All of lore.kernel.org
 help / color / mirror / Atom feed
From: Pavel Begunkov <asml.silence@gmail.com>
To: io-uring@vger.kernel.org
Cc: linux-block@vger.kernel.org, Jens Axboe <axboe@kernel.dk>,
	asml.silence@gmail.com, Kanchan Joshi <joshi.k@samsung.com>,
	Ming Lei <ming.lei@redhat.com>
Subject: [PATCH v3 03/13] io_uring/cmd: fix tw <-> issue_flags conversion
Date: Mon, 18 Mar 2024 22:00:25 +0000	[thread overview]
Message-ID: <aef76d34fe9410df8ecc42a14544fd76cd9d8b9e.1710799188.git.asml.silence@gmail.com> (raw)
In-Reply-To: <cover.1710799188.git.asml.silence@gmail.com>

!IO_URING_F_UNLOCKED does not translate to availability of the deferred
completion infra, IO_URING_F_COMPLETE_DEFER does, that what we should
pass and look for to use io_req_complete_defer() and other variants.

Luckily, it's not a real problem as two wrongs actually made it right,
at least as far as io_uring_cmd_work() goes.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
---
 io_uring/uring_cmd.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/io_uring/uring_cmd.c b/io_uring/uring_cmd.c
index 7c1c58c5837e..759f919b14a9 100644
--- a/io_uring/uring_cmd.c
+++ b/io_uring/uring_cmd.c
@@ -36,7 +36,8 @@ bool io_uring_try_cancel_uring_cmd(struct io_ring_ctx *ctx,
 			/* ->sqe isn't available if no async data */
 			if (!req_has_async_data(req))
 				cmd->sqe = NULL;
-			file->f_op->uring_cmd(cmd, IO_URING_F_CANCEL);
+			file->f_op->uring_cmd(cmd, IO_URING_F_CANCEL |
+						   IO_URING_F_COMPLETE_DEFER);
 			ret = true;
 		}
 	}
@@ -86,7 +87,11 @@ EXPORT_SYMBOL_GPL(io_uring_cmd_mark_cancelable);
 static void io_uring_cmd_work(struct io_kiocb *req, struct io_tw_state *ts)
 {
 	struct io_uring_cmd *ioucmd = io_kiocb_to_cmd(req, struct io_uring_cmd);
-	unsigned issue_flags = ts->locked ? 0 : IO_URING_F_UNLOCKED;
+	unsigned issue_flags = IO_URING_F_UNLOCKED;
+
+	/* locked task_work executor checks the deffered list completion */
+	if (ts->locked)
+		issue_flags = IO_URING_F_COMPLETE_DEFER;
 
 	ioucmd->task_work_cb(ioucmd, issue_flags);
 }
@@ -130,7 +135,9 @@ void io_uring_cmd_done(struct io_uring_cmd *ioucmd, ssize_t ret, ssize_t res2,
 	if (req->ctx->flags & IORING_SETUP_IOPOLL) {
 		/* order with io_iopoll_req_issued() checking ->iopoll_complete */
 		smp_store_release(&req->iopoll_completed, 1);
-	} else if (!(issue_flags & IO_URING_F_UNLOCKED)) {
+	} else if (issue_flags & IO_URING_F_COMPLETE_DEFER) {
+		if (WARN_ON_ONCE(issue_flags & IO_URING_F_UNLOCKED))
+			return;
 		io_req_complete_defer(req);
 	} else {
 		req->io_task_work.func = io_req_task_complete;
-- 
2.44.0


  parent reply	other threads:[~2024-03-18 22:02 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-03-18 22:00 [PATCH v3 00/13] Remove aux CQE caches Pavel Begunkov
2024-03-18 22:00 ` [PATCH v3 01/13] io_uring/cmd: move io_uring_try_cancel_uring_cmd() Pavel Begunkov
2024-03-19  1:29   ` Ming Lei
2024-03-18 22:00 ` [PATCH v3 02/13] io_uring/cmd: kill one issue_flags to tw conversion Pavel Begunkov
2024-03-19  1:33   ` Ming Lei
2024-03-18 22:00 ` Pavel Begunkov [this message]
2024-03-19  1:37   ` [PATCH v3 03/13] io_uring/cmd: fix tw <-> issue_flags conversion Ming Lei
2024-03-18 22:00 ` [PATCH v3 04/13] io_uring/cmd: introduce io_uring_cmd_complete Pavel Begunkov
2024-03-19  1:39   ` Ming Lei
2024-03-18 22:00 ` [PATCH v3 05/13] nvme/io_uring: don't hard code IO_URING_F_UNLOCKED Pavel Begunkov
2024-03-19  1:40   ` Ming Lei
2024-03-18 22:00 ` [PATCH v3 06/13] io_uring/rw: avoid punting to io-wq directly Pavel Begunkov
2024-03-18 22:00 ` [PATCH v3 07/13] io_uring: force tw ctx locking Pavel Begunkov
2024-03-18 22:00 ` [PATCH v3 08/13] io_uring: remove struct io_tw_state::locked Pavel Begunkov
2024-03-18 22:00 ` [PATCH v3 09/13] io_uring: refactor io_fill_cqe_req_aux Pavel Begunkov
2024-03-18 22:00 ` [PATCH v3 10/13] io_uring: get rid of intermediate aux cqe caches Pavel Begunkov
2024-03-18 22:00 ` [PATCH v3 11/13] io_uring: remove current check from complete_post Pavel Begunkov
2024-03-18 22:00 ` [PATCH v3 12/13] io_uring: refactor io_req_complete_post() Pavel Begunkov
2024-03-18 22:00 ` [PATCH v3 13/13] io_uring: clean up io_lockdep_assert_cq_locked Pavel Begunkov
2024-03-19  1:42 ` [PATCH v3 00/13] Remove aux CQE caches Ming Lei
2024-03-19  2:19 ` Jens Axboe

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=aef76d34fe9410df8ecc42a14544fd76cd9d8b9e.1710799188.git.asml.silence@gmail.com \
    --to=asml.silence@gmail.com \
    --cc=axboe@kernel.dk \
    --cc=io-uring@vger.kernel.org \
    --cc=joshi.k@samsung.com \
    --cc=linux-block@vger.kernel.org \
    --cc=ming.lei@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.