6.1-stable review patch. If anyone has any objections, please let me know. ------------------ From: Pavel Begunkov <asml.silence@xxxxxxxxx> commit 3181e22fb79910c7071e84a43af93ac89e8a7106 upstream. Flush completions is done either from the submit syscall or by the task_work, both are in the context of the submitter task, and when it goes for a single threaded rings like implied by ->task_complete, there won't be any waiters on ->cq_wait but the master task. That means that there can be no tasks sleeping on cq_wait while we run __io_submit_flush_completions() and so waking up can be skipped. Signed-off-by: Pavel Begunkov <asml.silence@xxxxxxxxx> Link: https://lore.kernel.org/r/60ad9768ec74435a0ddaa6eec0ffa7729474f69f.1673274244.git.asml.silence@xxxxxxxxx Signed-off-by: Jens Axboe <axboe@xxxxxxxxx> Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> --- io_uring/io_uring.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -582,6 +582,16 @@ static inline void __io_cq_unlock_post(s io_cqring_ev_posted(ctx); } +static inline void __io_cq_unlock_post_flush(struct io_ring_ctx *ctx) + __releases(ctx->completion_lock) +{ + io_commit_cqring(ctx); + spin_unlock(&ctx->completion_lock); + io_commit_cqring_flush(ctx); + if (!(ctx->flags & IORING_SETUP_DEFER_TASKRUN)) + __io_cqring_wake(ctx); +} + void io_cq_unlock_post(struct io_ring_ctx *ctx) { __io_cq_unlock_post(ctx); @@ -1339,7 +1349,7 @@ static void __io_submit_flush_completion if (!(req->flags & REQ_F_CQE_SKIP)) __io_fill_cqe_req(ctx, req); } - __io_cq_unlock_post(ctx); + __io_cq_unlock_post_flush(ctx); io_free_batch_list(ctx, state->compl_reqs.first); INIT_WQ_LIST(&state->compl_reqs);
