On 04/13/2016 10:25 PM, Ilia Mirkin wrote:
Reviewed-by: Ilia Mirkin <imir...@alum.mit.edu> If you wanted to play around with it to figure out which bit was which, I wouldn't object. I think someone pushed (or has sent) a more complete mem barrier test. Although such guessing can't be conclusive, so perhaps might as well just leave it like this...
REing those barriers bits doesn't seem to be really easy. :) I prefer to leave it like this as you suggest.
Oh, but one thing, instead please do if (flags & (PIPE_A | PIPE_B | PIPE_C)) that way it's a single operation instead of 20.
Fine by me.
-ilia On Tue, Apr 12, 2016 at 7:56 PM, Samuel Pitoiset <samuel.pitoi...@gmail.com> wrote:This fixes a bunch of subtests of arb_shader_image_load_store-host-mem-barrier. Signed-off-by: Samuel Pitoiset <samuel.pitoi...@gmail.com> --- src/gallium/drivers/nouveau/nvc0/nvc0_context.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/src/gallium/drivers/nouveau/nvc0/nvc0_context.c b/src/gallium/drivers/nouveau/nvc0/nvc0_context.c index 3e25572..3b20196 100644 --- a/src/gallium/drivers/nouveau/nvc0/nvc0_context.c +++ b/src/gallium/drivers/nouveau/nvc0/nvc0_context.c @@ -91,7 +91,14 @@ nvc0_memory_barrier(struct pipe_context *pipe, unsigned flags) } } } - if (flags & PIPE_BARRIER_SHADER_BUFFER) { + + if ((flags & PIPE_BARRIER_SHADER_BUFFER) || + (flags & PIPE_BARRIER_CONSTANT_BUFFER) || + (flags & PIPE_BARRIER_INDEX_BUFFER) || + (flags & PIPE_BARRIER_IMAGE) || + (flags & PIPE_BARRIER_TEXTURE) || + (flags & PIPE_BARRIER_VERTEX_BUFFER) || + (flags & PIPE_BARRIER_STREAMOUT_BUFFER)) { IMMED_NVC0(push, NVC0_3D(MEM_BARRIER), 0x1011); } } -- 2.8.0 _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
_______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
