From: Pavel Begunkov <asml.silence@gmail.com> mainline inclusion from mainline-v6.9-rc4 commit e5c12945be5016d681ff305ea7306fef5902219d category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/ICJPON CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i... -------------------------------- The restriction on multishot execution context disallowing io-wq is driven by rules of io_fill_cqe_req_aux(), it should only be called in the master task context, either from the syscall path or in task_work. Since task_work now always takes the ctx lock implying IO_URING_F_COMPLETE_DEFER, we can just assume that the function is always called with its defer argument set to true. Kill the argument. Also rename the function for more consistency as "fill" in CQE related functions was usually meant for raw interfaces only copying data into the CQ without any locking, waking the user and other accounting "post" functions take care of. Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Tested-by: Ming Lei <ming.lei@redhat.com> Link: https://lore.kernel.org/r/93423d106c33116c7d06bf277f651aa68b427328.171079918... Signed-off-by: Jens Axboe <axboe@kernel.dk> Conflicts: io_uring/io_uring.c io_uring/net.c io_uring/rw.c [Context differences in first two files. And in rw.c, IORING_OP_READ_MULTISHOT is not supported yet.] Signed-off-by: Yifan Qiao <qiaoyifan4@huawei.com> Signed-off-by: Long Li <leo.lilong@huawei.com> --- io_uring/io_uring.c | 15 +++------------ io_uring/io_uring.h | 2 +- io_uring/net.c | 6 ++---- io_uring/poll.c | 3 +-- io_uring/timeout.c | 2 +- 5 files changed, 8 insertions(+), 20 deletions(-) diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index 088d7cf4f77a..c3e11fcd5170 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -941,38 +941,29 @@ static void __io_flush_post_cqes(struct io_ring_ctx *ctx) state->cqes_count = 0; } -static bool __io_post_aux_cqe(struct io_ring_ctx *ctx, u64 user_data, s32 res, u32 cflags, - bool allow_overflow) +bool io_post_aux_cqe(struct io_ring_ctx *ctx, u64 user_data, s32 res, u32 cflags) { bool filled; io_cq_lock(ctx); filled = io_fill_cqe_aux(ctx, user_data, res, cflags); - if (!filled && allow_overflow) + if (!filled) filled = io_cqring_event_overflow(ctx, user_data, res, cflags, 0, 0); io_cq_unlock_post(ctx); return filled; } -bool io_post_aux_cqe(struct io_ring_ctx *ctx, u64 user_data, s32 res, u32 cflags) -{ - return __io_post_aux_cqe(ctx, user_data, res, cflags, true); -} - /* * A helper for multishot requests posting additional CQEs. * Should only be used from a task_work including IO_URING_F_MULTISHOT. */ -bool io_fill_cqe_req_aux(struct io_kiocb *req, bool defer, s32 res, u32 cflags) +bool io_req_post_cqe(struct io_kiocb *req, s32 res, u32 cflags) { struct io_ring_ctx *ctx = req->ctx; u64 user_data = req->cqe.user_data; struct io_uring_cqe *cqe; - if (!defer) - return __io_post_aux_cqe(ctx, user_data, res, cflags, false); - lockdep_assert_held(&ctx->uring_lock); if (ctx->submit_state.cqes_count == ARRAY_SIZE(ctx->completion_cqes)) { diff --git a/io_uring/io_uring.h b/io_uring/io_uring.h index 0c0dc4abd6d0..ed7d66e6786b 100644 --- a/io_uring/io_uring.h +++ b/io_uring/io_uring.h @@ -41,7 +41,7 @@ int io_run_task_work_sig(struct io_ring_ctx *ctx); void io_req_defer_failed(struct io_kiocb *req, s32 res); void io_req_complete_post(struct io_kiocb *req, unsigned issue_flags); bool io_post_aux_cqe(struct io_ring_ctx *ctx, u64 user_data, s32 res, u32 cflags); -bool io_fill_cqe_req_aux(struct io_kiocb *req, bool defer, s32 res, u32 cflags); +bool io_req_post_cqe(struct io_kiocb *req, s32 res, u32 cflags); void __io_commit_cqring_flush(struct io_ring_ctx *ctx); struct page **io_pin_pages(unsigned long ubuf, unsigned long len, int *npages); diff --git a/io_uring/net.c b/io_uring/net.c index 1a0e98e19dc0..4906cace45b0 100644 --- a/io_uring/net.c +++ b/io_uring/net.c @@ -697,8 +697,7 @@ static inline bool io_recv_finish(struct io_kiocb *req, int *ret, * Fill CQE for this receive and see if we should keep trying to * receive from this socket. */ - if (io_fill_cqe_req_aux(req, issue_flags & IO_URING_F_COMPLETE_DEFER, - *ret, cflags | IORING_CQE_F_MORE)) { + if (io_req_post_cqe(req, *ret, cflags | IORING_CQE_F_MORE)) { struct io_sr_msg *sr = io_kiocb_to_cmd(req, struct io_sr_msg); int mshot_retry_ret = IOU_ISSUE_SKIP_COMPLETE; @@ -1433,8 +1432,7 @@ int io_accept(struct io_kiocb *req, unsigned int issue_flags) if (ret < 0) return ret; - if (io_fill_cqe_req_aux(req, issue_flags & IO_URING_F_COMPLETE_DEFER, - ret, IORING_CQE_F_MORE)) + if (io_req_post_cqe(req, ret, IORING_CQE_F_MORE)) goto retry; io_req_set_res(req, ret, 0); diff --git a/io_uring/poll.c b/io_uring/poll.c index b975b05c3f4e..35c6550b3ed2 100644 --- a/io_uring/poll.c +++ b/io_uring/poll.c @@ -318,8 +318,7 @@ static int io_poll_check_events(struct io_kiocb *req, struct io_tw_state *ts) __poll_t mask = mangle_poll(req->cqe.res & req->apoll_events); - if (!io_fill_cqe_req_aux(req, true, mask, - IORING_CQE_F_MORE)) { + if (!io_req_post_cqe(req, mask, IORING_CQE_F_MORE)) { io_req_set_res(req, mask, 0); return IOU_POLL_REMOVE_POLL_USE_RES; } diff --git a/io_uring/timeout.c b/io_uring/timeout.c index 7d31b70e4d57..63c99962b9f3 100644 --- a/io_uring/timeout.c +++ b/io_uring/timeout.c @@ -72,7 +72,7 @@ static void io_timeout_complete(struct io_kiocb *req, struct io_tw_state *ts) struct io_ring_ctx *ctx = req->ctx; if (!io_timeout_finish(timeout, data)) { - if (io_fill_cqe_req_aux(req, true, -ETIME, IORING_CQE_F_MORE)) { + if (io_req_post_cqe(req, -ETIME, IORING_CQE_F_MORE)) { /* re-arm timer */ spin_lock_irq(&ctx->timeout_lock); list_add(&timeout->list, ctx->timeout_list.prev); -- 2.39.2