* [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-21 14:45 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Jeremy Kerr, Arnd Bergmann, Michael Ellerman, Nicholas Piggin,
Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Tom Talpey, Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
I've been working on a patchset to change how the inode->i_ctime is
accessed in order to give us conditional, high-res timestamps for the
ctime and mtime. struct timespec64 has unused bits in it that we can use
to implement this. In order to do that however, we need to wrap all
accesses of inode->i_ctime to ensure that bits used as flags are
appropriately handled.
This patchset first adds some new inode_ctime_* accessor functions. It
then converts all in-tree accesses of inode->i_ctime to use those new
functions and then renames the i_ctime field to __i_ctime to help ensure
that use of the accessors.
Most of this conversion was done via coccinelle, with a few of the more
non-standard accesses done by hand. There should be no behavioral
changes with this set. That will come later, as we convert individual
filesystems to use multigrain timestamps.
Some of these patches depend on the set I sent recently to add missing
ctime updates in various subsystems:
https://lore.kernel.org/linux-fsdevel/20230612104524.17058-1-jlayton@kernel.org/T/#m25399f903cc9526e46b2e0f5a35713c80b52fde9
Since this patchset is so large, I'm only going to send individual
conversion patches to the appropriate maintainers. Please send
Acked-by's or Reviewed-by's if you can. The intent is to merge these as
a set (probably in v6.6). Let me know if that causes conflicts though,
and we can work it out.
This is based on top of linux-next as of yesterday.
Jeff Layton (79):
fs: add ctime accessors infrastructure
spufs: switch to new ctime accessors
s390: switch to new ctime accessors
binderfs: switch to new ctime accessors
qib_fs: switch to new ctime accessors
ibm: switch to new ctime accessors
usb: switch to new ctime accessors
9p: switch to new ctime accessors
adfs: switch to new ctime accessors
affs: switch to new ctime accessors
afs: switch to new ctime accessors
fs: switch to new ctime accessors
autofs: switch to new ctime accessors
befs: switch to new ctime accessors
bfs: switch to new ctime accessors
btrfs: switch to new ctime accessors
ceph: switch to new ctime accessors
coda: switch to new ctime accessors
configfs: switch to new ctime accessors
cramfs: switch to new ctime accessors
debugfs: switch to new ctime accessors
devpts: switch to new ctime accessors
ecryptfs: switch to new ctime accessors
efivarfs: switch to new ctime accessors
efs: switch to new ctime accessors
erofs: switch to new ctime accessors
exfat: switch to new ctime accessors
ext2: switch to new ctime accessors
ext4: switch to new ctime accessors
f2fs: switch to new ctime accessors
fat: switch to new ctime accessors
freevxfs: switch to new ctime accessors
fuse: switch to new ctime accessors
gfs2: switch to new ctime accessors
hfs: switch to new ctime accessors
hfsplus: switch to new ctime accessors
hostfs: switch to new ctime accessors
hpfs: switch to new ctime accessors
hugetlbfs: switch to new ctime accessors
isofs: switch to new ctime accessors
jffs2: switch to new ctime accessors
jfs: switch to new ctime accessors
kernfs: switch to new ctime accessors
minix: switch to new ctime accessors
nfs: switch to new ctime accessors
nfsd: switch to new ctime accessors
nilfs2: switch to new ctime accessors
ntfs: switch to new ctime accessors
ntfs3: switch to new ctime accessors
ocfs2: switch to new ctime accessors
omfs: switch to new ctime accessors
openpromfs: switch to new ctime accessors
orangefs: switch to new ctime accessors
overlayfs: switch to new ctime accessors
proc: switch to new ctime accessors
pstore: switch to new ctime accessors
qnx4: switch to new ctime accessors
qnx6: switch to new ctime accessors
ramfs: switch to new ctime accessors
reiserfs: switch to new ctime accessors
romfs: switch to new ctime accessors
smb: switch to new ctime accessors
squashfs: switch to new ctime accessors
sysv: switch to new ctime accessors
tracefs: switch to new ctime accessors
ubifs: switch to new ctime accessors
udf: switch to new ctime accessors
ufs: switch to new ctime accessors
vboxsf: switch to new ctime accessors
xfs: switch to new ctime accessors
zonefs: switch to new ctime accessors
mqueue: switch to new ctime accessors
bpf: switch to new ctime accessors
shmem: switch to new ctime accessors
rpc_pipefs: switch to new ctime accessors
apparmor: switch to new ctime accessors
security: switch to new ctime accessors
selinux: switch to new ctime accessors
fs: rename i_ctime field to __i_ctime
arch/powerpc/platforms/cell/spufs/inode.c | 2 +-
arch/s390/hypfs/inode.c | 4 +-
drivers/android/binderfs.c | 8 +--
drivers/infiniband/hw/qib/qib_fs.c | 4 +-
drivers/misc/ibmasm/ibmasmfs.c | 2 +-
drivers/misc/ibmvmc.c | 2 +-
drivers/usb/core/devio.c | 16 +++---
drivers/usb/gadget/function/f_fs.c | 6 +--
drivers/usb/gadget/legacy/inode.c | 3 +-
fs/9p/vfs_inode.c | 6 ++-
fs/9p/vfs_inode_dotl.c | 11 ++--
fs/adfs/inode.c | 4 +-
fs/affs/amigaffs.c | 6 +--
fs/affs/inode.c | 17 +++---
fs/afs/dynroot.c | 2 +-
fs/afs/inode.c | 6 +--
fs/attr.c | 2 +-
fs/autofs/inode.c | 2 +-
fs/autofs/root.c | 6 +--
fs/bad_inode.c | 3 +-
fs/befs/linuxvfs.c | 2 +-
fs/bfs/dir.c | 16 +++---
fs/bfs/inode.c | 6 +--
fs/binfmt_misc.c | 3 +-
fs/btrfs/delayed-inode.c | 10 ++--
fs/btrfs/file.c | 24 +++------
fs/btrfs/inode.c | 66 +++++++++--------------
fs/btrfs/ioctl.c | 2 +-
fs/btrfs/reflink.c | 7 ++-
fs/btrfs/transaction.c | 3 +-
fs/btrfs/tree-log.c | 4 +-
fs/btrfs/xattr.c | 4 +-
fs/ceph/acl.c | 2 +-
fs/ceph/caps.c | 2 +-
fs/ceph/inode.c | 17 +++---
fs/ceph/snap.c | 2 +-
fs/ceph/xattr.c | 2 +-
fs/coda/coda_linux.c | 2 +-
fs/coda/dir.c | 2 +-
fs/coda/file.c | 2 +-
fs/coda/inode.c | 2 +-
fs/configfs/inode.c | 6 +--
fs/cramfs/inode.c | 2 +-
fs/debugfs/inode.c | 2 +-
fs/devpts/inode.c | 6 +--
fs/ecryptfs/inode.c | 2 +-
fs/efivarfs/file.c | 2 +-
fs/efivarfs/inode.c | 2 +-
fs/efs/inode.c | 5 +-
fs/erofs/inode.c | 16 +++---
fs/exfat/file.c | 4 +-
fs/exfat/inode.c | 6 +--
fs/exfat/namei.c | 29 +++++-----
fs/exfat/super.c | 4 +-
fs/ext2/acl.c | 2 +-
fs/ext2/dir.c | 6 +--
fs/ext2/ialloc.c | 2 +-
fs/ext2/inode.c | 11 ++--
fs/ext2/ioctl.c | 4 +-
fs/ext2/namei.c | 8 +--
fs/ext2/super.c | 2 +-
fs/ext2/xattr.c | 2 +-
fs/ext4/acl.c | 2 +-
fs/ext4/ext4.h | 20 +++++++
fs/ext4/extents.c | 12 ++---
fs/ext4/ialloc.c | 2 +-
fs/ext4/inline.c | 4 +-
fs/ext4/inode.c | 16 +++---
fs/ext4/ioctl.c | 9 ++--
fs/ext4/namei.c | 26 +++++----
fs/ext4/super.c | 2 +-
fs/ext4/xattr.c | 6 +--
fs/f2fs/dir.c | 8 +--
fs/f2fs/f2fs.h | 5 +-
fs/f2fs/file.c | 16 +++---
fs/f2fs/inline.c | 2 +-
fs/f2fs/inode.c | 10 ++--
fs/f2fs/namei.c | 12 ++---
fs/f2fs/recovery.c | 4 +-
fs/f2fs/super.c | 2 +-
fs/f2fs/xattr.c | 2 +-
fs/fat/inode.c | 8 +--
fs/fat/misc.c | 7 ++-
fs/freevxfs/vxfs_inode.c | 4 +-
fs/fuse/control.c | 2 +-
fs/fuse/dir.c | 8 +--
fs/fuse/inode.c | 18 ++++---
fs/gfs2/acl.c | 2 +-
fs/gfs2/bmap.c | 11 ++--
fs/gfs2/dir.c | 15 +++---
fs/gfs2/file.c | 2 +-
fs/gfs2/glops.c | 4 +-
fs/gfs2/inode.c | 8 +--
fs/gfs2/super.c | 4 +-
fs/gfs2/xattr.c | 8 +--
fs/hfs/catalog.c | 8 +--
fs/hfs/dir.c | 2 +-
fs/hfs/inode.c | 13 +++--
fs/hfs/sysdep.c | 2 +-
fs/hfsplus/catalog.c | 8 +--
fs/hfsplus/dir.c | 6 +--
fs/hfsplus/inode.c | 14 ++---
fs/hostfs/hostfs_kern.c | 4 +-
fs/hpfs/dir.c | 8 +--
fs/hpfs/inode.c | 6 +--
fs/hpfs/namei.c | 26 +++++----
fs/hpfs/super.c | 5 +-
fs/hugetlbfs/inode.c | 12 ++---
fs/inode.c | 26 +++++++--
fs/isofs/inode.c | 4 +-
fs/isofs/rock.c | 16 +++---
fs/jffs2/dir.c | 19 +++----
fs/jffs2/file.c | 3 +-
fs/jffs2/fs.c | 10 ++--
fs/jffs2/os-linux.h | 2 +-
fs/jfs/acl.c | 2 +-
fs/jfs/inode.c | 2 +-
fs/jfs/ioctl.c | 2 +-
fs/jfs/jfs_imap.c | 8 +--
fs/jfs/jfs_inode.c | 4 +-
fs/jfs/namei.c | 25 ++++-----
fs/jfs/super.c | 2 +-
fs/jfs/xattr.c | 2 +-
fs/kernfs/inode.c | 4 +-
fs/libfs.c | 32 +++++------
fs/minix/bitmap.c | 2 +-
fs/minix/dir.c | 6 +--
fs/minix/inode.c | 11 ++--
fs/minix/itree_common.c | 4 +-
fs/minix/namei.c | 6 +--
fs/nfs/callback_proc.c | 2 +-
fs/nfs/fscache.h | 4 +-
fs/nfs/inode.c | 21 ++++----
fs/nfsd/nfsctl.c | 2 +-
fs/nfsd/vfs.c | 2 +-
fs/nilfs2/dir.c | 6 +--
fs/nilfs2/inode.c | 12 ++---
fs/nilfs2/ioctl.c | 2 +-
fs/nilfs2/namei.c | 8 +--
fs/nsfs.c | 2 +-
fs/ntfs/inode.c | 15 +++---
fs/ntfs/mft.c | 3 +-
fs/ntfs3/file.c | 6 +--
fs/ntfs3/frecord.c | 4 +-
fs/ntfs3/inode.c | 14 ++---
fs/ntfs3/namei.c | 10 ++--
fs/ntfs3/xattr.c | 4 +-
fs/ocfs2/acl.c | 6 +--
fs/ocfs2/alloc.c | 6 +--
fs/ocfs2/aops.c | 2 +-
fs/ocfs2/dir.c | 8 +--
fs/ocfs2/dlmfs/dlmfs.c | 4 +-
fs/ocfs2/dlmglue.c | 10 ++--
fs/ocfs2/file.c | 16 +++---
fs/ocfs2/inode.c | 14 ++---
fs/ocfs2/move_extents.c | 6 +--
fs/ocfs2/namei.c | 22 ++++----
fs/ocfs2/refcounttree.c | 14 ++---
fs/ocfs2/xattr.c | 6 +--
fs/omfs/dir.c | 4 +-
fs/omfs/inode.c | 10 ++--
fs/openpromfs/inode.c | 4 +-
fs/orangefs/namei.c | 2 +-
fs/orangefs/orangefs-utils.c | 6 +--
fs/overlayfs/file.c | 7 ++-
fs/overlayfs/util.c | 2 +-
fs/pipe.c | 2 +-
fs/posix_acl.c | 2 +-
fs/proc/base.c | 2 +-
fs/proc/inode.c | 2 +-
fs/proc/proc_sysctl.c | 2 +-
fs/proc/self.c | 2 +-
fs/proc/thread_self.c | 2 +-
fs/pstore/inode.c | 4 +-
fs/qnx4/inode.c | 4 +-
fs/qnx6/inode.c | 4 +-
fs/ramfs/inode.c | 6 +--
fs/reiserfs/inode.c | 14 ++---
fs/reiserfs/ioctl.c | 4 +-
fs/reiserfs/namei.c | 21 ++++----
fs/reiserfs/stree.c | 4 +-
fs/reiserfs/super.c | 2 +-
fs/reiserfs/xattr.c | 5 +-
fs/reiserfs/xattr_acl.c | 2 +-
fs/romfs/super.c | 4 +-
fs/smb/client/file.c | 4 +-
fs/smb/client/fscache.h | 5 +-
fs/smb/client/inode.c | 15 +++---
fs/smb/client/smb2ops.c | 2 +-
fs/smb/server/smb2pdu.c | 8 +--
fs/squashfs/inode.c | 2 +-
fs/stack.c | 2 +-
fs/stat.c | 2 +-
fs/sysv/dir.c | 6 +--
fs/sysv/ialloc.c | 2 +-
fs/sysv/inode.c | 6 +--
fs/sysv/itree.c | 4 +-
fs/sysv/namei.c | 6 +--
fs/tracefs/inode.c | 2 +-
fs/ubifs/debug.c | 4 +-
fs/ubifs/dir.c | 39 +++++++-------
fs/ubifs/file.c | 16 +++---
fs/ubifs/ioctl.c | 2 +-
fs/ubifs/journal.c | 4 +-
fs/ubifs/super.c | 4 +-
fs/ubifs/xattr.c | 6 +--
fs/udf/ialloc.c | 2 +-
fs/udf/inode.c | 17 +++---
fs/udf/namei.c | 24 ++++-----
fs/ufs/dir.c | 6 +--
fs/ufs/ialloc.c | 2 +-
fs/ufs/inode.c | 23 ++++----
fs/ufs/namei.c | 8 +--
fs/vboxsf/utils.c | 4 +-
fs/xfs/libxfs/xfs_inode_buf.c | 4 +-
fs/xfs/libxfs/xfs_trans_inode.c | 2 +-
fs/xfs/xfs_acl.c | 2 +-
fs/xfs/xfs_bmap_util.c | 6 ++-
fs/xfs/xfs_inode.c | 2 +-
fs/xfs/xfs_inode_item.c | 2 +-
fs/xfs/xfs_iops.c | 4 +-
fs/xfs/xfs_itable.c | 4 +-
fs/zonefs/super.c | 8 +--
include/linux/fs.h | 55 ++++++++++++++++++-
include/linux/fs_stack.h | 2 +-
ipc/mqueue.c | 20 ++++---
kernel/bpf/inode.c | 4 +-
mm/shmem.c | 28 +++++-----
net/sunrpc/rpc_pipe.c | 2 +-
security/apparmor/apparmorfs.c | 6 +--
security/apparmor/policy_unpack.c | 4 +-
security/inode.c | 2 +-
security/selinux/selinuxfs.c | 2 +-
233 files changed, 914 insertions(+), 808 deletions(-)
--
2.41.0
^ permalink raw reply [flat|nested] 229+ messages in thread
* [f2fs-dev] [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-21 14:45 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Jeremy Kerr, Arnd Bergmann, Michael Ellerman, Nicholas Piggin,
Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Tom Talpey, Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
I've been working on a patchset to change how the inode->i_ctime is
accessed in order to give us conditional, high-res timestamps for the
ctime and mtime. struct timespec64 has unused bits in it that we can use
to implement this. In order to do that however, we need to wrap all
accesses of inode->i_ctime to ensure that bits used as flags are
appropriately handled.
This patchset first adds some new inode_ctime_* accessor functions. It
then converts all in-tree accesses of inode->i_ctime to use those new
functions and then renames the i_ctime field to __i_ctime to help ensure
that use of the accessors.
Most of this conversion was done via coccinelle, with a few of the more
non-standard accesses done by hand. There should be no behavioral
changes with this set. That will come later, as we convert individual
filesystems to use multigrain timestamps.
Some of these patches depend on the set I sent recently to add missing
ctime updates in various subsystems:
https://lore.kernel.org/linux-fsdevel/20230612104524.17058-1-jlayton@kernel.org/T/#m25399f903cc9526e46b2e0f5a35713c80b52fde9
Since this patchset is so large, I'm only going to send individual
conversion patches to the appropriate maintainers. Please send
Acked-by's or Reviewed-by's if you can. The intent is to merge these as
a set (probably in v6.6). Let me know if that causes conflicts though,
and we can work it out.
This is based on top of linux-next as of yesterday.
Jeff Layton (79):
fs: add ctime accessors infrastructure
spufs: switch to new ctime accessors
s390: switch to new ctime accessors
binderfs: switch to new ctime accessors
qib_fs: switch to new ctime accessors
ibm: switch to new ctime accessors
usb: switch to new ctime accessors
9p: switch to new ctime accessors
adfs: switch to new ctime accessors
affs: switch to new ctime accessors
afs: switch to new ctime accessors
fs: switch to new ctime accessors
autofs: switch to new ctime accessors
befs: switch to new ctime accessors
bfs: switch to new ctime accessors
btrfs: switch to new ctime accessors
ceph: switch to new ctime accessors
coda: switch to new ctime accessors
configfs: switch to new ctime accessors
cramfs: switch to new ctime accessors
debugfs: switch to new ctime accessors
devpts: switch to new ctime accessors
ecryptfs: switch to new ctime accessors
efivarfs: switch to new ctime accessors
efs: switch to new ctime accessors
erofs: switch to new ctime accessors
exfat: switch to new ctime accessors
ext2: switch to new ctime accessors
ext4: switch to new ctime accessors
f2fs: switch to new ctime accessors
fat: switch to new ctime accessors
freevxfs: switch to new ctime accessors
fuse: switch to new ctime accessors
gfs2: switch to new ctime accessors
hfs: switch to new ctime accessors
hfsplus: switch to new ctime accessors
hostfs: switch to new ctime accessors
hpfs: switch to new ctime accessors
hugetlbfs: switch to new ctime accessors
isofs: switch to new ctime accessors
jffs2: switch to new ctime accessors
jfs: switch to new ctime accessors
kernfs: switch to new ctime accessors
minix: switch to new ctime accessors
nfs: switch to new ctime accessors
nfsd: switch to new ctime accessors
nilfs2: switch to new ctime accessors
ntfs: switch to new ctime accessors
ntfs3: switch to new ctime accessors
ocfs2: switch to new ctime accessors
omfs: switch to new ctime accessors
openpromfs: switch to new ctime accessors
orangefs: switch to new ctime accessors
overlayfs: switch to new ctime accessors
proc: switch to new ctime accessors
pstore: switch to new ctime accessors
qnx4: switch to new ctime accessors
qnx6: switch to new ctime accessors
ramfs: switch to new ctime accessors
reiserfs: switch to new ctime accessors
romfs: switch to new ctime accessors
smb: switch to new ctime accessors
squashfs: switch to new ctime accessors
sysv: switch to new ctime accessors
tracefs: switch to new ctime accessors
ubifs: switch to new ctime accessors
udf: switch to new ctime accessors
ufs: switch to new ctime accessors
vboxsf: switch to new ctime accessors
xfs: switch to new ctime accessors
zonefs: switch to new ctime accessors
mqueue: switch to new ctime accessors
bpf: switch to new ctime accessors
shmem: switch to new ctime accessors
rpc_pipefs: switch to new ctime accessors
apparmor: switch to new ctime accessors
security: switch to new ctime accessors
selinux: switch to new ctime accessors
fs: rename i_ctime field to __i_ctime
arch/powerpc/platforms/cell/spufs/inode.c | 2 +-
arch/s390/hypfs/inode.c | 4 +-
drivers/android/binderfs.c | 8 +--
drivers/infiniband/hw/qib/qib_fs.c | 4 +-
drivers/misc/ibmasm/ibmasmfs.c | 2 +-
drivers/misc/ibmvmc.c | 2 +-
drivers/usb/core/devio.c | 16 +++---
drivers/usb/gadget/function/f_fs.c | 6 +--
drivers/usb/gadget/legacy/inode.c | 3 +-
fs/9p/vfs_inode.c | 6 ++-
fs/9p/vfs_inode_dotl.c | 11 ++--
fs/adfs/inode.c | 4 +-
fs/affs/amigaffs.c | 6 +--
fs/affs/inode.c | 17 +++---
fs/afs/dynroot.c | 2 +-
fs/afs/inode.c | 6 +--
fs/attr.c | 2 +-
fs/autofs/inode.c | 2 +-
fs/autofs/root.c | 6 +--
fs/bad_inode.c | 3 +-
fs/befs/linuxvfs.c | 2 +-
fs/bfs/dir.c | 16 +++---
fs/bfs/inode.c | 6 +--
fs/binfmt_misc.c | 3 +-
fs/btrfs/delayed-inode.c | 10 ++--
fs/btrfs/file.c | 24 +++------
fs/btrfs/inode.c | 66 +++++++++--------------
fs/btrfs/ioctl.c | 2 +-
fs/btrfs/reflink.c | 7 ++-
fs/btrfs/transaction.c | 3 +-
fs/btrfs/tree-log.c | 4 +-
fs/btrfs/xattr.c | 4 +-
fs/ceph/acl.c | 2 +-
fs/ceph/caps.c | 2 +-
fs/ceph/inode.c | 17 +++---
fs/ceph/snap.c | 2 +-
fs/ceph/xattr.c | 2 +-
fs/coda/coda_linux.c | 2 +-
fs/coda/dir.c | 2 +-
fs/coda/file.c | 2 +-
fs/coda/inode.c | 2 +-
fs/configfs/inode.c | 6 +--
fs/cramfs/inode.c | 2 +-
fs/debugfs/inode.c | 2 +-
fs/devpts/inode.c | 6 +--
fs/ecryptfs/inode.c | 2 +-
fs/efivarfs/file.c | 2 +-
fs/efivarfs/inode.c | 2 +-
fs/efs/inode.c | 5 +-
fs/erofs/inode.c | 16 +++---
fs/exfat/file.c | 4 +-
fs/exfat/inode.c | 6 +--
fs/exfat/namei.c | 29 +++++-----
fs/exfat/super.c | 4 +-
fs/ext2/acl.c | 2 +-
fs/ext2/dir.c | 6 +--
fs/ext2/ialloc.c | 2 +-
fs/ext2/inode.c | 11 ++--
fs/ext2/ioctl.c | 4 +-
fs/ext2/namei.c | 8 +--
fs/ext2/super.c | 2 +-
fs/ext2/xattr.c | 2 +-
fs/ext4/acl.c | 2 +-
fs/ext4/ext4.h | 20 +++++++
fs/ext4/extents.c | 12 ++---
fs/ext4/ialloc.c | 2 +-
fs/ext4/inline.c | 4 +-
fs/ext4/inode.c | 16 +++---
fs/ext4/ioctl.c | 9 ++--
fs/ext4/namei.c | 26 +++++----
fs/ext4/super.c | 2 +-
fs/ext4/xattr.c | 6 +--
fs/f2fs/dir.c | 8 +--
fs/f2fs/f2fs.h | 5 +-
fs/f2fs/file.c | 16 +++---
fs/f2fs/inline.c | 2 +-
fs/f2fs/inode.c | 10 ++--
fs/f2fs/namei.c | 12 ++---
fs/f2fs/recovery.c | 4 +-
fs/f2fs/super.c | 2 +-
fs/f2fs/xattr.c | 2 +-
fs/fat/inode.c | 8 +--
fs/fat/misc.c | 7 ++-
fs/freevxfs/vxfs_inode.c | 4 +-
fs/fuse/control.c | 2 +-
fs/fuse/dir.c | 8 +--
fs/fuse/inode.c | 18 ++++---
fs/gfs2/acl.c | 2 +-
fs/gfs2/bmap.c | 11 ++--
fs/gfs2/dir.c | 15 +++---
fs/gfs2/file.c | 2 +-
fs/gfs2/glops.c | 4 +-
fs/gfs2/inode.c | 8 +--
fs/gfs2/super.c | 4 +-
fs/gfs2/xattr.c | 8 +--
fs/hfs/catalog.c | 8 +--
fs/hfs/dir.c | 2 +-
fs/hfs/inode.c | 13 +++--
fs/hfs/sysdep.c | 2 +-
fs/hfsplus/catalog.c | 8 +--
fs/hfsplus/dir.c | 6 +--
fs/hfsplus/inode.c | 14 ++---
fs/hostfs/hostfs_kern.c | 4 +-
fs/hpfs/dir.c | 8 +--
fs/hpfs/inode.c | 6 +--
fs/hpfs/namei.c | 26 +++++----
fs/hpfs/super.c | 5 +-
fs/hugetlbfs/inode.c | 12 ++---
fs/inode.c | 26 +++++++--
fs/isofs/inode.c | 4 +-
fs/isofs/rock.c | 16 +++---
fs/jffs2/dir.c | 19 +++----
fs/jffs2/file.c | 3 +-
fs/jffs2/fs.c | 10 ++--
fs/jffs2/os-linux.h | 2 +-
fs/jfs/acl.c | 2 +-
fs/jfs/inode.c | 2 +-
fs/jfs/ioctl.c | 2 +-
fs/jfs/jfs_imap.c | 8 +--
fs/jfs/jfs_inode.c | 4 +-
fs/jfs/namei.c | 25 ++++-----
fs/jfs/super.c | 2 +-
fs/jfs/xattr.c | 2 +-
fs/kernfs/inode.c | 4 +-
fs/libfs.c | 32 +++++------
fs/minix/bitmap.c | 2 +-
fs/minix/dir.c | 6 +--
fs/minix/inode.c | 11 ++--
fs/minix/itree_common.c | 4 +-
fs/minix/namei.c | 6 +--
fs/nfs/callback_proc.c | 2 +-
fs/nfs/fscache.h | 4 +-
fs/nfs/inode.c | 21 ++++----
fs/nfsd/nfsctl.c | 2 +-
fs/nfsd/vfs.c | 2 +-
fs/nilfs2/dir.c | 6 +--
fs/nilfs2/inode.c | 12 ++---
fs/nilfs2/ioctl.c | 2 +-
fs/nilfs2/namei.c | 8 +--
fs/nsfs.c | 2 +-
fs/ntfs/inode.c | 15 +++---
fs/ntfs/mft.c | 3 +-
fs/ntfs3/file.c | 6 +--
fs/ntfs3/frecord.c | 4 +-
fs/ntfs3/inode.c | 14 ++---
fs/ntfs3/namei.c | 10 ++--
fs/ntfs3/xattr.c | 4 +-
fs/ocfs2/acl.c | 6 +--
fs/ocfs2/alloc.c | 6 +--
fs/ocfs2/aops.c | 2 +-
fs/ocfs2/dir.c | 8 +--
fs/ocfs2/dlmfs/dlmfs.c | 4 +-
fs/ocfs2/dlmglue.c | 10 ++--
fs/ocfs2/file.c | 16 +++---
fs/ocfs2/inode.c | 14 ++---
fs/ocfs2/move_extents.c | 6 +--
fs/ocfs2/namei.c | 22 ++++----
fs/ocfs2/refcounttree.c | 14 ++---
fs/ocfs2/xattr.c | 6 +--
fs/omfs/dir.c | 4 +-
fs/omfs/inode.c | 10 ++--
fs/openpromfs/inode.c | 4 +-
fs/orangefs/namei.c | 2 +-
fs/orangefs/orangefs-utils.c | 6 +--
fs/overlayfs/file.c | 7 ++-
fs/overlayfs/util.c | 2 +-
fs/pipe.c | 2 +-
fs/posix_acl.c | 2 +-
fs/proc/base.c | 2 +-
fs/proc/inode.c | 2 +-
fs/proc/proc_sysctl.c | 2 +-
fs/proc/self.c | 2 +-
fs/proc/thread_self.c | 2 +-
fs/pstore/inode.c | 4 +-
fs/qnx4/inode.c | 4 +-
fs/qnx6/inode.c | 4 +-
fs/ramfs/inode.c | 6 +--
fs/reiserfs/inode.c | 14 ++---
fs/reiserfs/ioctl.c | 4 +-
fs/reiserfs/namei.c | 21 ++++----
fs/reiserfs/stree.c | 4 +-
fs/reiserfs/super.c | 2 +-
fs/reiserfs/xattr.c | 5 +-
fs/reiserfs/xattr_acl.c | 2 +-
fs/romfs/super.c | 4 +-
fs/smb/client/file.c | 4 +-
fs/smb/client/fscache.h | 5 +-
fs/smb/client/inode.c | 15 +++---
fs/smb/client/smb2ops.c | 2 +-
fs/smb/server/smb2pdu.c | 8 +--
fs/squashfs/inode.c | 2 +-
fs/stack.c | 2 +-
fs/stat.c | 2 +-
fs/sysv/dir.c | 6 +--
fs/sysv/ialloc.c | 2 +-
fs/sysv/inode.c | 6 +--
fs/sysv/itree.c | 4 +-
fs/sysv/namei.c | 6 +--
fs/tracefs/inode.c | 2 +-
fs/ubifs/debug.c | 4 +-
fs/ubifs/dir.c | 39 +++++++-------
fs/ubifs/file.c | 16 +++---
fs/ubifs/ioctl.c | 2 +-
fs/ubifs/journal.c | 4 +-
fs/ubifs/super.c | 4 +-
fs/ubifs/xattr.c | 6 +--
fs/udf/ialloc.c | 2 +-
fs/udf/inode.c | 17 +++---
fs/udf/namei.c | 24 ++++-----
fs/ufs/dir.c | 6 +--
fs/ufs/ialloc.c | 2 +-
fs/ufs/inode.c | 23 ++++----
fs/ufs/namei.c | 8 +--
fs/vboxsf/utils.c | 4 +-
fs/xfs/libxfs/xfs_inode_buf.c | 4 +-
fs/xfs/libxfs/xfs_trans_inode.c | 2 +-
fs/xfs/xfs_acl.c | 2 +-
fs/xfs/xfs_bmap_util.c | 6 ++-
fs/xfs/xfs_inode.c | 2 +-
fs/xfs/xfs_inode_item.c | 2 +-
fs/xfs/xfs_iops.c | 4 +-
fs/xfs/xfs_itable.c | 4 +-
fs/zonefs/super.c | 8 +--
include/linux/fs.h | 55 ++++++++++++++++++-
include/linux/fs_stack.h | 2 +-
ipc/mqueue.c | 20 ++++---
kernel/bpf/inode.c | 4 +-
mm/shmem.c | 28 +++++-----
net/sunrpc/rpc_pipe.c | 2 +-
security/apparmor/apparmorfs.c | 6 +--
security/apparmor/policy_unpack.c | 4 +-
security/inode.c | 2 +-
security/selinux/selinuxfs.c | 2 +-
233 files changed, 914 insertions(+), 808 deletions(-)
--
2.41.0
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* [Ocfs2-devel] [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-21 14:45 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton via Ocfs2-devel @ 2023-06-21 14:45 UTC (permalink / raw)
To: Jeremy Kerr, Arnd Bergmann, Michael Ellerman, Nicholas Piggin,
Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Tom Talpey, Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
I've been working on a patchset to change how the inode->i_ctime is
accessed in order to give us conditional, high-res timestamps for the
ctime and mtime. struct timespec64 has unused bits in it that we can use
to implement this. In order to do that however, we need to wrap all
accesses of inode->i_ctime to ensure that bits used as flags are
appropriately handled.
