[Mesa-dev] [PATCH] radv: force compute flush at end of command stream.

[Dave Airlie]

From: Dave Airlie <airlied at redhat.com>

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.

Signed-off-by: Dave Airlie <airlied at redhat.com>
---
 src/amd/vulkan/radv_cmd_buffer.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/src/amd/vulkan/radv_cmd_buffer.c b/src/amd/vulkan/radv_cmd_buffer.c
index 4415e36..d185c00 100644
--- a/src/amd/vulkan/radv_cmd_buffer.c
+++ b/src/amd/vulkan/radv_cmd_buffer.c
@@ -2233,8 +2233,10 @@ VkResult radv_EndCommandBuffer(
 {
 	RADV_FROM_HANDLE(radv_cmd_buffer, cmd_buffer, commandBuffer);
 
-	if (cmd_buffer->queue_family_index != RADV_QUEUE_TRANSFER)
+	if (cmd_buffer->queue_family_index != RADV_QUEUE_TRANSFER) {
+		cmd_buffer->state.flush_bits |= RADV_CMD_FLAG_CS_PARTIAL_FLUSH;
 		si_emit_cache_flush(cmd_buffer);
+	}
 
 	if (!cmd_buffer->device->ws->cs_finalize(cmd_buffer->cs) ||
 	    cmd_buffer->record_fail)
-- 
2.9.4