All of lore.kernel.org
 help / color / mirror / Atom feed
From: Eric Wheeler <linux-block@lists.ewheeler.net>
To: linux-block@vger.kernel.org
Subject: loop: it looks like REQ_OP_FLUSH could return before IO completion.
Date: Sat, 19 Mar 2022 10:14:29 -0700 (PDT)	[thread overview]
Message-ID: <af3e552a-6c77-b295-19e1-d7a1e39b31f3@ewheeler.net> (raw)

Hello all,

In loop.c do_req_filebacked() for REQ_OP_FLUSH, lo_req_flush() is called: 
it does not appear that lo_req_flush() does anything to make sure 
ki_complete has been called for pending work, it just calls vfs_fsync().

Is this a consistency problem?

For example, if the loop driver has queued async kiocb's to the filesystem 
via .read_iter/.write_iter, is it the filesystem's responsibility to 
complete that work before returning from vfs_sync() or is it possible that 
the vfs_sync() completes before ->ki_complete() is called?

Relatedly, does vfs_fsync() do anything for direct IO?  Ultimately 
f_op->fsync() is called so maybe the FS is told to commit its structures 
like sparse allocations that may not be on disk yet.


--
Eric Wheeler

             reply	other threads:[~2022-03-19 17:14 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-03-19 17:14 Eric Wheeler [this message]
2022-03-21  0:21 ` loop: it looks like REQ_OP_FLUSH could return before IO completion Ming Lei
2022-03-23  6:14   ` Eric Wheeler
2022-04-13 22:49     ` Eric Wheeler
2022-04-14  0:15       ` Ming Lei
2022-04-14  0:28       ` Ming Lei
2022-04-15  2:10         ` Eric Wheeler
2022-04-15 14:29           ` Ming Lei
2022-04-16  5:18             ` Christoph Hellwig
2022-04-16 20:05               ` Eric Wheeler
2022-04-17  0:59                 ` Jens Axboe
2022-04-17 16:32                   ` Eric Wheeler

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=af3e552a-6c77-b295-19e1-d7a1e39b31f3@ewheeler.net \
    --to=linux-block@lists.ewheeler.net \
    --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 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.