This patchset first adds some new inode_ctime_* accessor functions. It
then converts all in-tree accesses of inode->i_ctime to use those new
functions and then renames the i_ctime field to __i_ctime to help ensure
that use of the accessors.
Most of this conversion was done via coccinelle, with a few of the more
non-standard accesses done by hand. There should be no behavioral
changes with this set. That will come later, as we convert individual
filesystems to use multigrain timestamps.
Some of these patches depend on the set I sent recently to add missing
ctime updates in various subsystems:
https://lore.kernel.org/linux-fsdevel/20230612104524.17058-1-jlayton@kernel.org/T/#m25399f903cc9526e46b2e0f5a35713c80b52fde9
Since this patchset is so large, I'm only going to send individual
conversion patches to the appropriate maintainers. Please send
Acked-by's or Reviewed-by's if you can. The intent is to merge these as
a set (probably in v6.6). Let me know if that causes conflicts though,
and we can work it out.
This is based on top of linux-next as of yesterday.
Jeff Layton (79):
fs: add ctime accessors infrastructure
spufs: switch to new ctime accessors
s390: switch to new ctime accessors
binderfs: switch to new ctime accessors
qib_fs: switch to new ctime accessors
ibm: switch to new ctime accessors
usb: switch to new ctime accessors
9p: switch to new ctime accessors
adfs: switch to new ctime accessors
affs: switch to new ctime accessors
afs: switch to new ctime accessors
fs: switch to new ctime accessors
autofs: switch to new ctime accessors
befs: switch to new ctime accessors
bfs: switch to new ctime accessors
btrfs: switch to new ctime accessors
ceph: switch to new ctime accessors
coda: switch to new ctime accessors
configfs: switch to new ctime accessors
cramfs: switch to new ctime accessors
debugfs: switch to new ctime accessors
devpts: switch to new ctime accessors
ecryptfs: switch to new ctime accessors
efivarfs: switch to new ctime accessors
efs: switch to new ctime accessors
erofs: switch to new ctime accessors
exfat: switch to new ctime accessors
ext2: switch to new ctime accessors
ext4: switch to new ctime accessors
f2fs: switch to new ctime accessors
fat: switch to new ctime accessors
freevxfs: switch to new ctime accessors
fuse: switch to new ctime accessors
gfs2: switch to new ctime accessors
hfs: switch to new ctime accessors
hfsplus: switch to new ctime accessors
hostfs: switch to new ctime accessors
hpfs: switch to new ctime accessors
hugetlbfs: switch to new ctime accessors
isofs: switch to new ctime accessors
jffs2: switch to new ctime accessors
jfs: switch to new ctime accessors
kernfs: switch to new ctime accessors
minix: switch to new ctime accessors
nfs: switch to new ctime accessors
nfsd: switch to new ctime accessors
nilfs2: switch to new ctime accessors
ntfs: switch to new ctime accessors
ntfs3: switch to new ctime accessors
ocfs2: switch to new ctime accessors
omfs: switch to new ctime accessors
openpromfs: switch to new ctime accessors
orangefs: switch to new ctime accessors
overlayfs: switch to new ctime accessors
proc: switch to new ctime accessors
pstore: switch to new ctime accessors
qnx4: switch to new ctime accessors
qnx6: switch to new ctime accessors
ramfs: switch to new ctime accessors
reiserfs: switch to new ctime accessors
romfs: switch to new ctime accessors
smb: switch to new ctime accessors
squashfs: switch to new ctime accessors
sysv: switch to new ctime accessors
tracefs: switch to new ctime accessors
ubifs: switch to new ctime accessors
udf: switch to new ctime accessors
ufs: switch to new ctime accessors
vboxsf: switch to new ctime accessors
xfs: switch to new ctime accessors
zonefs: switch to new ctime accessors
mqueue: switch to new ctime accessors
bpf: switch to new ctime accessors
shmem: switch to new ctime accessors
rpc_pipefs: switch to new ctime accessors
apparmor: switch to new ctime accessors
security: switch to new ctime accessors
selinux: switch to new ctime accessors
fs: rename i_ctime field to __i_ctime
arch/powerpc/platforms/cell/spufs/inode.c | 2 +-
arch/s390/hypfs/inode.c | 4 +-
drivers/android/binderfs.c | 8 +--
drivers/infiniband/hw/qib/qib_fs.c | 4 +-
drivers/misc/ibmasm/ibmasmfs.c | 2 +-
drivers/misc/ibmvmc.c | 2 +-
drivers/usb/core/devio.c | 16 +++---
drivers/usb/gadget/function/f_fs.c | 6 +--
drivers/usb/gadget/legacy/inode.c | 3 +-
fs/9p/vfs_inode.c | 6 ++-
fs/9p/vfs_inode_dotl.c | 11 ++--
fs/adfs/inode.c | 4 +-
fs/affs/amigaffs.c | 6 +--
fs/affs/inode.c | 17 +++---
fs/afs/dynroot.c | 2 +-
fs/afs/inode.c | 6 +--
fs/attr.c | 2 +-
fs/autofs/inode.c | 2 +-
fs/autofs/root.c | 6 +--
fs/bad_inode.c | 3 +-
fs/befs/linuxvfs.c | 2 +-
fs/bfs/dir.c | 16 +++---
fs/bfs/inode.c | 6 +--
fs/binfmt_misc.c | 3 +-
fs/btrfs/delayed-inode.c | 10 ++--
fs/btrfs/file.c | 24 +++------
fs/btrfs/inode.c | 66 +++++++++--------------
fs/btrfs/ioctl.c | 2 +-
fs/btrfs/reflink.c | 7 ++-
fs/btrfs/transaction.c | 3 +-
fs/btrfs/tree-log.c | 4 +-
fs/btrfs/xattr.c | 4 +-
fs/ceph/acl.c | 2 +-
fs/ceph/caps.c | 2 +-
fs/ceph/inode.c | 17 +++---
fs/ceph/snap.c | 2 +-
fs/ceph/xattr.c | 2 +-
fs/coda/coda_linux.c | 2 +-
fs/coda/dir.c | 2 +-
fs/coda/file.c | 2 +-
fs/coda/inode.c | 2 +-
fs/configfs/inode.c | 6 +--
fs/cramfs/inode.c | 2 +-
fs/debugfs/inode.c | 2 +-
fs/devpts/inode.c | 6 +--
fs/ecryptfs/inode.c | 2 +-
fs/efivarfs/file.c | 2 +-
fs/efivarfs/inode.c | 2 +-
fs/efs/inode.c | 5 +-
fs/erofs/inode.c | 16 +++---
fs/exfat/file.c | 4 +-
fs/exfat/inode.c | 6 +--
fs/exfat/namei.c | 29 +++++-----
fs/exfat/super.c | 4 +-
fs/ext2/acl.c | 2 +-
fs/ext2/dir.c | 6 +--
fs/ext2/ialloc.c | 2 +-
fs/ext2/inode.c | 11 ++--
fs/ext2/ioctl.c | 4 +-
fs/ext2/namei.c | 8 +--
fs/ext2/super.c | 2 +-
fs/ext2/xattr.c | 2 +-
fs/ext4/acl.c | 2 +-
fs/ext4/ext4.h | 20 +++++++
fs/ext4/extents.c | 12 ++---
fs/ext4/ialloc.c | 2 +-
fs/ext4/inline.c | 4 +-
fs/ext4/inode.c | 16 +++---
fs/ext4/ioctl.c | 9 ++--
fs/ext4/namei.c | 26 +++++----
fs/ext4/super.c | 2 +-
fs/ext4/xattr.c | 6 +--
fs/f2fs/dir.c | 8 +--
fs/f2fs/f2fs.h | 5 +-
fs/f2fs/file.c | 16 +++---
fs/f2fs/inline.c | 2 +-
fs/f2fs/inode.c | 10 ++--
fs/f2fs/namei.c | 12 ++---
fs/f2fs/recovery.c | 4 +-
fs/f2fs/super.c | 2 +-
fs/f2fs/xattr.c | 2 +-
fs/fat/inode.c | 8 +--
fs/fat/misc.c | 7 ++-
fs/freevxfs/vxfs_inode.c | 4 +-
fs/fuse/control.c | 2 +-
fs/fuse/dir.c | 8 +--
fs/fuse/inode.c | 18 ++++---
fs/gfs2/acl.c | 2 +-
fs/gfs2/bmap.c | 11 ++--
fs/gfs2/dir.c | 15 +++---
fs/gfs2/file.c | 2 +-
fs/gfs2/glops.c | 4 +-
fs/gfs2/inode.c | 8 +--
fs/gfs2/super.c | 4 +-
fs/gfs2/xattr.c | 8 +--
fs/hfs/catalog.c | 8 +--
fs/hfs/dir.c | 2 +-
fs/hfs/inode.c | 13 +++--
fs/hfs/sysdep.c | 2 +-
fs/hfsplus/catalog.c | 8 +--
fs/hfsplus/dir.c | 6 +--
fs/hfsplus/inode.c | 14 ++---
fs/hostfs/hostfs_kern.c | 4 +-
fs/hpfs/dir.c | 8 +--
fs/hpfs/inode.c | 6 +--
fs/hpfs/namei.c | 26 +++++----
fs/hpfs/super.c | 5 +-
fs/hugetlbfs/inode.c | 12 ++---
fs/inode.c | 26 +++++++--
fs/isofs/inode.c | 4 +-
fs/isofs/rock.c | 16 +++---
fs/jffs2/dir.c | 19 +++----
fs/jffs2/file.c | 3 +-
fs/jffs2/fs.c | 10 ++--
fs/jffs2/os-linux.h | 2 +-
fs/jfs/acl.c | 2 +-
fs/jfs/inode.c | 2 +-
fs/jfs/ioctl.c | 2 +-
fs/jfs/jfs_imap.c | 8 +--
fs/jfs/jfs_inode.c | 4 +-
fs/jfs/namei.c | 25 ++++-----
fs/jfs/super.c | 2 +-
fs/jfs/xattr.c | 2 +-
fs/kernfs/inode.c | 4 +-
fs/libfs.c | 32 +++++------
fs/minix/bitmap.c | 2 +-
fs/minix/dir.c | 6 +--
fs/minix/inode.c | 11 ++--
fs/minix/itree_common.c | 4 +-
fs/minix/namei.c | 6 +--
fs/nfs/callback_proc.c | 2 +-
fs/nfs/fscache.h | 4 +-
fs/nfs/inode.c | 21 ++++----
fs/nfsd/nfsctl.c | 2 +-
fs/nfsd/vfs.c | 2 +-
fs/nilfs2/dir.c | 6 +--
fs/nilfs2/inode.c | 12 ++---
fs/nilfs2/ioctl.c | 2 +-
fs/nilfs2/namei.c | 8 +--
fs/nsfs.c | 2 +-
fs/ntfs/inode.c | 15 +++---
fs/ntfs/mft.c | 3 +-
fs/ntfs3/file.c | 6 +--
fs/ntfs3/frecord.c | 4 +-
fs/ntfs3/inode.c | 14 ++---
fs/ntfs3/namei.c | 10 ++--
fs/ntfs3/xattr.c | 4 +-
fs/ocfs2/acl.c | 6 +--
fs/ocfs2/alloc.c | 6 +--
fs/ocfs2/aops.c | 2 +-
fs/ocfs2/dir.c | 8 +--
fs/ocfs2/dlmfs/dlmfs.c | 4 +-
fs/ocfs2/dlmglue.c | 10 ++--
fs/ocfs2/file.c | 16 +++---
fs/ocfs2/inode.c | 14 ++---
fs/ocfs2/move_extents.c | 6 +--
fs/ocfs2/namei.c | 22 ++++----
fs/ocfs2/refcounttree.c | 14 ++---
fs/ocfs2/xattr.c | 6 +--
fs/omfs/dir.c | 4 +-
fs/omfs/inode.c | 10 ++--
fs/openpromfs/inode.c | 4 +-
fs/orangefs/namei.c | 2 +-
fs/orangefs/orangefs-utils.c | 6 +--
fs/overlayfs/file.c | 7 ++-
fs/overlayfs/util.c | 2 +-
fs/pipe.c | 2 +-
fs/posix_acl.c | 2 +-
fs/proc/base.c | 2 +-
fs/proc/inode.c | 2 +-
fs/proc/proc_sysctl.c | 2 +-
fs/proc/self.c | 2 +-
fs/proc/thread_self.c | 2 +-
fs/pstore/inode.c | 4 +-
fs/qnx4/inode.c | 4 +-
fs/qnx6/inode.c | 4 +-
fs/ramfs/inode.c | 6 +--
fs/reiserfs/inode.c | 14 ++---
fs/reiserfs/ioctl.c | 4 +-
fs/reiserfs/namei.c | 21 ++++----
fs/reiserfs/stree.c | 4 +-
fs/reiserfs/super.c | 2 +-
fs/reiserfs/xattr.c | 5 +-
fs/reiserfs/xattr_acl.c | 2 +-
fs/romfs/super.c | 4 +-
fs/smb/client/file.c | 4 +-
fs/smb/client/fscache.h | 5 +-
fs/smb/client/inode.c | 15 +++---
fs/smb/client/smb2ops.c | 2 +-
fs/smb/server/smb2pdu.c | 8 +--
fs/squashfs/inode.c | 2 +-
fs/stack.c | 2 +-
fs/stat.c | 2 +-
fs/sysv/dir.c | 6 +--
fs/sysv/ialloc.c | 2 +-
fs/sysv/inode.c | 6 +--
fs/sysv/itree.c | 4 +-
fs/sysv/namei.c | 6 +--
fs/tracefs/inode.c | 2 +-
fs/ubifs/debug.c | 4 +-
fs/ubifs/dir.c | 39 +++++++-------
fs/ubifs/file.c | 16 +++---
fs/ubifs/ioctl.c | 2 +-
fs/ubifs/journal.c | 4 +-
fs/ubifs/super.c | 4 +-
fs/ubifs/xattr.c | 6 +--
fs/udf/ialloc.c | 2 +-
fs/udf/inode.c | 17 +++---
fs/udf/namei.c | 24 ++++-----
fs/ufs/dir.c | 6 +--
fs/ufs/ialloc.c | 2 +-
fs/ufs/inode.c | 23 ++++----
fs/ufs/namei.c | 8 +--
fs/vboxsf/utils.c | 4 +-
fs/xfs/libxfs/xfs_inode_buf.c | 4 +-
fs/xfs/libxfs/xfs_trans_inode.c | 2 +-
fs/xfs/xfs_acl.c | 2 +-
fs/xfs/xfs_bmap_util.c | 6 ++-
fs/xfs/xfs_inode.c | 2 +-
fs/xfs/xfs_inode_item.c | 2 +-
fs/xfs/xfs_iops.c | 4 +-
fs/xfs/xfs_itable.c | 4 +-
fs/zonefs/super.c | 8 +--
include/linux/fs.h | 55 ++++++++++++++++++-
include/linux/fs_stack.h | 2 +-
ipc/mqueue.c | 20 ++++---
kernel/bpf/inode.c | 4 +-
mm/shmem.c | 28 +++++-----
net/sunrpc/rpc_pipe.c | 2 +-
security/apparmor/apparmorfs.c | 6 +--
security/apparmor/policy_unpack.c | 4 +-
security/inode.c | 2 +-
security/selinux/selinuxfs.c | 2 +-
233 files changed, 914 insertions(+), 808 deletions(-)
--
2.41.0
_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* [Cluster-devel] [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-21 14:45 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: cluster-devel.redhat.com
I've been working on a patchset to change how the inode->i_ctime is
accessed in order to give us conditional, high-res timestamps for the
ctime and mtime. struct timespec64 has unused bits in it that we can use
to implement this. In order to do that however, we need to wrap all
accesses of inode->i_ctime to ensure that bits used as flags are
appropriately handled.
This patchset first adds some new inode_ctime_* accessor functions. It
then converts all in-tree accesses of inode->i_ctime to use those new
functions and then renames the i_ctime field to __i_ctime to help ensure
that use of the accessors.
Most of this conversion was done via coccinelle, with a few of the more
non-standard accesses done by hand. There should be no behavioral
changes with this set. That will come later, as we convert individual
filesystems to use multigrain timestamps.
Some of these patches depend on the set I sent recently to add missing
ctime updates in various subsystems:
https://lore.kernel.org/linux-fsdevel/20230612104524.17058-1-jlayton at kernel.org/T/#m25399f903cc9526e46b2e0f5a35713c80b52fde9
Since this patchset is so large, I'm only going to send individual
conversion patches to the appropriate maintainers. Please send
Acked-by's or Reviewed-by's if you can. The intent is to merge these as
a set (probably in v6.6). Let me know if that causes conflicts though,
and we can work it out.
This is based on top of linux-next as of yesterday.
