mask;
>
>        clip.ViewportZClipTestEnable = !(ctx->Transform.DepthClampNear &&
>                                         ctx->Transform.DepthClampFar);
> @@ -1564,6 +1575,7 @@ static const struct brw_tracked_state genX(clip_state) = {
>        .brw   = BRW_NEW_BLORP |
>                 BRW_NEW_CONTEXT |
>                 BRW_NEW_FS_PROG_DATA |
> +               BRW_NEW_TES_PROG_DATA |
>                 BRW_NEW_GS_PROG_DATA |
>                 BRW_NEW_VS_PROG_DATA |
>                 BRW_NEW_META_IN_PROGRESS |
> --
I'm suspecting this (alongside 2/2 a second ago) is missing review?
Perhaps the CI team can add it through the system to see if the flaky
glescts tests improve?

HTH
Emil