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

2017-07-27 Thread Dave Airlie
On 27 July 2017 at 07:37, Dave Airlie wrote: > On 27 July 2017 at 00:06, Nicolai Hähnle wrote: >> On 26.07.2017 05:42, Dave Airlie wrote: >>> >>> From: Dave Airlie >>> >>> This seems like a workaround, but we don't see the bug on CIK/VI. >>> >>> On SI with the dEQP-VK.memory.pipeline_barrier.hos

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

2017-07-26 Thread Dave Airlie
On 27 July 2017 at 00:06, Nicolai Hähnle wrote: > On 26.07.2017 05:42, Dave Airlie wrote: >> >> From: Dave Airlie >> >> 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

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

2017-07-26 Thread Nicolai Hähnle
On 26.07.2017 05:42, Dave Airlie wrote: From: Dave Airlie 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

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

2017-07-25 Thread Dave Airlie
From: Dave Airlie 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 flu