Jeff Layton (79):
fs: add ctime accessors infrastructure
spufs: switch to new ctime accessors
s390: switch to new ctime accessors
binderfs: switch to new ctime accessors
qib_fs: switch to new ctime accessors
ibm: switch to new ctime accessors
usb: switch to new ctime accessors
9p: switch to new ctime accessors
adfs: switch to new ctime accessors
affs: switch to new ctime accessors
afs: switch to new ctime accessors
fs: switch to new ctime accessors
autofs: switch to new ctime accessors
befs: switch to new ctime accessors
bfs: switch to new ctime accessors
btrfs: switch to new ctime accessors
ceph: switch to new ctime accessors
coda: switch to new ctime accessors
configfs: switch to new ctime accessors
cramfs: switch to new ctime accessors
debugfs: switch to new ctime accessors
devpts: switch to new ctime accessors
ecryptfs: switch to new ctime accessors
efivarfs: switch to new ctime accessors
efs: switch to new ctime accessors
erofs: switch to new ctime accessors
exfat: switch to new ctime accessors
ext2: switch to new ctime accessors
ext4: switch to new ctime accessors
f2fs: switch to new ctime accessors
fat: switch to new ctime accessors
freevxfs: switch to new ctime accessors
fuse: switch to new ctime accessors
gfs2: switch to new ctime accessors
hfs: switch to new ctime accessors
hfsplus: switch to new ctime accessors
hostfs: switch to new ctime accessors
hpfs: switch to new ctime accessors
hugetlbfs: switch to new ctime accessors
isofs: switch to new ctime accessors
jffs2: switch to new ctime accessors
jfs: switch to new ctime accessors
kernfs: switch to new ctime accessors
minix: switch to new ctime accessors
nfs: switch to new ctime accessors
nfsd: switch to new ctime accessors
nilfs2: switch to new ctime accessors
ntfs: switch to new ctime accessors
ntfs3: switch to new ctime accessors
ocfs2: switch to new ctime accessors
omfs: switch to new ctime accessors
openpromfs: switch to new ctime accessors
orangefs: switch to new ctime accessors
overlayfs: switch to new ctime accessors
proc: switch to new ctime accessors
pstore: switch to new ctime accessors
qnx4: switch to new ctime accessors
qnx6: switch to new ctime accessors
ramfs: switch to new ctime accessors
reiserfs: switch to new ctime accessors
romfs: switch to new ctime accessors
smb: switch to new ctime accessors
squashfs: switch to new ctime accessors
sysv: switch to new ctime accessors
tracefs: switch to new ctime accessors
ubifs: switch to new ctime accessors
udf: switch to new ctime accessors
ufs: switch to new ctime accessors
vboxsf: switch to new ctime accessors
xfs: switch to new ctime accessors
zonefs: switch to new ctime accessors
mqueue: switch to new ctime accessors
bpf: switch to new ctime accessors
shmem: switch to new ctime accessors
rpc_pipefs: switch to new ctime accessors
apparmor: switch to new ctime accessors
security: switch to new ctime accessors
selinux: switch to new ctime accessors
fs: rename i_ctime field to __i_ctime
arch/powerpc/platforms/cell/spufs/inode.c | 2 +-
arch/s390/hypfs/inode.c | 4 +-
drivers/android/binderfs.c | 8 +--
drivers/infiniband/hw/qib/qib_fs.c | 4 +-
drivers/misc/ibmasm/ibmasmfs.c | 2 +-
drivers/misc/ibmvmc.c | 2 +-
drivers/usb/core/devio.c | 16 +++---
drivers/usb/gadget/function/f_fs.c | 6 +--
drivers/usb/gadget/legacy/inode.c | 3 +-
fs/9p/vfs_inode.c | 6 ++-
fs/9p/vfs_inode_dotl.c | 11 ++--
fs/adfs/inode.c | 4 +-
fs/affs/amigaffs.c | 6 +--
fs/affs/inode.c | 17 +++---
fs/afs/dynroot.c | 2 +-
fs/afs/inode.c | 6 +--
fs/attr.c | 2 +-
fs/autofs/inode.c | 2 +-
fs/autofs/root.c | 6 +--
fs/bad_inode.c | 3 +-
fs/befs/linuxvfs.c | 2 +-
fs/bfs/dir.c | 16 +++---
fs/bfs/inode.c | 6 +--
fs/binfmt_misc.c | 3 +-
fs/btrfs/delayed-inode.c | 10 ++--
fs/btrfs/file.c | 24 +++------
fs/btrfs/inode.c | 66 +++++++++--------------
fs/btrfs/ioctl.c | 2 +-
fs/btrfs/reflink.c | 7 ++-
fs/btrfs/transaction.c | 3 +-
fs/btrfs/tree-log.c | 4 +-
fs/btrfs/xattr.c | 4 +-
fs/ceph/acl.c | 2 +-
fs/ceph/caps.c | 2 +-
fs/ceph/inode.c | 17 +++---
fs/ceph/snap.c | 2 +-
fs/ceph/xattr.c | 2 +-
fs/coda/coda_linux.c | 2 +-
fs/coda/dir.c | 2 +-
fs/coda/file.c | 2 +-
fs/coda/inode.c | 2 +-
fs/configfs/inode.c | 6 +--
fs/cramfs/inode.c | 2 +-
fs/debugfs/inode.c | 2 +-
fs/devpts/inode.c | 6 +--
fs/ecryptfs/inode.c | 2 +-
fs/efivarfs/file.c | 2 +-
fs/efivarfs/inode.c | 2 +-
fs/efs/inode.c | 5 +-
fs/erofs/inode.c | 16 +++---
fs/exfat/file.c | 4 +-
fs/exfat/inode.c | 6 +--
fs/exfat/namei.c | 29 +++++-----
fs/exfat/super.c | 4 +-
fs/ext2/acl.c | 2 +-
fs/ext2/dir.c | 6 +--
fs/ext2/ialloc.c | 2 +-
fs/ext2/inode.c | 11 ++--
fs/ext2/ioctl.c | 4 +-
fs/ext2/namei.c | 8 +--
fs/ext2/super.c | 2 +-
fs/ext2/xattr.c | 2 +-
fs/ext4/acl.c | 2 +-
fs/ext4/ext4.h | 20 +++++++
fs/ext4/extents.c | 12 ++---
fs/ext4/ialloc.c | 2 +-
fs/ext4/inline.c | 4 +-
fs/ext4/inode.c | 16 +++---
fs/ext4/ioctl.c | 9 ++--
fs/ext4/namei.c | 26 +++++----
fs/ext4/super.c | 2 +-
fs/ext4/xattr.c | 6 +--
fs/f2fs/dir.c | 8 +--
fs/f2fs/f2fs.h | 5 +-
fs/f2fs/file.c | 16 +++---
fs/f2fs/inline.c | 2 +-
fs/f2fs/inode.c | 10 ++--
fs/f2fs/namei.c | 12 ++---
fs/f2fs/recovery.c | 4 +-
fs/f2fs/super.c | 2 +-
fs/f2fs/xattr.c | 2 +-
fs/fat/inode.c | 8 +--
fs/fat/misc.c | 7 ++-
fs/freevxfs/vxfs_inode.c | 4 +-
fs/fuse/control.c | 2 +-
fs/fuse/dir.c | 8 +--
fs/fuse/inode.c | 18 ++++---
fs/gfs2/acl.c | 2 +-
fs/gfs2/bmap.c | 11 ++--
fs/gfs2/dir.c | 15 +++---
fs/gfs2/file.c | 2 +-
fs/gfs2/glops.c | 4 +-
fs/gfs2/inode.c | 8 +--
fs/gfs2/super.c | 4 +-
fs/gfs2/xattr.c | 8 +--
fs/hfs/catalog.c | 8 +--
fs/hfs/dir.c | 2 +-
fs/hfs/inode.c | 13 +++--
fs/hfs/sysdep.c | 2 +-
fs/hfsplus/catalog.c | 8 +--
fs/hfsplus/dir.c | 6 +--
fs/hfsplus/inode.c | 14 ++---
fs/hostfs/hostfs_kern.c | 4 +-
fs/hpfs/dir.c | 8 +--
fs/hpfs/inode.c | 6 +--
fs/hpfs/namei.c | 26 +++++----
fs/hpfs/super.c | 5 +-
fs/hugetlbfs/inode.c | 12 ++---
fs/inode.c | 26 +++++++--
fs/isofs/inode.c | 4 +-
fs/isofs/rock.c | 16 +++---
fs/jffs2/dir.c | 19 +++----
fs/jffs2/file.c | 3 +-
fs/jffs2/fs.c | 10 ++--
fs/jffs2/os-linux.h | 2 +-
fs/jfs/acl.c | 2 +-
fs/jfs/inode.c | 2 +-
fs/jfs/ioctl.c | 2 +-
fs/jfs/jfs_imap.c | 8 +--
fs/jfs/jfs_inode.c | 4 +-
fs/jfs/namei.c | 25 ++++-----
fs/jfs/super.c | 2 +-
fs/jfs/xattr.c | 2 +-
fs/kernfs/inode.c | 4 +-
fs/libfs.c | 32 +++++------
fs/minix/bitmap.c | 2 +-
fs/minix/dir.c | 6 +--
fs/minix/inode.c | 11 ++--
fs/minix/itree_common.c | 4 +-
fs/minix/namei.c | 6 +--
fs/nfs/callback_proc.c | 2 +-
fs/nfs/fscache.h | 4 +-
fs/nfs/inode.c | 21 ++++----
fs/nfsd/nfsctl.c | 2 +-
fs/nfsd/vfs.c | 2 +-
fs/nilfs2/dir.c | 6 +--
fs/nilfs2/inode.c | 12 ++---
fs/nilfs2/ioctl.c | 2 +-
fs/nilfs2/namei.c | 8 +--
fs/nsfs.c | 2 +-
fs/ntfs/inode.c | 15 +++---
fs/ntfs/mft.c | 3 +-
fs/ntfs3/file.c | 6 +--
fs/ntfs3/frecord.c | 4 +-
fs/ntfs3/inode.c | 14 ++---
fs/ntfs3/namei.c | 10 ++--
fs/ntfs3/xattr.c | 4 +-
fs/ocfs2/acl.c | 6 +--
fs/ocfs2/alloc.c | 6 +--
fs/ocfs2/aops.c | 2 +-
fs/ocfs2/dir.c | 8 +--
fs/ocfs2/dlmfs/dlmfs.c | 4 +-
fs/ocfs2/dlmglue.c | 10 ++--
fs/ocfs2/file.c | 16 +++---
fs/ocfs2/inode.c | 14 ++---
fs/ocfs2/move_extents.c | 6 +--
fs/ocfs2/namei.c | 22 ++++----
fs/ocfs2/refcounttree.c | 14 ++---
fs/ocfs2/xattr.c | 6 +--
fs/omfs/dir.c | 4 +-
fs/omfs/inode.c | 10 ++--
fs/openpromfs/inode.c | 4 +-
fs/orangefs/namei.c | 2 +-
fs/orangefs/orangefs-utils.c | 6 +--
fs/overlayfs/file.c | 7 ++-
fs/overlayfs/util.c | 2 +-
fs/pipe.c | 2 +-
fs/posix_acl.c | 2 +-
fs/proc/base.c | 2 +-
fs/proc/inode.c | 2 +-
fs/proc/proc_sysctl.c | 2 +-
fs/proc/self.c | 2 +-
fs/proc/thread_self.c | 2 +-
fs/pstore/inode.c | 4 +-
fs/qnx4/inode.c | 4 +-
fs/qnx6/inode.c | 4 +-
fs/ramfs/inode.c | 6 +--
fs/reiserfs/inode.c | 14 ++---
fs/reiserfs/ioctl.c | 4 +-
fs/reiserfs/namei.c | 21 ++++----
fs/reiserfs/stree.c | 4 +-
fs/reiserfs/super.c | 2 +-
fs/reiserfs/xattr.c | 5 +-
fs/reiserfs/xattr_acl.c | 2 +-
fs/romfs/super.c | 4 +-
fs/smb/client/file.c | 4 +-
fs/smb/client/fscache.h | 5 +-
fs/smb/client/inode.c | 15 +++---
fs/smb/client/smb2ops.c | 2 +-
fs/smb/server/smb2pdu.c | 8 +--
fs/squashfs/inode.c | 2 +-
fs/stack.c | 2 +-
fs/stat.c | 2 +-
fs/sysv/dir.c | 6 +--
fs/sysv/ialloc.c | 2 +-
fs/sysv/inode.c | 6 +--
fs/sysv/itree.c | 4 +-
fs/sysv/namei.c | 6 +--
fs/tracefs/inode.c | 2 +-
fs/ubifs/debug.c | 4 +-
fs/ubifs/dir.c | 39 +++++++-------
fs/ubifs/file.c | 16 +++---
fs/ubifs/ioctl.c | 2 +-
fs/ubifs/journal.c | 4 +-
fs/ubifs/super.c | 4 +-
fs/ubifs/xattr.c | 6 +--
fs/udf/ialloc.c | 2 +-
fs/udf/inode.c | 17 +++---
fs/udf/namei.c | 24 ++++-----
fs/ufs/dir.c | 6 +--
fs/ufs/ialloc.c | 2 +-
fs/ufs/inode.c | 23 ++++----
fs/ufs/namei.c | 8 +--
fs/vboxsf/utils.c | 4 +-
fs/xfs/libxfs/xfs_inode_buf.c | 4 +-
fs/xfs/libxfs/xfs_trans_inode.c | 2 +-
fs/xfs/xfs_acl.c | 2 +-
fs/xfs/xfs_bmap_util.c | 6 ++-
fs/xfs/xfs_inode.c | 2 +-
fs/xfs/xfs_inode_item.c | 2 +-
fs/xfs/xfs_iops.c | 4 +-
fs/xfs/xfs_itable.c | 4 +-
fs/zonefs/super.c | 8 +--
include/linux/fs.h | 55 ++++++++++++++++++-
include/linux/fs_stack.h | 2 +-
ipc/mqueue.c | 20 ++++---
kernel/bpf/inode.c | 4 +-
mm/shmem.c | 28 +++++-----
net/sunrpc/rpc_pipe.c | 2 +-
security/apparmor/apparmorfs.c | 6 +--
security/apparmor/policy_unpack.c | 4 +-
security/inode.c | 2 +-
security/selinux/selinuxfs.c | 2 +-
233 files changed, 914 insertions(+), 808 deletions(-)
--
2.41.0
^ permalink raw reply [flat|nested] 229+ messages in thread
* [PATCH 01/79] fs: add ctime accessors infrastructure
2023-06-21 14:45 ` [f2fs-dev] " Jeff Layton
(?)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Jeremy Kerr, Arnd Bergmann, Michael Ellerman, Nicholas Piggin,
Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Tom Talpey, Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
struct timespec64 has unused bits in the tv_nsec field that can be used
for other purposes. In future patches, we're going to change how the
inode->i_ctime is accessed in certain inodes in order to make use of
them. In order to do that safely though, we'll need to eradicate raw
accesses of the inode->i_ctime field from the kernel.
Add new accessor functions for the ctime that we can use to replace them.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/inode.c | 16 ++++++++++++++
include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
2 files changed, 68 insertions(+), 1 deletion(-)
diff --git a/fs/inode.c b/fs/inode.c
index d37fad91c8da..c005e7328fbb 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
}
EXPORT_SYMBOL(current_time);
+/**
+ * inode_ctime_set_current - set the ctime to current_time
+ * @inode: inode
+ *
+ * Set the inode->i_ctime to the current value for the inode. Returns
+ * the current value that was assigned to i_ctime.
+ */
+struct timespec64 inode_ctime_set_current(struct inode *inode)
+{
+ struct timespec64 now = current_time(inode);
+
+ inode_set_ctime(inode, now);
+ return now;
+}
+EXPORT_SYMBOL(inode_ctime_set_current);
+
/**
* in_group_or_capable - check whether caller is CAP_FSETID privileged
* @idmap: idmap of the mount @inode was found from
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 6867512907d6..9afb30606373 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
kgid_has_mapping(fs_userns, kgid);
}
-extern struct timespec64 current_time(struct inode *inode);
+struct timespec64 current_time(struct inode *inode);
+struct timespec64 inode_ctime_set_current(struct inode *inode);
+
+/**
+ * inode_ctime_peek - fetch the current ctime from the inode
+ * @inode: inode from which to fetch ctime
+ *
+ * Grab the current ctime from the inode and return it.
+ */
+static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
+{
+ return inode->i_ctime;
+}
+
+/**
+ * inode_ctime_set - set the ctime in the inode to the given value
+ * @inode: inode in which to set the ctime
+ * @ts: timespec value to set the ctime
+ *
+ * Set the ctime in @inode to @ts.
+ */
+static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
+{
+ inode->i_ctime = ts;
+ return ts;
+}
+
+/**
+ * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
+ * @inode: inode in which to set the ctime
+ * @sec: value to set the tv_sec field
+ *
+ * Set the sec field in the ctime. Returns @sec.
+ */
+static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
+{
+ inode->i_ctime.tv_sec = sec;
+ return sec;
+}
+
+/**
+ * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
+ * @inode: inode in which to set the ctime
+ * @nsec: value to set the tv_nsec field
+ *
+ * Set the nsec field in the ctime. Returns @nsec.
+ */
+static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
+{
+ inode->i_ctime.tv_nsec = nsec;
+ return nsec;
+}
/*
* Snapshotting support.
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [f2fs-dev] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-21 14:45 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Jeremy Kerr, Arnd Bergmann, Michael Ellerman, Nicholas Piggin,
Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Tom Talpey, Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
struct timespec64 has unused bits in the tv_nsec field that can be used
for other purposes. In future patches, we're going to change how the
inode->i_ctime is accessed in certain inodes in order to make use of
them. In order to do that safely though, we'll need to eradicate raw
accesses of the inode->i_ctime field from the kernel.
Add new accessor functions for the ctime that we can use to replace them.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/inode.c | 16 ++++++++++++++
include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
2 files changed, 68 insertions(+), 1 deletion(-)
diff --git a/fs/inode.c b/fs/inode.c
index d37fad91c8da..c005e7328fbb 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
}
EXPORT_SYMBOL(current_time);
+/**
+ * inode_ctime_set_current - set the ctime to current_time
+ * @inode: inode
+ *
+ * Set the inode->i_ctime to the current value for the inode. Returns
+ * the current value that was assigned to i_ctime.
+ */
+struct timespec64 inode_ctime_set_current(struct inode *inode)
+{
+ struct timespec64 now = current_time(inode);
+
+ inode_set_ctime(inode, now);
+ return now;
+}
+EXPORT_SYMBOL(inode_ctime_set_current);
+
/**
* in_group_or_capable - check whether caller is CAP_FSETID privileged
* @idmap: idmap of the mount @inode was found from
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 6867512907d6..9afb30606373 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
kgid_has_mapping(fs_userns, kgid);
}
-extern struct timespec64 current_time(struct inode *inode);
+struct timespec64 current_time(struct inode *inode);
+struct timespec64 inode_ctime_set_current(struct inode *inode);
+
+/**
+ * inode_ctime_peek - fetch the current ctime from the inode
+ * @inode: inode from which to fetch ctime
+ *
+ * Grab the current ctime from the inode and return it.
+ */
+static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
+{
+ return inode->i_ctime;
+}
+
+/**
+ * inode_ctime_set - set the ctime in the inode to the given value
+ * @inode: inode in which to set the ctime
+ * @ts: timespec value to set the ctime
+ *
+ * Set the ctime in @inode to @ts.
+ */
+static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
+{
+ inode->i_ctime = ts;
+ return ts;
+}
+
+/**
+ * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
+ * @inode: inode in which to set the ctime
+ * @sec: value to set the tv_sec field
+ *
+ * Set the sec field in the ctime. Returns @sec.
+ */
+static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
+{
+ inode->i_ctime.tv_sec = sec;
+ return sec;
+}
+
+/**
+ * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
+ * @inode: inode in which to set the ctime
+ * @nsec: value to set the tv_nsec field
+ *
+ * Set the nsec field in the ctime. Returns @nsec.
+ */
+static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
+{
+ inode->i_ctime.tv_nsec = nsec;
+ return nsec;
+}
/*
* Snapshotting support.
--
2.41.0
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [Ocfs2-devel] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-21 14:45 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton via Ocfs2-devel @ 2023-06-21 14:45 UTC (permalink / raw)
To: Jeremy Kerr, Arnd Bergmann, Michael Ellerman, Nicholas Piggin,
Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Tom Talpey, Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
struct timespec64 has unused bits in the tv_nsec field that can be used
for other purposes. In future patches, we're going to change how the
inode->i_ctime is accessed in certain inodes in order to make use of
them. In order to do that safely though, we'll need to eradicate raw
accesses of the inode->i_ctime field from the kernel.
Add new accessor functions for the ctime that we can use to replace them.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/inode.c | 16 ++++++++++++++
include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
2 files changed, 68 insertions(+), 1 deletion(-)
diff --git a/fs/inode.c b/fs/inode.c
index d37fad91c8da..c005e7328fbb 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
}
EXPORT_SYMBOL(current_time);
+/**
+ * inode_ctime_set_current - set the ctime to current_time
+ * @inode: inode
+ *
+ * Set the inode->i_ctime to the current value for the inode. Returns
+ * the current value that was assigned to i_ctime.
+ */
+struct timespec64 inode_ctime_set_current(struct inode *inode)
+{
+ struct timespec64 now = current_time(inode);
+
+ inode_set_ctime(inode, now);
+ return now;
+}
+EXPORT_SYMBOL(inode_ctime_set_current);
+
/**
* in_group_or_capable - check whether caller is CAP_FSETID privileged
* @idmap: idmap of the mount @inode was found from
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 6867512907d6..9afb30606373 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
kgid_has_mapping(fs_userns, kgid);
}
-extern struct timespec64 current_time(struct inode *inode);
+struct timespec64 current_time(struct inode *inode);
+struct timespec64 inode_ctime_set_current(struct inode *inode);
+
+/**
+ * inode_ctime_peek - fetch the current ctime from the inode
+ * @inode: inode from which to fetch ctime
+ *
+ * Grab the current ctime from the inode and return it.
+ */
+static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
+{
+ return inode->i_ctime;
+}
+
+/**
+ * inode_ctime_set - set the ctime in the inode to the given value
+ * @inode: inode in which to set the ctime
+ * @ts: timespec value to set the ctime
+ *
+ * Set the ctime in @inode to @ts.
+ */
+static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
+{
+ inode->i_ctime = ts;
+ return ts;
+}
+
+/**
+ * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
+ * @inode: inode in which to set the ctime
+ * @sec: value to set the tv_sec field
+ *
+ * Set the sec field in the ctime. Returns @sec.
+ */
+static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
+{
+ inode->i_ctime.tv_sec = sec;
+ return sec;
+}
+
+/**
+ * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
+ * @inode: inode in which to set the ctime
+ * @nsec: value to set the tv_nsec field
+ *
+ * Set the nsec field in the ctime. Returns @nsec.
+ */
+static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
+{
+ inode->i_ctime.tv_nsec = nsec;
+ return nsec;
+}
/*
* Snapshotting support.
--
2.41.0
_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [Cluster-devel] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-21 14:45 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: cluster-devel.redhat.com
struct timespec64 has unused bits in the tv_nsec field that can be used
for other purposes. In future patches, we're going to change how the
inode->i_ctime is accessed in certain inodes in order to make use of
them. In order to do that safely though, we'll need to eradicate raw
accesses of the inode->i_ctime field from the kernel.
Add new accessor functions for the ctime that we can use to replace them.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/inode.c | 16 ++++++++++++++
include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
2 files changed, 68 insertions(+), 1 deletion(-)
diff --git a/fs/inode.c b/fs/inode.c
index d37fad91c8da..c005e7328fbb 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
}
EXPORT_SYMBOL(current_time);
+/**
+ * inode_ctime_set_current - set the ctime to current_time
+ * @inode: inode
+ *
+ * Set the inode->i_ctime to the current value for the inode. Returns
+ * the current value that was assigned to i_ctime.
+ */
+struct timespec64 inode_ctime_set_current(struct inode *inode)
+{
+ struct timespec64 now = current_time(inode);
+
+ inode_set_ctime(inode, now);
+ return now;
+}
+EXPORT_SYMBOL(inode_ctime_set_current);
+
/**
* in_group_or_capable - check whether caller is CAP_FSETID privileged
* @idmap: idmap of the mount @inode was found from
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 6867512907d6..9afb30606373 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
kgid_has_mapping(fs_userns, kgid);
}
-extern struct timespec64 current_time(struct inode *inode);
+struct timespec64 current_time(struct inode *inode);
+struct timespec64 inode_ctime_set_current(struct inode *inode);
+
+/**
+ * inode_ctime_peek - fetch the current ctime from the inode
+ * @inode: inode from which to fetch ctime
+ *
+ * Grab the current ctime from the inode and return it.
+ */
+static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
+{
+ return inode->i_ctime;
+}
+
+/**
+ * inode_ctime_set - set the ctime in the inode to the given value
+ * @inode: inode in which to set the ctime
+ * @ts: timespec value to set the ctime
+ *
+ * Set the ctime in @inode to @ts.
+ */
+static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
+{
+ inode->i_ctime = ts;
+ return ts;
+}
+
+/**
+ * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
+ * @inode: inode in which to set the ctime
+ * @sec: value to set the tv_sec field
+ *
+ * Set the sec field in the ctime. Returns @sec.
+ */
+static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
+{
+ inode->i_ctime.tv_sec = sec;
+ return sec;
+}
+
+/**
+ * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
+ * @inode: inode in which to set the ctime
+ * @nsec: value to set the tv_nsec field
+ *
+ * Set the nsec field in the ctime. Returns @nsec.
+ */
+static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
+{
+ inode->i_ctime.tv_nsec = nsec;
+ return nsec;
+}
/*
* Snapshotting support.
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 02/79] spufs: switch to new ctime accessors
2023-06-21 14:45 ` [f2fs-dev] " Jeff Layton
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy
Cc: Al Viro, Jan Kara, linuxppc-dev, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
arch/powerpc/platforms/cell/spufs/inode.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c
index ea807aa0c31a..55418395bd9a 100644
--- a/arch/powerpc/platforms/cell/spufs/inode.c
+++ b/arch/powerpc/platforms/cell/spufs/inode.c
@@ -86,7 +86,7 @@ spufs_new_inode(struct super_block *sb, umode_t mode)
inode->i_mode = mode;
inode->i_uid = current_fsuid();
inode->i_gid = current_fsgid();
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
out:
return inode;
}
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 02/79] spufs: switch to new ctime accessors
@ 2023-06-21 14:45 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy
Cc: linuxppc-dev, Jan Kara, Al Viro, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
arch/powerpc/platforms/cell/spufs/inode.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c
index ea807aa0c31a..55418395bd9a 100644
--- a/arch/powerpc/platforms/cell/spufs/inode.c
+++ b/arch/powerpc/platforms/cell/spufs/inode.c
@@ -86,7 +86,7 @@ spufs_new_inode(struct super_block *sb, umode_t mode)
inode->i_mode = mode;
inode->i_uid = current_fsuid();
inode->i_gid = current_fsgid();
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
out:
return inode;
}
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 03/79] s390: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
(?)
