Lines Matching refs:compute
230 * Long-running compute command buffers, while still using traditional end of
232 fences which get reattached when the compute job is rescheduled.
279 userspace is allowed to use userspace fencing or long running compute
302 faults on GPUs are limited to pure compute workloads.
305 compute side, like compute units or command submission engines. If both a 3D
306 job with a DMA fence and a compute workload using recoverable page faults are
309 - The 3D workload might need to wait for the compute job to finish and release
312 - The compute workload might be stuck in a page fault, because the memory
323 achieved through e.g. through dedicated engines and minimal compute unit
334 fences must complete before a compute job with page fault handling can be
336 made visible anywhere in the system, all compute workloads must be preempted
353 Fences` discussions: Infinite fences from compute workloads are allowed to