public inbox for linux-xfs@vger.kernel.org
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@infradead.org>
To: "Darrick J. Wong" <djwong@kernel.org>
Cc: Dave Chinner <david@fromorbit.com>,
	linux-xfs@vger.kernel.org, Christoph Hellwig <hch@infradead.org>
Subject: Re: [PATCH] xfs: _{attr,data}_map_shared should take ILOCK_EXCL until iread_extents is completely done
Date: Mon, 10 Apr 2023 22:06:46 -0700	[thread overview]
Message-ID: <ZDTq5mdvNErn+JrE@infradead.org> (raw)
In-Reply-To: <20230411010638.GF360889@frogsfrogsfrogs>

On Mon, Apr 10, 2023 at 06:06:38PM -0700, Darrick J. Wong wrote:
> +++ b/fs/xfs/libxfs/xfs_bmap.c
> @@ -1171,6 +1171,8 @@ xfs_iread_extents(
>  		goto out;
>  	}
>  	ASSERT(ir.loaded == xfs_iext_count(ifp));
> +	smp_mb();
> +	ifp->if_needextents = 0;

Shouldn't this be a WRITE_ONCE?

> diff --git a/fs/xfs/libxfs/xfs_inode_fork.c b/fs/xfs/libxfs/xfs_inode_fork.c
> index 6b21760184d9..eadae924dc42 100644
> --- a/fs/xfs/libxfs/xfs_inode_fork.c
> +++ b/fs/xfs/libxfs/xfs_inode_fork.c
> @@ -174,6 +174,8 @@ xfs_iformat_btree(
>  	int			level;
>  
>  	ifp = xfs_ifork_ptr(ip, whichfork);
> +	ifp->if_needextents = 1;

Same here.

>  	dfp = (xfs_bmdr_block_t *)XFS_DFORK_PTR(dip, whichfork);
>  	size = XFS_BMAP_BROOT_SPACE(mp, dfp);
>  	nrecs = be16_to_cpu(dfp->bb_numrecs);
> diff --git a/fs/xfs/libxfs/xfs_inode_fork.h b/fs/xfs/libxfs/xfs_inode_fork.h
> index d3943d6ad0b9..e69c78c35c96 100644
> --- a/fs/xfs/libxfs/xfs_inode_fork.h
> +++ b/fs/xfs/libxfs/xfs_inode_fork.h
> @@ -24,6 +24,7 @@ struct xfs_ifork {
>  	xfs_extnum_t		if_nextents;	/* # of extents in this fork */
>  	short			if_broot_bytes;	/* bytes allocated for root */
>  	int8_t			if_format;	/* format of this fork */
> +	int8_t			if_needextents;	/* extents have not been read */
>  };

I don't think this should be signed.

>  static inline bool xfs_need_iread_extents(struct xfs_ifork *ifp)
>  {
> -	return ifp->if_format == XFS_DINODE_FMT_BTREE && ifp->if_height == 0;
> +	return ifp->if_format == XFS_DINODE_FMT_BTREE && ifp->if_needextents;

.. and this should use READ_ONCE and drop the if_format check given
that if_needextents is only set in xfs_iformat_btree.

      parent reply	other threads:[~2023-04-11  5:07 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-04-11  1:06 [PATCH] xfs: _{attr,data}_map_shared should take ILOCK_EXCL until iread_extents is completely done Darrick J. Wong
2023-04-11  3:20 ` Dave Chinner
2023-04-11  4:18   ` Darrick J. Wong
2023-04-11 21:58     ` Dave Chinner
2023-04-11  5:11   ` Christoph Hellwig
2023-04-11  5:06 ` Christoph Hellwig [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=ZDTq5mdvNErn+JrE@infradead.org \
    --to=hch@infradead.org \
    --cc=david@fromorbit.com \
    --cc=djwong@kernel.org \
    --cc=linux-xfs@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