@ 2023-06-21 14:45 ` Jeff Layton
2023-06-21 16:34 ` Jan Kara
2023-06-22 17:35 ` Alexander Gordeev
-1 siblings, 2 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle
Cc: Al Viro, Jan Kara, linux-s390, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
arch/s390/hypfs/inode.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c
index ee919bfc8186..30fa336ec63e 100644
--- a/arch/s390/hypfs/inode.c
+++ b/arch/s390/hypfs/inode.c
@@ -53,7 +53,7 @@ static void hypfs_update_update(struct super_block *sb)
struct inode *inode = d_inode(sb_info->update_file);
sb_info->last_update = ktime_get_seconds();
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
}
/* directory tree removal functions */
@@ -101,7 +101,7 @@ static struct inode *hypfs_make_inode(struct super_block *sb, umode_t mode)
ret->i_mode = mode;
ret->i_uid = hypfs_info->uid;
ret->i_gid = hypfs_info->gid;
- ret->i_atime = ret->i_mtime = ret->i_ctime = current_time(ret);
+ ret->i_atime = ret->i_mtime = inode_ctime_set_current(ret);
if (S_ISDIR(mode))
set_nlink(ret, 2);
}
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 04/79] binderfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
(?)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
2023-06-21 15:29 ` Greg Kroah-Hartman
2023-06-21 16:35 ` Jan Kara
-1 siblings, 2 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Greg Kroah-Hartman, Arve Hjønnevåg,
Todd Kjos, Martijn Coenen, Joel Fernandes, Carlos Llamas,
Suren Baghdasaryan
Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
drivers/android/binderfs.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/android/binderfs.c b/drivers/android/binderfs.c
index 76e7d6676657..b6b1220e1e55 100644
--- a/drivers/android/binderfs.c
+++ b/drivers/android/binderfs.c
@@ -153,7 +153,7 @@ static int binderfs_binder_device_create(struct inode *ref_inode,
goto err;
inode->i_ino = minor + INODE_OFFSET;
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
init_special_inode(inode, S_IFCHR | 0600,
MKDEV(MAJOR(binderfs_dev), minor));
inode->i_fop = &binder_fops;
@@ -432,7 +432,7 @@ static int binderfs_binder_ctl_create(struct super_block *sb)
}
inode->i_ino = SECOND_INODE;
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
init_special_inode(inode, S_IFCHR | 0600,
MKDEV(MAJOR(binderfs_dev), minor));
inode->i_fop = &binder_ctl_fops;
@@ -474,7 +474,7 @@ static struct inode *binderfs_make_inode(struct super_block *sb, int mode)
if (ret) {
ret->i_ino = iunique(sb, BINDERFS_MAX_MINOR + INODE_OFFSET);
ret->i_mode = mode;
- ret->i_atime = ret->i_mtime = ret->i_ctime = current_time(ret);
+ ret->i_atime = ret->i_mtime = inode_ctime_set_current(ret);
}
return ret;
}
@@ -703,7 +703,7 @@ static int binderfs_fill_super(struct super_block *sb, struct fs_context *fc)
inode->i_ino = FIRST_INODE;
inode->i_fop = &simple_dir_operations;
inode->i_mode = S_IFDIR | 0755;
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
inode->i_op = &binderfs_dir_inode_operations;
set_nlink(inode, 2);
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 05/79] qib_fs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (2 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
2023-06-21 16:35 ` Jan Kara
2023-06-21 19:37 ` Dennis Dalessandro
-1 siblings, 2 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Dennis Dalessandro, Jason Gunthorpe, Leon Romanovsky
Cc: Al Viro, Jan Kara, linux-rdma, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
drivers/infiniband/hw/qib/qib_fs.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/drivers/infiniband/hw/qib/qib_fs.c b/drivers/infiniband/hw/qib/qib_fs.c
index a973905afd13..db008f2884e4 100644
--- a/drivers/infiniband/hw/qib/qib_fs.c
+++ b/drivers/infiniband/hw/qib/qib_fs.c
@@ -64,9 +64,7 @@ static int qibfs_mknod(struct inode *dir, struct dentry *dentry,
inode->i_uid = GLOBAL_ROOT_UID;
inode->i_gid = GLOBAL_ROOT_GID;
inode->i_blocks = 0;
- inode->i_atime = current_time(inode);
- inode->i_mtime = inode->i_atime;
- inode->i_ctime = inode->i_atime;
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
inode->i_private = data;
if (S_ISDIR(mode)) {
inode->i_op = &simple_dir_inode_operations;
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 06/79] ibm: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (3 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
2023-06-21 15:29 ` Greg Kroah-Hartman
2023-06-21 16:35 ` Jan Kara
-1 siblings, 2 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Arnd Bergmann, Greg Kroah-Hartman,
Brad Warrum, Ritu Agarwal
Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
drivers/misc/ibmasm/ibmasmfs.c | 2 +-
drivers/misc/ibmvmc.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/misc/ibmasm/ibmasmfs.c b/drivers/misc/ibmasm/ibmasmfs.c
index 35fec1bf1b3d..34d41b2a0d89 100644
--- a/drivers/misc/ibmasm/ibmasmfs.c
+++ b/drivers/misc/ibmasm/ibmasmfs.c
@@ -139,7 +139,7 @@ static struct inode *ibmasmfs_make_inode(struct super_block *sb, int mode)
if (ret) {
ret->i_ino = get_next_ino();
ret->i_mode = mode;
- ret->i_atime = ret->i_mtime = ret->i_ctime = current_time(ret);
+ ret->i_atime = ret->i_mtime = inode_ctime_set_current(ret);
}
return ret;
}
diff --git a/drivers/misc/ibmvmc.c b/drivers/misc/ibmvmc.c
index d7c7f0305257..b69c978c2590 100644
--- a/drivers/misc/ibmvmc.c
+++ b/drivers/misc/ibmvmc.c
@@ -1124,7 +1124,7 @@ static ssize_t ibmvmc_write(struct file *file, const char *buffer,
goto out;
inode = file_inode(file);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
mark_inode_dirty(inode);
dev_dbg(adapter->dev, "write: file = 0x%lx, count = 0x%lx\n",
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 07/79] usb: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (4 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
2023-06-21 15:29 ` Greg Kroah-Hartman
2023-06-21 16:36 ` Jan Kara
-1 siblings, 2 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Greg Kroah-Hartman
Cc: Al Viro, Jan Kara, linux-usb, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
drivers/usb/core/devio.c | 16 ++++++++--------
drivers/usb/gadget/function/f_fs.c | 6 +-----
drivers/usb/gadget/legacy/inode.c | 3 +--
3 files changed, 10 insertions(+), 15 deletions(-)
diff --git a/drivers/usb/core/devio.c b/drivers/usb/core/devio.c
index 1a16a8bdea60..02f718e0deaf 100644
--- a/drivers/usb/core/devio.c
+++ b/drivers/usb/core/devio.c
@@ -2642,21 +2642,21 @@ static long usbdev_do_ioctl(struct file *file, unsigned int cmd,
snoop(&dev->dev, "%s: CONTROL\n", __func__);
ret = proc_control(ps, p);
if (ret >= 0)
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
break;
case USBDEVFS_BULK:
snoop(&dev->dev, "%s: BULK\n", __func__);
ret = proc_bulk(ps, p);
if (ret >= 0)
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
break;
case USBDEVFS_RESETEP:
snoop(&dev->dev, "%s: RESETEP\n", __func__);
ret = proc_resetep(ps, p);
if (ret >= 0)
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
break;
case USBDEVFS_RESET:
@@ -2668,7 +2668,7 @@ static long usbdev_do_ioctl(struct file *file, unsigned int cmd,
snoop(&dev->dev, "%s: CLEAR_HALT\n", __func__);
ret = proc_clearhalt(ps, p);
if (ret >= 0)
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
break;
case USBDEVFS_GETDRIVER:
@@ -2695,7 +2695,7 @@ static long usbdev_do_ioctl(struct file *file, unsigned int cmd,
snoop(&dev->dev, "%s: SUBMITURB\n", __func__);
ret = proc_submiturb(ps, p);
if (ret >= 0)
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
break;
#ifdef CONFIG_COMPAT
@@ -2703,14 +2703,14 @@ static long usbdev_do_ioctl(struct file *file, unsigned int cmd,
snoop(&dev->dev, "%s: CONTROL32\n", __func__);
ret = proc_control_compat(ps, p);
if (ret >= 0)
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
break;
case USBDEVFS_BULK32:
snoop(&dev->dev, "%s: BULK32\n", __func__);
ret = proc_bulk_compat(ps, p);
if (ret >= 0)
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
break;
case USBDEVFS_DISCSIGNAL32:
@@ -2722,7 +2722,7 @@ static long usbdev_do_ioctl(struct file *file, unsigned int cmd,
snoop(&dev->dev, "%s: SUBMITURB32\n", __func__);
ret = proc_submiturb_compat(ps, p);
if (ret >= 0)
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
break;
case USBDEVFS_IOCTL32:
diff --git a/drivers/usb/gadget/function/f_fs.c b/drivers/usb/gadget/function/f_fs.c
index f41a385a5c42..756c78043a04 100644
--- a/drivers/usb/gadget/function/f_fs.c
+++ b/drivers/usb/gadget/function/f_fs.c
@@ -1377,16 +1377,12 @@ ffs_sb_make_inode(struct super_block *sb, void *data,
inode = new_inode(sb);
if (inode) {
- struct timespec64 ts = current_time(inode);
-
inode->i_ino = get_next_ino();
inode->i_mode = perms->mode;
inode->i_uid = perms->uid;
inode->i_gid = perms->gid;
- inode->i_atime = ts;
- inode->i_mtime = ts;
- inode->i_ctime = ts;
inode->i_private = data;
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
if (fops)
inode->i_fop = fops;
if (iops)
diff --git a/drivers/usb/gadget/legacy/inode.c b/drivers/usb/gadget/legacy/inode.c
index 28249d0bf062..b83a68feb316 100644
--- a/drivers/usb/gadget/legacy/inode.c
+++ b/drivers/usb/gadget/legacy/inode.c
@@ -1969,8 +1969,7 @@ gadgetfs_make_inode (struct super_block *sb,
inode->i_mode = mode;
inode->i_uid = make_kuid(&init_user_ns, default_uid);
inode->i_gid = make_kgid(&init_user_ns, default_gid);
- inode->i_atime = inode->i_mtime = inode->i_ctime
- = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
inode->i_private = data;
inode->i_fop = fops;
}
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 08/79] 9p: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (5 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
2023-06-21 16:37 ` Jan Kara
-1 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Eric Van Hensbergen, Latchesar Ionkov,
Dominique Martinet, Christian Schoenebeck
Cc: Al Viro, Jan Kara, v9fs, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/9p/vfs_inode.c | 6 ++++--
fs/9p/vfs_inode_dotl.c | 11 +++++------
2 files changed, 9 insertions(+), 8 deletions(-)
diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c
index 36b466e35887..098e4c7160a8 100644
--- a/fs/9p/vfs_inode.c
+++ b/fs/9p/vfs_inode.c
@@ -261,7 +261,7 @@ int v9fs_init_inode(struct v9fs_session_info *v9ses,
inode_init_owner(&nop_mnt_idmap, inode, NULL, mode);
inode->i_blocks = 0;
inode->i_rdev = rdev;
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
inode->i_mapping->a_ops = &v9fs_addr_operations;
inode->i_private = NULL;
@@ -1153,12 +1153,14 @@ v9fs_stat2inode(struct p9_wstat *stat, struct inode *inode,
umode_t mode;
struct v9fs_session_info *v9ses = sb->s_fs_info;
struct v9fs_inode *v9inode = V9FS_I(inode);
+ struct timespec64 ctime = { .tv_sec = stat->mtime,
+ .tv_nsec = 0 };
set_nlink(inode, 1);
inode->i_atime.tv_sec = stat->atime;
inode->i_mtime.tv_sec = stat->mtime;
- inode->i_ctime.tv_sec = stat->mtime;
+ inode_ctime_set(inode, ctime);
inode->i_uid = v9ses->dfltuid;
inode->i_gid = v9ses->dfltgid;
diff --git a/fs/9p/vfs_inode_dotl.c b/fs/9p/vfs_inode_dotl.c
index 5361cd2d7996..0041b5fc4407 100644
--- a/fs/9p/vfs_inode_dotl.c
+++ b/fs/9p/vfs_inode_dotl.c
@@ -640,14 +640,15 @@ v9fs_stat2inode_dotl(struct p9_stat_dotl *stat, struct inode *inode,
{
umode_t mode;
struct v9fs_inode *v9inode = V9FS_I(inode);
+ struct timespec64 ctime = { .tv_sec = stat->st_ctime_sec,
+ .tv_nsec = stat->st_ctime_nsec };
if ((stat->st_result_mask & P9_STATS_BASIC) == P9_STATS_BASIC) {
inode->i_atime.tv_sec = stat->st_atime_sec;
inode->i_atime.tv_nsec = stat->st_atime_nsec;
inode->i_mtime.tv_sec = stat->st_mtime_sec;
inode->i_mtime.tv_nsec = stat->st_mtime_nsec;
- inode->i_ctime.tv_sec = stat->st_ctime_sec;
- inode->i_ctime.tv_nsec = stat->st_ctime_nsec;
+ inode_ctime_set(inode, ctime);
inode->i_uid = stat->st_uid;
inode->i_gid = stat->st_gid;
set_nlink(inode, stat->st_nlink);
@@ -668,10 +669,8 @@ v9fs_stat2inode_dotl(struct p9_stat_dotl *stat, struct inode *inode,
inode->i_mtime.tv_sec = stat->st_mtime_sec;
inode->i_mtime.tv_nsec = stat->st_mtime_nsec;
}
- if (stat->st_result_mask & P9_STATS_CTIME) {
- inode->i_ctime.tv_sec = stat->st_ctime_sec;
- inode->i_ctime.tv_nsec = stat->st_ctime_nsec;
- }
+ if (stat->st_result_mask & P9_STATS_CTIME)
+ inode_ctime_set(inode, ctime);
if (stat->st_result_mask & P9_STATS_UID)
inode->i_uid = stat->st_uid;
if (stat->st_result_mask & P9_STATS_GID)
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 09/79] adfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (6 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
2023-06-21 16:38 ` Jan Kara
-1 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner; +Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/adfs/inode.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/adfs/inode.c b/fs/adfs/inode.c
index c3ac613d0975..266191e6b80c 100644
--- a/fs/adfs/inode.c
+++ b/fs/adfs/inode.c
@@ -270,7 +270,7 @@ adfs_iget(struct super_block *sb, struct object_info *obj)
inode->i_mode = adfs_atts2mode(sb, inode);
adfs_adfs2unix_time(&inode->i_mtime, inode);
inode->i_atime = inode->i_mtime;
- inode->i_ctime = inode->i_mtime;
+ inode_ctime_set(inode, inode->i_mtime);
if (S_ISDIR(inode->i_mode)) {
inode->i_op = &adfs_dir_inode_operations;
@@ -331,7 +331,7 @@ adfs_notify_change(struct mnt_idmap *idmap, struct dentry *dentry,
if (ia_valid & ATTR_ATIME)
inode->i_atime = attr->ia_atime;
if (ia_valid & ATTR_CTIME)
- inode->i_ctime = attr->ia_ctime;
+ inode_ctime_set(inode, attr->ia_ctime);
if (ia_valid & ATTR_MODE) {
ADFS_I(inode)->attr = adfs_mode2atts(sb, inode, attr->ia_mode);
inode->i_mode = adfs_atts2mode(sb, inode);
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 10/79] affs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (7 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
2023-06-21 16:39 ` Jan Kara
2023-06-22 11:37 ` David Sterba
-1 siblings, 2 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, David Sterba
Cc: Al Viro, Jan Kara, linux-fsdevel, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/affs/amigaffs.c | 6 +++---
fs/affs/inode.c | 17 +++++++++--------
2 files changed, 12 insertions(+), 11 deletions(-)
diff --git a/fs/affs/amigaffs.c b/fs/affs/amigaffs.c
index 29f11e10a7c7..2b508aa6707e 100644
--- a/fs/affs/amigaffs.c
+++ b/fs/affs/amigaffs.c
@@ -60,7 +60,7 @@ affs_insert_hash(struct inode *dir, struct buffer_head *bh)
mark_buffer_dirty_inode(dir_bh, dir);
affs_brelse(dir_bh);
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
inode_inc_iversion(dir);
mark_inode_dirty(dir);
@@ -114,7 +114,7 @@ affs_remove_hash(struct inode *dir, struct buffer_head *rem_bh)
affs_brelse(bh);
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
inode_inc_iversion(dir);
mark_inode_dirty(dir);
@@ -315,7 +315,7 @@ affs_remove_header(struct dentry *dentry)
else
clear_nlink(inode);
affs_unlock_link(inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
mark_inode_dirty(inode);
done:
diff --git a/fs/affs/inode.c b/fs/affs/inode.c
index 27f77a52c5c8..177bac4def5e 100644
--- a/fs/affs/inode.c
+++ b/fs/affs/inode.c
@@ -19,6 +19,7 @@ struct inode *affs_iget(struct super_block *sb, unsigned long ino)
{
struct affs_sb_info *sbi = AFFS_SB(sb);
struct buffer_head *bh;
+ struct timespec64 ctime;
struct affs_tail *tail;
struct inode *inode;
u32 block;
@@ -149,13 +150,13 @@ struct inode *affs_iget(struct super_block *sb, unsigned long ino)
break;
}
- inode->i_mtime.tv_sec = inode->i_atime.tv_sec = inode->i_ctime.tv_sec
- = (be32_to_cpu(tail->change.days) * 86400LL +
- be32_to_cpu(tail->change.mins) * 60 +
- be32_to_cpu(tail->change.ticks) / 50 +
- AFFS_EPOCH_DELTA) +
- sys_tz.tz_minuteswest * 60;
- inode->i_mtime.tv_nsec = inode->i_ctime.tv_nsec = inode->i_atime.tv_nsec = 0;
+ ctime.tv_sec = (be32_to_cpu(tail->change.days) * 86400LL +
+ be32_to_cpu(tail->change.mins) * 60 +
+ be32_to_cpu(tail->change.ticks) / 50 +
+ AFFS_EPOCH_DELTA) +
+ sys_tz.tz_minuteswest * 60;
+ ctime.tv_nsec = 0;
+ inode->i_atime = inode->i_mtime = inode_ctime_set(inode, ctime);
affs_brelse(bh);
unlock_new_inode(inode);
return inode;
@@ -314,7 +315,7 @@ affs_new_inode(struct inode *dir)
inode->i_gid = current_fsgid();
inode->i_ino = block;
set_nlink(inode, 1);
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
atomic_set(&AFFS_I(inode)->i_opencnt, 0);
AFFS_I(inode)->i_blkcnt = 0;
AFFS_I(inode)->i_lc = NULL;
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 11/79] afs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (8 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
2023-06-21 16:43 ` Jan Kara
-1 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, David Howells, Marc Dionne
Cc: Al Viro, Jan Kara, linux-afs, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/afs/dynroot.c | 2 +-
fs/afs/inode.c | 6 +++---
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/fs/afs/dynroot.c b/fs/afs/dynroot.c
index d7d9402ff718..ee36375de337 100644
--- a/fs/afs/dynroot.c
+++ b/fs/afs/dynroot.c
@@ -88,9 +88,9 @@ struct inode *afs_iget_pseudo_dir(struct super_block *sb, bool root)
set_nlink(inode, 2);
inode->i_uid = GLOBAL_ROOT_UID;
inode->i_gid = GLOBAL_ROOT_GID;
- inode->i_ctime = inode->i_atime = inode->i_mtime = current_time(inode);
inode->i_blocks = 0;
inode->i_generation = 0;
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
set_bit(AFS_VNODE_PSEUDODIR, &vnode->flags);
if (!root) {
diff --git a/fs/afs/inode.c b/fs/afs/inode.c
index 866bab860a88..0b59903c2a43 100644
--- a/fs/afs/inode.c
+++ b/fs/afs/inode.c
@@ -90,7 +90,7 @@ static int afs_inode_init_from_status(struct afs_operation *op,
vnode->status = *status;
t = status->mtime_client;
- inode->i_ctime = t;
+ inode_ctime_set(inode, t);
inode->i_mtime = t;
inode->i_atime = t;
inode->i_flags |= S_NOATIME;
@@ -206,7 +206,7 @@ static void afs_apply_status(struct afs_operation *op,
t = status->mtime_client;
inode->i_mtime = t;
if (vp->update_ctime)
- inode->i_ctime = op->ctime;
+ inode_ctime_set(inode, op->ctime);
if (vnode->status.data_version != status->data_version)
data_changed = true;
@@ -252,7 +252,7 @@ static void afs_apply_status(struct afs_operation *op,
vnode->netfs.remote_i_size = status->size;
if (change_size) {
afs_set_i_size(vnode, status->size);
- inode->i_ctime = t;
+ inode_ctime_set(inode, t);
inode->i_atime = t;
}
}
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 12/79] fs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (9 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
2023-06-21 16:42 ` Jan Kara
-1 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Alexander Viro, Eric Biederman, Kees Cook
Cc: Jan Kara, linux-fsdevel, linux-kernel, linux-mm
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/attr.c | 2 +-
fs/bad_inode.c | 3 +--
fs/binfmt_misc.c | 3 +--
fs/inode.c | 12 ++++++++----
fs/libfs.c | 32 +++++++++++++++++---------------
fs/nsfs.c | 2 +-
fs/pipe.c | 2 +-
fs/posix_acl.c | 2 +-
fs/stack.c | 2 +-
fs/stat.c | 2 +-
include/linux/fs_stack.h | 2 +-
11 files changed, 34 insertions(+), 30 deletions(-)
diff --git a/fs/attr.c b/fs/attr.c
index d60dc1edb526..2750e5f98dfb 100644
--- a/fs/attr.c
+++ b/fs/attr.c
@@ -312,7 +312,7 @@ void setattr_copy(struct mnt_idmap *idmap, struct inode *inode,
if (ia_valid & ATTR_MTIME)
inode->i_mtime = attr->ia_mtime;
if (ia_valid & ATTR_CTIME)
- inode->i_ctime = attr->ia_ctime;
+ inode_ctime_set(inode, attr->ia_ctime);
if (ia_valid & ATTR_MODE) {
umode_t mode = attr->ia_mode;
if (!in_group_or_capable(idmap, inode,
diff --git a/fs/bad_inode.c b/fs/bad_inode.c
index db649487d58c..bd3762e1b670 100644
--- a/fs/bad_inode.c
+++ b/fs/bad_inode.c
@@ -209,8 +209,7 @@ void make_bad_inode(struct inode *inode)
remove_inode_hash(inode);
inode->i_mode = S_IFREG;
- inode->i_atime = inode->i_mtime = inode->i_ctime =
- current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
inode->i_op = &bad_inode_ops;
inode->i_opflags &= ~IOP_XATTR;
inode->i_fop = &bad_file_ops;
diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c
index bb202ad369d5..6af92eb1b871 100644
--- a/fs/binfmt_misc.c
+++ b/fs/binfmt_misc.c
@@ -547,8 +547,7 @@ static struct inode *bm_get_inode(struct super_block *sb, int mode)
if (inode) {
inode->i_ino = get_next_ino();
inode->i_mode = mode;
- inode->i_atime = inode->i_mtime = inode->i_ctime =
- current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
}
return inode;
}
diff --git a/fs/inode.c b/fs/inode.c
index c005e7328fbb..a7f484e9e7c1 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -1851,6 +1851,7 @@ EXPORT_SYMBOL(bmap);
static int relatime_need_update(struct vfsmount *mnt, struct inode *inode,
struct timespec64 now)
{
+ struct timespec64 ctime;
if (!(mnt->mnt_flags & MNT_RELATIME))
return 1;
@@ -1862,7 +1863,8 @@ static int relatime_need_update(struct vfsmount *mnt, struct inode *inode,
/*
* Is ctime younger than or equal to atime? If yes, update atime:
*/
- if (timespec64_compare(&inode->i_ctime, &inode->i_atime) >= 0)
+ ctime = inode_ctime_peek(inode);
+ if (timespec64_compare(&ctime, &inode->i_atime) >= 0)
return 1;
/*
@@ -1885,7 +1887,7 @@ int generic_update_time(struct inode *inode, struct timespec64 *time, int flags)
if (flags & S_ATIME)
inode->i_atime = *time;
if (flags & S_CTIME)
- inode->i_ctime = *time;
+ inode_ctime_set(inode, *time);
if (flags & S_MTIME)
inode->i_mtime = *time;
@@ -2071,6 +2073,7 @@ EXPORT_SYMBOL(file_remove_privs);
static int inode_needs_update_time(struct inode *inode, struct timespec64 *now)
{
int sync_it = 0;
+ struct timespec64 ctime;
/* First try to exhaust all avenues to not sync */
if (IS_NOCMTIME(inode))
@@ -2079,7 +2082,8 @@ static int inode_needs_update_time(struct inode *inode, struct timespec64 *now)
if (!timespec64_equal(&inode->i_mtime, now))
sync_it = S_MTIME;
- if (!timespec64_equal(&inode->i_ctime, now))
+ ctime = inode_ctime_peek(inode);
+ if (!timespec64_equal(&ctime, now))
sync_it |= S_CTIME;
if (IS_I_VERSION(inode) && inode_iversion_need_inc(inode))
@@ -2510,7 +2514,7 @@ struct timespec64 inode_ctime_set_current(struct inode *inode)
{
struct timespec64 now = current_time(inode);
- inode_set_ctime(inode, now);
+ inode_ctime_set(inode, now);
return now;
}
EXPORT_SYMBOL(inode_ctime_set_current);
diff --git a/fs/libfs.c b/fs/libfs.c
index 5b851315eeed..4a914f09fa87 100644
--- a/fs/libfs.c
+++ b/fs/libfs.c
@@ -275,7 +275,7 @@ void simple_recursive_removal(struct dentry *dentry,
while ((child = find_next_child(this, victim)) == NULL) {
// kill and ascend
// update metadata while it's still locked
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
clear_nlink(inode);
inode_unlock(inode);
victim = this;
@@ -293,8 +293,7 @@ void simple_recursive_removal(struct dentry *dentry,
dput(victim); // unpin it
}
if (victim == dentry) {
- inode->i_ctime = inode->i_mtime =
- current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
if (d_is_dir(dentry))
drop_nlink(inode);
inode_unlock(inode);
@@ -335,7 +334,7 @@ static int pseudo_fs_fill_super(struct super_block *s, struct fs_context *fc)
*/
root->i_ino = 1;
root->i_mode = S_IFDIR | S_IRUSR | S_IWUSR;
- root->i_atime = root->i_mtime = root->i_ctime = current_time(root);
+ root->i_atime = root->i_mtime = inode_ctime_set_current(root);
