Re: [PATCH 02/14] xfs: introduce new file range exchange ioctls

["Darrick J. Wong" <djwong@kernel.org>]

On Wed, Feb 28, 2024 at 07:44:32AM -0800, Christoph Hellwig wrote:
> On Mon, Feb 26, 2024 at 06:21:23PM -0800, Darrick J. Wong wrote:
> > From: Darrick J. Wong <djwong@kernel.org>
> > 
> > Introduce a pair of new ioctls to handle exchanging ranges of bytes
> > between files.  The goal here is to perform the exchange atomically with
> > respect to applications -- either they see the file contents before the
> > exchange or they see that A-B is now B-A, even if the kernel crashes.
> > 
> > The simpler of the two ioctls is XFS_IOC_EXCHANGE_RANGE, which performs
> > the exchange unconditionally.  XFS_IOC_COMMIT_RANGE, on the other hand,
> > requires the caller to sample the file attributes of one of the files
> > participating in the exchange, and aborts the exchange if that file has
> > changed in the meantime (presumably by another thread).
> 
> So per all the discussions, wouldn't it make sense to separate out
> XFS_IOC_COMMIT_RANGE (plus the new start commit one later), and if
> discussions are still going on just get XFS_IOC_EXCHANGE_RANGE
> done ASAP to go on with online repair, and give XFS_IOC_COMMIT_RANGE
> enough time to discuss the finer details?

Done.  All of the COMMIT_RANGE code (and the _CHECK_FRESH2 code) have
been moved to a separate patch and combined with the patch[1] from
yesterday.

[1] https://lore.kernel.org/linux-xfs/CAOQ4uxiPfno-Hx+fH3LEN_4D6HQgyMAySRNCU=O2R_-ksrxSDQ@mail.gmail.com/

> > +struct xfs_exch_range {
> > +	__s64		file1_fd;
> 
> I should have noticed this last time, by why are we passing a fd
> as a 64-bit value when it actually just is a 32-bit value in syscalls?
> (same for commit).

I'll change that.

> > +	if (((fxr->file1->f_flags | fxr->file2->f_flags) & (__O_SYNC | O_DSYNC)) ||
> 
> Nit: overly long line here.

I'll replace the __O_SYNC | O_DSYNC with O_SYNC, since they're the same
thing.

> > +	if (fxr->flags & ~(XFS_EXCHRANGE_ALL_FLAGS | __XFS_EXCHRANGE_CHECK_FRESH2))
> 
> .. and here

Fixed, thanks.

> > +	/*
> > +	 * The ioctl enforces that src and dest files are on the same mount.
> > +	 * However, they only need to be on the same file system.
> > +	 */
> > +	if (inode1->i_sb != inode2->i_sb)
> > +		return -EXDEV;
> 
> How about only doing this checks once further up?  As the same sb also
> applies the same mount.

I'll remove this check entirely, since we've already checked that the
vfsmnt are the same.  Assuming that's what you meant-- I was slightly
confused by "same sb also applies the same mount" and decided to
interpret that as "same sb implies the same mount".

--D