There is no reason not to use __io_cq_unlock_post_flush for intermediate
aux CQE flushing, all ->task_complete should apply there, i.e. if set it
should be the submitter task. Combine them, get rid of of
__io_cq_unlock_post() and rename the left function.
This place was also taking a couple percents of CPU according to
profiles for max throughput net benchmarks due to multishot recv
flooding it with completions.
Signed-off-by: Pavel Begunkov <asml.silence@xxxxxxxxx>
---
io_uring/io_uring.c | 13 +------------
1 file changed, 1 insertion(+), 12 deletions(-)
diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
index 70fffed83e95..1b53a2ab0a27 100644
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -637,18 +637,7 @@ static inline void io_cq_lock(struct io_ring_ctx *ctx)
spin_lock(&ctx->completion_lock);
}
-/* keep it inlined for io_submit_flush_completions() */
static inline void __io_cq_unlock_post(struct io_ring_ctx *ctx)
-{
- io_commit_cqring(ctx);
- if (!ctx->task_complete)
- spin_unlock(&ctx->completion_lock);
-
- io_commit_cqring_flush(ctx);
- io_cqring_wake(ctx);
-}
-
-static void __io_cq_unlock_post_flush(struct io_ring_ctx *ctx)
{
io_commit_cqring(ctx);
@@ -1568,7 +1557,7 @@ static void __io_submit_flush_completions(struct io_ring_ctx *ctx)
}
}
}
- __io_cq_unlock_post_flush(ctx);
+ __io_cq_unlock_post(ctx);
if (!wq_list_empty(&ctx->submit_state.compl_reqs)) {
io_free_batch_list(ctx, state->compl_reqs.first);
--
2.40.0