s->s_root = d_make_root(root);
if (!s->s_root)
return -ENOMEM;
@@ -391,7 +390,8 @@ int simple_link(struct dentry *old_dentry, struct inode *dir, struct dentry *den
{
struct inode *inode = d_inode(old_dentry);
- inode->i_ctime = dir->i_ctime = dir->i_mtime = current_time(inode);
+ inode_ctime_set_current(inode);
+ inode->i_mtime = inode_ctime_set_current(dir);
inc_nlink(inode);
ihold(inode);
dget(dentry);
@@ -425,7 +425,8 @@ int simple_unlink(struct inode *dir, struct dentry *dentry)
{
struct inode *inode = d_inode(dentry);
- inode->i_ctime = dir->i_ctime = dir->i_mtime = current_time(inode);
+ inode_ctime_set_current(inode);
+ dir->i_mtime = inode_ctime_set_current(dir);
drop_nlink(inode);
dput(dentry);
return 0;
@@ -459,10 +460,10 @@ int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry,
inc_nlink(old_dir);
}
}
- old_dir->i_ctime = old_dir->i_mtime =
- new_dir->i_ctime = new_dir->i_mtime =
- d_inode(old_dentry)->i_ctime =
- d_inode(new_dentry)->i_ctime = current_time(old_dir);
+ old_dir->i_mtime = inode_ctime_set_current(old_dir);
+ new_dir->i_mtime = inode_ctime_set_current(new_dir);
+ inode_ctime_set_current(d_inode(old_dentry));
+ inode_ctime_set_current(d_inode(new_dentry));
return 0;
}
@@ -495,8 +496,9 @@ int simple_rename(struct mnt_idmap *idmap, struct inode *old_dir,
inc_nlink(new_dir);
}
- old_dir->i_ctime = old_dir->i_mtime = new_dir->i_ctime =
- new_dir->i_mtime = inode->i_ctime = current_time(old_dir);
+ old_dir->i_mtime = inode_ctime_set_current(old_dir);
+ new_dir->i_mtime = inode_ctime_set_current(new_dir);
+ inode_ctime_set_current(inode);
return 0;
}
@@ -659,7 +661,7 @@ int simple_fill_super(struct super_block *s, unsigned long magic,
*/
inode->i_ino = 1;
inode->i_mode = S_IFDIR | 0755;
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
inode->i_op = &simple_dir_inode_operations;
inode->i_fop = &simple_dir_operations;
set_nlink(inode, 2);
@@ -685,7 +687,7 @@ int simple_fill_super(struct super_block *s, unsigned long magic,
goto out;
}
inode->i_mode = S_IFREG | files->mode;
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
inode->i_fop = files->ops;
inode->i_ino = i;
d_add(dentry, inode);
@@ -1253,7 +1255,7 @@ struct inode *alloc_anon_inode(struct super_block *s)
inode->i_uid = current_fsuid();
inode->i_gid = current_fsgid();
inode->i_flags |= S_PRIVATE;
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
return inode;
}
EXPORT_SYMBOL(alloc_anon_inode);
diff --git a/fs/nsfs.c b/fs/nsfs.c
index f602a96a1afe..c052cc55eacd 100644
--- a/fs/nsfs.c
+++ b/fs/nsfs.c
@@ -84,7 +84,7 @@ static int __ns_get_path(struct path *path, struct ns_common *ns)
return -ENOMEM;
}
inode->i_ino = ns->inum;
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
inode->i_flags |= S_IMMUTABLE;
inode->i_mode = S_IFREG | S_IRUGO;
inode->i_fop = &ns_file_operations;
diff --git a/fs/pipe.c b/fs/pipe.c
index 2d88f73f585a..bb90b6fc4a96 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -899,7 +899,7 @@ static struct inode * get_pipe_inode(void)
inode->i_mode = S_IFIFO | S_IRUSR | S_IWUSR;
inode->i_uid = current_fsuid();
inode->i_gid = current_fsgid();
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
return inode;
diff --git a/fs/posix_acl.c b/fs/posix_acl.c
index 7fa1b738bbab..cc9c390fd2af 100644
--- a/fs/posix_acl.c
+++ b/fs/posix_acl.c
@@ -1027,7 +1027,7 @@ int simple_set_acl(struct mnt_idmap *idmap, struct dentry *dentry,
return error;
}
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
if (IS_I_VERSION(inode))
inode_inc_iversion(inode);
set_cached_acl(inode, type, acl);
diff --git a/fs/stack.c b/fs/stack.c
index c9830924eb12..efd0de85bace 100644
--- a/fs/stack.c
+++ b/fs/stack.c
@@ -68,7 +68,7 @@ void fsstack_copy_attr_all(struct inode *dest, const struct inode *src)
dest->i_rdev = src->i_rdev;
dest->i_atime = src->i_atime;
dest->i_mtime = src->i_mtime;
- dest->i_ctime = src->i_ctime;
+ inode_ctime_set(dest, inode_ctime_peek(src));
dest->i_blkbits = src->i_blkbits;
dest->i_flags = src->i_flags;
set_nlink(dest, src->i_nlink);
diff --git a/fs/stat.c b/fs/stat.c
index 7c238da22ef0..5d87e34d6dd5 100644
--- a/fs/stat.c
+++ b/fs/stat.c
@@ -58,7 +58,7 @@ void generic_fillattr(struct mnt_idmap *idmap, struct inode *inode,
stat->size = i_size_read(inode);
stat->atime = inode->i_atime;
stat->mtime = inode->i_mtime;
- stat->ctime = inode->i_ctime;
+ stat->ctime = inode_ctime_peek(inode);
stat->blksize = i_blocksize(inode);
stat->blocks = inode->i_blocks;
}
diff --git a/include/linux/fs_stack.h b/include/linux/fs_stack.h
index 54210a42c30d..1488a118fe91 100644
--- a/include/linux/fs_stack.h
+++ b/include/linux/fs_stack.h
@@ -24,7 +24,7 @@ static inline void fsstack_copy_attr_times(struct inode *dest,
{
dest->i_atime = src->i_atime;
dest->i_mtime = src->i_mtime;
- dest->i_ctime = src->i_ctime;
+ inode_ctime_set(dest, inode_ctime_peek(src));
}
#endif /* _LINUX_FS_STACK_H */
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 13/79] autofs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (10 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
2023-06-21 16:43 ` Jan Kara
2023-06-27 1:48 ` Ian Kent
-1 siblings, 2 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Ian Kent; +Cc: Al Viro, Jan Kara, autofs, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/autofs/inode.c | 2 +-
fs/autofs/root.c | 6 +++---
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/fs/autofs/inode.c b/fs/autofs/inode.c
index affa70360b1f..47e3054b29dc 100644
--- a/fs/autofs/inode.c
+++ b/fs/autofs/inode.c
@@ -370,7 +370,7 @@ struct inode *autofs_get_inode(struct super_block *sb, umode_t mode)
inode->i_uid = d_inode(sb->s_root)->i_uid;
inode->i_gid = d_inode(sb->s_root)->i_gid;
}
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
inode->i_ino = get_next_ino();
if (S_ISDIR(mode)) {
diff --git a/fs/autofs/root.c b/fs/autofs/root.c
index 93046c9dc461..4c0fc0f8d688 100644
--- a/fs/autofs/root.c
+++ b/fs/autofs/root.c
@@ -600,7 +600,7 @@ static int autofs_dir_symlink(struct mnt_idmap *idmap,
p_ino = autofs_dentry_ino(dentry->d_parent);
p_ino->count++;
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
return 0;
}
@@ -633,7 +633,7 @@ static int autofs_dir_unlink(struct inode *dir, struct dentry *dentry)
d_inode(dentry)->i_size = 0;
clear_nlink(d_inode(dentry));
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
spin_lock(&sbi->lookup_lock);
__autofs_add_expiring(dentry);
@@ -749,7 +749,7 @@ static int autofs_dir_mkdir(struct mnt_idmap *idmap,
p_ino = autofs_dentry_ino(dentry->d_parent);
p_ino->count++;
inc_nlink(dir);
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
return 0;
}
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 14/79] befs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (11 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
2023-06-21 16:44 ` Jan Kara
-1 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Luis de Bethencourt, Salah Triki
Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/befs/linuxvfs.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c
index eee9237386e2..1eee7baa808b 100644
--- a/fs/befs/linuxvfs.c
+++ b/fs/befs/linuxvfs.c
@@ -363,7 +363,7 @@ static struct inode *befs_iget(struct super_block *sb, unsigned long ino)
inode->i_mtime.tv_sec =
fs64_to_cpu(sb, raw_inode->last_modified_time) >> 16;
inode->i_mtime.tv_nsec = 0; /* lower 16 bits are not a time */
- inode->i_ctime = inode->i_mtime;
+ inode_ctime_set(inode, inode->i_mtime);
inode->i_atime = inode->i_mtime;
befs_ino->i_inode_num = fsrun_to_cpu(sb, raw_inode->inode_num);
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 15/79] bfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (12 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
2023-06-21 16:48 ` Jan Kara
-1 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Tigran A. Aivazian; +Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/bfs/dir.c | 16 ++++++++--------
fs/bfs/inode.c | 6 +++---
2 files changed, 11 insertions(+), 11 deletions(-)
diff --git a/fs/bfs/dir.c b/fs/bfs/dir.c
index d2e8a2a56b05..5bcfc6e481bc 100644
--- a/fs/bfs/dir.c
+++ b/fs/bfs/dir.c
@@ -97,7 +97,7 @@ static int bfs_create(struct mnt_idmap *idmap, struct inode *dir,
set_bit(ino, info->si_imap);
info->si_freei--;
inode_init_owner(&nop_mnt_idmap, inode, dir, mode);
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
inode->i_blocks = 0;
inode->i_op = &bfs_file_inops;
inode->i_fop = &bfs_file_operations;
@@ -158,7 +158,7 @@ static int bfs_link(struct dentry *old, struct inode *dir,
return err;
}
inc_nlink(inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
mark_inode_dirty(inode);
ihold(inode);
d_instantiate(new, inode);
@@ -187,9 +187,9 @@ static int bfs_unlink(struct inode *dir, struct dentry *dentry)
}
de->ino = 0;
mark_buffer_dirty_inode(bh, dir);
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
mark_inode_dirty(dir);
- inode->i_ctime = dir->i_ctime;
+ inode_ctime_set(inode, inode_ctime_peek(dir));
inode_dec_link_count(inode);
error = 0;
@@ -240,10 +240,10 @@ static int bfs_rename(struct mnt_idmap *idmap, struct inode *old_dir,
goto end_rename;
}
old_de->ino = 0;
- old_dir->i_ctime = old_dir->i_mtime = current_time(old_dir);
+ old_dir->i_mtime = inode_ctime_set_current(old_dir);
mark_inode_dirty(old_dir);
if (new_inode) {
- new_inode->i_ctime = current_time(new_inode);
+ inode_ctime_set_current(new_inode);
inode_dec_link_count(new_inode);
}
mark_buffer_dirty_inode(old_bh, old_dir);
@@ -292,9 +292,9 @@ static int bfs_add_entry(struct inode *dir, const struct qstr *child, int ino)
pos = (block - sblock) * BFS_BSIZE + off;
if (pos >= dir->i_size) {
dir->i_size += BFS_DIRENT_SIZE;
- dir->i_ctime = current_time(dir);
+ inode_ctime_set_current(dir);
}
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
mark_inode_dirty(dir);
de->ino = cpu_to_le16((u16)ino);
for (i = 0; i < BFS_NAMELEN; i++)
diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c
index 1926bec2c850..c964316be32b 100644
--- a/fs/bfs/inode.c
+++ b/fs/bfs/inode.c
@@ -82,10 +82,10 @@ struct inode *bfs_iget(struct super_block *sb, unsigned long ino)
inode->i_blocks = BFS_FILEBLOCKS(di);
inode->i_atime.tv_sec = le32_to_cpu(di->i_atime);
inode->i_mtime.tv_sec = le32_to_cpu(di->i_mtime);
- inode->i_ctime.tv_sec = le32_to_cpu(di->i_ctime);
+ inode_ctime_set_sec(inode, le32_to_cpu(di->i_ctime));
inode->i_atime.tv_nsec = 0;
inode->i_mtime.tv_nsec = 0;
- inode->i_ctime.tv_nsec = 0;
+ inode_ctime_set_nsec(inode, 0);
brelse(bh);
unlock_new_inode(inode);
@@ -143,7 +143,7 @@ static int bfs_write_inode(struct inode *inode, struct writeback_control *wbc)
di->i_nlink = cpu_to_le32(inode->i_nlink);
di->i_atime = cpu_to_le32(inode->i_atime.tv_sec);
di->i_mtime = cpu_to_le32(inode->i_mtime.tv_sec);
- di->i_ctime = cpu_to_le32(inode->i_ctime.tv_sec);
+ di->i_ctime = cpu_to_le32(inode_ctime_peek(inode).tv_sec);
i_sblock = BFS_I(inode)->i_sblock;
di->i_sblock = cpu_to_le32(i_sblock);
di->i_eblock = cpu_to_le32(BFS_I(inode)->i_eblock);
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 16/79] btrfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (13 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
2023-06-22 11:42 ` David Sterba
-1 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Chris Mason, Josef Bacik, David Sterba
Cc: Al Viro, Jan Kara, linux-btrfs, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/btrfs/delayed-inode.c | 10 +++---
fs/btrfs/file.c | 24 +++++----------
fs/btrfs/inode.c | 66 ++++++++++++++++------------------------
fs/btrfs/ioctl.c | 2 +-
fs/btrfs/reflink.c | 7 ++---
fs/btrfs/transaction.c | 3 +-
fs/btrfs/tree-log.c | 4 +--
fs/btrfs/xattr.c | 4 +--
8 files changed, 48 insertions(+), 72 deletions(-)
diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c
index c0a6a1784697..ae493a4dc206 100644
--- a/fs/btrfs/delayed-inode.c
+++ b/fs/btrfs/delayed-inode.c
@@ -1808,9 +1808,9 @@ static void fill_stack_inode_item(struct btrfs_trans_handle *trans,
inode->i_mtime.tv_nsec);
btrfs_set_stack_timespec_sec(&inode_item->ctime,
- inode->i_ctime.tv_sec);
+ inode_ctime_peek(inode).tv_sec);
btrfs_set_stack_timespec_nsec(&inode_item->ctime,
- inode->i_ctime.tv_nsec);
+ inode_ctime_peek(inode).tv_nsec);
btrfs_set_stack_timespec_sec(&inode_item->otime,
BTRFS_I(inode)->i_otime.tv_sec);
@@ -1861,8 +1861,10 @@ int btrfs_fill_inode(struct inode *inode, u32 *rdev)
inode->i_mtime.tv_sec = btrfs_stack_timespec_sec(&inode_item->mtime);
inode->i_mtime.tv_nsec = btrfs_stack_timespec_nsec(&inode_item->mtime);
- inode->i_ctime.tv_sec = btrfs_stack_timespec_sec(&inode_item->ctime);
- inode->i_ctime.tv_nsec = btrfs_stack_timespec_nsec(&inode_item->ctime);
+ inode_ctime_set_sec(inode,
+ btrfs_stack_timespec_sec(&inode_item->ctime));
+ inode_ctime_set_nsec(inode,
+ btrfs_stack_timespec_nsec(&inode_item->ctime));
BTRFS_I(inode)->i_otime.tv_sec =
btrfs_stack_timespec_sec(&inode_item->otime);
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index fd03e689a6be..b4082b322b41 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -1108,17 +1108,10 @@ void btrfs_check_nocow_unlock(struct btrfs_inode *inode)
static void update_time_for_write(struct inode *inode)
{
- struct timespec64 now;
-
if (IS_NOCMTIME(inode))
return;
- now = current_time(inode);
- if (!timespec64_equal(&inode->i_mtime, &now))
- inode->i_mtime = now;
-
- if (!timespec64_equal(&inode->i_ctime, &now))
- inode->i_ctime = now;
+ inode->i_mtime = inode_ctime_set_current(inode);
if (IS_I_VERSION(inode))
inode_inc_iversion(inode);
@@ -2460,8 +2453,9 @@ int btrfs_replace_file_extents(struct btrfs_inode *inode,
inode_inc_iversion(&inode->vfs_inode);
if (!extent_info || extent_info->update_times) {
- inode->vfs_inode.i_mtime = current_time(&inode->vfs_inode);
- inode->vfs_inode.i_ctime = inode->vfs_inode.i_mtime;
+ struct inode *vinode = &inode->vfs_inode;
+
+ vinode->i_mtime = inode_ctime_set_current(vinode);
}
ret = btrfs_update_inode(trans, root, inode);
@@ -2703,8 +2697,7 @@ static int btrfs_punch_hole(struct file *file, loff_t offset, loff_t len)
ASSERT(trans != NULL);
inode_inc_iversion(inode);
- inode->i_mtime = current_time(inode);
- inode->i_ctime = inode->i_mtime;
+ inode->i_mtime = inode_ctime_set_current(inode);
ret = btrfs_update_inode(trans, root, BTRFS_I(inode));
updated_inode = true;
btrfs_end_transaction(trans);
@@ -2721,11 +2714,8 @@ static int btrfs_punch_hole(struct file *file, loff_t offset, loff_t len)
* for detecting, at fsync time, if the inode isn't yet in the
* log tree or it's there but not up to date.
*/
- struct timespec64 now = current_time(inode);
-
inode_inc_iversion(inode);
- inode->i_mtime = now;
- inode->i_ctime = now;
+ inode->i_mtime = inode_ctime_set_current(inode);
trans = btrfs_start_transaction(root, 1);
if (IS_ERR(trans)) {
ret = PTR_ERR(trans);
@@ -2796,7 +2786,7 @@ static int btrfs_fallocate_update_isize(struct inode *inode,
if (IS_ERR(trans))
return PTR_ERR(trans);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
i_size_write(inode, end);
btrfs_inode_safe_disk_i_size_write(BTRFS_I(inode), 0);
ret = btrfs_update_inode(trans, root, BTRFS_I(inode));
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 601fdc956484..4bbb6c6a7516 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -3901,8 +3901,10 @@ static int btrfs_read_locked_inode(struct inode *inode,
inode->i_mtime.tv_sec = btrfs_timespec_sec(leaf, &inode_item->mtime);
inode->i_mtime.tv_nsec = btrfs_timespec_nsec(leaf, &inode_item->mtime);
- inode->i_ctime.tv_sec = btrfs_timespec_sec(leaf, &inode_item->ctime);
- inode->i_ctime.tv_nsec = btrfs_timespec_nsec(leaf, &inode_item->ctime);
+ inode_ctime_set_sec(inode,
+ btrfs_timespec_sec(leaf, &inode_item->ctime));
+ inode_ctime_set_nsec(inode,
+ btrfs_timespec_nsec(leaf, &inode_item->ctime));
BTRFS_I(inode)->i_otime.tv_sec =
btrfs_timespec_sec(leaf, &inode_item->otime);
@@ -4073,9 +4075,9 @@ static void fill_inode_item(struct btrfs_trans_handle *trans,
inode->i_mtime.tv_nsec);
btrfs_set_token_timespec_sec(&token, &item->ctime,
- inode->i_ctime.tv_sec);
+ inode_ctime_peek(inode).tv_sec);
btrfs_set_token_timespec_nsec(&token, &item->ctime,
- inode->i_ctime.tv_nsec);
+ inode_ctime_peek(inode).tv_nsec);
btrfs_set_token_timespec_sec(&token, &item->otime,
BTRFS_I(inode)->i_otime.tv_sec);
@@ -4273,9 +4275,8 @@ static int __btrfs_unlink_inode(struct btrfs_trans_handle *trans,
btrfs_i_size_write(dir, dir->vfs_inode.i_size - name->len * 2);
inode_inc_iversion(&inode->vfs_inode);
inode_inc_iversion(&dir->vfs_inode);
- inode->vfs_inode.i_ctime = current_time(&inode->vfs_inode);
- dir->vfs_inode.i_mtime = inode->vfs_inode.i_ctime;
- dir->vfs_inode.i_ctime = inode->vfs_inode.i_ctime;
+ inode_ctime_set_current(&inode->vfs_inode);
+ dir->vfs_inode.i_mtime = inode_ctime_set_current(&dir->vfs_inode);
ret = btrfs_update_inode(trans, root, dir);
out:
return ret;
@@ -4448,8 +4449,7 @@ static int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
btrfs_i_size_write(dir, dir->vfs_inode.i_size - fname.disk_name.len * 2);
inode_inc_iversion(&dir->vfs_inode);
- dir->vfs_inode.i_mtime = current_time(&dir->vfs_inode);
- dir->vfs_inode.i_ctime = dir->vfs_inode.i_mtime;
+ dir->vfs_inode.i_mtime = inode_ctime_set_current(&dir->vfs_inode);
ret = btrfs_update_inode_fallback(trans, root, dir);
if (ret)
btrfs_abort_transaction(trans, ret);
@@ -5090,10 +5090,8 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr)
*/
if (newsize != oldsize) {
inode_inc_iversion(inode);
- if (!(mask & (ATTR_CTIME | ATTR_MTIME))) {
- inode->i_mtime = current_time(inode);
- inode->i_ctime = inode->i_mtime;
- }
+ if (!(mask & (ATTR_CTIME | ATTR_MTIME)))
+ inode->i_mtime = inode_ctime_set_current(inode);
}
if (newsize > oldsize) {
@@ -5736,9 +5734,7 @@ static struct inode *new_simple_dir(struct super_block *s,
inode->i_opflags &= ~IOP_XATTR;
inode->i_fop = &simple_dir_operations;
inode->i_mode = S_IFDIR | S_IRUGO | S_IWUSR | S_IXUGO;
- inode->i_mtime = current_time(inode);
- inode->i_atime = inode->i_mtime;
- inode->i_ctime = inode->i_mtime;
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
BTRFS_I(inode)->i_otime = inode->i_mtime;
return inode;
@@ -6075,7 +6071,7 @@ static int btrfs_update_time(struct inode *inode, struct timespec64 *now,
if (flags & S_VERSION)
dirty |= inode_maybe_inc_iversion(inode, dirty);
if (flags & S_CTIME)
- inode->i_ctime = *now;
+ inode_ctime_set(inode, *now);
if (flags & S_MTIME)
inode->i_mtime = *now;
if (flags & S_ATIME)
@@ -6378,9 +6374,7 @@ int btrfs_create_new_inode(struct btrfs_trans_handle *trans,
goto discard;
}
- inode->i_mtime = current_time(inode);
- inode->i_atime = inode->i_mtime;
- inode->i_ctime = inode->i_mtime;
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
BTRFS_I(inode)->i_otime = inode->i_mtime;
/*
@@ -6545,12 +6539,9 @@ int btrfs_add_link(struct btrfs_trans_handle *trans,
* log replay procedure is responsible for setting them to their correct
* values (the ones it had when the fsync was done).
*/
- if (!test_bit(BTRFS_FS_LOG_RECOVERING, &root->fs_info->flags)) {
- struct timespec64 now = current_time(&parent_inode->vfs_inode);
+ if (!test_bit(BTRFS_FS_LOG_RECOVERING, &root->fs_info->flags))
+ parent_inode->vfs_inode.i_mtime = inode_ctime_set_current(&parent_inode->vfs_inode);
- parent_inode->vfs_inode.i_mtime = now;
- parent_inode->vfs_inode.i_ctime = now;
- }
ret = btrfs_update_inode(trans, root, parent_inode);
if (ret)
btrfs_abort_transaction(trans, ret);
@@ -6690,7 +6681,7 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
BTRFS_I(inode)->dir_index = 0ULL;
inc_nlink(inode);
inode_inc_iversion(inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
ihold(inode);
set_bit(BTRFS_INODE_COPY_EVERYTHING, &BTRFS_I(inode)->runtime_flags);
@@ -8778,7 +8769,6 @@ static int btrfs_rename_exchange(struct inode *old_dir,
struct btrfs_root *dest = BTRFS_I(new_dir)->root;
struct inode *new_inode = new_dentry->d_inode;
struct inode *old_inode = old_dentry->d_inode;
- struct timespec64 ctime = current_time(old_inode);
struct btrfs_rename_ctx old_rename_ctx;
struct btrfs_rename_ctx new_rename_ctx;
u64 old_ino = btrfs_ino(BTRFS_I(old_inode));
@@ -8909,12 +8899,10 @@ static int btrfs_rename_exchange(struct inode *old_dir,
inode_inc_iversion(new_dir);
inode_inc_iversion(old_inode);
inode_inc_iversion(new_inode);
- old_dir->i_mtime = ctime;
- old_dir->i_ctime = ctime;
- new_dir->i_mtime = ctime;
- new_dir->i_ctime = ctime;
- old_inode->i_ctime = ctime;
- new_inode->i_ctime = ctime;
+ old_dir->i_mtime = inode_ctime_set_current(old_dir);
+ new_dir->i_mtime = inode_ctime_set_current(new_dir);
+ inode_ctime_set_current(old_inode);
+ inode_ctime_set_current(new_inode);
if (old_dentry->d_parent != new_dentry->d_parent) {
btrfs_record_unlink_dir(trans, BTRFS_I(old_dir),
@@ -9178,11 +9166,9 @@ static int btrfs_rename(struct mnt_idmap *idmap,
inode_inc_iversion(old_dir);
inode_inc_iversion(new_dir);
inode_inc_iversion(old_inode);
- old_dir->i_mtime = current_time(old_dir);
- old_dir->i_ctime = old_dir->i_mtime;
- new_dir->i_mtime = old_dir->i_mtime;
- new_dir->i_ctime = old_dir->i_mtime;
- old_inode->i_ctime = old_dir->i_mtime;
+ old_dir->i_mtime = inode_ctime_set_current(old_dir);
+ new_dir->i_mtime = inode_ctime_set_current(new_dir);
+ inode_ctime_set_current(old_inode);
if (old_dentry->d_parent != new_dentry->d_parent)
btrfs_record_unlink_dir(trans, BTRFS_I(old_dir),
@@ -9204,7 +9190,7 @@ static int btrfs_rename(struct mnt_idmap *idmap,
if (new_inode) {
inode_inc_iversion(new_inode);
- new_inode->i_ctime = current_time(new_inode);
+ inode_ctime_set_current(new_inode);
if (unlikely(btrfs_ino(BTRFS_I(new_inode)) ==
BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)) {
ret = btrfs_unlink_subvol(trans, BTRFS_I(new_dir), new_dentry);
@@ -9744,7 +9730,7 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
*alloc_hint = ins.objectid + ins.offset;
inode_inc_iversion(inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
BTRFS_I(inode)->flags |= BTRFS_INODE_PREALLOC;
if (!(mode & FALLOC_FL_KEEP_SIZE) &&
(actual_len > inode->i_size) &&
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index a895d105464b..3d50bd67ec85 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -384,7 +384,7 @@ int btrfs_fileattr_set(struct mnt_idmap *idmap,
binode->flags = binode_flags;
btrfs_sync_inode_flags_to_i_flags(inode);
inode_inc_iversion(inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
ret = btrfs_update_inode(trans, root, BTRFS_I(inode));
out_end_trans:
diff --git a/fs/btrfs/reflink.c b/fs/btrfs/reflink.c
index 0474bbe39da7..3cf4716dea46 100644
--- a/fs/btrfs/reflink.c
+++ b/fs/btrfs/reflink.c
@@ -29,10 +29,9 @@ static int clone_finish_inode_update(struct btrfs_trans_handle *trans,
int ret;
inode_inc_iversion(inode);
- if (!no_time_update) {
- inode->i_mtime = current_time(inode);
- inode->i_ctime = inode->i_mtime;
- }
+ if (!no_time_update)
+ inode->i_mtime = inode_ctime_set_current(inode);
+
/*
* We round up to the block size at eof when determining which
* extents to clone above, but shouldn't round up the file size.
