FAILED: patch "[PATCH] io_uring/rw: end write accounting from ->ki_complete" failed to apply to 5.15-stable tree

The patch below does not apply to the 5.15-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.

To reproduce the conflict and resubmit, you may use the following commands:

git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.15.y
git checkout FETCH_HEAD
git cherry-pick -x 796aa0547557e63338657ed1c487906f9fac4c73
# <resolve conflicts, build, test, etc.>
git commit -s
git send-email --to '<[email protected]>' --in-reply-to '2026091626-pacify-aluminum-bcbd@gregkh' --subject-prefix 'PATCH 5.15.y' 'HEAD^..'

Possible dependencies:

thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From 796aa0547557e63338657ed1c487906f9fac4c73 Mon Sep 17 00:00:00 2001
From: Jens Axboe <[email protected]>
Date: Tue, 8 Sep 2026 17:53:20 -0600
Subject: [PATCH] io_uring/rw: end write accounting from ->ki_complete

Commit b000145e9907 moved both the fsnotify calls and the write accounting out of the kiocb completion handler and into the io_req_rw_complete() task_work. However, only the fsnotify part actually needed to move as it may sleep. Ending the write accounting is just a percpu_up_read() on the superblock writers sem.

Deferring it is a problem, because it makes dropping SB_FREEZE_WRITE protection depend on the ring owner getting to running task_work. But the task may be blocked in freeze_super(), causing it to never get to that:

task io-wq worker
--------------------------------------------------------------
io_write()
io_kiocb_start_write() (takes sb_writers, hidden from
lockdep by __sb_writers_release)
write_iter() -> -EIOCBQUEUED
ioctl(FS_IOC_SHUTDOWN)
bdev_freeze()
freeze_super()
percpu_down_write() <- waits for the reader above
io_write()
kiocb_start_write()
percpu_down_read() <- queued
behind the writer
<bio completes>
io_complete_rw()
queues io_req_rw_complete() <- never runs, task is in D state

End the write from io_complete_rw() instead, and leave only the fsnotify calls in task_work.

index 95106dd1d7eb..3e22f294bdf2 100644
@@ -517,20 +517,25 @@ static void io_req_end_write(struct io_kiocb *req)
 	}
 }
 
-/*
- * Trigger the notifications after having done some IO, and finish the write
- * accounting, if any.
- */
+/* Trigger the notifications after having done some IO. */
+static void io_req_io_notify(struct io_kiocb *req)
+{
+	struct io_rw *rw = io_kiocb_to_cmd(req, struct io_rw);
+
+	if (rw->kiocb.ki_flags & IOCB_WRITE)
+		fsnotify_modify(req->file);
+	else
+		fsnotify_access(req->file);
+}
+
+/* Finish write accounting and notify, for inline completions only. */
 static void io_req_io_end(struct io_kiocb *req)
 {
 	struct io_rw *rw = io_kiocb_to_cmd(req, struct io_rw);
 
-	if (rw->kiocb.ki_flags & IOCB_WRITE) {
+	if (rw->kiocb.ki_flags & IOCB_WRITE)
 		io_req_end_write(req);
-		fsnotify_modify(req->file);
-	} else {
-		fsnotify_access(req->file);
-	}
+	io_req_io_notify(req);
 }
 
 static void __io_complete_rw_common(struct io_kiocb *req, long res)

@@ -563,7 +568,7 @@ void io_req_rw_complete(struct io_tw_req tw_req, io_tw_token_t tw)
 {
 	struct io_kiocb *req = tw_req.req;
 
-	io_req_io_end(req);
+	io_req_io_notify(req);
 
 	if (req->flags & (REQ_F_BUFFER_SELECTED|REQ_F_BUFFER_RING))
 		req->cqe.flags |= io_put_kbuf(req, max(req->cqe.res, 0), NULL);

@@ -577,6 +582,10 @@ static void io_complete_rw(struct kiocb *kiocb, long res)
 	struct io_rw *rw = container_of(kiocb, struct io_rw, kiocb);
 	struct io_kiocb *req = cmd_to_io_kiocb(rw);
 
+	/* ring owner may block in freeze_super() before task_work runs */
+	if (kiocb->ki_flags & IOCB_WRITE)
+		io_req_end_write(req);
+
 	__io_complete_rw_common(req, res);
 	io_req_set_res(req, io_fixup_rw_res(req, res), 0);
 	req->io_task_work.func = io_req_rw_complete;

From: Jens Axboe <[email protected]>

[ Upstream commit 796aa0547557e63338657ed1c487906f9fac4c73 ]

Commit b000145e9907 moved both the fsnotify calls and the write accounting out of the kiocb completion handler and into the io_req_rw_complete() task_work. However, only the fsnotify part actually needed to move as it may sleep. Ending the write accounting is just a percpu_up_read() on the superblock writers sem.

Deferring it is a problem, because it makes dropping SB_FREEZE_WRITE protection depend on the ring owner getting to running task_work. But the task may be blocked in freeze_super(), causing it to never get to that:

task io-wq worker
--------------------------------------------------------------
io_write()
io_kiocb_start_write() (takes sb_writers, hidden from
lockdep by __sb_writers_release)
write_iter() -> -EIOCBQUEUED
ioctl(FS_IOC_SHUTDOWN)
bdev_freeze()
freeze_super()
percpu_down_write() <- waits for the reader above
io_write()
kiocb_start_write()
percpu_down_read() <- queued
behind the writer
<bio completes>
io_complete_rw()
queues io_req_rw_complete() <- never runs, task is in D state

End the write from io_complete_rw() instead, and leave only the fsnotify calls in task_work.

---
 io_uring/io_uring.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

index 8747c585a0770..8d5791d8f4031 100644
@@ -2797,7 +2797,10 @@ static void io_req_task_complete(struct io_kiocb *req, bool *locked)
 
 static void io_req_rw_complete(struct io_kiocb *req, bool *locked)
 {
-	io_req_io_end(req);
+	if (req->rw.kiocb.ki_flags & IOCB_WRITE)
+		fsnotify_modify(req->file);
+	else
+		fsnotify_access(req->file);
 	io_req_task_complete(req, locked);
 }
 

@@ -2807,6 +2810,10 @@ static void io_complete_rw(struct kiocb *kiocb, long res, long res2)
 
 	if (__io_complete_rw_common(req, res))
 		return;
+	/* ring owner may block in freeze_super() before task_work runs */
+	if (kiocb->ki_flags & IOCB_WRITE)
+		io_req_end_write(req);
+
 	req->result = io_fixup_rw_res(req, res);
 	req->io_task_work.func = io_req_rw_complete;
 	io_req_task_work_add(req);

2.53.0