linux-xfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: hch <hch@lst.de>
To: Johannes Thumshirn <Johannes.Thumshirn@wdc.com>
Cc: "Darrick J. Wong" <djwong@kernel.org>, hch <hch@lst.de>,
	Jan Kara <jack@suse.cz>, Keith Busch <kbusch@kernel.org>,
	Dave Chinner <david@fromorbit.com>,
	Carlos Maiolino <cem@kernel.org>,
	Christian Brauner <brauner@kernel.org>,
	"Martin K. Petersen" <martin.petersen@oracle.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"linux-xfs@vger.kernel.org" <linux-xfs@vger.kernel.org>,
	"linux-fsdevel@vger.kernel.org" <linux-fsdevel@vger.kernel.org>,
	"linux-raid@vger.kernel.org" <linux-raid@vger.kernel.org>,
	"linux-block@vger.kernel.org" <linux-block@vger.kernel.org>
Subject: Re: fall back from direct to buffered I/O when stable writes are required
Date: Thu, 6 Nov 2025 13:49:00 +0100	[thread overview]
Message-ID: <20251106124900.GA6144@lst.de> (raw)
In-Reply-To: <9530fca4-418d-4415-b365-cad04a06449b@wdc.com>

On Thu, Nov 06, 2025 at 09:50:10AM +0000, Johannes Thumshirn wrote:
> On 11/5/25 10:44 PM, Darrick J. Wong wrote:
> > Just out of curiosity -- is qemu itself mutating the buffers that it is
> > passing down to the lower levels via dio?  Or is it a program in the
> > guest that's mutating buffers that are submitted for dio, which then get
> > zerocopied all the way down to the hypervisor?
> 
> If my memory serves me right it is the guest (or at least can be). I 
> remember a bug report on btrfs where a Windows guest had messed up 
> checksums because of modifying inflight I/O.

qemu passes I/O through, so yes it is guest controller.  Windows is most
famous, but the Linux swap code can trigger it easily too.


  reply	other threads:[~2025-11-06 12:49 UTC|newest]

Thread overview: 61+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-10-29  7:15 fall back from direct to buffered I/O when stable writes are required Christoph Hellwig
2025-10-29  7:15 ` [PATCH 1/4] fs: replace FOP_DIO_PARALLEL_WRITE with a fmode bits Christoph Hellwig
2025-10-29 16:01   ` Darrick J. Wong
2025-11-04  7:00   ` Nirjhar Roy (IBM)
2025-11-05 14:04     ` Christoph Hellwig
2025-11-11  9:44   ` Christian Brauner
2025-10-29  7:15 ` [PATCH 2/4] fs: return writeback errors for IOCB_DONTCACHE in generic_write_sync Christoph Hellwig
2025-10-29 16:01   ` Darrick J. Wong
2025-10-29 16:37     ` Christoph Hellwig
2025-10-29 18:12       ` Darrick J. Wong
2025-10-30  5:59         ` Christoph Hellwig
2025-11-04 12:04       ` Nirjhar Roy (IBM)
2025-11-04 15:53         ` Christoph Hellwig
2025-10-29  7:15 ` [PATCH 3/4] xfs: use IOCB_DONTCACHE when falling back to buffered writes Christoph Hellwig
2025-10-29 15:57   ` Darrick J. Wong
2025-11-04 12:33   ` Nirjhar Roy (IBM)
2025-11-04 15:52     ` Christoph Hellwig
2025-10-29  7:15 ` [PATCH 4/4] xfs: fallback to buffered I/O for direct I/O when stable writes are required Christoph Hellwig
2025-10-29 15:53   ` Darrick J. Wong
2025-10-29 16:35     ` Christoph Hellwig
2025-10-29 21:23       ` Qu Wenruo
2025-10-30  5:58         ` Christoph Hellwig
2025-10-30  6:37           ` Qu Wenruo
2025-10-30  6:49             ` Christoph Hellwig
2025-10-30  6:53               ` Qu Wenruo
2025-10-30  6:55                 ` Christoph Hellwig
2025-10-30  7:14                   ` Qu Wenruo
2025-10-30  7:17                     ` Christoph Hellwig
2025-11-10 13:38   ` Nirjhar Roy (IBM)
2025-11-10 13:59     ` Christoph Hellwig
2025-11-12  7:13       ` Nirjhar Roy (IBM)
2025-10-29 15:58 ` fall back from direct to buffered " Bart Van Assche
2025-10-29 16:14   ` Darrick J. Wong
2025-10-29 16:33   ` Christoph Hellwig
2025-10-30 11:20 ` Dave Chinner
2025-10-30 12:00   ` Geoff Back
2025-10-30 12:54     ` Jan Kara
2025-10-30 14:35     ` Christoph Hellwig
2025-10-30 22:02     ` Dave Chinner
2025-10-30 14:33   ` Christoph Hellwig
2025-10-30 23:18     ` Dave Chinner
2025-10-31 13:00       ` Christoph Hellwig
2025-10-31 15:57         ` Keith Busch
2025-10-31 16:47           ` Christoph Hellwig
2025-11-03 11:14             ` Jan Kara
2025-11-03 12:21               ` Christoph Hellwig
2025-11-03 22:47                 ` Keith Busch
2025-11-04 23:38                 ` Darrick J. Wong
2025-11-05 14:11                   ` Christoph Hellwig
2025-11-05 21:44                     ` Darrick J. Wong
2025-11-06  9:50                       ` Johannes Thumshirn
2025-11-06 12:49                         ` hch [this message]
2025-11-12 14:18                           ` Ming Lei
2025-11-12 14:38                             ` hch
2025-11-13 17:39                 ` Kevin Wolf
2025-11-14  5:39                   ` Christoph Hellwig
2025-11-14  9:29                     ` Kevin Wolf
2025-11-14 12:01                       ` Christoph Hellwig
2025-11-14 12:31                         ` Kevin Wolf
2025-11-14 15:36                           ` Christoph Hellwig
2025-11-14 16:55                             ` Kevin Wolf

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=20251106124900.GA6144@lst.de \
    --to=hch@lst.de \
    --cc=Johannes.Thumshirn@wdc.com \
    --cc=brauner@kernel.org \
    --cc=cem@kernel.org \
    --cc=david@fromorbit.com \
    --cc=djwong@kernel.org \
    --cc=jack@suse.cz \
    --cc=kbusch@kernel.org \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-raid@vger.kernel.org \
    --cc=linux-xfs@vger.kernel.org \
    --cc=martin.petersen@oracle.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).