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index e7cfc992e02a..138f919646e2 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -1831,8 +1831,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
btrfs_i_size_write(BTRFS_I(parent_inode), parent_inode->i_size +
fname.disk_name.len * 2);
- parent_inode->i_mtime = current_time(parent_inode);
- parent_inode->i_ctime = parent_inode->i_mtime;
+ parent_inode->i_mtime = inode_ctime_set_current(parent_inode);
ret = btrfs_update_inode_fallback(trans, parent_root, BTRFS_I(parent_inode));
if (ret) {
btrfs_abort_transaction(trans, ret);
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 365a1cc0a3c3..c862d1450365 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -4148,9 +4148,9 @@ static void fill_inode_item(struct btrfs_trans_handle *trans,
inode->i_mtime.tv_nsec);
btrfs_set_token_timespec_sec(&token, &item->ctime,
- inode->i_ctime.tv_sec);
+ inode_ctime_peek(inode).tv_sec);
btrfs_set_token_timespec_nsec(&token, &item->ctime,
- inode->i_ctime.tv_nsec);
+ inode_ctime_peek(inode).tv_nsec);
/*
* We do not need to set the nbytes field, in fact during a fast fsync
diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c
index fc4b20c2688a..a2d331436963 100644
--- a/fs/btrfs/xattr.c
+++ b/fs/btrfs/xattr.c
@@ -264,7 +264,7 @@ int btrfs_setxattr_trans(struct inode *inode, const char *name,
goto out;
inode_inc_iversion(inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
ret = btrfs_update_inode(trans, root, BTRFS_I(inode));
if (ret)
btrfs_abort_transaction(trans, ret);
@@ -407,7 +407,7 @@ static int btrfs_xattr_handler_set_prop(const struct xattr_handler *handler,
ret = btrfs_set_prop(trans, inode, name, value, size, flags);
if (!ret) {
inode_inc_iversion(inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
ret = btrfs_update_inode(trans, root, BTRFS_I(inode));
if (ret)
btrfs_abort_transaction(trans, ret);
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 17/79] ceph: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (14 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
2023-06-26 0:56 ` Xiubo Li
-1 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Xiubo Li, Ilya Dryomov
Cc: Al Viro, Jan Kara, ceph-devel, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/ceph/acl.c | 2 +-
fs/ceph/caps.c | 2 +-
fs/ceph/inode.c | 17 ++++++++++-------
fs/ceph/snap.c | 2 +-
fs/ceph/xattr.c | 2 +-
5 files changed, 14 insertions(+), 11 deletions(-)
diff --git a/fs/ceph/acl.c b/fs/ceph/acl.c
index 6945a938d396..a3de2b9c3a68 100644
--- a/fs/ceph/acl.c
+++ b/fs/ceph/acl.c
@@ -93,7 +93,7 @@ int ceph_set_acl(struct mnt_idmap *idmap, struct dentry *dentry,
char *value = NULL;
struct iattr newattrs;
struct inode *inode = d_inode(dentry);
- struct timespec64 old_ctime = inode->i_ctime;
+ struct timespec64 old_ctime = inode_ctime_peek(inode);
umode_t new_mode = inode->i_mode, old_mode = inode->i_mode;
if (ceph_snap(inode) != CEPH_NOSNAP) {
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index 2321e5ddb664..c144a07e334e 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -1400,7 +1400,7 @@ static void __prep_cap(struct cap_msg_args *arg, struct ceph_cap *cap,
arg->mtime = inode->i_mtime;
arg->atime = inode->i_atime;
- arg->ctime = inode->i_ctime;
+ arg->ctime = inode_ctime_peek(inode);
arg->btime = ci->i_btime;
arg->change_attr = inode_peek_iversion_raw(inode);
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index 8e5f41d45283..f0b3b11d695e 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -100,7 +100,7 @@ struct inode *ceph_get_snapdir(struct inode *parent)
inode->i_uid = parent->i_uid;
inode->i_gid = parent->i_gid;
inode->i_mtime = parent->i_mtime;
- inode->i_ctime = parent->i_ctime;
+ inode_ctime_set(inode, inode_ctime_peek(parent));
inode->i_atime = parent->i_atime;
ci->i_rbytes = 0;
ci->i_btime = ceph_inode(parent)->i_btime;
@@ -695,12 +695,14 @@ void ceph_fill_file_time(struct inode *inode, int issued,
CEPH_CAP_FILE_BUFFER|
CEPH_CAP_AUTH_EXCL|
CEPH_CAP_XATTR_EXCL)) {
+ struct timespec64 ictime = inode_ctime_peek(inode);
+
if (ci->i_version == 0 ||
- timespec64_compare(ctime, &inode->i_ctime) > 0) {
+ timespec64_compare(ctime, &ictime) > 0) {
dout("ctime %lld.%09ld -> %lld.%09ld inc w/ cap\n",
- inode->i_ctime.tv_sec, inode->i_ctime.tv_nsec,
+ ictime.tv_sec, ictime.tv_nsec,
ctime->tv_sec, ctime->tv_nsec);
- inode->i_ctime = *ctime;
+ inode_ctime_set(inode, *ctime);
}
if (ci->i_version == 0 ||
ceph_seq_cmp(time_warp_seq, ci->i_time_warp_seq) > 0) {
@@ -738,7 +740,7 @@ void ceph_fill_file_time(struct inode *inode, int issued,
} else {
/* we have no write|excl caps; whatever the MDS says is true */
if (ceph_seq_cmp(time_warp_seq, ci->i_time_warp_seq) >= 0) {
- inode->i_ctime = *ctime;
+ inode_ctime_set(inode, *ctime);
inode->i_mtime = *mtime;
inode->i_atime = *atime;
ci->i_time_warp_seq = time_warp_seq;
@@ -2166,7 +2168,8 @@ int __ceph_setattr(struct inode *inode, struct iattr *attr)
bool only = (ia_valid & (ATTR_SIZE|ATTR_MTIME|ATTR_ATIME|
ATTR_MODE|ATTR_UID|ATTR_GID)) == 0;
dout("setattr %p ctime %lld.%ld -> %lld.%ld (%s)\n", inode,
- inode->i_ctime.tv_sec, inode->i_ctime.tv_nsec,
+ inode_ctime_peek(inode).tv_sec,
+ inode_ctime_peek(inode).tv_nsec,
attr->ia_ctime.tv_sec, attr->ia_ctime.tv_nsec,
only ? "ctime only" : "ignored");
if (only) {
@@ -2191,7 +2194,7 @@ int __ceph_setattr(struct inode *inode, struct iattr *attr)
if (dirtied) {
inode_dirty_flags = __ceph_mark_dirty_caps(ci, dirtied,
&prealloc_cf);
- inode->i_ctime = attr->ia_ctime;
+ inode_ctime_set(inode, attr->ia_ctime);
inode_inc_iversion_raw(inode);
}
diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c
index 2e73ba62bd7a..f02df070fa84 100644
--- a/fs/ceph/snap.c
+++ b/fs/ceph/snap.c
@@ -660,7 +660,7 @@ int __ceph_finish_cap_snap(struct ceph_inode_info *ci,
capsnap->size = i_size_read(inode);
capsnap->mtime = inode->i_mtime;
capsnap->atime = inode->i_atime;
- capsnap->ctime = inode->i_ctime;
+ capsnap->ctime = inode_ctime_peek(inode);
capsnap->btime = ci->i_btime;
capsnap->change_attr = inode_peek_iversion_raw(inode);
capsnap->time_warp_seq = ci->i_time_warp_seq;
diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c
index 806183959c47..8e217f7f58bd 100644
--- a/fs/ceph/xattr.c
+++ b/fs/ceph/xattr.c
@@ -1238,7 +1238,7 @@ int __ceph_setxattr(struct inode *inode, const char *name,
dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_XATTR_EXCL,
&prealloc_cf);
ci->i_xattrs.dirty = true;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
}
spin_unlock(&ci->i_ceph_lock);
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 18/79] coda: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (15 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Jan Harkes, coda
Cc: Al Viro, Jan Kara, codalist, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/coda/coda_linux.c | 2 +-
fs/coda/dir.c | 2 +-
fs/coda/file.c | 2 +-
fs/coda/inode.c | 2 +-
4 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/fs/coda/coda_linux.c b/fs/coda/coda_linux.c
index 903ca8fa4b9b..2a6187ca52d9 100644
--- a/fs/coda/coda_linux.c
+++ b/fs/coda/coda_linux.c
@@ -127,7 +127,7 @@ void coda_vattr_to_iattr(struct inode *inode, struct coda_vattr *attr)
if (attr->va_mtime.tv_sec != -1)
inode->i_mtime = coda_to_timespec64(attr->va_mtime);
if (attr->va_ctime.tv_sec != -1)
- inode->i_ctime = coda_to_timespec64(attr->va_ctime);
+ inode_ctime_set(inode, coda_to_timespec64(attr->va_ctime));
}
diff --git a/fs/coda/dir.c b/fs/coda/dir.c
index 8450b1bd354b..b158a505570c 100644
--- a/fs/coda/dir.c
+++ b/fs/coda/dir.c
@@ -111,7 +111,7 @@ static inline void coda_dir_update_mtime(struct inode *dir)
/* optimistically we can also act as if our nose bleeds. The
* granularity of the mtime is coarse anyways so we might actually be
* right most of the time. Note: we only do this for directories. */
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
#endif
}
diff --git a/fs/coda/file.c b/fs/coda/file.c
index 12b26bd13564..18684d73d085 100644
--- a/fs/coda/file.c
+++ b/fs/coda/file.c
@@ -84,7 +84,7 @@ coda_file_write_iter(struct kiocb *iocb, struct iov_iter *to)
ret = vfs_iter_write(cfi->cfi_container, to, &iocb->ki_pos, 0);
coda_inode->i_size = file_inode(host_file)->i_size;
coda_inode->i_blocks = (coda_inode->i_size + 511) >> 9;
- coda_inode->i_mtime = coda_inode->i_ctime = current_time(coda_inode);
+ coda_inode->i_mtime = inode_ctime_set_current(coda_inode);
inode_unlock(coda_inode);
file_end_write(host_file);
diff --git a/fs/coda/inode.c b/fs/coda/inode.c
index d661e6cf17ac..996729c3041a 100644
--- a/fs/coda/inode.c
+++ b/fs/coda/inode.c
@@ -269,7 +269,7 @@ int coda_setattr(struct mnt_idmap *idmap, struct dentry *de,
memset(&vattr, 0, sizeof(vattr));
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
coda_iattr_to_vattr(iattr, &vattr);
vattr.va_type = C_VNON; /* cannot set type */
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 19/79] configfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (16 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Joel Becker, Christoph Hellwig
Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/configfs/inode.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/fs/configfs/inode.c b/fs/configfs/inode.c
index 1c15edbe70ff..21466f7f655f 100644
--- a/fs/configfs/inode.c
+++ b/fs/configfs/inode.c
@@ -89,7 +89,7 @@ static inline void set_default_inode_attr(struct inode * inode, umode_t mode)
{
inode->i_mode = mode;
inode->i_atime = inode->i_mtime =
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
}
static inline void set_inode_attr(struct inode * inode, struct iattr * iattr)
@@ -99,7 +99,7 @@ static inline void set_inode_attr(struct inode * inode, struct iattr * iattr)
inode->i_gid = iattr->ia_gid;
inode->i_atime = iattr->ia_atime;
inode->i_mtime = iattr->ia_mtime;
- inode->i_ctime = iattr->ia_ctime;
+ inode_ctime_set(inode, iattr->ia_ctime);
}
struct inode *configfs_new_inode(umode_t mode, struct configfs_dirent *sd,
@@ -172,7 +172,7 @@ struct inode *configfs_create(struct dentry *dentry, umode_t mode)
return ERR_PTR(-ENOMEM);
p_inode = d_inode(dentry->d_parent);
- p_inode->i_mtime = p_inode->i_ctime = current_time(p_inode);
+ p_inode->i_mtime = inode_ctime_set_current(p_inode);
configfs_set_inode_lock_class(sd, inode);
return inode;
}
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 20/79] cramfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (17 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
2023-06-21 15:29 ` Nicolas Pitre
-1 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Nicolas Pitre; +Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/cramfs/inode.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c
index 27c6597aa1be..039ce0f8858b 100644
--- a/fs/cramfs/inode.c
+++ b/fs/cramfs/inode.c
@@ -133,7 +133,7 @@ static struct inode *get_cramfs_inode(struct super_block *sb,
}
/* Struct copy intentional */
- inode->i_mtime = inode->i_atime = inode->i_ctime = zerotime;
+ inode->i_mtime = inode->i_atime = inode_ctime_set(inode, zerotime);
/* inode->i_nlink is left 1 - arguably wrong for directories,
but it's the best we can do without reading the directory
contents. 1 yields the right result in GNU find, even
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 21/79] debugfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (18 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
2023-06-21 15:29 ` Greg Kroah-Hartman
-1 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Greg Kroah-Hartman, Rafael J. Wysocki
Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/debugfs/inode.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c
index 3f81f73c241a..d4932a128c7d 100644
--- a/fs/debugfs/inode.c
+++ b/fs/debugfs/inode.c
@@ -73,7 +73,7 @@ static struct inode *debugfs_get_inode(struct super_block *sb)
if (inode) {
inode->i_ino = get_next_ino();
inode->i_atime = inode->i_mtime =
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
}
return inode;
}
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 22/79] devpts: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (19 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner; +Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/devpts/inode.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c
index fe3db0eda8e4..e07ad406cd52 100644
--- a/fs/devpts/inode.c
+++ b/fs/devpts/inode.c
@@ -338,7 +338,7 @@ static int mknod_ptmx(struct super_block *sb)
}
inode->i_ino = 2;
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
mode = S_IFCHR|opts->ptmxmode;
init_special_inode(inode, mode, MKDEV(TTYAUX_MAJOR, 2));
@@ -451,7 +451,7 @@ devpts_fill_super(struct super_block *s, void *data, int silent)
if (!inode)
goto fail;
inode->i_ino = 1;
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO | S_IWUSR;
inode->i_op = &simple_dir_inode_operations;
inode->i_fop = &simple_dir_operations;
@@ -560,7 +560,7 @@ struct dentry *devpts_pty_new(struct pts_fs_info *fsi, int index, void *priv)
inode->i_ino = index + 3;
inode->i_uid = opts->setuid ? opts->uid : current_fsuid();
inode->i_gid = opts->setgid ? opts->gid : current_fsgid();
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
init_special_inode(inode, S_IFCHR|opts->mode, MKDEV(UNIX98_PTY_SLAVE_MAJOR, index));
sprintf(s, "%d", index);
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 23/79] ecryptfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (20 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Tyler Hicks; +Cc: Al Viro, Jan Kara, ecryptfs, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/ecryptfs/inode.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c
index 83274915ba6d..3171214fdffd 100644
--- a/fs/ecryptfs/inode.c
+++ b/fs/ecryptfs/inode.c
@@ -148,7 +148,7 @@ static int ecryptfs_do_unlink(struct inode *dir, struct dentry *dentry,
}
fsstack_copy_attr_times(dir, lower_dir);
set_nlink(inode, ecryptfs_inode_to_lower(inode)->i_nlink);
- inode->i_ctime = dir->i_ctime;
+ inode_ctime_set(inode, inode_ctime_peek(dir));
out_unlock:
dput(lower_dentry);
inode_unlock(lower_dir);
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 24/79] efivarfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (21 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Jeremy Kerr, Ard Biesheuvel
Cc: Al Viro, Jan Kara, linux-efi, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/efivarfs/file.c | 2 +-
fs/efivarfs/inode.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/efivarfs/file.c b/fs/efivarfs/file.c
index 375576111dc3..46b65dfca153 100644
--- a/fs/efivarfs/file.c
+++ b/fs/efivarfs/file.c
@@ -51,7 +51,7 @@ static ssize_t efivarfs_file_write(struct file *file,
} else {
inode_lock(inode);
i_size_write(inode, datasize + sizeof(attributes));
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
inode_unlock(inode);
}
diff --git a/fs/efivarfs/inode.c b/fs/efivarfs/inode.c
index b973a2c03dde..a47dded39be8 100644
--- a/fs/efivarfs/inode.c
+++ b/fs/efivarfs/inode.c
@@ -25,7 +25,7 @@ struct inode *efivarfs_get_inode(struct super_block *sb,
if (inode) {
inode->i_ino = get_next_ino();
inode->i_mode = mode;
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
inode->i_flags = is_removable ? 0 : S_IMMUTABLE;
switch (mode & S_IFMT) {
case S_IFREG:
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 25/79] efs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (22 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner; +Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/efs/inode.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/fs/efs/inode.c b/fs/efs/inode.c
index 3ba94bb005a6..206978b490e6 100644
--- a/fs/efs/inode.c
+++ b/fs/efs/inode.c
@@ -105,8 +105,9 @@ struct inode *efs_iget(struct super_block *super, unsigned long ino)
inode->i_size = be32_to_cpu(efs_inode->di_size);
inode->i_atime.tv_sec = be32_to_cpu(efs_inode->di_atime);
inode->i_mtime.tv_sec = be32_to_cpu(efs_inode->di_mtime);
- inode->i_ctime.tv_sec = be32_to_cpu(efs_inode->di_ctime);
- inode->i_atime.tv_nsec = inode->i_mtime.tv_nsec = inode->i_ctime.tv_nsec = 0;
+ inode_ctime_set_sec(inode, be32_to_cpu(efs_inode->di_ctime));
+ inode->i_atime.tv_nsec = inode->i_mtime.tv_nsec = inode_ctime_set_nsec(inode,
+ 0);
/* this is the number of blocks in the file */
if (inode->i_size == 0) {
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 26/79] erofs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu
Cc: linux-erofs, Jan Kara, Al Viro, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/erofs/inode.c | 16 ++++++++--------
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/fs/erofs/inode.c b/fs/erofs/inode.c
index d70b12b81507..8af56d6d0ff3 100644
--- a/fs/erofs/inode.c
+++ b/fs/erofs/inode.c
@@ -105,8 +105,8 @@ static void *erofs_read_inode(struct erofs_buf *buf,
set_nlink(inode, le32_to_cpu(die->i_nlink));
/* extended inode has its own timestamp */
- inode->i_ctime.tv_sec = le64_to_cpu(die->i_mtime);
- inode->i_ctime.tv_nsec = le32_to_cpu(die->i_mtime_nsec);
+ inode_ctime_set_sec(inode, le64_to_cpu(die->i_mtime));
+ inode_ctime_set_nsec(inode, le32_to_cpu(die->i_mtime_nsec));
inode->i_size = le64_to_cpu(die->i_size);
@@ -148,8 +148,8 @@ static void *erofs_read_inode(struct erofs_buf *buf,
set_nlink(inode, le16_to_cpu(dic->i_nlink));
/* use build time for compact inodes */
- inode->i_ctime.tv_sec = sbi->build_time;
- inode->i_ctime.tv_nsec = sbi->build_time_nsec;
+ inode_ctime_set_sec(inode, sbi->build_time);
+ inode_ctime_set_nsec(inode, sbi->build_time_nsec);
inode->i_size = le32_to_cpu(dic->i_size);
if (erofs_inode_is_data_compressed(vi->datalayout))
@@ -176,10 +176,10 @@ static void *erofs_read_inode(struct erofs_buf *buf,
vi->chunkbits = sb->s_blocksize_bits +
(vi->chunkformat & EROFS_CHUNK_FORMAT_BLKBITS_MASK);
}
- inode->i_mtime.tv_sec = inode->i_ctime.tv_sec;
- inode->i_atime.tv_sec = inode->i_ctime.tv_sec;
- inode->i_mtime.tv_nsec = inode->i_ctime.tv_nsec;
- inode->i_atime.tv_nsec = inode->i_ctime.tv_nsec;
+ inode->i_mtime.tv_sec = inode_ctime_peek(inode).tv_sec;
+ inode->i_atime.tv_sec = inode_ctime_peek(inode).tv_sec;
+ inode->i_mtime.tv_nsec = inode_ctime_peek(inode).tv_nsec;
+ inode->i_atime.tv_nsec = inode_ctime_peek(inode).tv_nsec;
inode->i_flags &= ~S_DAX;
if (test_opt(&sbi->opt, DAX_ALWAYS) && S_ISREG(inode->i_mode) &&
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 26/79] erofs: switch to new ctime accessors
@ 2023-06-21 14:45 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu
Cc: Al Viro, Jan Kara, linux-erofs, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/erofs/inode.c | 16 ++++++++--------
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/fs/erofs/inode.c b/fs/erofs/inode.c
index d70b12b81507..8af56d6d0ff3 100644
--- a/fs/erofs/inode.c
+++ b/fs/erofs/inode.c
@@ -105,8 +105,8 @@ static void *erofs_read_inode(struct erofs_buf *buf,
set_nlink(inode, le32_to_cpu(die->i_nlink));
/* extended inode has its own timestamp */
- inode->i_ctime.tv_sec = le64_to_cpu(die->i_mtime);
- inode->i_ctime.tv_nsec = le32_to_cpu(die->i_mtime_nsec);
+ inode_ctime_set_sec(inode, le64_to_cpu(die->i_mtime));
+ inode_ctime_set_nsec(inode, le32_to_cpu(die->i_mtime_nsec));
inode->i_size = le64_to_cpu(die->i_size);
@@ -148,8 +148,8 @@ static void *erofs_read_inode(struct erofs_buf *buf,
set_nlink(inode, le16_to_cpu(dic->i_nlink));
/* use build time for compact inodes */
- inode->i_ctime.tv_sec = sbi->build_time;
- inode->i_ctime.tv_nsec = sbi->build_time_nsec;
+ inode_ctime_set_sec(inode, sbi->build_time);
+ inode_ctime_set_nsec(inode, sbi->build_time_nsec);
inode->i_size = le32_to_cpu(dic->i_size);
if (erofs_inode_is_data_compressed(vi->datalayout))
@@ -176,10 +176,10 @@ static void *erofs_read_inode(struct erofs_buf *buf,
vi->chunkbits = sb->s_blocksize_bits +
(vi->chunkformat & EROFS_CHUNK_FORMAT_BLKBITS_MASK);
}
- inode->i_mtime.tv_sec = inode->i_ctime.tv_sec;
- inode->i_atime.tv_sec = inode->i_ctime.tv_sec;
- inode->i_mtime.tv_nsec = inode->i_ctime.tv_nsec;
- inode->i_atime.tv_nsec = inode->i_ctime.tv_nsec;
+ inode->i_mtime.tv_sec = inode_ctime_peek(inode).tv_sec;
+ inode->i_atime.tv_sec = inode_ctime_peek(inode).tv_sec;
+ inode->i_mtime.tv_nsec = inode_ctime_peek(inode).tv_nsec;
+ inode->i_atime.tv_nsec = inode_ctime_peek(inode).tv_nsec;
inode->i_flags &= ~S_DAX;
if (test_opt(&sbi->opt, DAX_ALWAYS) && S_ISREG(inode->i_mode) &&
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 27/79] exfat: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (24 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Namjae Jeon, Sungjong Seo
Cc: Al Viro, Jan Kara, linux-fsdevel, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/exfat/file.c | 4 ++--
fs/exfat/inode.c | 6 +++---
fs/exfat/namei.c | 29 ++++++++++++++---------------
fs/exfat/super.c | 4 ++--
4 files changed, 21 insertions(+), 22 deletions(-)
diff --git a/fs/exfat/file.c b/fs/exfat/file.c
index 3cbd270e0cba..853ba8ec4095 100644
--- a/fs/exfat/file.c
+++ b/fs/exfat/file.c
@@ -22,7 +22,7 @@ static int exfat_cont_expand(struct inode *inode, loff_t size)
if (err)
return err;
- inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
mark_inode_dirty(inode);
if (!IS_SYNC(inode))
@@ -290,7 +290,7 @@ int exfat_setattr(struct mnt_idmap *idmap, struct dentry *dentry,
}
if (attr->ia_valid & ATTR_SIZE)
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
setattr_copy(&nop_mnt_idmap, inode, attr);
exfat_truncate_atime(&inode->i_atime);
diff --git a/fs/exfat/inode.c b/fs/exfat/inode.c
index 481dd338f2b8..b06b40b7c7b4 100644
--- a/fs/exfat/inode.c
+++ b/fs/exfat/inode.c
@@ -355,7 +355,7 @@ static void exfat_write_failed(struct address_space *mapping, loff_t to)
if (to > i_size_read(inode)) {
truncate_pagecache(inode, i_size_read(inode));
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
exfat_truncate(inode);
}
}
@@ -398,7 +398,7 @@ static int exfat_write_end(struct file *file, struct address_space *mapping,
exfat_write_failed(mapping, pos+len);
if (!(err < 0) && !(ei->attr & ATTR_ARCHIVE)) {
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
ei->attr |= ATTR_ARCHIVE;
mark_inode_dirty(inode);
}
@@ -577,7 +577,7 @@ static int exfat_fill_inode(struct inode *inode, struct exfat_dir_entry *info)
inode->i_blocks = round_up(i_size_read(inode), sbi->cluster_size) >> 9;
inode->i_mtime = info->mtime;
- inode->i_ctime = info->mtime;
+ inode_ctime_set(inode, info->mtime);
ei->i_crtime = info->crtime;
inode->i_atime = info->atime;
diff --git a/fs/exfat/namei.c b/fs/exfat/namei.c
index d9b46fa36bff..a8e6a84e6009 100644
--- a/fs/exfat/namei.c
+++ b/fs/exfat/namei.c
@@ -569,7 +569,7 @@ static int exfat_create(struct mnt_idmap *idmap, struct inode *dir,
goto unlock;
inode_inc_iversion(dir);
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
if (IS_DIRSYNC(dir))
exfat_sync_inode(dir);
else
@@ -582,8 +582,8 @@ static int exfat_create(struct mnt_idmap *idmap, struct inode *dir,
goto unlock;
inode_inc_iversion(inode);
- inode->i_mtime = inode->i_atime = inode->i_ctime =
- EXFAT_I(inode)->i_crtime = current_time(inode);
+ inode->i_mtime = inode->i_atime = EXFAT_I(inode)->i_crtime =
+ inode_ctime_set_current(inode);
exfat_truncate_atime(&inode->i_atime);
/* timestamp is already written, so mark_inode_dirty() is unneeded. */
@@ -817,7 +817,7 @@ static int exfat_unlink(struct inode *dir, struct dentry *dentry)
ei->dir.dir = DIR_DELETED;
inode_inc_iversion(dir);
- dir->i_mtime = dir->i_atime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = dir->i_atime = inode_ctime_set_current(dir);
exfat_truncate_atime(&dir->i_atime);
if (IS_DIRSYNC(dir))
exfat_sync_inode(dir);
@@ -825,7 +825,7 @@ static int exfat_unlink(struct inode *dir, struct dentry *dentry)
mark_inode_dirty(dir);
clear_nlink(inode);
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
exfat_truncate_atime(&inode->i_atime);
exfat_unhash_inode(inode);
exfat_d_version_set(dentry, inode_query_iversion(dir));
@@ -852,7 +852,7 @@ static int exfat_mkdir(struct mnt_idmap *idmap, struct inode *dir,
goto unlock;
inode_inc_iversion(dir);
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
if (IS_DIRSYNC(dir))
exfat_sync_inode(dir);
else
@@ -866,8 +866,8 @@ static int exfat_mkdir(struct mnt_idmap *idmap, struct inode *dir,
goto unlock;
inode_inc_iversion(inode);
- inode->i_mtime = inode->i_atime = inode->i_ctime =
- EXFAT_I(inode)->i_crtime = current_time(inode);
+ inode->i_mtime = inode->i_atime = EXFAT_I(inode)->i_crtime =
+ inode_ctime_set_current(inode);
exfat_truncate_atime(&inode->i_atime);
/* timestamp is already written, so mark_inode_dirty() is unneeded. */
@@ -979,7 +979,7 @@ static int exfat_rmdir(struct inode *dir, struct dentry *dentry)
ei->dir.dir = DIR_DELETED;
inode_inc_iversion(dir);
- dir->i_mtime = dir->i_atime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = dir->i_atime = inode_ctime_set_current(dir);
exfat_truncate_atime(&dir->i_atime);
if (IS_DIRSYNC(dir))
exfat_sync_inode(dir);
@@ -988,7 +988,7 @@ static int exfat_rmdir(struct inode *dir, struct dentry *dentry)
drop_nlink(dir);
clear_nlink(inode);
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
exfat_truncate_atime(&inode->i_atime);
exfat_unhash_inode(inode);
exfat_d_version_set(dentry, inode_query_iversion(dir));
@@ -1312,8 +1312,8 @@ static int exfat_rename(struct mnt_idmap *idmap,
goto unlock;
inode_inc_iversion(new_dir);
- new_dir->i_ctime = new_dir->i_mtime = new_dir->i_atime =
- EXFAT_I(new_dir)->i_crtime = current_time(new_dir);
+ new_dir->i_mtime = new_dir->i_atime = EXFAT_I(new_dir)->i_crtime =
+ inode_ctime_set_current(new_dir);
exfat_truncate_atime(&new_dir->i_atime);
if (IS_DIRSYNC(new_dir))
exfat_sync_inode(new_dir);
@@ -1336,7 +1336,7 @@ static int exfat_rename(struct mnt_idmap *idmap,
}
inode_inc_iversion(old_dir);
- old_dir->i_ctime = old_dir->i_mtime = current_time(old_dir);
+ old_dir->i_mtime = inode_ctime_set_current(old_dir);
if (IS_DIRSYNC(old_dir))
exfat_sync_inode(old_dir);
else
@@ -1354,8 +1354,7 @@ static int exfat_rename(struct mnt_idmap *idmap,
exfat_warn(sb, "abnormal access to an inode dropped");
WARN_ON(new_inode->i_nlink == 0);
}
- new_inode->i_ctime = EXFAT_I(new_inode)->i_crtime =
- current_time(new_inode);
+ EXFAT_I(new_inode)->i_crtime = inode_ctime_set_current(new_inode);
}
unlock:
diff --git a/fs/exfat/super.c b/fs/exfat/super.c
index 8c32460e031e..f4f3ccedf1bc 100644
--- a/fs/exfat/super.c
+++ b/fs/exfat/super.c
@@ -379,8 +379,8 @@ static int exfat_read_root(struct inode *inode)
ei->i_size_ondisk = i_size_read(inode);
exfat_save_attr(inode, ATTR_SUBDIR);
- inode->i_mtime = inode->i_atime = inode->i_ctime = ei->i_crtime =
- current_time(inode);
+ inode->i_mtime = inode->i_atime = ei->i_crtime =
+ inode_ctime_set_current(inode);
exfat_truncate_atime(&inode->i_atime);
return 0;
}
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 28/79] ext2: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (25 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Jan Kara; +Cc: Al Viro, Jan Kara, linux-ext4, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/ext2/acl.c | 2 +-
fs/ext2/dir.c | 6 +++---
fs/ext2/ialloc.c | 2 +-
fs/ext2/inode.c | 11 ++++++-----
fs/ext2/ioctl.c | 4 ++--
fs/ext2/namei.c | 8 ++++----
fs/ext2/super.c | 2 +-
fs/ext2/xattr.c | 2 +-
8 files changed, 19 insertions(+), 18 deletions(-)
diff --git a/fs/ext2/acl.c b/fs/ext2/acl.c
index 82b17d7fc93f..8f0fc476c423 100644
--- a/fs/ext2/acl.c
+++ b/fs/ext2/acl.c
@@ -237,7 +237,7 @@ ext2_set_acl(struct mnt_idmap *idmap, struct dentry *dentry,
error = __ext2_set_acl(inode, acl, type);
if (!error && update_mode) {
inode->i_mode = mode;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
mark_inode_dirty(inode);
}
return error;
diff --git a/fs/ext2/dir.c b/fs/ext2/dir.c
index 42db804794bd..b449599cbf6a 100644
--- a/fs/ext2/dir.c
+++ b/fs/ext2/dir.c
@@ -468,7 +468,7 @@ int ext2_set_link(struct inode *dir, struct ext2_dir_entry_2 *de,
ext2_set_de_type(de, inode);
ext2_commit_chunk(page, pos, len);
if (update_times)
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
EXT2_I(dir)->i_flags &= ~EXT2_BTREE_FL;
mark_inode_dirty(dir);
return ext2_handle_dirsync(dir);
@@ -555,7 +555,7 @@ int ext2_add_link (struct dentry *dentry, struct inode *inode)
de->inode = cpu_to_le32(inode->i_ino);
ext2_set_de_type (de, inode);
ext2_commit_chunk(page, pos, rec_len);
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
EXT2_I(dir)->i_flags &= ~EXT2_BTREE_FL;
mark_inode_dirty(dir);
err = ext2_handle_dirsync(dir);
@@ -606,7 +606,7 @@ int ext2_delete_entry(struct ext2_dir_entry_2 *dir, struct page *page)
pde->rec_len = ext2_rec_len_to_disk(to - from);
dir->inode = 0;
ext2_commit_chunk(page, pos, to - from);
- inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
EXT2_I(inode)->i_flags &= ~EXT2_BTREE_FL;
mark_inode_dirty(inode);
return ext2_handle_dirsync(inode);
diff --git a/fs/ext2/ialloc.c b/fs/ext2/ialloc.c
index a4e1d7a9c544..3a5bd92ba907 100644
--- a/fs/ext2/ialloc.c
+++ b/fs/ext2/ialloc.c
@@ -549,7 +549,7 @@ struct inode *ext2_new_inode(struct inode *dir, umode_t mode,
inode->i_ino = ino;
inode->i_blocks = 0;
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
memset(ei->i_data, 0, sizeof(ei->i_data));
ei->i_flags =
ext2_mask_flags(mode, EXT2_I(dir)->i_flags & EXT2_FL_INHERITED);
diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c
index 75983215c7a1..318b0b5e8d60 100644
--- a/fs/ext2/inode.c
+++ b/fs/ext2/inode.c
@@ -595,7 +595,7 @@ static void ext2_splice_branch(struct inode *inode,
if (where->bh)
mark_buffer_dirty_inode(where->bh, inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
mark_inode_dirty(inode);
}
@@ -1287,7 +1287,7 @@ static int ext2_setsize(struct inode *inode, loff_t newsize)
__ext2_truncate_blocks(inode, newsize);
filemap_invalidate_unlock(inode->i_mapping);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
if (inode_needs_sync(inode)) {
sync_mapping_buffers(inode->i_mapping);
sync_inode_metadata(inode, 1);
@@ -1409,9 +1409,10 @@ struct inode *ext2_iget (struct super_block *sb, unsigned long ino)
set_nlink(inode, le16_to_cpu(raw_inode->i_links_count));
inode->i_size = le32_to_cpu(raw_inode->i_size);
inode->i_atime.tv_sec = (signed)le32_to_cpu(raw_inode->i_atime);
- inode->i_ctime.tv_sec = (signed)le32_to_cpu(raw_inode->i_ctime);
+ inode_ctime_set_sec(inode, (signed)le32_to_cpu(raw_inode->i_ctime));
inode->i_mtime.tv_sec = (signed)le32_to_cpu(raw_inode->i_mtime);
- inode->i_atime.tv_nsec = inode->i_mtime.tv_nsec = inode->i_ctime.tv_nsec = 0;
+ inode->i_atime.tv_nsec = inode->i_mtime.tv_nsec = inode_ctime_set_nsec(inode,
+ 0);
ei->i_dtime = le32_to_cpu(raw_inode->i_dtime);
/* We now have enough fields to check if the inode was active or not.
