Linux XFS filesystem development
 help / color / mirror / Atom feed
From: "Darrick J. Wong" <djwong@kernel.org>
To: Eric Sandeen <sandeen@redhat.com>
Cc: Eric Sandeen <sandeen@sandeen.net>,
	linux-xfs@vger.kernel.org, cem@kernel.org
Subject: Re: [PATCH V2 2/3] xfs: factor out xrep_reset_fork_to_extents helper for scrub/repair
Date: Mon, 5 Oct 2026 14:17:41 -0700	[thread overview]
Message-ID: <20261005211741.GF1615495@frogsfrogsfrogs> (raw)
In-Reply-To: <d68afc86-6510-45c4-a225-784e7f629a33@redhat.com>

On Mon, Oct 05, 2026 at 03:43:08PM -0500, Eric Sandeen wrote:
> On 10/5/26 3:20 PM, Darrick J. Wong wrote:
> > On Mon, Oct 05, 2026 at 02:11:37PM -0500, Eric Sandeen wrote:
> >> The open-coded ~10 line series of steps to reset a fork to empty extents
> >> format is repeated 3 times; factor this out into a helper to eliminate
> >> cut and paste.
> 
> ...
> 
> >> diff --git a/fs/xfs/scrub/repair.c b/fs/xfs/scrub/repair.c
> >> index c2a437416227..8385308c852f 100644
> >> --- a/fs/xfs/scrub/repair.c
> >> +++ b/fs/xfs/scrub/repair.c
> >> @@ -882,6 +882,32 @@ xrep_ino_ensure_extent_count(
> >>  	return 0;
> >>  }
> >>  
> >> +/* Discard the contents of this fork and initialize as empty extent-format. */
> >> +void
> >> +xrep_reset_fork_to_extents(
> >> +	struct xfs_scrub	*sc,
> >> +	int			whichfork)
> >> +{
> >> +	struct xfs_ifork	*ifp = xfs_ifork_ptr(sc->ip, whichfork);
> >> +	uint			ilog_flags = XFS_ILOG_CORE;
> >> +
> >> +	ASSERT(whichfork == XFS_DATA_FORK || whichfork == XFS_ATTR_FORK);
> >> +
> >> +	if (whichfork == XFS_DATA_FORK)
> >> +		ilog_flags |= XFS_ILOG_DDATA;
> >> +	else
> >> +		ilog_flags |= XFS_ILOG_ADATA;
> > 
> > Nit pick:
> 
> I thought about a case switch but it seemed excessive for 2 choices.
> My ASSERT does make sure we have one or the other valid value... *shrug*
> 
> > 	switch (whichfork) {
> > 	case XFS_DATA_FORK:
> > 		ilog_flags |= XFS_ILOG_DDATA;
> > 		break;
> > 	case XFS_ATTR_FORK:
> > 		ilog_flags |= XFS_ILOG_ADATA;
> > 		break;
> > 	default:
> > 		ASSERT(0);
> > 		return;
> 
> I guess this has the advantage of kinda handling an unknown whichfork or
> XFS_COW_FORK, though who knows what happens to callers that get a do-nothing
> return at that point?
> > 	}

There shouldn't be any, I just thought it looks cleaner. :)

