Dave Airlie 1e11687029 radv: force cs/ps/l2 flush at end of command stream. (v2) ... This seems like a workaround, but we don't see the bug on CIK/VI. On SI with the dEQP-VK.memory.pipeline_barrier.host_read_transfer_dst.* tests, when one tests complete, the first flush at the start of the next test causes a VM fault as we've destroyed the VM, but we end up flushing the compute shader then, and it must still be in the process of doing something. Could also be a kernel difference between SI and CIK. v2: hit this with a bigger hammer. This fixes a bunch of hangs in the vk cts with the robustness tests. Fixes: f4e499ec79 ("radv: add initial non-conformant radv vulkan driver") Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=101334 Acked-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl> Signed-off-by: Dave Airlie <airlied@redhat.com> (cherry picked from commit 82ba384c10)		2017-08-11 21:00:49 +01:00
..
addrlib	amd/addrlib: fix typo in api name.	2017-07-17 01:44:14 +01:00
common	radv: fix f16->f32 denorm handling for SI/CIK. (v2)	2017-08-11 21:00:43 +01:00
vulkan	radv: force cs/ps/l2 flush at end of command stream. (v2)	2017-08-11 21:00:49 +01:00
Android.addrlib.mk	Android: amd: use exported include dirs instead of explicit includes	2017-05-11 13:52:21 +01:00
Android.common.mk	android: ac: add missing libdrm_amdgpu shared dependency	2017-06-17 18:38:31 +01:00
Android.mk	android: amd/common: add support for libmesa_amd_common	2016-11-05 18:42:29 +01:00
Makefile.addrlib.am	amd/addrlib: import gfx9 support	2017-03-30 14:44:33 +02:00
Makefile.am	ac: automake: rework sid_tables.h generation	2017-01-12 19:29:28 +00:00
Makefile.common.am	Revert "amd/common: add missing libdrm include path"	2017-06-16 12:41:44 +01:00
Makefile.sources	ac/radeonsi: move radeon_info initialization to amd/common	2017-05-18 11:48:52 +02:00