[Public]
Hello Alex,
Should not. Almost original codes set sched.ready before amdgpu_ring_test_helper in each ring initialization. The new codes drop that and leave sched.ready set in amdgpu_ring_test_helper.
I tested it both on gfx10 and gfx11 chip.
Regards,
Guchun
From: Deucher, Alexander <Alexander.Deucher@xxxxxxx>
Sent: Thursday, May 18, 2023 10:47 PM
To: Chen, Guchun <Guchun.Chen@xxxxxxx>; amd-gfx@xxxxxxxxxxxxxxxxxxxxx; Zhang, Hawking <Hawking.Zhang@xxxxxxx>; Koenig, Christian <Christian.Koenig@xxxxxxx>; Pan, Xinhui <Xinhui.Pan@xxxxxxx>
Subject: Re: [PATCH 0/3] cleanup abuse of sched.ready in each ring init
[AMD Official Use Only - General]
Threre are a lot of places in the code where we check sched.ready. Will this break any of those?
This series intend to fix the abuse of flag sched.ready in
each ring init process. Its status needs to be updated only after
respective ring/IB test in init/resume.
Guchun Chen (3):
drm/amdgpu/sdma: set sched.ready status after ring/IB test in sdma
drm/amdgpu/gfx: set sched.ready status after ring/IB test in gfx
drm/amdgpu: switch to unified amdgpu_ring_test_helper
drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c | 9 ++-------
drivers/gpu/drm/amd/amdgpu/cik_sdma.c | 2 --
drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 24 ++++-------------------
drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 25 ++++--------------------
drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 2 --
drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 2 --
drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 1 -
drivers/gpu/drm/amd/amdgpu/mes_v10_1.c | 8 +-------
drivers/gpu/drm/amd/amdgpu/mes_v11_0.c | 7 +------
drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c | 2 --
drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c | 2 --
drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c | 4 ----
drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c | 4 ----
drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c | 2 --
drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c | 8 ++------
drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c | 6 +-----
drivers/gpu/drm/amd/amdgpu/si_dma.c | 2 --
17 files changed, 15 insertions(+), 95 deletions(-)
--
2.25.1
|