linux/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c, branch v4.20Mirror of https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/
https://git.shady.money/linux/atom?h=v4.202018-09-19T17:38:48Zdrm/amdgpu: fix up GDS/GWS/OA shifting2018-09-19T17:38:48ZChristian Königchristian.koenig@amd.com2018-09-14T14:06:31Zurn:sha1:77a2faa55c1a497f4e7e89eabd11830f0e3cb3dd
That only worked by pure coincident. Completely remove the shifting and
always apply correct PAGE_SHIFT.
Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: fix shadow BO restoring2018-09-19T17:38:41ZChristian Königchristian.koenig@amd.com2018-09-11T09:50:57Zurn:sha1:403009bfba45163887398652762ed1fc6645181c
Don't grab the reservation lock any more and simplify the handling quite
a bit.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: shadow BOs don't need any alignment2018-09-19T17:38:29ZChristian Königchristian.koenig@amd.com2018-09-11T08:31:54Zurn:sha1:7fcb0657ffa93544142f6a0bb20b7fecf1cc1dca
They aren't directly used by the hardware.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: always enable shadow BOs v22018-09-19T17:38:23ZChristian Königchristian.koenig@amd.com2018-09-11T08:30:31Zurn:sha1:16688677427687d6e3796bfa3aa72c68f71572df
Even when GPU recovery is disabled we could run into a manually
triggered recovery.
v2: keep accidental removed comments
Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Emily Deng <Emily.Deng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: stop pipelining VM PDs/PTs moves2018-09-19T17:38:17ZChristian Königchristian.koenig@amd.com2018-09-11T07:30:46Zurn:sha1:4947b2f248f867626c7a3797fc3a314bd93aeac5
We are going to need this for recoverable page fault handling and it
makes shadow handling during GPU reset much more easier.
Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Junwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amd/amdgpu: Avoid fault when allocating an empty buffer object2018-09-19T17:37:04ZTom St Denistom.stdenis@amd.com2018-09-17T18:07:00Zurn:sha1:ddaf501347e21a19d15977029af1d229cd800379
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: don't allocate zero sized kernel BOs2018-09-19T17:36:50ZChristian Königchristian.koenig@amd.com2018-09-14T19:03:37Zurn:sha1:21a7e77f77b348e73c9c67064a545e724c4a0c9a
Just free the BO if the size should be zero.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: use a single linked list for amdgpu_vm_bo_base2018-09-13T20:14:12ZChristian Königchristian.koenig@amd.com2018-09-10T18:02:46Zurn:sha1:646b90259842faa8341b076a3488a227927d84a2
Instead of the double linked list. Gets the size of amdgpu_vm_pt down to
64 bytes again.
We could even reduce it down to 32 bytes, but that would require some
rather extreme hacks.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: manually map the shadow BOs again2018-09-11T03:41:46ZChristian Königchristian.koenig@amd.com2018-08-29T12:52:50Zurn:sha1:3d5fe658b5b3f3e0d5605a540d6c1a67d9b15735
Otherwise we won't be able to use the AGP aperture.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: correctly sign extend 48bit addresses v32018-09-11T03:41:24ZChristian Königchristian.koenig@amd.com2018-08-27T16:22:31Zurn:sha1:ad9a5b78f585e9a9bd5ad06dfaf1269659a99f43
Correct sign extend the GMC addresses to 48bit.
v2: sign extending turned out easier than thought.
v3: clean up the defines and move them into amdgpu_gmc.h as well
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>