[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ZqkJC5vPKRUkIH6m@infradead.org>
Date: Tue, 30 Jul 2024 08:38:51 -0700
From: Christoph Hellwig <hch@...radead.org>
To: Dave Chinner <david@...morbit.com>
Cc: Theodore Ts'o <tytso@....edu>, Mateusz Guzik <mjguzik@...il.com>,
Florian Weimer <fweimer@...hat.com>, linux-fsdevel@...r.kernel.org,
linux-kernel@...r.kernel.org, linux-api@...r.kernel.org,
Dave Chinner <dchinner@...hat.com>
Subject: Re: Testing if two open descriptors refer to the same inode
On Tue, Jul 30, 2024 at 12:31:57PM +1000, Dave Chinner wrote:
> There are at least two different "is this inode identical"
> use cases that {st_dev,st_ino} is being used for.
>
> The first, as Florian described, is to determine if two open fds
> refer to the same inode for collision avoidance.
>
> This works on traditional filesystems like ext4 and XFS, but isn't
> reliable on filesystems with integrated snapshot/subvolume
> functionality.
It's not about snapshot, it's about file systems being broken. Even
btrfs for example always has a unique st_dev,st_ino pair, it can
just unexpectly change at any subvolume root and not just at a mount
point.
> That is our long term challenge: replacing the use of {dev,ino} for
> data uniqueness disambiguation. Making the identification of owners
> of non-unique/shared data simple for applications to use and fast
> for filesystems to resolve will be a challenge.
I don't think there is any way to provide such a guarantee as there
is so many levels of cloning or dedup, many of which are totally
invisible to the high level file system interface.
Powered by blists - more mailing lists