From: Christoph Hellwig <hch@infradead.org>
To: "Darrick J. Wong" <darrick.wong@oracle.com>
Cc: david@fromorbit.com, hch@infradead.org,
linux-xfs@vger.kernel.org, amir73il@gmail.com,
sandeen@sandeen.net
Subject: Re: [PATCH 08/11] xfs: widen ondisk inode timestamps to deal with y2038+
Date: Thu, 27 Aug 2020 07:58:16 +0100 [thread overview]
Message-ID: <20200827065816.GB17534@infradead.org> (raw)
In-Reply-To: <159847954997.2601708.12578930799217289682.stgit@magnolia>
> - tv.tv_sec = (int64_t)t >> 32;
> - tv.tv_nsec = (int32_t)(t & 0xffffffff);
> + if (xfs_dinode_has_bigtime(dip)) {
> + s = xfs_bigtime_to_unix(div_u64_rem(t, NSEC_PER_SEC, &n));
> + } else {
> + s = (int64_t)t >> 32;
> + n = (int32_t)(t & 0xffffffff);
Move this branche into a xfs_legacytime_to_unix helper just to be
symmetric?
This also made me think of the encoding: another sensible option
would be to always read the time stamps as two 32-bit values using the
struct type, and just add them up for the bigtime case.
> + if (xfs_inode_has_bigtime(ip))
> + t = xfs_inode_encode_bigtime(tv);
> + else
> + t = ((int64_t)tv.tv_sec << 32) | (tv.tv_nsec & 0xffffffff);
> +
Same here.
> @@ -305,9 +320,9 @@ xfs_inode_to_disk(
> to->di_projid_hi = cpu_to_be16(from->di_projid >> 16);
>
> memset(to->di_pad, 0, sizeof(to->di_pad));
> - to->di_atime = xfs_inode_to_disk_ts(inode->i_atime);
> - to->di_mtime = xfs_inode_to_disk_ts(inode->i_mtime);
> - to->di_ctime = xfs_inode_to_disk_ts(inode->i_ctime);
> + to->di_atime = xfs_inode_to_disk_ts(ip, inode->i_atime);
> + to->di_mtime = xfs_inode_to_disk_ts(ip, inode->i_mtime);
> + to->di_ctime = xfs_inode_to_disk_ts(ip, inode->i_ctime);
> to->di_nlink = cpu_to_be32(inode->i_nlink);
> to->di_gen = cpu_to_be32(inode->i_generation);
> to->di_mode = cpu_to_be16(inode->i_mode);
> @@ -326,7 +341,7 @@ xfs_inode_to_disk(
> if (xfs_sb_version_has_v3inode(&ip->i_mount->m_sb)) {
> to->di_version = 3;
> to->di_changecount = cpu_to_be64(inode_peek_iversion(inode));
> - to->di_crtime = xfs_inode_to_disk_ts(from->di_crtime);
> + to->di_crtime = xfs_inode_to_disk_ts(ip, from->di_crtime);
> to->di_flags2 = cpu_to_be64(from->di_flags2);
> to->di_cowextsize = cpu_to_be32(from->di_cowextsize);
> to->di_ino = cpu_to_be64(ip->i_ino);
> @@ -546,6 +561,11 @@ xfs_dinode_verify(
> if (fa)
> return fa;
>
> + /* bigtime iflag can only happen on bigtime filesystems */
> + if (xfs_dinode_has_bigtime(dip) &&
> + !xfs_sb_version_hasbigtime(&mp->m_sb))
> + return __this_address;
> +
> return NULL;
> }
>
> diff --git a/fs/xfs/libxfs/xfs_inode_buf.h b/fs/xfs/libxfs/xfs_inode_buf.h
> index 3060ecd24a2e..e05bfe52fd8f 100644
> --- a/fs/xfs/libxfs/xfs_inode_buf.h
> +++ b/fs/xfs/libxfs/xfs_inode_buf.h
> @@ -32,6 +32,11 @@ struct xfs_icdinode {
> struct timespec64 di_crtime; /* time created */
> };
>
> +static inline bool xfs_icdinode_has_bigtime(const struct xfs_icdinode *icd)
> +{
> + return icd->di_flags2 & XFS_DIFLAG2_BIGTIME;
> +}
> +
> /*
> * Inode location information. Stored in the inode and passed to
> * xfs_imap_to_bp() to get a buffer and dinode for a given inode.
> @@ -58,6 +63,14 @@ xfs_failaddr_t xfs_inode_validate_cowextsize(struct xfs_mount *mp,
> uint32_t cowextsize, uint16_t mode, uint16_t flags,
> uint64_t flags2);
>
> -struct timespec64 xfs_inode_from_disk_ts(const xfs_timestamp_t ts);
> +static inline uint64_t xfs_inode_encode_bigtime(struct timespec64 tv)
> +{
> + uint64_t t = xfs_unix_to_bigtime(tv.tv_sec) * NSEC_PER_SEC;
> +
> + return t + tv.tv_nsec;
Why not:
return xfs_unix_to_bigtime(tv.tv_sec) * NSEC_PER_SEC + tv.tv_nsec;
?
