1
0
linux/debian/patches/patchset-zen/fixes/0001-Partially-revert-drm-amd-amdgpu-add-pipe1-hardware-s.patch
Konstantin Demin 8cbaf1dea2 add 3rd party/custom patches
3rd patchs (in alphabetical order):
- bbr3
- ntsync5
- openwrt
- pf-kernel
- xanmod
- zen

no configuration changes for now
2024-10-29 05:12:06 +03:00

39 lines
1.5 KiB
Diff

From 2b70fecae5eabc439190c435be020db8afd3bd33 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 7 Aug 2024 17:04:02 -0400
Subject: Partially revert "drm/amd/amdgpu: add pipe1 hardware support"
This partially reverts commit b7a1a0ef12b81957584fef7b61e2d5ec049c7209.
A user reported stuttering under heavy gfx load with this commit.
I suspect it's due to the fact that the gfx contexts are shared
between the pipes so if there is alot of load on one pipe, we could
end up stalling waiting for a context.
That said, having both pipes is useful in some contexts and
this patch was actually enabled mainly to support some SR-IOV
use cases, so leave it enabled for SR-IOV.
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3519
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: ZhenGuo Yin <zhenguo.yin@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -4711,7 +4711,10 @@ static int gfx_v10_0_sw_init(void *handl
case IP_VERSION(10, 3, 3):
case IP_VERSION(10, 3, 7):
adev->gfx.me.num_me = 1;
- adev->gfx.me.num_pipe_per_me = 2;
+ if (amdgpu_sriov_vf(adev))
+ adev->gfx.me.num_pipe_per_me = 2;
+ else
+ adev->gfx.me.num_pipe_per_me = 1;
adev->gfx.me.num_queue_per_pipe = 1;
adev->gfx.mec.num_mec = 2;
adev->gfx.mec.num_pipe_per_mec = 4;