From: "Darrick J. Wong" <djwong@kernel.org>
To: Christoph Hellwig <hch@lst.de>
Cc: linux-xfs@vger.kernel.org
Subject: Re: [PATCH 13/18] xfs: use a union for i_cowextsize and i_flushiter
Date: Wed, 24 Mar 2021 11:33:11 -0700 [thread overview]
Message-ID: <20210324183311.GL22100@magnolia> (raw)
In-Reply-To: <20210324142129.1011766-14-hch@lst.de>
On Wed, Mar 24, 2021 at 03:21:24PM +0100, Christoph Hellwig wrote:
> The i_cowextsize field is only used for v3 inodes, and the i_flushiter
> field is only used for v1/v2 inodes. Use a union to pack the inode a
> littler better after adding a few missing guards around their usage.
>
> Signed-off-by: Christoph Hellwig <hch@lst.de>
Yes, that's a nice compression opportunity.
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
--D
> ---
> fs/xfs/libxfs/xfs_inode_buf.c | 3 ++-
> fs/xfs/xfs_inode.c | 6 ++++--
> fs/xfs/xfs_inode.h | 7 +++++--
> fs/xfs/xfs_ioctl.c | 6 +++++-
> 4 files changed, 16 insertions(+), 6 deletions(-)
>
> diff --git a/fs/xfs/libxfs/xfs_inode_buf.c b/fs/xfs/libxfs/xfs_inode_buf.c
> index d090274fb8a152..96db2649f6b2fe 100644
> --- a/fs/xfs/libxfs/xfs_inode_buf.c
> +++ b/fs/xfs/libxfs/xfs_inode_buf.c
> @@ -193,7 +193,8 @@ xfs_inode_from_disk(
> * inode. If the inode is unused, mode is zero and we shouldn't mess
> * with the uninitialized part of it.
> */
> - ip->i_flushiter = be16_to_cpu(from->di_flushiter);
> + if (!xfs_sb_version_has_v3inode(&ip->i_mount->m_sb))
> + ip->i_flushiter = be16_to_cpu(from->di_flushiter);
> inode->i_generation = be32_to_cpu(from->di_gen);
> inode->i_mode = be16_to_cpu(from->di_mode);
> if (!inode->i_mode)
> diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
> index e951ea48b3a276..b4b6fddccd1ca0 100644
> --- a/fs/xfs/xfs_inode.c
> +++ b/fs/xfs/xfs_inode.c
> @@ -3459,8 +3459,10 @@ xfs_iflush(
> xfs_inode_to_disk(ip, dip, iip->ili_item.li_lsn);
>
> /* Wrap, we never let the log put out DI_MAX_FLUSH */
> - if (ip->i_flushiter == DI_MAX_FLUSH)
> - ip->i_flushiter = 0;
> + if (!xfs_sb_version_has_v3inode(&mp->m_sb)) {
> + if (ip->i_flushiter == DI_MAX_FLUSH)
> + ip->i_flushiter = 0;
> + }
>
> xfs_iflush_fork(ip, dip, iip, XFS_DATA_FORK);
> if (XFS_IFORK_Q(ip))
> diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
> index 6246ee8a4359ab..7ba0ffa50ede20 100644
> --- a/fs/xfs/xfs_inode.h
> +++ b/fs/xfs/xfs_inode.h
> @@ -58,8 +58,11 @@ typedef struct xfs_inode {
> xfs_rfsblock_t i_nblocks; /* # of direct & btree blocks */
> uint32_t i_projid; /* owner's project id */
> xfs_extlen_t i_extsize; /* basic/minimum extent size */
> - xfs_extlen_t i_cowextsize; /* basic cow extent size */
> - uint16_t i_flushiter; /* incremented on flush */
> + /* cowextsize is only used for v3 inodes, flushiter for v1/2 */
> + union {
> + xfs_extlen_t i_cowextsize; /* basic cow extent size */
> + uint16_t i_flushiter; /* incremented on flush */
> + };
>
> struct xfs_icdinode i_d; /* most of ondisk inode */
>
> diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c
> index e45bce9b11082c..3405a5f5bacfda 100644
> --- a/fs/xfs/xfs_ioctl.c
> +++ b/fs/xfs/xfs_ioctl.c
> @@ -1121,7 +1121,11 @@ xfs_fill_fsxattr(
>
> simple_fill_fsxattr(fa, xfs_ip2xflags(ip));
> fa->fsx_extsize = ip->i_extsize << ip->i_mount->m_sb.sb_blocklog;
> - fa->fsx_cowextsize = ip->i_cowextsize << ip->i_mount->m_sb.sb_blocklog;
> + if (xfs_sb_version_has_v3inode(&ip->i_mount->m_sb) &&
