Samuel Pitoiset 517600b4d5 Revert "radv: stop using VM_ALWAYS_VALID on APUs" ...

Disabling VM_ALWAYS_VALID actually hurts more than it helps
after doing more testing. Managing the global BO list in userspace
is really costly and make a bunch of games CPU bound.

I think re-enabling VM_ALWAYS_VALID is a step in the right direction.

This reverts commit 6ac6e2fbfb.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9341>

2021-03-04 09:37:59 +00:00

..

addrlib

amd: update addrlib

2021-02-15 09:15:10 +00:00

ci

ci/piglit: Stop including the test counts at the end of expectations.

2021-02-24 18:55:02 +00:00

common

ac/sqtt: fix determining if the trace is complete on GFX10+

2021-03-03 08:40:32 +01:00

compiler

aco: add missing usable_read2 check

2021-03-02 13:13:59 +00:00

llvm

ac/llvm: open code fpow on LLVM 12 using fmul.legacy

2021-03-03 20:06:09 +00:00

registers

radeonsi: skip s_sendmsg(gs_alloc_req) for NGG passthrough on new chips

2021-02-13 04:56:05 +00:00

vulkan

Revert "radv: stop using VM_ALWAYS_VALID on APUs"

2021-03-04 09:37:59 +00:00

Android.addrlib.mk

android: amd/addrlib: add gfx10 support

2019-07-10 09:03:55 +02:00

Android.common.mk

android: amd/registers: switch to new generated register definitions

2020-09-06 20:20:34 +02:00

Android.compiler.mk

android: aco/isel: Move context initialization code to a dedicated file

2020-09-14 21:26:53 +02:00

Android.mk

android: aco: add support for libmesa_aco

2019-09-28 15:56:34 +02:00

Makefile.sources

ac/rgp: add helper function to write rgp elf oject

2021-02-19 14:29:47 +00:00

meson.build

aco: add framework for unit testing

2020-07-30 16:13:08 +00:00