We apparently don't have anything else making sure that it's flushed in
between use as a render target and use as a texture source, so bypass-mode
depth texture sampling could get stale data.
Fixes consistent (as far as I could see) failures in FD_MESA_DEBUG=nogmem
on:
dEQP-GLES31.functional.texture.multisample.samples_*.use_texture_depth_2d
dEQP-GLES31.functional.stencil_texturing.render.depth24_stencil8_draw
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8146>
# Non-sysmem flakes
dEQP-VK.pipeline.spec_constant.compute.composite.matrix.mat3x2
-# Sysmem flakes
-dEQP-GLES31.functional.texture.multisample.samples_3.use_texture_depth_2d
-dEQP-GLES31.functional.texture.multisample.samples_4.use_texture_depth_2d_array
-dEQP-GLES31.functional.stencil_texturing.render.depth24_stencil8_draw
-
# Fails NIR_VALIDATE so probably flaky
dEQP-VK.memory_model.write_after_read.core11.u32.coherent.fence_fence.atomicwrite.workgroup.payload_nonlocal.workgroup.guard_local.buffer.comp
fd6_emit_lrz_flush(ring);
fd6_event_write(batch, ring, PC_CCU_FLUSH_COLOR_TS, true);
+ fd6_event_write(batch, ring, PC_CCU_FLUSH_DEPTH_TS, true);
}
void