> +static inline bool xfs_inode_want_bigtime_upgrade(struct xfs_inode *ip)
> +{
> + return xfs_sb_version_hasbigtime(&ip->i_mount->m_sb) &&
> + !xfs_inode_has_bigtime(ip);
> +}
> +
> /*
> * This is called to mark the fields indicated in fieldmask as needing to be
> * logged when the transaction is committed. The inode must already be
> @@ -131,6 +137,16 @@ xfs_trans_log_inode(
> iversion_flags = XFS_ILOG_CORE;
> }
>
> + /*
> + * If we're updating the inode core or the timestamps and it's possible
> + * to upgrade this inode to bigtime format, do so now.
> + */
> + if ((flags & (XFS_ILOG_CORE | XFS_ILOG_TIMESTAMP)) &&
> + xfs_inode_want_bigtime_upgrade(ip)) {
> + ip->i_d.di_flags2 |= XFS_DIFLAG2_BIGTIME;
> + flags |= XFS_ILOG_CORE;
> + }
I find the way to directly set XFS_DIFLAG2_BIGTIME but using a helper
to check it here rather confusing.
Why not:
if (xfs_sb_version_hasbigtime(&ip->i_mount->m_sb) &&
(flags & (XFS_ILOG_CORE | XFS_ILOG_TIMESTAMP)) &&
!(ip->i_d.di_flags2 & XFS_DIFLAG2_BIGTIME)) {
ip->i_d.di_flags2 |= XFS_DIFLAG2_BIGTIME;
flags |= XFS_ILOG_CORE;
}
?
next prev parent reply other threads:[~2020-08-27 6:58 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-08-26 22:04 [PATCH v4 00/11] xfs: widen timestamps to deal with y2038 Darrick J. Wong
2020-08-26 22:05 ` [PATCH 01/11] xfs: explicitly define inode timestamp range Darrick J. Wong
2020-08-27 6:42 ` Christoph Hellwig
2020-08-28 4:08 ` Allison Collins
2020-08-26 22:05 ` [PATCH 02/11] xfs: refactor quota expiration timer modification Darrick J. Wong
2020-08-27 6:43 ` Christoph Hellwig
2020-08-28 4:08 ` Allison Collins
2020-08-26 22:05 ` [PATCH 03/11] xfs: refactor default quota grace period setting code Darrick J. Wong
2020-08-27 6:44 ` Christoph Hellwig
2020-08-28 4:08 ` Allison Collins
2020-08-26 22:05 ` [PATCH 04/11] xfs: refactor quota timestamp coding Darrick J. Wong
2020-08-27 6:44 ` Christoph Hellwig
2020-08-28 4:08 ` Allison Collins
2020-08-26 22:05 ` [PATCH 05/11] xfs: move xfs_log_dinode_to_disk to the log recovery code Darrick J. Wong
2020-08-27 6:45 ` Christoph Hellwig
2020-08-28 4:08 ` Allison Collins
2020-08-26 22:05 ` [PATCH 06/11] xfs: kill struct xfs_timestamp Darrick J. Wong
2020-08-28 4:08 ` Allison Collins
2020-08-26 22:05 ` [PATCH 07/11] xfs: kill struct xfs_ictimestamp Darrick J. Wong
2020-08-27 6:51 ` Christoph Hellwig
2020-08-27 8:17 ` Amir Goldstein
2020-08-27 8:18 ` Christoph Hellwig
2020-08-27 8:56 ` Amir Goldstein
2020-08-27 15:31 ` Darrick J. Wong
2020-08-26 22:05 ` [PATCH 08/11] xfs: widen ondisk inode timestamps to deal with y2038+ Darrick J. Wong
2020-08-27 6:58 ` Christoph Hellwig [this message]
2020-08-27 15:38 ` Darrick J. Wong
2020-08-26 22:05 ` [PATCH 09/11] xfs: widen ondisk quota expiration timestamps to handle y2038+ Darrick J. Wong
2020-08-27 7:00 ` Christoph Hellwig
2020-08-27 17:49 ` Darrick J. Wong
2020-08-28 6:08 ` Allison Collins
2020-08-26 22:06 ` [PATCH 10/11] xfs: trace timestamp limits Darrick J. Wong
2020-08-27 7:01 ` Christoph Hellwig
2020-08-28 6:08 ` Allison Collins
2020-08-26 22:06 ` [PATCH 11/11] xfs: enable big timestamps Darrick J. Wong
2020-08-28 6:09 ` Allison Collins
2020-08-27 6:41 ` [PATCH v4 00/11] xfs: widen timestamps to deal with y2038 Christoph Hellwig
-- strict thread matches above, loose matches on Subject: below --
2020-08-31 6:06 [PATCH v5 " Darrick J. Wong
2020-08-31 6:07 ` [PATCH 08/11] xfs: widen ondisk inode timestamps to deal with y2038+ Darrick J. Wong
2020-08-31 16:08 ` Christoph Hellwig
2020-08-31 19:44 ` Darrick J. Wong
2020-09-01 11:44 ` Gao Xiang
2020-09-01 17:53 ` Darrick J. Wong
2020-09-02 2:56 [PATCH v6 00/11] xfs: widen timestamps to deal with y2038 Darrick J. Wong
2020-09-02 2:57 ` [PATCH 08/11] xfs: widen ondisk inode timestamps to deal with y2038+ 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=20200827065816.GB17534@infradead.org \
--to=hch@infradead.org \
--cc=amir73il@gmail.com \
--cc=darrick.wong@oracle.com \
--cc=david@fromorbit.com \
--cc=linux-xfs@vger.kernel.org \
--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