Am 11.01.19 um 15:17 schrieb Michel Dänzer:
On 2019-01-11 2:15 p.m., Christian König wrote:Move the BO on the LRU only when it is actually moved by a DMA operation. [...] @@ -177,6 +175,13 @@ static vm_fault_t ttm_bo_vm_fault(struct vm_fault *vmf) ret = VM_FAULT_SIGBUS; goto out_unlock; } + + if (bo->moving != moving) {Hmm, could a driver just update the existing fence instead of attaching a new one?
Mhm, not as far as I know. That would violate similar checks elsewhere. Christian. _______________________________________________ amd-gfx mailing list amd-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/amd-gfx
