linux-stable.git/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c, branch v4.6.2

drm/amdgpu: Revert "add mutex for ba_va->valids/invalids"

2016-03-09T18:04:02+00:00

Not needed any more because we need to protect the elements on the list anyway.

This reverts commit 38bf516c75b4ef0f5c716e05fa9baab7c52d6c39.

Signed-off-by: Christian König 
Reviewed-by: Chunming Zhou 
Acked-by: Alex Deucher 
Signed-off-by: Alex Deucher

drm/amdgpu: Revert "add lock for interval tree in vm"

2016-03-09T18:04:02+00:00

Not needed any more because we need to protect the elements on the list anyway.

This reverts commit fe237ed7efec8ac147a4572fdf81173a7f8ddda7.

Signed-off-by: Christian König 
Reviewed-by: Chunming Zhou 
Acked-by: Alex Deucher 
Signed-off-by: Alex Deucher

drm/amdgpu: Revert "add spin lock to protect freed list in vm (v3)"

2016-03-09T18:04:01+00:00

Not needed any more because we need to protect the elements on the list anyway.

This reverts commit dae6ecf9e6c9b677e577826c3ac665c6dd9c490b.

Signed-off-by: Christian König 
Reviewed-by: Chunming Zhou 
Acked-by: Alex Deucher 
Signed-off-by: Alex Deucher

drm/amdgpu: Fix two bugs in amdgpu_vm_bo_split_mapping

2016-03-09T18:04:00+00:00

Off-by-one: last is inclusive, so the maximum is start + max_size - 1
Wrong unit: addr is in bytes, max_size is in pages

Signed-off-by: Felix Kuehling 
Reviewed-by: Christian König 
Signed-off-by: Alex Deucher

drm/amdgpu: move get_user_pages out of amdgpu_ttm_tt_pin_userptr v6

2016-03-08T16:01:50+00:00

That avoids lock inversion between the BO reservation lock
and the anon_vma lock.

v2:
* Changed amdgpu_bo_list_entry.user_pages to an array of pointers
* Lock mmap_sem only for get_user_pages
* Added invalidation of unbound userpointer BOs
* Fixed memory leak and page reference leak

v3 (chk):
* Revert locking mmap_sem only for_get user_pages
* Revert adding invalidation of unbound userpointer BOs
* Sanitize and fix error handling

v4 (chk):
* Init userpages pointer everywhere.
* Fix error handling when get_user_pages() fails.
* Add invalidation of unbound userpointer BOs again.

v5 (chk):
* Add maximum number of tries.

v6 (chk):
* Fix error handling when we run out of tries.

Signed-off-by: Christian König 
Signed-off-by: Felix Kuehling 
Reviewed-by: Felix Kuehling  (v4)
Acked-by: Alex Deucher

drm/amdgpu: if a GDS switch is needed emit a pipeline sync as well

2016-03-08T16:01:50+00:00

Otherwise we might change the GDS settings while they are still in use.

Signed-off-by: Christian König 
Reviewed-by: Alex Deucher 
Reviewed-by: Chunming Zhou

drm/amdgpu: split pipeline sync and vm flush

2016-03-08T16:01:49+00:00

This allows us to use the pipeline sync for other tasks as well.

Signed-off-by: Christian König 
Reviewed-by: Alex Deucher 
Reviewed-by: Chunming Zhou

drm/amdgpu: switch the GDS only on demand v2

2016-03-08T16:01:49+00:00

Switching the GDS space to often seems to be problematic.

This patch together with the following can avoid VM faults on context switch.

v2: extend commit message a bit

Signed-off-by: Christian König 
Reviewed-by: Alex Deucher  (v1)
Reviewed-by: Chunming Zhou  (v1)

drm/amdgpu: move the GDS switch into vm flush as well

2016-03-08T16:01:48+00:00

After all it's an operation on the VMID.

Signed-off-by: Christian König 
Reviewed-by: Alex Deucher 
Reviewed-by: Chunming Zhou

drm/amdgpu: sync to the active user on reusing a VMID

2016-03-08T16:01:45+00:00

Signed-off-by: Christian König 
Reviewed-by: Chunming Zhou 
Signed-off-by: Alex Deucher