Commit 10dcab3e authored Dec 12, 2016 by Ben Skeggs

drm/nouveau/ttm: wait for bo fence to signal before unmapping vmas

TTM was changed a while back to allow for pipelining of buffer moves, and
part of this was the removal of waiting for a BO to idle before calling
move(), placing the responsibility on the driver to do this if required.

That's all well and good, except, we make use of move_notify() to handle
mapping/unmapping from the GPU VMM as move() isn't called on all paths.

This commit adds a wait before unmapping from a VMM in move_notify(), to
prevent GPU page faults where a buffer is still being accessed.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Cc: stable@vger.kernel.org [v4.8+]

parent 732be807

drivers/gpu/drm/nouveau/nouveau_bo.c

+1 −0

Original line number	Original line	Diff line number	Diff line
	@@ -1209,6 +1209,7 @@ nouveau_bo_move_ntfy(struct ttm_buffer_object bo, struct ttm_mem_reg new_mem)
	nvbo->page_shift != vma->vm->mmu->lpg_shift)) {		nvbo->page_shift != vma->vm->mmu->lpg_shift)) {
	nvkm_vm_map(vma, new_mem->mm_node);		nvkm_vm_map(vma, new_mem->mm_node);
	} else {		} else {
			WARN_ON(ttm_bo_wait(bo, false, false));
	nvkm_vm_unmap(vma);		nvkm_vm_unmap(vma);
	}		}
	}		}