Lines Matching defs:completion
3 * Shared application/kernel submission and completion ring pairs, for
174 * Number of completion events lost because the queue was full;
177 * the completion queue.
183 * As completion events come in out of order this counter is not
188 * Ring buffer of completion events.
190 * The kernel writes completion events fresh every time they are
254 struct completion done;
295 struct completion exited;
320 * Batch completion logic
324 /* inline/task_work completion list, under ->uring_lock */
386 /* IRQ completion list, under ->completion_lock */
458 struct completion ref_comp;
762 /* doesn't sever on completion < 0 */
785 /* completion is deferred through io_comp_state */
1192 static void io_refs_resurrect(struct percpu_ref *ref, struct completion *compl)
1978 * handlers and io_issue_sqe() are done with it, e.g. inline completion path.
3544 * will register a callback when the page is unlocked at IO completion. Through
4080 * IORING_OP_NOP just posts a completion event, nothing else.
7028 * race with the completion of the linked work.
9471 struct completion completion;
9489 complete(&work->completion);
9545 init_completion(&exit.completion);
9573 wait_for_completion_interruptible(&exit.completion);
9913 * avoids a race where a completion comes in before we did
10134 * space applications don't need to do io completion events