Linux block layer
 help / color / mirror / Atom feed
From: Ming Lei <tom.leiming@gmail.com>
To: linux-block@vger.kernel.org
Cc: Ming Lei <tom.leiming@gmail.com>, Jens Axboe <axboe@kernel.dk>,
	Caleb Sander Mateos <csander@purestorage.com>,
	Josef Bacik <josef@toxicpanda.com>
Subject: [PATCH 2/8] ublk: mark the io command cancelable before publishing it
Date: Thu,  1 Oct 2026 07:54:16 -0500	[thread overview]
Message-ID: <20261001125422.1364260-3-tom.leiming@gmail.com> (raw)
In-Reply-To: <20261001125422.1364260-1-tom.leiming@gmail.com>

FETCH, COMMIT_AND_FETCH and NEED_GET_DATA publish the command in io->cmd
before marking it cancelable. A cancel from the control path (STOP_DEV,
QUIESCE_DEV) can complete it in between:

  issue path                       control-path cancel
  io->cmd = C
                                   take C, io_uring_cmd_done(C):
                                     C not marked, nothing to unlink
  ublk_prep_cancel(C)
    the completed C is linked on the cancelable list
  ring exit: the cancel walk hits it → GPF (KASAN, with a delay added)

Mark the command before ublk_fill_io_cmd() publishes it. The handlers
run with uring_lock held, as io_uring's cancel walk does, so marking
takes no lock and the walk can't see a marked command before it is
published. smp_wmb() orders the mark before the io->cmd store; the
cancel loads cmd with READ_ONCE(io->cmd) before reading cmd->flags.

A marked command has to be completed by io_uring_cmd_done(), so a failed
FETCH and the inline UBLK_IO_RES_OK of NEED_GET_DATA now do that. The CQE
is the same.

Fixes: 216c8f5ef0f2 ("ublk: replace monitor with cancelable uring_cmd")
Cc: stable@vger.kernel.org
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
---
 drivers/block/ublk_drv.c | 27 +++++++++++++++++++++------
 1 file changed, 21 insertions(+), 6 deletions(-)

diff --git a/drivers/block/ublk_drv.c b/drivers/block/ublk_drv.c
index 38ed7d0e3979..6015fb2fb925 100644
--- a/drivers/block/ublk_drv.c
+++ b/drivers/block/ublk_drv.c
@@ -2800,7 +2800,8 @@ static void ublk_cancel_cmd(struct ublk_queue *ubq, u16 tag,
 	done = !!(io->flags & UBLK_IO_FLAG_CANCELED);
 	if (!done) {
 		io->flags |= UBLK_IO_FLAG_CANCELED;
-		cmd = io->cmd;
+		/* dependency ordered against smp_wmb() in ublk_prep_cancel() */
+		cmd = READ_ONCE(io->cmd);
 		io->cmd = NULL;
 	}
 	spin_unlock(&ubq->cancel_lock);
@@ -3163,6 +3164,12 @@ ublk_fill_io_cmd(struct ublk_io *io, struct io_uring_cmd *cmd)
 	return req;
 }
 
+/*
+ * Call before ublk_fill_io_cmd() publishes @cmd in io->cmd: a control-path
+ * cancel may complete any command found there, and io_uring_cmd_done() only
+ * takes it off the cancelable list if it is marked already. The handlers
+ * hold uring_lock, so marking takes no lock.
+ */
 static inline void ublk_prep_cancel(struct io_uring_cmd *cmd,
 				    unsigned int issue_flags,
 				    struct ublk_queue *ubq, u16 tag)
@@ -3176,6 +3183,8 @@ static inline void ublk_prep_cancel(struct io_uring_cmd *cmd,
 	pdu->ubq = ubq;
 	pdu->tag = tag;
 	io_uring_cmd_mark_cancelable(cmd, issue_flags);
+	/* pairs with the cancel loading cmd from io->cmd, then cmd->flags */
+	smp_wmb();
 }
 
 static void ublk_io_release(void *priv)
@@ -3423,11 +3432,11 @@ static int ublk_ch_uring_cmd_local(struct io_uring_cmd *cmd,
 		ret = ublk_check_fetch_buf(ub, addr);
 		if (ret)
 			goto out;
+		/* before ublk_fetch() publishes io->cmd, see ublk_prep_cancel() */
+		ublk_prep_cancel(cmd, issue_flags, ubq, tag);
 		ret = ublk_fetch(cmd, ub, io, addr, q_id);
 		if (ret)
-			goto out;
-
-		ublk_prep_cancel(cmd, issue_flags, ubq, tag);
+			goto out_done;
 		return -EIOCBQUEUED;
 	}
 
