linux-api.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Florian Weimer <fweimer@redhat.com>
To: Jeff Layton <jlayton@kernel.org>
Cc: linux-fsdevel@vger.kernel.org,  linux-kernel@vger.kernel.org,
	linux-api@vger.kernel.org,  Dave Chinner <dchinner@redhat.com>
Subject: Re: Testing if two open descriptors refer to the same inode
Date: Mon, 29 Jul 2024 17:39:28 +0200	[thread overview]
Message-ID: <87frrsmclr.fsf@oldenburg.str.redhat.com> (raw)
In-Reply-To: <a13bff5812cb36adf3fed80093cbe1de601ec506.camel@kernel.org> (Jeff Layton's message of "Mon, 29 Jul 2024 11:24:41 -0400")

* Jeff Layton:

> On Mon, 2024-07-29 at 08:55 +0200, Florian Weimer wrote:
>> It was pointed out to me that inode numbers on Linux are no longer
>> expected to be unique per file system, even for local file systems.
>> Applications sometimes need to check if two (open) files are the
>> same.
>> For example, a program may want to use a temporary file if is invoked
>> with input and output files referring to the same file.
>> 
>> How can we check for this?  The POSIX way is to compare st_ino and
>> st_dev in stat output, but if inode numbers are not unique, that will
>> result in files falsely being reported as identical.  It's harmless
>> in
>> the temporary file case, but it in other scenarios, it may result in
>> data loss.
>> 
>
> I believe this is the problem that STATX_SUBVOL was intended to solve.
>
> Both bcachefs and btrfs will provide this attribute if requested. So,
> basically to uniquely ID an inode using statx, you need a tuple of:
>
> stx_dev_major/minor
> stx_subvol
> stx_ino
>
> If the filesystem doesn't provide STATX_SUBVOL, then one can (likely)
> conclude that stx_dev_* and stx_ino are enough.

Does this really work for the virtiofs case, though?  It has to pass
through all three *and* make things unique relative to the host, I
think.

Thanks,
Florian


      reply	other threads:[~2024-07-29 15:39 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-07-29  6:55 Testing if two open descriptors refer to the same inode Florian Weimer
2024-07-29  9:09 ` Aleksa Sarai
2024-07-29  9:29   ` Florian Weimer
2024-07-29 10:18 ` Mateusz Guzik
2024-07-29 10:40   ` Florian Weimer
2024-07-29 10:50     ` Mateusz Guzik
2024-07-29 10:56       ` Mateusz Guzik
2024-07-29 10:57       ` Florian Weimer
2024-07-29 11:06         ` Mateusz Guzik
2024-07-29 11:36           ` Florian Weimer
2024-07-29 12:00             ` Mateusz Guzik
2024-07-29 11:40           ` Aleksa Sarai
2024-07-31 18:07             ` David Sterba
2024-07-29 11:47           ` Aleksa Sarai
2024-07-29 12:12             ` Mateusz Guzik
2024-07-29 23:19               ` Dave Chinner
2024-07-29 23:08         ` Dave Chinner
2024-07-29 12:26   ` Christian Brauner
2024-07-29 13:36   ` Theodore Ts'o
2024-07-30  2:31     ` Dave Chinner
2024-07-30  4:19       ` Theodore Ts'o
2024-07-30 15:38       ` Christoph Hellwig
2024-07-29 15:24 ` Jeff Layton
2024-07-29 15:39   ` Florian Weimer [this message]

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=87frrsmclr.fsf@oldenburg.str.redhat.com \
    --to=fweimer@redhat.com \
    --cc=dchinner@redhat.com \
    --cc=jlayton@kernel.org \
    --cc=linux-api@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@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;
as well as URLs for NNTP newsgroup(s).