Re: [Lsf-pc] [LSF TOPIC] statx extensions for subvol/snapshot filesystems & more

From: Miklos Szeredi <miklos@szeredi.hu>
To: Jan Kara <jack@suse.cz>
Cc: Kent Overstreet <kent.overstreet@linux.dev>,
	Josef Bacik <josef@toxicpanda.com>,
	 linux-kernel@vger.kernel.org, linux-bcachefs@vger.kernel.org,
	 linux-fsdevel@vger.kernel.org,
	lsf-pc@lists.linux-foundation.org,  linux-btrfs@vger.kernel.org
Subject: Re: [Lsf-pc] [LSF TOPIC] statx extensions for subvol/snapshot filesystems & more
Date: Mon, 26 Feb 2024 09:27:17 +0100	[thread overview]
Message-ID: <CAJfpegvvuzXUDusbsJ1VO0CQf5iZO=TZ8kK7V3-k738oi5RM5w@mail.gmail.com> (raw)
In-Reply-To: <20240222160823.pclx6isoyaf7l64r@quack3>

On Thu, 22 Feb 2024 at 17:08, Jan Kara <jack@suse.cz> wrote:

> > If we are going to start fixing userspace, then we better make sure to
> > use the right interfaces, that won't have issues in the future.
>
> I agree we should give this a good thought which identification of a
> filesystem is the best.

To find mount boundaries statx.stx_mnt_id (especially with
STATX_MNT_ID_UNIQUE) is perfect.

By supplying stx_mnt_id to statmount(2) it's possible to get the
device number associated with that filesystem (statmount.sb_dev_*).  I
think it's what Josef wants btrfs to return as st_dev.

And statx could return that in stx_dev_*, with an interface feature
flag, same as we've done with stx_mnt_id.  I.e. STATX_DEV_NOHACK would
force the vfs to replace anything the filesystem put in kstat.dev with
sb->s_dev.

Thanks,
Miklos