--D

> > 
> > Either way this looks like a good hoist to me, so
> > Reviewed-by: "Darrick J. Wong" <djwong@kernel.org>
> > 
> > --D
> > 
> >> +
> >> +	xfs_idestroy_fork(ifp);
> >> +	ifp->if_format = XFS_DINODE_FMT_EXTENTS;
> >> +	ifp->if_nextents = 0;
> >> +	ifp->if_bytes = 0;
> >> +	ifp->if_data = NULL;
> >> +	ifp->if_height = 0;
> >> +
> >> +	xfs_trans_log_inode(sc->tp, sc->ip, ilog_flags);
> >> +}
> >> +
> >>  /*
> >>   * Initialize all the btree cursors for an AG repair except for the btree that
> >>   * we're rebuilding.
> >> diff --git a/fs/xfs/scrub/repair.h b/fs/xfs/scrub/repair.h
> >> index 2bb125c4f9bf..c1ba462e4426 100644
> >> --- a/fs/xfs/scrub/repair.h
> >> +++ b/fs/xfs/scrub/repair.h
> >> @@ -81,6 +81,7 @@ int xrep_setup_xfbtree(struct xfs_scrub *sc, const char *descr);
> >>  int xrep_ino_ensure_extent_count(struct xfs_scrub *sc, int whichfork,
> >>  		xfs_extnum_t nextents);
> >>  int xrep_reset_perag_resv(struct xfs_scrub *sc);
> >> +void xrep_reset_fork_to_extents(struct xfs_scrub *sc, int whichfork);
> >>  int xrep_bmap(struct xfs_scrub *sc, int whichfork, bool allow_unwritten);
> >>  int xrep_metadata_inode_forks(struct xfs_scrub *sc);
> >>  int xrep_setup_ag_rmapbt(struct xfs_scrub *sc);
> >> diff --git a/fs/xfs/scrub/symlink_repair.c b/fs/xfs/scrub/symlink_repair.c
> >> index 181961364233..9f6901905bfd 100644
> >> --- a/fs/xfs/scrub/symlink_repair.c
> >> +++ b/fs/xfs/scrub/symlink_repair.c
> >> @@ -303,20 +303,8 @@ xrep_symlink_swap_prep(
> >>  	 * to an empty extent list in preparation for the atomic mapping
> >>  	 * exchange.
> >>  	 */
> >> -	if (ip_local) {
> >> -		struct xfs_ifork	*ifp;
> >> -
> >> -		ifp = xfs_ifork_ptr(sc->ip, XFS_DATA_FORK);
> >> -		xfs_idestroy_fork(ifp);
> >> -		ifp->if_format = XFS_DINODE_FMT_EXTENTS;
> >> -		ifp->if_nextents = 0;
> >> -		ifp->if_bytes = 0;
> >> -		ifp->if_data = NULL;
> >> -		ifp->if_height = 0;
> >> -
> >> -		xfs_trans_log_inode(sc->tp, sc->ip,
> >> -				XFS_ILOG_CORE | XFS_ILOG_DDATA);
> >> -	}
> >> +	if (ip_local)
> >> +		xrep_reset_fork_to_extents(sc, XFS_DATA_FORK);
> >>  
> >>  	return 0;
> >>  }
> >> -- 
> >> 2.55.0
> >>
> >>
> > 
> 

  reply	other threads:[~2026-10-05 21:17 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-02 21:08 [PATCH 0/3] xfs: more misc code deduplication Eric Sandeen
2026-10-02 21:08 ` [PATCH 1/3] xfs: factor out xfs_dir2_sf_copy_entries helper Eric Sandeen
2026-10-05 12:14   ` Carlos Maiolino
2026-10-07 13:50   ` Christoph Hellwig
2026-10-02 21:08 ` [PATCH 2/3] xfs: factor out xrep_reset_fork_to_extents helper for scrub/repair Eric Sandeen
2026-10-05 13:07   ` Carlos Maiolino
2026-10-05 14:12     ` Eric Sandeen
2026-10-05 19:11   ` [PATCH V2 " Eric Sandeen
2026-10-05 20:20     ` Darrick J. Wong
2026-10-05 20:43       ` Eric Sandeen
2026-10-05 21:17         ` Darrick J. Wong [this message]
2026-10-06  8:32           ` Carlos Maiolino
2026-10-07 13:52             ` Christoph Hellwig
2026-10-07 15:25     ` [PATCH V3 " Eric Sandeen
2026-10-08 11:02       ` Carlos Maiolino
2026-10-02 21:08 ` [PATCH 3/3] xfs: re-use refcount scrub/repair code for rtrefcount Eric Sandeen
2026-10-05 20:38   ` Darrick J. Wong
2026-10-06  8:25   ` Carlos Maiolino
2026-10-07 13:51   ` Christoph Hellwig
2026-10-07 15:32     ` Eric Sandeen
2026-10-07 15:37       ` Darrick J. Wong
2026-10-08 11:15         ` Carlos Maiolino
2026-10-08 11:09   ` Carlos Maiolino
2026-10-08 13:32 ` [PATCH 0/3] xfs: more misc code deduplication Carlos Maiolino

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=20261005211741.GF1615495@frogsfrogsfrogs \
    --to=djwong@kernel.org \
    --cc=cem@kernel.org \
    --cc=linux-xfs@vger.kernel.org \
    --cc=sandeen@redhat.com \
    --cc=sandeen@sandeen.net \
    /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