public inbox for [email protected]
 help / color / mirror / Atom feed
From: Ming Lei <[email protected]>
To: Pavel Begunkov <[email protected]>
Cc: Jens Axboe <[email protected]>,
	[email protected], [email protected],
	Kanchan Joshi <[email protected]>,
	[email protected]
Subject: Re: (subset) [PATCH 00/11] remove aux CQE caches
Date: Sat, 16 Mar 2024 11:54:38 +0800	[thread overview]
Message-ID: <ZfUX/kSYOW6we1SB@fedora> (raw)
In-Reply-To: <[email protected]>

On Sat, Mar 16, 2024 at 02:54:19AM +0000, Pavel Begunkov wrote:
> On 3/16/24 02:24, Ming Lei wrote:
> > On Sat, Mar 16, 2024 at 10:04 AM Ming Lei <[email protected]> wrote:
> > > 
> > > On Fri, Mar 15, 2024 at 04:53:21PM -0600, Jens Axboe wrote:
> > > > 
> > > > On Fri, 15 Mar 2024 15:29:50 +0000, Pavel Begunkov wrote:
> > > > > Patch 1 is a fix.
> > > > > 
> > > > > Patches 2-7 are cleanups mainly dealing with issue_flags conversions,
> > > > > misundertsandings of the flags and of the tw state. It'd be great to have
> > > > > even without even w/o the rest.
> > > > > 
> > > > > 8-11 mandate ctx locking for task_work and finally removes the CQE
> > > > > caches, instead we post directly into the CQ. Note that the cache is
> > > > > used by multishot auxiliary completions.
> > > > > 
> > > > > [...]
> > > > 
> > > > Applied, thanks!
> > > 
> > > Hi Jens and Pavel,
> > > 
> > > Looks this patch causes hang when running './check ublk/002' in blktests.
> > 
> > Not take close look, and  I guess it hangs in
> > 
> > io_uring_cmd_del_cancelable() -> io_ring_submit_lock
> 
> Thanks, the trace doesn't completely explains it, but my blind spot
> was io_uring_cmd_done() potentially grabbing the mutex. They're
> supposed to be irq safe mimicking io_req_task_work_add(), that's how
> nvme passthrough uses it as well (but at least it doesn't need the
> cancellation bits).
> 
> One option is to replace it with a spinlock, the other is to delay
> the io_uring_cmd_del_cancelable() call to the task_work callback.
> The latter would be cleaner and more preferable, but I'm lacking
> context to tell if that would be correct. Ming, what do you think?

I prefer to the latter approach because the two cancelable helpers are
run in fast path.

Looks all new io_uring_cmd_complete() in ublk have this issue, and the
following patch should avoid them all.

diff --git a/drivers/block/ublk_drv.c b/drivers/block/ublk_drv.c
index 97dceecadab2..1f54da0e655c 100644
--- a/drivers/block/ublk_drv.c
+++ b/drivers/block/ublk_drv.c
@@ -1417,6 +1417,12 @@ static bool ublk_abort_requests(struct ublk_device *ub, struct ublk_queue *ubq)
 	return true;
 }
 
+static void ublk_cancel_cmd_cb(struct io_uring_cmd *cmd,
+		unsigned int issue_flags)
+{
+	io_uring_cmd_done(cmd, UBLK_IO_RES_ABORT, 0, issue_flags);
+}
+
 static void ublk_cancel_cmd(struct ublk_queue *ubq, struct ublk_io *io)
 {
 	bool done;
@@ -1431,7 +1437,7 @@ static void ublk_cancel_cmd(struct ublk_queue *ubq, struct ublk_io *io)
 	spin_unlock(&ubq->cancel_lock);
 
 	if (!done)
-		io_uring_cmd_complete(io->cmd, UBLK_IO_RES_ABORT, 0);
+		io_uring_cmd_complete_in_task(io->cmd, ublk_cancel_cmd_cb);
 }
 
 /*
@@ -1775,10 +1781,9 @@ static int __ublk_ch_uring_cmd(struct io_uring_cmd *cmd,
 	return -EIOCBQUEUED;
 
  out:
-	io_uring_cmd_complete(cmd, ret, 0);
 	pr_devel("%s: complete: cmd op %d, tag %d ret %x io_flags %x\n",
 			__func__, cmd_op, tag, ret, io->flags);
-	return -EIOCBQUEUED;
+	return ret;
 }
 
 static inline struct request *__ublk_check_and_get_req(struct ublk_device *ub,
@@ -2928,10 +2933,9 @@ static int ublk_ctrl_uring_cmd(struct io_uring_cmd *cmd,
 	if (ub)
 		ublk_put_device(ub);
  out:
-	io_uring_cmd_complete(cmd, ret, 0);
 	pr_devel("%s: cmd done ret %d cmd_op %x, dev id %d qid %d\n",
 			__func__, ret, cmd->cmd_op, header->dev_id, header->queue_id);
-	return -EIOCBQUEUED;
+	return ret;
 }
 
 static const struct file_operations ublk_ctl_fops = {



Thanks,
Ming


  reply	other threads:[~2024-03-16  3:54 UTC|newest]