* This is needed because nfsd might try to access dead inodes
@@ -1541,7 +1542,7 @@ static int __ext2_write_inode(struct inode *inode, int do_sync)
raw_inode->i_links_count = cpu_to_le16(inode->i_nlink);
raw_inode->i_size = cpu_to_le32(inode->i_size);
raw_inode->i_atime = cpu_to_le32(inode->i_atime.tv_sec);
- raw_inode->i_ctime = cpu_to_le32(inode->i_ctime.tv_sec);
+ raw_inode->i_ctime = cpu_to_le32(inode_ctime_peek(inode).tv_sec);
raw_inode->i_mtime = cpu_to_le32(inode->i_mtime.tv_sec);
raw_inode->i_blocks = cpu_to_le32(inode->i_blocks);
diff --git a/fs/ext2/ioctl.c b/fs/ext2/ioctl.c
index cc87d413eb43..7e1db518d104 100644
--- a/fs/ext2/ioctl.c
+++ b/fs/ext2/ioctl.c
@@ -44,7 +44,7 @@ int ext2_fileattr_set(struct mnt_idmap *idmap,
(fa->flags & EXT2_FL_USER_MODIFIABLE);
ext2_set_inode_flags(inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
mark_inode_dirty(inode);
return 0;
@@ -77,7 +77,7 @@ long ext2_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
}
inode_lock(inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
inode->i_generation = generation;
inode_unlock(inode);
diff --git a/fs/ext2/namei.c b/fs/ext2/namei.c
index 937dd8f60f96..2559974afbff 100644
--- a/fs/ext2/namei.c
+++ b/fs/ext2/namei.c
@@ -211,7 +211,7 @@ static int ext2_link (struct dentry * old_dentry, struct inode * dir,
if (err)
return err;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
inode_inc_link_count(inode);
ihold(inode);
@@ -291,7 +291,7 @@ static int ext2_unlink(struct inode *dir, struct dentry *dentry)
if (err)
goto out;
- inode->i_ctime = dir->i_ctime;
+ inode_ctime_set(inode, inode_ctime_peek(dir));
inode_dec_link_count(inode);
err = 0;
out:
@@ -367,7 +367,7 @@ static int ext2_rename (struct mnt_idmap * idmap,
ext2_put_page(new_page, new_de);
if (err)
goto out_dir;
- new_inode->i_ctime = current_time(new_inode);
+ inode_ctime_set_current(new_inode);
if (dir_de)
drop_nlink(new_inode);
inode_dec_link_count(new_inode);
@@ -383,7 +383,7 @@ static int ext2_rename (struct mnt_idmap * idmap,
* Like most other Unix systems, set the ctime for inodes on a
* rename.
*/
- old_inode->i_ctime = current_time(old_inode);
+ inode_ctime_set_current(old_inode);
mark_inode_dirty(old_inode);
err = ext2_delete_entry(old_de, old_page);
diff --git a/fs/ext2/super.c b/fs/ext2/super.c
index f342f347a695..dfb0b95cf4cb 100644
--- a/fs/ext2/super.c
+++ b/fs/ext2/super.c
@@ -1587,7 +1587,7 @@ static ssize_t ext2_quota_write(struct super_block *sb, int type,
if (inode->i_size < off+len-towrite)
i_size_write(inode, off+len-towrite);
inode_inc_iversion(inode);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
mark_inode_dirty(inode);
return len - towrite;
}
diff --git a/fs/ext2/xattr.c b/fs/ext2/xattr.c
index 8906ba479aaf..b05645ba7b1d 100644
--- a/fs/ext2/xattr.c
+++ b/fs/ext2/xattr.c
@@ -773,7 +773,7 @@ ext2_xattr_set2(struct inode *inode, struct buffer_head *old_bh,
/* Update the inode. */
EXT2_I(inode)->i_file_acl = new_bh ? new_bh->b_blocknr : 0;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
if (IS_SYNC(inode)) {
error = sync_inode_metadata(inode, 1);
/* In case sync failed due to ENOSPC the inode was actually
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 29/79] ext4: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (26 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Theodore Ts'o, Andreas Dilger
Cc: Al Viro, Jan Kara, linux-ext4, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/ext4/acl.c | 2 +-
fs/ext4/ext4.h | 20 ++++++++++++++++++++
fs/ext4/extents.c | 12 ++++++------
fs/ext4/ialloc.c | 2 +-
fs/ext4/inline.c | 4 ++--
fs/ext4/inode.c | 16 +++++++---------
fs/ext4/ioctl.c | 9 +++++----
fs/ext4/namei.c | 26 ++++++++++++--------------
fs/ext4/super.c | 2 +-
fs/ext4/xattr.c | 6 +++---
10 files changed, 58 insertions(+), 41 deletions(-)
diff --git a/fs/ext4/acl.c b/fs/ext4/acl.c
index 27fcbddfb148..4db1a1d5e552 100644
--- a/fs/ext4/acl.c
+++ b/fs/ext4/acl.c
@@ -259,7 +259,7 @@ ext4_set_acl(struct mnt_idmap *idmap, struct dentry *dentry,
error = __ext4_set_acl(handle, inode, type, acl, 0 /* xattr_flags */);
if (!error && update_mode) {
inode->i_mode = mode;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
error = ext4_mark_inode_dirty(handle, inode);
}
out_stop:
diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index 0a2d55faa095..4262d31f74d8 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -3823,6 +3823,26 @@ static inline int ext4_buffer_uptodate(struct buffer_head *bh)
return buffer_uptodate(bh);
}
+static inline void ext4_inode_set_ctime(struct inode *inode, struct ext4_inode *raw_inode)
+{
+ struct timespec64 ctime = inode_ctime_peek(inode);
+
+ if (EXT4_FITS_IN_INODE(raw_inode, EXT4_I(inode), i_ctime_extra)) {
+ raw_inode->i_ctime = cpu_to_le32(ctime.tv_sec);
+ raw_inode->i_ctime_extra = ext4_encode_extra_time(&ctime);
+ } else {
+ raw_inode->i_ctime = cpu_to_le32(clamp_t(int32_t, ctime.tv_sec, S32_MIN, S32_MAX));
+ }
+}
+
+static inline void ext4_inode_get_ctime(struct inode *inode, const struct ext4_inode *raw_inode)
+{
+ struct timespec64 ctime = { .tv_sec = (signed)le32_to_cpu(raw_inode->i_ctime) };
+
+ if (EXT4_FITS_IN_INODE(raw_inode, EXT4_I(inode), i_ctime_extra))
+ ext4_decode_extra_time(&ctime, raw_inode->i_ctime_extra);
+ inode_ctime_set(inode, ctime);
+}
#endif /* __KERNEL__ */
#define EFSBADCRC EBADMSG /* Bad CRC detected */
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index e4115d338f10..213178a31111 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -4476,12 +4476,12 @@ static int ext4_alloc_file_blocks(struct file *file, ext4_lblk_t offset,
map.m_lblk += ret;
map.m_len = len = len - ret;
epos = (loff_t)map.m_lblk << inode->i_blkbits;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
if (new_size) {
if (epos > new_size)
epos = new_size;
if (ext4_update_inode_size(inode, epos) & 0x1)
- inode->i_mtime = inode->i_ctime;
+ inode->i_mtime = inode_ctime_peek(inode);
}
ret2 = ext4_mark_inode_dirty(handle, inode);
ext4_update_inode_fsync_trans(handle, inode, 1);
@@ -4617,7 +4617,7 @@ static long ext4_zero_range(struct file *file, loff_t offset,
/* Now release the pages and zero block aligned part of pages */
truncate_pagecache_range(inode, start, end - 1);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
ret = ext4_alloc_file_blocks(file, lblk, max_blocks, new_size,
flags);
@@ -4642,7 +4642,7 @@ static long ext4_zero_range(struct file *file, loff_t offset,
goto out_mutex;
}
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
if (new_size)
ext4_update_inode_size(inode, new_size);
ret = ext4_mark_inode_dirty(handle, inode);
@@ -5378,7 +5378,7 @@ static int ext4_collapse_range(struct file *file, loff_t offset, loff_t len)
up_write(&EXT4_I(inode)->i_data_sem);
if (IS_SYNC(inode))
ext4_handle_sync(handle);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
ret = ext4_mark_inode_dirty(handle, inode);
ext4_update_inode_fsync_trans(handle, inode, 1);
@@ -5488,7 +5488,7 @@ static int ext4_insert_range(struct file *file, loff_t offset, loff_t len)
/* Expand file to avoid data loss if there is error while shifting */
inode->i_size += len;
EXT4_I(inode)->i_disksize += len;
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
ret = ext4_mark_inode_dirty(handle, inode);
if (ret)
goto out_stop;
diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
index 754f961cd9fd..03fcfcaf8dd6 100644
--- a/fs/ext4/ialloc.c
+++ b/fs/ext4/ialloc.c
@@ -1250,7 +1250,7 @@ struct inode *__ext4_new_inode(struct mnt_idmap *idmap,
inode->i_ino = ino + group * EXT4_INODES_PER_GROUP(sb);
/* This is the optimal IO size (for stat), not the fs block size */
inode->i_blocks = 0;
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
ei->i_crtime = inode->i_mtime;
memset(ei->i_data, 0, sizeof(ei->i_data));
diff --git a/fs/ext4/inline.c b/fs/ext4/inline.c
index a4b7e4bc32d4..678dc459d7f7 100644
--- a/fs/ext4/inline.c
+++ b/fs/ext4/inline.c
@@ -1037,7 +1037,7 @@ static int ext4_add_dirent_to_inline(handle_t *handle,
* happen is that the times are slightly out of date
* and/or different from the directory change time.
*/
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
ext4_update_dx_flag(dir);
inode_inc_iversion(dir);
return 1;
@@ -1991,7 +1991,7 @@ int ext4_inline_data_truncate(struct inode *inode, int *has_inline)
ext4_orphan_del(handle, inode);
if (err == 0) {
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
err = ext4_mark_inode_dirty(handle, inode);
if (IS_SYNC(inode))
ext4_handle_sync(handle);
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 43775a6ca505..ee9586e71246 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -3986,7 +3986,7 @@ int ext4_punch_hole(struct file *file, loff_t offset, loff_t length)
if (IS_SYNC(inode))
ext4_handle_sync(handle);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
ret2 = ext4_mark_inode_dirty(handle, inode);
if (unlikely(ret2))
ret = ret2;
@@ -4146,7 +4146,7 @@ int ext4_truncate(struct inode *inode)
if (inode->i_nlink)
ext4_orphan_del(handle, inode);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
err2 = ext4_mark_inode_dirty(handle, inode);
if (unlikely(err2 && !err))
err = err2;
@@ -4249,7 +4249,7 @@ static int ext4_fill_raw_inode(struct inode *inode, struct ext4_inode *raw_inode
}
raw_inode->i_links_count = cpu_to_le16(inode->i_nlink);
- EXT4_INODE_SET_XTIME(i_ctime, inode, raw_inode);
+ ext4_inode_set_ctime(inode, raw_inode);
EXT4_INODE_SET_XTIME(i_mtime, inode, raw_inode);
EXT4_INODE_SET_XTIME(i_atime, inode, raw_inode);
EXT4_EINODE_SET_XTIME(i_crtime, ei, raw_inode);
@@ -4858,7 +4858,7 @@ struct inode *__ext4_iget(struct super_block *sb, unsigned long ino,
}
}
- EXT4_INODE_GET_XTIME(i_ctime, inode, raw_inode);
+ ext4_inode_get_ctime(inode, raw_inode);
EXT4_INODE_GET_XTIME(i_mtime, inode, raw_inode);
EXT4_INODE_GET_XTIME(i_atime, inode, raw_inode);
EXT4_EINODE_GET_XTIME(i_crtime, ei, raw_inode);
@@ -4981,7 +4981,7 @@ static void __ext4_update_other_inode_time(struct super_block *sb,
spin_unlock(&inode->i_lock);
spin_lock(&ei->i_raw_lock);
- EXT4_INODE_SET_XTIME(i_ctime, inode, raw_inode);
+ ext4_inode_set_ctime(inode, raw_inode);
EXT4_INODE_SET_XTIME(i_mtime, inode, raw_inode);
EXT4_INODE_SET_XTIME(i_atime, inode, raw_inode);
ext4_inode_csum_set(inode, raw_inode, ei);
@@ -5376,10 +5376,8 @@ int ext4_setattr(struct mnt_idmap *idmap, struct dentry *dentry,
* Update c/mtime on truncate up, ext4_truncate() will
* update c/mtime in shrink case below
*/
- if (!shrink) {
- inode->i_mtime = current_time(inode);
- inode->i_ctime = inode->i_mtime;
- }
+ if (!shrink)
+ inode->i_mtime = inode_ctime_set_current(inode);
if (shrink)
ext4_fc_track_range(handle, inode,
diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c
index 961284cc9b65..c212371cf003 100644
--- a/fs/ext4/ioctl.c
+++ b/fs/ext4/ioctl.c
@@ -449,7 +449,8 @@ static long swap_inode_boot_loader(struct super_block *sb,
diff = size - size_bl;
swap_inode_data(inode, inode_bl);
- inode->i_ctime = inode_bl->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
+ inode_ctime_set_current(inode_bl);
inode_inc_iversion(inode);
inode->i_generation = get_random_u32();
@@ -663,7 +664,7 @@ static int ext4_ioctl_setflags(struct inode *inode,
ext4_set_inode_flags(inode, false);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
inode_inc_iversion(inode);
err = ext4_mark_iloc_dirty(handle, inode, &iloc);
@@ -774,7 +775,7 @@ static int ext4_ioctl_setproject(struct inode *inode, __u32 projid)
}
EXT4_I(inode)->i_projid = kprojid;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
inode_inc_iversion(inode);
out_dirty:
rc = ext4_mark_iloc_dirty(handle, inode, &iloc);
@@ -1263,7 +1264,7 @@ static long __ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
}
err = ext4_reserve_inode_write(handle, inode, &iloc);
if (err == 0) {
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
inode_inc_iversion(inode);
inode->i_generation = generation;
err = ext4_mark_iloc_dirty(handle, inode, &iloc);
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c
index 0caf6c730ce3..63392b06fa5f 100644
--- a/fs/ext4/namei.c
+++ b/fs/ext4/namei.c
@@ -2203,7 +2203,7 @@ static int add_dirent_to_buf(handle_t *handle, struct ext4_filename *fname,
* happen is that the times are slightly out of date
* and/or different from the directory change time.
*/
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
ext4_update_dx_flag(dir);
inode_inc_iversion(dir);
err2 = ext4_mark_inode_dirty(handle, dir);
@@ -3197,7 +3197,8 @@ static int ext4_rmdir(struct inode *dir, struct dentry *dentry)
* recovery. */
inode->i_size = 0;
ext4_orphan_add(handle, inode);
- inode->i_ctime = dir->i_ctime = dir->i_mtime = current_time(inode);
+ inode_ctime_set_current(inode);
+ dir->i_mtime = inode_ctime_set_current(inode);
retval = ext4_mark_inode_dirty(handle, inode);
if (retval)
goto end_rmdir;
@@ -3271,7 +3272,7 @@ int __ext4_unlink(struct inode *dir, const struct qstr *d_name,
retval = ext4_delete_entry(handle, dir, de, bh);
if (retval)
goto out_handle;
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
ext4_update_dx_flag(dir);
retval = ext4_mark_inode_dirty(handle, dir);
if (retval)
@@ -3286,7 +3287,7 @@ int __ext4_unlink(struct inode *dir, const struct qstr *d_name,
drop_nlink(inode);
if (!inode->i_nlink)
ext4_orphan_add(handle, inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
retval = ext4_mark_inode_dirty(handle, inode);
if (dentry && !retval)
ext4_fc_track_unlink(handle, dentry);
@@ -3463,7 +3464,7 @@ int __ext4_link(struct inode *dir, struct inode *inode, struct dentry *dentry)
if (IS_DIRSYNC(dir))
ext4_handle_sync(handle);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
ext4_inc_count(inode);
ihold(inode);
@@ -3641,8 +3642,7 @@ static int ext4_setent(handle_t *handle, struct ext4_renament *ent,
if (ext4_has_feature_filetype(ent->dir->i_sb))
ent->de->file_type = file_type;
inode_inc_iversion(ent->dir);
- ent->dir->i_ctime = ent->dir->i_mtime =
- current_time(ent->dir);
+ ent->dir->i_mtime = inode_ctime_set_current(ent->dir);
retval = ext4_mark_inode_dirty(handle, ent->dir);
BUFFER_TRACE(ent->bh, "call ext4_handle_dirty_metadata");
if (!ent->inlined) {
@@ -3941,7 +3941,7 @@ static int ext4_rename(struct mnt_idmap *idmap, struct inode *old_dir,
* Like most other Unix systems, set the ctime for inodes on a
* rename.
*/
- old.inode->i_ctime = current_time(old.inode);
+ inode_ctime_set_current(old.inode);
retval = ext4_mark_inode_dirty(handle, old.inode);
if (unlikely(retval))
goto end_rename;
@@ -3955,9 +3955,9 @@ static int ext4_rename(struct mnt_idmap *idmap, struct inode *old_dir,
if (new.inode) {
ext4_dec_count(new.inode);
- new.inode->i_ctime = current_time(new.inode);
+ inode_ctime_set_current(new.inode);
}
- old.dir->i_ctime = old.dir->i_mtime = current_time(old.dir);
+ old.dir->i_mtime = inode_ctime_set_current(old.dir);
ext4_update_dx_flag(old.dir);
if (old.dir_bh) {
retval = ext4_rename_dir_finish(handle, &old, new.dir->i_ino);
@@ -4053,7 +4053,6 @@ static int ext4_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
};
u8 new_file_type;
int retval;
- struct timespec64 ctime;
if ((ext4_test_inode_flag(new_dir, EXT4_INODE_PROJINHERIT) &&
!projid_eq(EXT4_I(new_dir)->i_projid,
@@ -4147,9 +4146,8 @@ static int ext4_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
* Like most other Unix systems, set the ctime for inodes on a
* rename.
