1io_uring_for_each_cqe(3) liburing Manual io_uring_for_each_cqe(3)
2
3
4
6 io_uring_for_each_cqe - iterate pending completion events
7
9 #include <liburing.h>
10
11 io_uring_for_each_cqe(struct io_uring *ring,
12 unsigned head,
13 struct io_uring_cqe *cqe) { }
14
16 The io_uring_for_each_cqe(3) is a macro helper that iterates completion
17 events belonging to the ring using head as a temporary iterator, and
18 points cqe to each pending event when iterating.
19
20 This helper provides an efficient way to iterate all pending events in
21 the ring, and then advancing the CQ ring by calling io_uring_cq_ad‐
22 vance(3) with the number of CQEs consumed when done. As updating the
23 kernel visible CQ ring state involves an ordered write, doing it once
24 for a number of events is more efficient than handling each completion
25 separately and calling io_uring_cqe_seen(3) for each of them.
26
27
29 void handle_cqes(struct io_uring *ring)
30 {
31 struct io_uring_cqe *cqe;
32 unsigned head;
33 unsigned i = 0;
34
35 io_uring_for_each_cqe(ring, head, cqe) {
36 /* handle completion */
37 printf("cqe: %d\n", cqe->res);
38 i++;
39 }
40
41 io_uring_cq_advance(ring, i);
42 }
43
44
46 None
47
49 io_uring_wait_cqe_timeout(3), io_uring_wait_cqe(3), io_ur‐
50 ing_wait_cqes(3), io_uring_cqe_seen(3), io_uring_buf_ring_cq_advance(3)
51
52
53
54liburing-2.4 June 04, 2023 io_uring_for_each_cqe(3)