drm/amdgpu: fix calltrace during kmd unload(v3)
authorMonk Liu <Monk.Liu@amd.com>
Tue, 26 Nov 2019 11:42:25 +0000 (19:42 +0800)
committerAlex Deucher <alexander.deucher@amd.com>
Tue, 3 Dec 2019 16:57:00 +0000 (11:57 -0500)
commit747d4f715fb5aca0002216355df28714cc20250c
treec0dc7daace2c6f585ccdc5611f0f347441ff46fd
parentfa2b93e39b1d167d342aecd6d3f53d9972405226
drm/amdgpu: fix calltrace during kmd unload(v3)

issue:
kernel would report a warning from a double unpin
during the driver unloading on the CSB bo

why:
we unpin it during hw_fini, and there will be another
unpin in sw_fini on CSB bo.

fix:
actually we don't need to pin/unpin it during
hw_init/fini since it is created with kernel pinned,
we only need to fullfill the CSB again during hw_init
to prevent CSB/VRAM lost after S3

v2:
get_csb in init_rlc so hw_init() will make CSIB content
back even after reset or s3

v3:
use bo_create_kernel instead of bo_create_reserved for CSB
otherwise the bo_free_kernel() on CSB is not aligned and
would lead to its internal reserve pending there forever

take care of gfx7/8 as well

Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Xiaojie Yuan <xiaojie.yuan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c
drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c