Re: [PATCH 06/10] xfs: move transaction handling to xfs_bmapi_convert_delalloc

["Darrick J. Wong" <darrick.wong@oracle.com>]

On Mon, Feb 11, 2019 at 01:54:23PM +0100, Christoph Hellwig wrote:
> No need to deal with the transaction and the inode locking in the
> caller.  Also move to automatic unlocking on transaction commit or
> cancel to simplify the code a little more.
> 
> Note that we also switch to passing whichfork as the second paramters,
> matching what most related functions do.
> 
> Signed-off-by: Christoph Hellwig <hch@lst.de>
> ---
>  fs/xfs/libxfs/xfs_bmap.c | 35 ++++++++++++++++++++++++++++++-----
>  fs/xfs/libxfs/xfs_bmap.h |  5 +++--
>  fs/xfs/xfs_iomap.c       | 32 ++++----------------------------
>  3 files changed, 37 insertions(+), 35 deletions(-)
> 
> diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
> index be2cb5800e02..d9d66e1856d7 100644
> --- a/fs/xfs/libxfs/xfs_bmap.c
> +++ b/fs/xfs/libxfs/xfs_bmap.c
> @@ -4446,16 +4446,30 @@ xfs_bmapi_write(
>   */
>  int
>  xfs_bmapi_convert_delalloc(
> -	struct xfs_trans	*tp,
>  	struct xfs_inode	*ip,
> -	xfs_fileoff_t		offset_fsb,
>  	int			whichfork,
> -	struct xfs_bmbt_irec	*imap)
> +	xfs_fileoff_t		offset_fsb,
> +	struct xfs_bmbt_irec	*imap,
> +	unsigned int		*seq)
>  {
>  	struct xfs_ifork	*ifp = XFS_IFORK_PTR(ip, whichfork);
> +	struct xfs_mount	*mp = ip->i_mount;
>  	struct xfs_bmalloca	bma = { NULL };
> +	struct xfs_trans	*tp;
>  	int			error;
>  
> +	/*
> +	 * Space for the extent and indirect blocks was reserved when the
> +	 * delalloc extent was created so there's no need to do so here.
> +	 */
> +	error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write, 0, 0,
> +				XFS_TRANS_RESERVE, &tp);
> +	if (error)
> +		return error;
> +
> +	xfs_ilock(ip, XFS_ILOCK_EXCL);
> +	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);

Aha, this is the problem -- this operation involves deferred rmap
updates, which means that we have to retain the ILOCK until we've
finished processing the rmap updates so that another thread cannot jump
in and start modifying the file's bmap while we're still trying to
finish the rmap updates.

The patch below fixes generic/127 for me, for this configuration:

FSTYP=xfs
MKFS_OPTIONS="-m reflink=1,rmapbt=1 -i sparse=1"
MOUNT_OPTIONS="-o usrquota,grpquota,prjquota"

--D

diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
index d9d66e1856d7..fd7757b205a6 100644
--- a/fs/xfs/libxfs/xfs_bmap.c
+++ b/fs/xfs/libxfs/xfs_bmap.c
@@ -4468,7 +4468,7 @@ xfs_bmapi_convert_delalloc(
 		return error;
 
 	xfs_ilock(ip, XFS_ILOCK_EXCL);
-	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
+	xfs_trans_ijoin(tp, ip, 0);
 
 	if (!xfs_iext_lookup_extent(ip, ifp, offset_fsb, &bma.icur, &bma.got) ||
 	    bma.got.br_startoff > offset_fsb) {
@@ -4531,12 +4531,15 @@ xfs_bmapi_convert_delalloc(
 		goto out_finish;
 
 	xfs_bmapi_finish(&bma, whichfork, 0);
-	return xfs_trans_commit(tp);
+	error = xfs_trans_commit(tp);
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
+	return error;
 
 out_finish:
 	xfs_bmapi_finish(&bma, whichfork, error);
 out_trans_cancel:
 	xfs_trans_cancel(tp);
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 	return error;
 }
 