> + (ip->i_d.di_flags2 & XFS_DIFLAG2_COWEXTSIZE)) {
> + fa->fsx_cowextsize =
> + ip->i_cowextsize << ip->i_mount->m_sb.sb_blocklog;
> + }
> fa->fsx_projid = ip->i_projid;
> if (ifp && (ifp->if_flags & XFS_IFEXTENTS))
> fa->fsx_nextents = xfs_iext_count(ifp);
> --
> 2.30.1
>
next prev parent reply other threads:[~2021-03-24 18:34 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-24 14:21 xfs inode structure cleanups v3 Christoph Hellwig
2021-03-24 14:21 ` [PATCH 01/18] xfs: split xfs_imap_to_bp Christoph Hellwig
2021-03-24 18:07 ` Darrick J. Wong
2021-03-24 14:21 ` [PATCH 02/18] xfs: consistently initialize di_flags2 Christoph Hellwig
2021-03-24 18:08 ` Darrick J. Wong
2021-03-24 14:21 ` [PATCH 03/18] xfs: handle crtime more carefully in xfs_bulkstat_one_int Christoph Hellwig
2021-03-24 18:09 ` Darrick J. Wong
2021-03-24 14:21 ` [PATCH 04/18] xfs: remove the unused xfs_icdinode_has_bigtime helper Christoph Hellwig
2021-03-24 18:10 ` Darrick J. Wong
2021-03-24 14:21 ` [PATCH 05/18] xfs: remove the di_dmevmask and di_dmstate fields from struct xfs_icdinode Christoph Hellwig
2021-03-24 18:17 ` Darrick J. Wong
2021-03-24 18:20 ` Christoph Hellwig
2021-03-25 14:29 ` Anthony Iliopoulos
2021-03-24 14:21 ` [PATCH 06/18] xfs: don't clear the "dinode core" in xfs_inode_alloc Christoph Hellwig
2021-03-24 18:27 ` Darrick J. Wong
2021-03-24 18:28 ` Christoph Hellwig
2021-03-24 18:40 ` Darrick J. Wong
2021-03-24 14:21 ` [PATCH 07/18] xfs: move the di_projid field to struct xfs_inode Christoph Hellwig
2021-03-24 18:20 ` Darrick J. Wong
2021-03-25 8:39 ` Christoph Hellwig
2021-03-24 14:21 ` [PATCH 08/18] xfs: move the di_size " Christoph Hellwig
2021-03-24 18:21 ` Darrick J. Wong
2021-03-24 14:21 ` [PATCH 09/18] xfs: move the di_nblocks " Christoph Hellwig
2021-03-24 18:22 ` Darrick J. Wong
2021-03-25 8:39 ` Christoph Hellwig
2021-03-26 0:25 ` Darrick J. Wong
2021-03-24 14:21 ` [PATCH 10/18] xfs: move the di_extsize " Christoph Hellwig
2021-03-24 18:29 ` Darrick J. Wong
2021-03-24 14:21 ` [PATCH 11/18] xfs: move the di_cowextsize " Christoph Hellwig
2021-03-24 18:31 ` Darrick J. Wong
2021-03-24 18:33 ` Darrick J. Wong
2021-03-24 18:42 ` Darrick J. Wong
2021-03-25 8:41 ` Christoph Hellwig
2021-03-24 14:21 ` [PATCH 12/18] xfs: move the di_flushiter " Christoph Hellwig
2021-03-24 18:32 ` Darrick J. Wong
2021-03-24 14:21 ` [PATCH 13/18] xfs: use a union for i_cowextsize and i_flushiter Christoph Hellwig
2021-03-24 18:33 ` Darrick J. Wong [this message]
2021-03-25 3:06 ` Darrick J. Wong
2021-03-25 9:01 ` Christoph Hellwig
2021-03-24 14:21 ` [PATCH 14/18] xfs: cleanup xfs_fill_fsxattr Christoph Hellwig
2021-03-24 18:34 ` Darrick J. Wong
2021-03-24 14:21 ` [PATCH 15/18] xfs: move the di_forkoff field to struct xfs_inode Christoph Hellwig
2021-03-24 18:35 ` Darrick J. Wong
2021-03-24 14:21 ` [PATCH 16/18] xfs: move the di_flags " Christoph Hellwig
2021-03-24 18:36 ` Darrick J. Wong
2021-03-24 14:21 ` [PATCH 17/18] xfs: move the di_flags2 " Christoph Hellwig
2021-03-24 18:37 ` Darrick J. Wong
2021-03-24 14:21 ` [PATCH 18/18] xfs: move the di_crtime " Christoph Hellwig
2021-03-24 18:38 ` Darrick J. Wong
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=20210324183311.GL22100@magnolia \
--to=djwong@kernel.org \
--cc=hch@lst.de \
--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;
as well as URLs for NNTP newsgroup(s).