kernel/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c, branch linux-4.16.y

drm/amdgpu: fix another potential cause of VM faults

2018-01-30T04:15:26Z

The root PD can be evicted directly after allocating it, just validate it on first use. Signed-off-by: Christian König Acked-by: Felix Kuehling Signed-off-by: Alex Deucher

drm/amdgpu: fix CPU based VM updates

2018-01-16T20:35:27Z

That got accidentially removed. Signed-off-by: Christian König Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher

drm/amdgpu: fix amdgpu_vm_pasid_fault_credit

2018-01-16T20:35:26Z

As soon as the lock is dropped the VM pointer can be invalid. Signed-off-by: Christian König Reviewed-by: Chunming Zhou Signed-off-by: Alex Deucher

drm/amdgpu: optimize moved handling only when vm_debug is inactive

2018-01-10T20:44:54Z

Otherwise we would completely circumvent that debugging feature. Signed-off-by: Christian König Reviewed-by: Roger He Signed-off-by: Alex Deucher

drm/amdgpu: simplify huge page handling

2018-01-10T20:44:53Z

Update the PDEs after resetting the huge flag. Signed-off-by: Christian König Reviewed-by: Chunming Zhou Acked-by: Felix Kuehling Signed-off-by: Alex Deucher

drm/amdgpu: minor optimize VM moved handling v2

2018-01-10T20:44:53Z

Try to lock moved BOs if it's successful we can update the PTEs directly to the new location. v2: rebase Signed-off-by: Christian König Acked-by: Felix Kuehling Signed-off-by: Alex Deucher

drm/amdgpu: loosen the criteria for huge pages a bit

2018-01-10T20:44:52Z

We can actually handle invalid huge pages perfectly fine now. Signed-off-by: Christian König Reviewed-by: Chunming Zhou Acked-by: Felix Kuehling Signed-off-by: Alex Deucher

drm/amdgpu: Correct the IB size of bo update mapping.

2017-12-31T03:24:20Z

The amdgpu_vm_frag_ptes will call amdgpu_vm_update_ptes, and for buffer object that has shadow buffer, need twice commands. Signed-off-by: Emily Deng Reviewed-by: Chunming Zhou Signed-off-by: Alex Deucher

drm/amdgpu: fix VM faults with per VM BOs

2017-12-27T16:34:31Z

There was a small window between unreserve and second reserve where the freshly allocated BO could have been evicted without the VM noticing it. Signed-off-by: Christian König Reviewed-by: Chunming Zhou Signed-off-by: Alex Deucher

drm/amdgpu: drop client_id from VM

2017-12-27T16:34:09Z

Use the fence context from the scheduler entity. Signed-off-by: Christian König Reviewed-by: Chunming Zhou Signed-off-by: Alex Deucher