> +
>  	if (!xfs_iext_lookup_extent(ip, ifp, offset_fsb, &bma.icur, &bma.got) ||
>  	    bma.got.br_startoff > offset_fsb) {
>  		/*
> @@ -4464,7 +4478,8 @@ xfs_bmapi_convert_delalloc(
>  		 * might have moved the extent to the data fork in the meantime.
>  		 */
>  		WARN_ON_ONCE(whichfork != XFS_COW_FORK);
> -		return -EAGAIN;
> +		error = -EAGAIN;
> +		goto out_trans_cancel;
>  	}
>  
>  	/*
> @@ -4473,7 +4488,8 @@ xfs_bmapi_convert_delalloc(
>  	 */
>  	if (!isnullstartblock(bma.got.br_startblock)) {
>  		*imap = bma.got;
> -		return 0;
> +		*seq = READ_ONCE(ifp->if_seq);
> +		goto out_trans_cancel;
>  	}
>  
>  	bma.tp = tp;
> @@ -4500,6 +4516,7 @@ xfs_bmapi_convert_delalloc(
>  	ASSERT(!isnullstartblock(bma.got.br_startblock));
>  	ASSERT(bma.got.br_startblock || XFS_IS_REALTIME_INODE(ip));
>  	*imap = bma.got;
> +	*seq = READ_ONCE(ifp->if_seq);
>  
>  	if (whichfork == XFS_COW_FORK) {
>  		error = xfs_refcount_alloc_cow_extent(tp, bma.blkno,
> @@ -4510,8 +4527,16 @@ xfs_bmapi_convert_delalloc(
>  
>  	error = xfs_bmap_btree_to_extents(tp, ip, bma.cur, &bma.logflags,
>  			whichfork);
> +	if (error)
> +		goto out_finish;
> +
> +	xfs_bmapi_finish(&bma, whichfork, 0);
> +	return xfs_trans_commit(tp);
> +
>  out_finish:
>  	xfs_bmapi_finish(&bma, whichfork, error);
> +out_trans_cancel:
> +	xfs_trans_cancel(tp);
>  	return error;
>  }
>  
> diff --git a/fs/xfs/libxfs/xfs_bmap.h b/fs/xfs/libxfs/xfs_bmap.h
> index b5eca7a26949..78b190b6e908 100644
> --- a/fs/xfs/libxfs/xfs_bmap.h
> +++ b/fs/xfs/libxfs/xfs_bmap.h
> @@ -223,8 +223,9 @@ int	xfs_bmapi_reserve_delalloc(struct xfs_inode *ip, int whichfork,
>  		xfs_fileoff_t off, xfs_filblks_t len, xfs_filblks_t prealloc,
>  		struct xfs_bmbt_irec *got, struct xfs_iext_cursor *cur,
>  		int eof);
> -int	xfs_bmapi_convert_delalloc(struct xfs_trans *, struct xfs_inode *,
> -		xfs_fileoff_t, int, struct xfs_bmbt_irec *);
> +int	xfs_bmapi_convert_delalloc(struct xfs_inode *ip, int whichfork,
> +		xfs_fileoff_t offset_fsb, struct xfs_bmbt_irec *imap,
> +		unsigned int *seq);
>  
>  static inline void
>  xfs_bmap_add_free(
> diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
> index fd3aacd4bf02..39be741cac5a 100644
> --- a/fs/xfs/xfs_iomap.c
> +++ b/fs/xfs/xfs_iomap.c
> @@ -684,11 +684,9 @@ xfs_iomap_write_allocate(
>  	unsigned int		*seq)
>  {
>  	struct xfs_mount	*mp = ip->i_mount;
> -	struct xfs_ifork	*ifp = XFS_IFORK_PTR(ip, whichfork);
>  	xfs_fileoff_t		offset_fsb;
>  	xfs_fileoff_t		map_start_fsb;
>  	xfs_extlen_t		map_count_fsb;
> -	struct xfs_trans	*tp;
>  	int			error = 0;
>  
>  	/*
> @@ -716,17 +714,8 @@ xfs_iomap_write_allocate(
>  		/*
>  		 * Allocate in a loop because it may take several attempts to
>  		 * allocate real blocks for a contiguous delalloc extent if free
> -		 * space is sufficiently fragmented. Note that space for the
> -		 * extent and indirect blocks was reserved when the delalloc
> -		 * extent was created so there's no need to do so here.
> +		 * space is sufficiently fragmented.
>  		 */
> -		error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write, 0, 0,
> -					XFS_TRANS_RESERVE, &tp);
> -		if (error)
> -			return error;
> -
> -		xfs_ilock(ip, XFS_ILOCK_EXCL);
> -		xfs_trans_ijoin(tp, ip, 0);
>  
>  		/*
>  		 * ilock was dropped since imap was populated which means it
> @@ -737,17 +726,10 @@ xfs_iomap_write_allocate(
>  		 * caller. We'll trim it down to the caller's most recently
>  		 * validated range before we return.
>  		 */
> -		error = xfs_bmapi_convert_delalloc(tp, ip, offset_fsb,
> -						   whichfork, imap);
> -		if (error)
> -			goto trans_cancel;
> -
> -		error = xfs_trans_commit(tp);
> +		error = xfs_bmapi_convert_delalloc(ip, whichfork, offset_fsb,
> +				imap, seq);
>  		if (error)
> -			goto error0;
> -
> -		*seq = READ_ONCE(ifp->if_seq);
> -		xfs_iunlock(ip, XFS_ILOCK_EXCL);
> +			return error;
>  
>  		/*
>  		 * See if we were able to allocate an extent that covers at
> @@ -766,12 +748,6 @@ xfs_iomap_write_allocate(
>  			return 0;
>  		}
>  	}
> -
> -trans_cancel:
> -	xfs_trans_cancel(tp);
> -error0:
> -	xfs_iunlock(ip, XFS_ILOCK_EXCL);
> -	return error;
>  }
>  
>  int
> -- 
> 2.20.1
>