Your idea of renaming ST_NEW_SAMPLE_MASK and using that sounds good. (we
already have 64 flags)
Marek
On Wed, May 9, 2018 at 12:21 AM, Marek Olšák wrote:
> I'd like a separate state for this, because set_framebuffer_state is a
> memory and execution barrier, so it decreases performance, thus th
I'd like a separate state for this, because set_framebuffer_state is a
memory and execution barrier, so it decreases performance, thus there is a
high incentive not to couple other states with it. Since cso_context won't
be used for setting the state (as I suggested on patch 2), st/mesa will
have t
On 05/04/2018 05:09 AM, Rhys Perry wrote:
> Signed-off-by: Rhys Perry
> ---
> src/mesa/state_tracker/st_atom_framebuffer.c | 64
>
> src/mesa/state_tracker/st_cb_msaa.c | 22 ++
> src/mesa/state_tracker/st_extensions.c | 1 +
> 3 files changed
I can't remember why I memset them to 0x88. I don't think it's
necessary, so I'll remove it.
I think I will memset the pipe_sample_locations_state to zero though
so cso_set_sample_locations() works better.
On Mon, May 7, 2018 at 4:30 PM, Brian Paul wrote:
> More nit-picks below.
>
>
>
> On 05/04
More nit-picks below.
On 05/04/2018 06:09 AM, Rhys Perry wrote:
Signed-off-by: Rhys Perry
---
src/mesa/state_tracker/st_atom_framebuffer.c | 64
src/mesa/state_tracker/st_cb_msaa.c | 22 ++
src/mesa/state_tracker/st_extensions.c | 1 +
Signed-off-by: Rhys Perry
---
src/mesa/state_tracker/st_atom_framebuffer.c | 64
src/mesa/state_tracker/st_cb_msaa.c | 22 ++
src/mesa/state_tracker/st_extensions.c | 1 +
3 files changed, 87 insertions(+)
diff --git a/src/mesa/state_tracker/s