@@ -3471,6 +3480,7 @@ static int ublk_ch_uring_cmd_local(struct io_uring_cmd *cmd,
 		if (ret)
 			goto out;
 		io->res = result;
+		ublk_prep_cancel(cmd, issue_flags, ubq, tag);
 		req = ublk_fill_io_cmd(io, cmd);
 		ublk_apply_io_buf(ub, io, cmd, addr, &auto_buf, &buf_idx);
 		if (buf_idx != UBLK_INVALID_BUF_IDX)
@@ -3489,19 +3499,24 @@ static int ublk_ch_uring_cmd_local(struct io_uring_cmd *cmd,
 		 * uring_cmd active first and prepare for handling new requeued
 		 * request
 		 */
+		ublk_prep_cancel(cmd, issue_flags, ubq, tag);
 		req = ublk_fill_io_cmd(io, cmd);
 		io->buf.addr = addr;
 		if (likely(ublk_get_data(ubq, io, req))) {
 			__ublk_prep_compl_io_cmd(io, req);
-			return UBLK_IO_RES_OK;
+			ret = UBLK_IO_RES_OK;
+			goto out_done;
 		}
 		break;
 	default:
 		goto out;
 	}
-	ublk_prep_cancel(cmd, issue_flags, ubq, tag);
 	return -EIOCBQUEUED;
 
+ out_done:
+	/* marked cancelable: complete through io_uring_cmd_done() */
+	io_uring_cmd_done(cmd, ret, issue_flags);
+	return -EIOCBQUEUED;
  out:
 	pr_devel("%s: complete: cmd op %d, tag %d ret %x io_flags %x\n",
 			__func__, cmd_op, tag, ret, io ? io->flags : 0);
-- 
2.55.0


  parent reply	other threads:[~2026-10-01 12:54 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-01 12:54 [PATCH 0/8] ublk: don't dispatch to canceled io commands Ming Lei
2026-10-01 12:54 ` [PATCH 1/8] ublk: keep a canceled FETCH round canceling until the server is gone Ming Lei
2026-10-01 12:54 ` Ming Lei [this message]
2026-10-01 12:54 ` [PATCH 3/8] ublk: mark the batch fetch command cancelable before linking it Ming Lei
2026-10-01 12:54 ` [PATCH 4/8] ublk: reset the FETCH round in release also without a disk Ming Lei
2026-10-01 12:54 ` [PATCH 5/8] ublk: reset the FETCH round under ub->mutex Ming Lei
2026-10-01 12:54 ` [PATCH 6/8] ublk: let STOP_DEV cancel the server's commands before its release Ming Lei
2026-10-01 12:54 ` [PATCH 7/8] selftests: ublk: move the control command helpers into ctrl.c Ming Lei
2026-10-01 12:54 ` [PATCH 8/8] selftests: ublk: add test for going live over canceled io commands Ming Lei
2026-10-05 16:23 ` [PATCH] ublk: refuse to go live after an io command was canceled Josef Bacik
2026-10-06 14:14   ` Ming Lei
2026-10-05 16:23     ` [PATCH v2] " Josef Bacik
2026-10-05 18:50 ` [PATCH 0/8] ublk: don't dispatch to canceled io commands Josef Bacik
2026-10-06 16:10 ` [PATCH 0/4] ublk: fix UBLK_CMD_QUIESCE_DEV leaving commands behind Josef Bacik
2026-10-06 13:05   ` [PATCH 1/4] ublk: don't cancel commands in QUIESCE_DEV on a device that isn't live Josef Bacik
2026-10-06 14:49   ` [PATCH 2/4] ublk: drop QUIESCE_DEV's wait for an idle command Josef Bacik
2026-10-06 14:50   ` [PATCH 3/4] ublk: give the command back from COMMIT_AND_FETCH on a canceling queue Josef Bacik
2026-10-06 14:50   ` [PATCH 4/4] ublk: keep canceling in QUIESCE_DEV until the server's commands are taken Josef Bacik

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=20261001125422.1364260-3-tom.leiming@gmail.com \
    --to=tom.leiming@gmail.com \
    --cc=axboe@kernel.dk \
    --cc=csander@purestorage.com \
    --cc=josef@toxicpanda.com \
    --cc=linux-block@vger.kernel.org \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox