Lines Matching refs:fences

139    :doc: DMA fences overview
192 * Future fences, used in HWC1 to signal when a buffer isn't used by the display
196 * Proxy fences, proposed to handle &drm_syncobj for which the fence has not yet
199 * Userspace fences or gpu futexes, fine-grained locking within a command buffer
205   batch DMA fences for memory management instead of context preemption DMA
206   fences which get reattached when the compute job is rescheduled.
209 fences and controls when they fire. Mixing indefinite fences with normal
210 in-kernel DMA fences does not work, even when a fallback timeout is included to
216 * Only userspace knows about all dependencies in indefinite fences and when
220 for memory management needs, which means we must support indefinite fences being
221 dependent upon DMA fences. If the kernel also support indefinite fences in the
232       userspace [label="userspace controlled fences"]
247 fences in the kernel. This means:
249 * No future fences, proxy fences or userspace fences imported as DMA fences,
252 * No DMA fences that signal end of batchbuffer for command submission where