talos-obmc-linux/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c, branch dev-4.13

drm/syncobj: Rename fence_get to find_fence

2017-08-28T20:17:37+00:00

The function has far more in common with drm_syncobj_find than with any in the get/put functions. Signed-off-by: Jason Ekstrand Acked-by: Christian König (v1) Signed-off-by: Dave Airlie

drm/amdgpu: rename VM invalidated to moved

2017-08-17T19:46:08+00:00

That better describes what happens here with the BO. Signed-off-by: Christian König Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher

drm/amdgpu: separate bo_va structure

2017-08-17T19:46:07+00:00

Split that into vm_bo_base and bo_va to allow other uses as well. Signed-off-by: Christian König Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher

drm/amdgpu: cleanup static CSA handling

2017-08-17T19:46:05+00:00

Move the CSA bo_va from the VM to the fpriv structure. Signed-off-by: Christian König Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher

drm/amdgpu: move vram usage tracking into the vram manager v2

2017-08-17T19:46:03+00:00

Looks like a better place for this. v2: use atomic64_t members instead Signed-off-by: Christian König Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher

drm/amdgpu: only move VM BOs in the LRU during validation v2

2017-08-17T19:45:57+00:00

This should save us a bunch of command submission overhead. v2: move the LRU move to the right place to avoid the move for the root BO and handle the shadow BOs as well. This turned out to be a bug fix because the move needs to happen before the kmap. Signed-off-by: Christian König Reviewed-by: Felix Kuehling Acked-by: Chunming Zhou Signed-off-by: Alex Deucher

drm/amdgpu: Fix preferred typo

2017-08-15T18:46:16+00:00

Change "prefered" to "preferred" Signed-off-by: Kent Russell Reviewed-by: Christian König Signed-off-by: Alex Deucher

drm/amdgpu: switch to drm_*{get,put} helpers

2017-08-15T18:46:12+00:00

drm_*_reference() and drm_*_unreference() functions are just compatibility alias for drm_*_get() and drm_*_put() and should not be used by new code. So convert all users of compatibility functions to use the new APIs. Reviewed-by: Christian König Signed-off-by: Cihangir Akturk Signed-off-by: Alex Deucher

drm/amdgpu: consistent use u64_to_user_ptr

2017-08-15T18:45:53+00:00

Instead of open coding the conversion from u64 to pointers. Signed-off-by: Christian König Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher

drm/amdgpu: Throttle visible VRAM moves separately

2017-07-14T15:06:33+00:00

The BO move throttling code is designed to allow VRAM to fill quickly if it is relatively empty. However, this does not take into account situations where the visible VRAM is smaller than total VRAM, and total VRAM may not be close to full but the visible VRAM segment is under pressure. In such situations, visible VRAM would experience unrestricted swapping and performance would drop. Add a separate counter specifically for moves involving visible VRAM, and check it before moving BOs there. v2: Only perform calculations for separate counter if visible VRAM is smaller than total VRAM. (Michel Dänzer) v3: [Michel Dänzer] * Use BO's location rather than the AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED flag to determine whether to account a move for visible VRAM in most cases. * Use a single if (adev->mc.visible_vram_size < adev->mc.real_vram_size) { block in amdgpu_cs_get_threshold_for_moves. Fixes: 95844d20ae02 (drm/amdgpu: throttle buffer migrations at CS using a fixed MBps limit (v2)) Signed-off-by: John Brooks Reviewed-by: Christian König Signed-off-by: Michel Dänzer Signed-off-by: Alex Deucher