1io_uring_buf_ring_cq_advance(3) liburing Manualio_uring_buf_ring_cq_advance(3)
2
3
4
6 io_uring_buf_ring_cq_advance - advance index of provided buffer and CQ
7 ring
8
10 #include <liburing.h>
11
12 void io_uring_buf_ring_cq_advance(struct io_uring *ring,
13 struct io_uring_buf_ring *br,
14 int count);
15
16 void __io_uring_buf_ring_cq_advance(struct io_uring *ring,
17 struct io_uring_buf_ring *br,
18 int cq_count,
19 int buf_count);
20
22 The io_uring_buf_ring_cq_advance(3) commits count previously added buf‐
23 fers to the shared buffer ring br, making them visible to the kernel
24 and hence consumable. This passes ownership of the buffer to the ring.
25 At the same time, it advances the CQ ring of ring by count amount. This
26 effectively bundles both a io_uring_buf_ring_advance(3) call and a
27 io_uring_cq_advance(3) into one operation. Since updating either ring
28 index entails a store memory barrier, doing both at once is more effi‐
29 cient.
30
31 The __io_uring_buf_ring_cq_advance(3) function performs the same opera‐
32 tion, except it splits the counts into two separate values. It advances
33 the CQ ring by cq_count entries, and the buffer ring by buf_count en‐
34 tries rather than increment both by the same value.
35
36
38 None
39
41 io_uring_register_buf_ring(3), io_uring_buf_ring_add(3), io_ur‐
42 ing_buf_ring_advance(3)
43
44
45
46liburing-2.2 May 18, 2022 io_uring_buf_ring_cq_advance(3)