Thread overview: 54+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-03-15 15:29 [PATCH 00/11] remove aux CQE caches Pavel Begunkov
2024-03-15 15:29 ` [PATCH 01/11] io_uring: fix poll_remove stalled req completion Pavel Begunkov
2024-03-15 15:29 ` [PATCH 02/11] io_uring/cmd: kill one issue_flags to tw conversion Pavel Begunkov
2024-03-15 15:29 ` [PATCH 03/11] io_uring/cmd: fix tw <-> issue_flags conversion Pavel Begunkov
2024-03-15 15:29 ` [PATCH 04/11] io_uring/cmd: introduce io_uring_cmd_complete Pavel Begunkov
2024-03-15 15:29 ` [PATCH 05/11] ublk: don't hard code IO_URING_F_UNLOCKED Pavel Begunkov
2024-03-15 15:29 ` [PATCH 06/11] nvme/io_uring: " Pavel Begunkov
2024-03-15 15:29 ` [PATCH 07/11] io_uring/rw: avoid punting to io-wq directly Pavel Begunkov
2024-03-15 15:29 ` [PATCH 08/11] io_uring: force tw ctx locking Pavel Begunkov
2024-03-15 15:40   ` Jens Axboe
2024-03-15 16:14     ` Pavel Begunkov
2024-03-15 15:29 ` [PATCH 09/11] io_uring: remove struct io_tw_state::locked Pavel Begunkov
2024-03-15 15:30 ` [PATCH 10/11] io_uring: refactor io_fill_cqe_req_aux Pavel Begunkov
2024-03-15 15:30 ` [PATCH 11/11] io_uring: get rid of intermediate aux cqe caches Pavel Begunkov
2024-03-15 16:20   ` Jens Axboe
2024-03-15 16:23     ` Pavel Begunkov
2024-03-15 16:25       ` Jens Axboe
2024-03-15 16:27         ` Jens Axboe
2024-03-15 16:44           ` Pavel Begunkov
2024-03-15 16:49             ` Jens Axboe
2024-03-15 17:26               ` Pavel Begunkov
2024-03-15 18:26                 ` Jens Axboe
2024-03-15 18:51                   ` Pavel Begunkov
2024-03-15 19:02                     ` Jens Axboe
2024-03-15 16:29         ` Pavel Begunkov
2024-03-15 16:33           ` Jens Axboe
2024-03-15 15:42 ` [PATCH 00/11] remove aux CQE caches Jens Axboe
2024-03-15 16:00 ` Jens Axboe
2024-03-15 22:53 ` (subset) " Jens Axboe
2024-03-16  2:03   ` Ming Lei
2024-03-16  2:24     ` Ming Lei
2024-03-16  2:54       ` Pavel Begunkov
2024-03-16  3:54         ` Ming Lei [this message]
2024-03-16  4:13           ` Pavel Begunkov
2024-03-16  4:20             ` Pavel Begunkov
2024-03-16  9:53               ` Ming Lei
2024-03-16 11:52   ` Ming Lei
2024-03-16 13:27     ` Pavel Begunkov
2024-03-16 13:56       ` Ming Lei
2024-03-17 20:55         ` Pavel Begunkov
2024-03-17 21:24           ` Jens Axboe
2024-03-17 21:29             ` Pavel Begunkov
2024-03-17 21:32               ` Jens Axboe
2024-03-17 21:34                 ` Pavel Begunkov
2024-03-17 21:47                   ` Pavel Begunkov
2024-03-17 21:51                     ` Jens Axboe
2024-03-17 22:07                       ` Jens Axboe
2024-03-17 22:24                         ` Jens Axboe
2024-03-18  0:15                           ` Ming Lei
2024-03-18  1:34                             ` Jens Axboe
2024-03-18  1:44                               ` Jens Axboe
2024-03-18  1:49                               ` Ming Lei
2024-03-17 23:16                       ` Pavel Begunkov
2024-03-16 14:39       ` 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=ZfUX/kSYOW6we1SB@fedora \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    /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