*/
- ctime = current_time(old.inode);
- old.inode->i_ctime = ctime;
- new.inode->i_ctime = ctime;
+ inode_ctime_set_current(old.inode);
+ inode_ctime_set_current(new.inode);
retval = ext4_mark_inode_dirty(handle, old.inode);
if (unlikely(retval))
goto end_rename;
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 15ea8782222e..27f9a3b96a3e 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -7095,7 +7095,7 @@ static int ext4_quota_off(struct super_block *sb, int type)
}
EXT4_I(inode)->i_flags &= ~(EXT4_NOATIME_FL | EXT4_IMMUTABLE_FL);
inode_set_flags(inode, 0, S_NOATIME | S_IMMUTABLE);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
err = ext4_mark_inode_dirty(handle, inode);
ext4_journal_stop(handle);
out_unlock:
diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c
index 321e3a888c20..b405a4314814 100644
--- a/fs/ext4/xattr.c
+++ b/fs/ext4/xattr.c
@@ -356,13 +356,13 @@ ext4_xattr_inode_hash(struct ext4_sb_info *sbi, const void *buffer, size_t size)
static u64 ext4_xattr_inode_get_ref(struct inode *ea_inode)
{
- return ((u64)ea_inode->i_ctime.tv_sec << 32) |
+ return ((u64) inode_ctime_peek(ea_inode).tv_sec << 32) |
(u32) inode_peek_iversion_raw(ea_inode);
}
static void ext4_xattr_inode_set_ref(struct inode *ea_inode, u64 ref_count)
{
- ea_inode->i_ctime.tv_sec = (u32)(ref_count >> 32);
+ inode_ctime_set_sec(ea_inode, (u32)(ref_count >> 32));
inode_set_iversion_raw(ea_inode, ref_count & 0xffffffff);
}
@@ -2459,7 +2459,7 @@ ext4_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index,
}
if (!error) {
ext4_xattr_update_super_block(handle, inode->i_sb);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
inode_inc_iversion(inode);
if (!value)
no_expand = 0;
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [f2fs-dev] [PATCH 30/79] f2fs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Jaegeuk Kim, Chao Yu
Cc: linux-kernel, Jan Kara, Al Viro, linux-f2fs-devel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/f2fs/dir.c | 8 ++++----
fs/f2fs/f2fs.h | 5 ++++-
fs/f2fs/file.c | 16 ++++++++--------
fs/f2fs/inline.c | 2 +-
fs/f2fs/inode.c | 10 +++++-----
fs/f2fs/namei.c | 12 ++++++------
fs/f2fs/recovery.c | 4 ++--
fs/f2fs/super.c | 2 +-
fs/f2fs/xattr.c | 2 +-
9 files changed, 32 insertions(+), 29 deletions(-)
diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c
index 887e55988450..54fa7285e2d6 100644
--- a/fs/f2fs/dir.c
+++ b/fs/f2fs/dir.c
@@ -455,7 +455,7 @@ void f2fs_set_link(struct inode *dir, struct f2fs_dir_entry *de,
de->file_type = fs_umode_to_ftype(inode->i_mode);
set_page_dirty(page);
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
f2fs_mark_inode_dirty_sync(dir, false);
f2fs_put_page(page, 1);
}
@@ -609,7 +609,7 @@ void f2fs_update_parent_metadata(struct inode *dir, struct inode *inode,
f2fs_i_links_write(dir, true);
clear_inode_flag(inode, FI_NEW_INODE);
}
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
f2fs_mark_inode_dirty_sync(dir, false);
if (F2FS_I(dir)->i_current_depth != current_depth)
@@ -851,7 +851,7 @@ void f2fs_drop_nlink(struct inode *dir, struct inode *inode)
if (S_ISDIR(inode->i_mode))
f2fs_i_links_write(dir, false);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
f2fs_i_links_write(inode, false);
if (S_ISDIR(inode->i_mode)) {
@@ -912,7 +912,7 @@ void f2fs_delete_entry(struct f2fs_dir_entry *dentry, struct page *page,
}
f2fs_put_page(page, 1);
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
f2fs_mark_inode_dirty_sync(dir, false);
if (inode)
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index 7b9af2d51656..b0a0de41f823 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -3302,9 +3302,12 @@ static inline void clear_file(struct inode *inode, int type)
static inline bool f2fs_is_time_consistent(struct inode *inode)
{
+ struct timespec64 ctime;
+
if (!timespec64_equal(F2FS_I(inode)->i_disk_time, &inode->i_atime))
return false;
- if (!timespec64_equal(F2FS_I(inode)->i_disk_time + 1, &inode->i_ctime))
+ ctime = inode_ctime_peek(inode);
+ if (!timespec64_equal(F2FS_I(inode)->i_disk_time + 1, &ctime))
return false;
if (!timespec64_equal(F2FS_I(inode)->i_disk_time + 2, &inode->i_mtime))
return false;
diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
index 38688b5192ef..76c38cd89865 100644
--- a/fs/f2fs/file.c
+++ b/fs/f2fs/file.c
@@ -809,7 +809,7 @@ int f2fs_truncate(struct inode *inode)
if (err)
return err;
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
f2fs_mark_inode_dirty_sync(inode, false);
return 0;
}
@@ -920,7 +920,7 @@ static void __setattr_copy(struct mnt_idmap *idmap,
if (ia_valid & ATTR_MTIME)
inode->i_mtime = attr->ia_mtime;
if (ia_valid & ATTR_CTIME)
- inode->i_ctime = attr->ia_ctime;
+ inode_ctime_set(inode, attr->ia_ctime);
if (ia_valid & ATTR_MODE) {
umode_t mode = attr->ia_mode;
vfsgid_t vfsgid = i_gid_into_vfsgid(idmap, inode);
@@ -1023,7 +1023,7 @@ int f2fs_setattr(struct mnt_idmap *idmap, struct dentry *dentry,
return err;
spin_lock(&F2FS_I(inode)->i_size_lock);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
F2FS_I(inode)->last_disk_size = i_size_read(inode);
spin_unlock(&F2FS_I(inode)->i_size_lock);
}
@@ -1850,7 +1850,7 @@ static long f2fs_fallocate(struct file *file, int mode,
}
if (!ret) {
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
f2fs_mark_inode_dirty_sync(inode, false);
f2fs_update_time(F2FS_I_SB(inode), REQ_TIME);
}
@@ -1952,7 +1952,7 @@ static int f2fs_setflags_common(struct inode *inode, u32 iflags, u32 mask)
else
clear_inode_flag(inode, FI_PROJ_INHERIT);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
f2fs_set_inode_flags(inode);
f2fs_mark_inode_dirty_sync(inode, true);
return 0;
@@ -3077,7 +3077,7 @@ static int f2fs_ioc_setproject(struct inode *inode, __u32 projid)
goto out_unlock;
fi->i_projid = kprojid;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
f2fs_mark_inode_dirty_sync(inode, true);
out_unlock:
f2fs_unlock_op(sbi);
@@ -3510,7 +3510,7 @@ static int f2fs_release_compress_blocks(struct file *filp, unsigned long arg)
goto out;
set_inode_flag(inode, FI_COMPRESS_RELEASED);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
f2fs_mark_inode_dirty_sync(inode, true);
if (!atomic_read(&F2FS_I(inode)->i_compr_blocks))
@@ -3712,7 +3712,7 @@ static int f2fs_reserve_compress_blocks(struct file *filp, unsigned long arg)
if (ret >= 0) {
clear_inode_flag(inode, FI_COMPRESS_RELEASED);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
f2fs_mark_inode_dirty_sync(inode, true);
}
unlock_inode:
diff --git a/fs/f2fs/inline.c b/fs/f2fs/inline.c
index 4638fee16a91..72bb5340b088 100644
--- a/fs/f2fs/inline.c
+++ b/fs/f2fs/inline.c
@@ -698,7 +698,7 @@ void f2fs_delete_inline_entry(struct f2fs_dir_entry *dentry, struct page *page,
set_page_dirty(page);
f2fs_put_page(page, 1);
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
f2fs_mark_inode_dirty_sync(dir, false);
if (inode)
diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c
index 09e986b050c6..95b89e48a159 100644
--- a/fs/f2fs/inode.c
+++ b/fs/f2fs/inode.c
@@ -403,7 +403,7 @@ static void init_idisk_time(struct inode *inode)
struct f2fs_inode_info *fi = F2FS_I(inode);
fi->i_disk_time[0] = inode->i_atime;
- fi->i_disk_time[1] = inode->i_ctime;
+ fi->i_disk_time[1] = inode_ctime_peek(inode);
fi->i_disk_time[2] = inode->i_mtime;
}
@@ -434,10 +434,10 @@ static int do_read_inode(struct inode *inode)
inode->i_blocks = SECTOR_FROM_BLOCK(le64_to_cpu(ri->i_blocks) - 1);
inode->i_atime.tv_sec = le64_to_cpu(ri->i_atime);
- inode->i_ctime.tv_sec = le64_to_cpu(ri->i_ctime);
+ inode_ctime_set_sec(inode, le64_to_cpu(ri->i_ctime));
inode->i_mtime.tv_sec = le64_to_cpu(ri->i_mtime);
inode->i_atime.tv_nsec = le32_to_cpu(ri->i_atime_nsec);
- inode->i_ctime.tv_nsec = le32_to_cpu(ri->i_ctime_nsec);
+ inode_ctime_set_nsec(inode, le32_to_cpu(ri->i_ctime_nsec));
inode->i_mtime.tv_nsec = le32_to_cpu(ri->i_mtime_nsec);
inode->i_generation = le32_to_cpu(ri->i_generation);
if (S_ISDIR(inode->i_mode))
@@ -714,10 +714,10 @@ void f2fs_update_inode(struct inode *inode, struct page *node_page)
set_raw_inline(inode, ri);
ri->i_atime = cpu_to_le64(inode->i_atime.tv_sec);
- ri->i_ctime = cpu_to_le64(inode->i_ctime.tv_sec);
+ ri->i_ctime = cpu_to_le64(inode_ctime_peek(inode).tv_sec);
ri->i_mtime = cpu_to_le64(inode->i_mtime.tv_sec);
ri->i_atime_nsec = cpu_to_le32(inode->i_atime.tv_nsec);
- ri->i_ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
+ ri->i_ctime_nsec = cpu_to_le32(inode_ctime_peek(inode).tv_nsec);
ri->i_mtime_nsec = cpu_to_le32(inode->i_mtime.tv_nsec);
if (S_ISDIR(inode->i_mode))
ri->i_current_depth =
diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c
index 614c33f0f17e..c7876c5d74a6 100644
--- a/fs/f2fs/namei.c
+++ b/fs/f2fs/namei.c
@@ -233,7 +233,7 @@ static struct inode *f2fs_new_inode(struct mnt_idmap *idmap,
inode->i_ino = ino;
inode->i_blocks = 0;
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
F2FS_I(inode)->i_crtime = inode->i_mtime;
inode->i_generation = get_random_u32();
@@ -410,7 +410,7 @@ static int f2fs_link(struct dentry *old_dentry, struct inode *dir,
f2fs_balance_fs(sbi, true);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
ihold(inode);
set_inode_flag(inode, FI_INC_LINK);
@@ -1042,7 +1042,7 @@ static int f2fs_rename(struct mnt_idmap *idmap, struct inode *old_dir,
f2fs_set_link(new_dir, new_entry, new_page, old_inode);
new_page = NULL;
- new_inode->i_ctime = current_time(new_inode);
+ inode_ctime_set_current(new_inode);
f2fs_down_write(&F2FS_I(new_inode)->i_sem);
if (old_dir_entry)
f2fs_i_links_write(new_inode, false);
@@ -1076,7 +1076,7 @@ static int f2fs_rename(struct mnt_idmap *idmap, struct inode *old_dir,
f2fs_i_pino_write(old_inode, new_dir->i_ino);
f2fs_up_write(&F2FS_I(old_inode)->i_sem);
- old_inode->i_ctime = current_time(old_inode);
+ inode_ctime_set_current(old_inode);
f2fs_mark_inode_dirty_sync(old_inode, false);
f2fs_delete_entry(old_entry, old_page, old_dir, NULL);
@@ -1241,7 +1241,7 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
f2fs_i_pino_write(old_inode, new_dir->i_ino);
f2fs_up_write(&F2FS_I(old_inode)->i_sem);
- old_dir->i_ctime = current_time(old_dir);
+ inode_ctime_set_current(old_dir);
if (old_nlink) {
f2fs_down_write(&F2FS_I(old_dir)->i_sem);
f2fs_i_links_write(old_dir, old_nlink > 0);
@@ -1260,7 +1260,7 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
f2fs_i_pino_write(new_inode, old_dir->i_ino);
f2fs_up_write(&F2FS_I(new_inode)->i_sem);
- new_dir->i_ctime = current_time(new_dir);
+ inode_ctime_set_current(new_dir);
if (new_nlink) {
f2fs_down_write(&F2FS_I(new_dir)->i_sem);
f2fs_i_links_write(new_dir, new_nlink > 0);
diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c
index 4e7d4ceeb084..5a06dd15453d 100644
--- a/fs/f2fs/recovery.c
+++ b/fs/f2fs/recovery.c
@@ -321,10 +321,10 @@ static int recover_inode(struct inode *inode, struct page *page)
f2fs_i_size_write(inode, le64_to_cpu(raw->i_size));
inode->i_atime.tv_sec = le64_to_cpu(raw->i_atime);
- inode->i_ctime.tv_sec = le64_to_cpu(raw->i_ctime);
+ inode_ctime_set_sec(inode, le64_to_cpu(raw->i_ctime));
inode->i_mtime.tv_sec = le64_to_cpu(raw->i_mtime);
inode->i_atime.tv_nsec = le32_to_cpu(raw->i_atime_nsec);
- inode->i_ctime.tv_nsec = le32_to_cpu(raw->i_ctime_nsec);
+ inode_ctime_set_nsec(inode, le32_to_cpu(raw->i_ctime_nsec));
inode->i_mtime.tv_nsec = le32_to_cpu(raw->i_mtime_nsec);
F2FS_I(inode)->i_advise = raw->i_advise;
diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index bc15841fd33c..3c6bb358a7a3 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -2702,7 +2702,7 @@ static ssize_t f2fs_quota_write(struct super_block *sb, int type,
if (len == towrite)
return err;
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
f2fs_mark_inode_dirty_sync(inode, false);
return len - towrite;
}
diff --git a/fs/f2fs/xattr.c b/fs/f2fs/xattr.c
index bdc8a55085a2..9bc685b33f4d 100644
--- a/fs/f2fs/xattr.c
+++ b/fs/f2fs/xattr.c
@@ -749,7 +749,7 @@ static int __f2fs_setxattr(struct inode *inode, int index,
same:
if (is_inode_flag_set(inode, FI_ACL_MODE)) {
inode->i_mode = F2FS_I(inode)->i_acl_mode;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
clear_inode_flag(inode, FI_ACL_MODE);
}
--
2.41.0
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 30/79] f2fs: switch to new ctime accessors
@ 2023-06-21 14:45 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Jaegeuk Kim, Chao Yu
Cc: Al Viro, Jan Kara, linux-f2fs-devel, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/f2fs/dir.c | 8 ++++----
fs/f2fs/f2fs.h | 5 ++++-
fs/f2fs/file.c | 16 ++++++++--------
fs/f2fs/inline.c | 2 +-
fs/f2fs/inode.c | 10 +++++-----
fs/f2fs/namei.c | 12 ++++++------
fs/f2fs/recovery.c | 4 ++--
fs/f2fs/super.c | 2 +-
fs/f2fs/xattr.c | 2 +-
9 files changed, 32 insertions(+), 29 deletions(-)
diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c
index 887e55988450..54fa7285e2d6 100644
--- a/fs/f2fs/dir.c
+++ b/fs/f2fs/dir.c
@@ -455,7 +455,7 @@ void f2fs_set_link(struct inode *dir, struct f2fs_dir_entry *de,
de->file_type = fs_umode_to_ftype(inode->i_mode);
set_page_dirty(page);
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
f2fs_mark_inode_dirty_sync(dir, false);
f2fs_put_page(page, 1);
}
@@ -609,7 +609,7 @@ void f2fs_update_parent_metadata(struct inode *dir, struct inode *inode,
f2fs_i_links_write(dir, true);
clear_inode_flag(inode, FI_NEW_INODE);
}
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
f2fs_mark_inode_dirty_sync(dir, false);
if (F2FS_I(dir)->i_current_depth != current_depth)
@@ -851,7 +851,7 @@ void f2fs_drop_nlink(struct inode *dir, struct inode *inode)
if (S_ISDIR(inode->i_mode))
f2fs_i_links_write(dir, false);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
f2fs_i_links_write(inode, false);
if (S_ISDIR(inode->i_mode)) {
@@ -912,7 +912,7 @@ void f2fs_delete_entry(struct f2fs_dir_entry *dentry, struct page *page,
}
f2fs_put_page(page, 1);
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
f2fs_mark_inode_dirty_sync(dir, false);
if (inode)
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index 7b9af2d51656..b0a0de41f823 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -3302,9 +3302,12 @@ static inline void clear_file(struct inode *inode, int type)
static inline bool f2fs_is_time_consistent(struct inode *inode)
{
+ struct timespec64 ctime;
+
if (!timespec64_equal(F2FS_I(inode)->i_disk_time, &inode->i_atime))
return false;
- if (!timespec64_equal(F2FS_I(inode)->i_disk_time + 1, &inode->i_ctime))
+ ctime = inode_ctime_peek(inode);
+ if (!timespec64_equal(F2FS_I(inode)->i_disk_time + 1, &ctime))
return false;
if (!timespec64_equal(F2FS_I(inode)->i_disk_time + 2, &inode->i_mtime))
return false;
diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
index 38688b5192ef..76c38cd89865 100644
--- a/fs/f2fs/file.c
+++ b/fs/f2fs/file.c
@@ -809,7 +809,7 @@ int f2fs_truncate(struct inode *inode)
if (err)
return err;
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
f2fs_mark_inode_dirty_sync(inode, false);
return 0;
}
@@ -920,7 +920,7 @@ static void __setattr_copy(struct mnt_idmap *idmap,
if (ia_valid & ATTR_MTIME)
inode->i_mtime = attr->ia_mtime;
if (ia_valid & ATTR_CTIME)
- inode->i_ctime = attr->ia_ctime;
+ inode_ctime_set(inode, attr->ia_ctime);
if (ia_valid & ATTR_MODE) {
umode_t mode = attr->ia_mode;
vfsgid_t vfsgid = i_gid_into_vfsgid(idmap, inode);
@@ -1023,7 +1023,7 @@ int f2fs_setattr(struct mnt_idmap *idmap, struct dentry *dentry,
return err;
spin_lock(&F2FS_I(inode)->i_size_lock);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
F2FS_I(inode)->last_disk_size = i_size_read(inode);
spin_unlock(&F2FS_I(inode)->i_size_lock);
}
@@ -1850,7 +1850,7 @@ static long f2fs_fallocate(struct file *file, int mode,
}
if (!ret) {
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
f2fs_mark_inode_dirty_sync(inode, false);
f2fs_update_time(F2FS_I_SB(inode), REQ_TIME);
}
@@ -1952,7 +1952,7 @@ static int f2fs_setflags_common(struct inode *inode, u32 iflags, u32 mask)
else
clear_inode_flag(inode, FI_PROJ_INHERIT);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
f2fs_set_inode_flags(inode);
f2fs_mark_inode_dirty_sync(inode, true);
return 0;
@@ -3077,7 +3077,7 @@ static int f2fs_ioc_setproject(struct inode *inode, __u32 projid)
goto out_unlock;
fi->i_projid = kprojid;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
f2fs_mark_inode_dirty_sync(inode, true);
out_unlock:
f2fs_unlock_op(sbi);
@@ -3510,7 +3510,7 @@ static int f2fs_release_compress_blocks(struct file *filp, unsigned long arg)
goto out;
set_inode_flag(inode, FI_COMPRESS_RELEASED);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
f2fs_mark_inode_dirty_sync(inode, true);
if (!atomic_read(&F2FS_I(inode)->i_compr_blocks))
@@ -3712,7 +3712,7 @@ static int f2fs_reserve_compress_blocks(struct file *filp, unsigned long arg)
if (ret >= 0) {
clear_inode_flag(inode, FI_COMPRESS_RELEASED);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
f2fs_mark_inode_dirty_sync(inode, true);
}
unlock_inode:
diff --git a/fs/f2fs/inline.c b/fs/f2fs/inline.c
index 4638fee16a91..72bb5340b088 100644
--- a/fs/f2fs/inline.c
+++ b/fs/f2fs/inline.c
@@ -698,7 +698,7 @@ void f2fs_delete_inline_entry(struct f2fs_dir_entry *dentry, struct page *page,
set_page_dirty(page);
f2fs_put_page(page, 1);
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
f2fs_mark_inode_dirty_sync(dir, false);
if (inode)
diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c
index 09e986b050c6..95b89e48a159 100644
--- a/fs/f2fs/inode.c
+++ b/fs/f2fs/inode.c
@@ -403,7 +403,7 @@ static void init_idisk_time(struct inode *inode)
struct f2fs_inode_info *fi = F2FS_I(inode);
fi->i_disk_time[0] = inode->i_atime;
- fi->i_disk_time[1] = inode->i_ctime;
+ fi->i_disk_time[1] = inode_ctime_peek(inode);
fi->i_disk_time[2] = inode->i_mtime;
}
@@ -434,10 +434,10 @@ static int do_read_inode(struct inode *inode)
inode->i_blocks = SECTOR_FROM_BLOCK(le64_to_cpu(ri->i_blocks) - 1);
inode->i_atime.tv_sec = le64_to_cpu(ri->i_atime);
- inode->i_ctime.tv_sec = le64_to_cpu(ri->i_ctime);
+ inode_ctime_set_sec(inode, le64_to_cpu(ri->i_ctime));
inode->i_mtime.tv_sec = le64_to_cpu(ri->i_mtime);
inode->i_atime.tv_nsec = le32_to_cpu(ri->i_atime_nsec);
- inode->i_ctime.tv_nsec = le32_to_cpu(ri->i_ctime_nsec);
+ inode_ctime_set_nsec(inode, le32_to_cpu(ri->i_ctime_nsec));
inode->i_mtime.tv_nsec = le32_to_cpu(ri->i_mtime_nsec);
inode->i_generation = le32_to_cpu(ri->i_generation);
if (S_ISDIR(inode->i_mode))
@@ -714,10 +714,10 @@ void f2fs_update_inode(struct inode *inode, struct page *node_page)
set_raw_inline(inode, ri);
ri->i_atime = cpu_to_le64(inode->i_atime.tv_sec);
- ri->i_ctime = cpu_to_le64(inode->i_ctime.tv_sec);
+ ri->i_ctime = cpu_to_le64(inode_ctime_peek(inode).tv_sec);
ri->i_mtime = cpu_to_le64(inode->i_mtime.tv_sec);
ri->i_atime_nsec = cpu_to_le32(inode->i_atime.tv_nsec);
- ri->i_ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
+ ri->i_ctime_nsec = cpu_to_le32(inode_ctime_peek(inode).tv_nsec);
ri->i_mtime_nsec = cpu_to_le32(inode->i_mtime.tv_nsec);
if (S_ISDIR(inode->i_mode))
ri->i_current_depth =
diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c
index 614c33f0f17e..c7876c5d74a6 100644
--- a/fs/f2fs/namei.c
+++ b/fs/f2fs/namei.c
@@ -233,7 +233,7 @@ static struct inode *f2fs_new_inode(struct mnt_idmap *idmap,
inode->i_ino = ino;
inode->i_blocks = 0;
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
F2FS_I(inode)->i_crtime = inode->i_mtime;
inode->i_generation = get_random_u32();
@@ -410,7 +410,7 @@ static int f2fs_link(struct dentry *old_dentry, struct inode *dir,
f2fs_balance_fs(sbi, true);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
ihold(inode);
set_inode_flag(inode, FI_INC_LINK);
@@ -1042,7 +1042,7 @@ static int f2fs_rename(struct mnt_idmap *idmap, struct inode *old_dir,
f2fs_set_link(new_dir, new_entry, new_page, old_inode);
new_page = NULL;
- new_inode->i_ctime = current_time(new_inode);
+ inode_ctime_set_current(new_inode);
f2fs_down_write(&F2FS_I(new_inode)->i_sem);
if (old_dir_entry)
f2fs_i_links_write(new_inode, false);
@@ -1076,7 +1076,7 @@ static int f2fs_rename(struct mnt_idmap *idmap, struct inode *old_dir,
f2fs_i_pino_write(old_inode, new_dir->i_ino);
f2fs_up_write(&F2FS_I(old_inode)->i_sem);
- old_inode->i_ctime = current_time(old_inode);
+ inode_ctime_set_current(old_inode);
f2fs_mark_inode_dirty_sync(old_inode, false);
f2fs_delete_entry(old_entry, old_page, old_dir, NULL);
@@ -1241,7 +1241,7 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
f2fs_i_pino_write(old_inode, new_dir->i_ino);
f2fs_up_write(&F2FS_I(old_inode)->i_sem);
- old_dir->i_ctime = current_time(old_dir);
+ inode_ctime_set_current(old_dir);
if (old_nlink) {
f2fs_down_write(&F2FS_I(old_dir)->i_sem);
f2fs_i_links_write(old_dir, old_nlink > 0);
@@ -1260,7 +1260,7 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
f2fs_i_pino_write(new_inode, old_dir->i_ino);
f2fs_up_write(&F2FS_I(new_inode)->i_sem);
- new_dir->i_ctime = current_time(new_dir);
+ inode_ctime_set_current(new_dir);
if (new_nlink) {
f2fs_down_write(&F2FS_I(new_dir)->i_sem);
f2fs_i_links_write(new_dir, new_nlink > 0);
diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c
index 4e7d4ceeb084..5a06dd15453d 100644
--- a/fs/f2fs/recovery.c
+++ b/fs/f2fs/recovery.c
@@ -321,10 +321,10 @@ static int recover_inode(struct inode *inode, struct page *page)
f2fs_i_size_write(inode, le64_to_cpu(raw->i_size));
inode->i_atime.tv_sec = le64_to_cpu(raw->i_atime);
- inode->i_ctime.tv_sec = le64_to_cpu(raw->i_ctime);
+ inode_ctime_set_sec(inode, le64_to_cpu(raw->i_ctime));
inode->i_mtime.tv_sec = le64_to_cpu(raw->i_mtime);
inode->i_atime.tv_nsec = le32_to_cpu(raw->i_atime_nsec);
- inode->i_ctime.tv_nsec = le32_to_cpu(raw->i_ctime_nsec);
+ inode_ctime_set_nsec(inode, le32_to_cpu(raw->i_ctime_nsec));
inode->i_mtime.tv_nsec = le32_to_cpu(raw->i_mtime_nsec);
F2FS_I(inode)->i_advise = raw->i_advise;
diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index bc15841fd33c..3c6bb358a7a3 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -2702,7 +2702,7 @@ static ssize_t f2fs_quota_write(struct super_block *sb, int type,
if (len == towrite)
return err;
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
f2fs_mark_inode_dirty_sync(inode, false);
return len - towrite;
}
diff --git a/fs/f2fs/xattr.c b/fs/f2fs/xattr.c
index bdc8a55085a2..9bc685b33f4d 100644
--- a/fs/f2fs/xattr.c
+++ b/fs/f2fs/xattr.c
@@ -749,7 +749,7 @@ static int __f2fs_setxattr(struct inode *inode, int index,
same:
if (is_inode_flag_set(inode, FI_ACL_MODE)) {
inode->i_mode = F2FS_I(inode)->i_acl_mode;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
clear_inode_flag(inode, FI_ACL_MODE);
}
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 31/79] fat: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (28 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, OGAWA Hirofumi; +Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/fat/inode.c | 8 +++++---
fs/fat/misc.c | 7 +++++--
2 files changed, 10 insertions(+), 5 deletions(-)
diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index d99b8549ec8f..b34ae7298796 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -562,7 +562,7 @@ int fat_fill_inode(struct inode *inode, struct msdos_dir_entry *de)
& ~((loff_t)sbi->cluster_size - 1)) >> 9;
fat_time_fat2unix(sbi, &inode->i_mtime, de->time, de->date, 0);
- inode->i_ctime = inode->i_mtime;
+ inode_ctime_set(inode, inode->i_mtime);
if (sbi->options.isvfat) {
fat_time_fat2unix(sbi, &inode->i_atime, 0, de->adate, 0);
fat_time_fat2unix(sbi, &MSDOS_I(inode)->i_crtime, de->ctime,
@@ -1407,8 +1407,10 @@ static int fat_read_root(struct inode *inode)
MSDOS_I(inode)->mmu_private = inode->i_size;
fat_save_attrs(inode, ATTR_DIR);
- inode->i_mtime.tv_sec = inode->i_atime.tv_sec = inode->i_ctime.tv_sec = 0;
- inode->i_mtime.tv_nsec = inode->i_atime.tv_nsec = inode->i_ctime.tv_nsec = 0;
+ inode->i_mtime.tv_sec = inode->i_atime.tv_sec = inode_ctime_set_sec(inode,
+ 0);
+ inode->i_mtime.tv_nsec = inode->i_atime.tv_nsec = inode_ctime_set_nsec(inode,
+ 0);
set_nlink(inode, fat_subdirs(inode)+2);
return 0;
diff --git a/fs/fat/misc.c b/fs/fat/misc.c
index 7e5d6ae305f2..0d3ba6001ff0 100644
--- a/fs/fat/misc.c
+++ b/fs/fat/misc.c
@@ -331,8 +331,11 @@ int fat_truncate_time(struct inode *inode, struct timespec64 *now, int flags)
* identical in memory. all mtime updates will be applied to ctime,
* but ctime updates are ignored.
*/
- if (flags & S_MTIME)
- inode->i_mtime = inode->i_ctime = fat_truncate_mtime(sbi, now);
+ if (flags & S_MTIME) {
+ ts = fat_truncate_mtime(sbi, now);
+ inode->i_mtime = ts;
+ inode_ctime_set(inode, ts);
+ }
return 0;
}
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 32/79] freevxfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (29 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Christoph Hellwig; +Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/freevxfs/vxfs_inode.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/freevxfs/vxfs_inode.c b/fs/freevxfs/vxfs_inode.c
index ceb6a12649ba..ad0f0a627715 100644
--- a/fs/freevxfs/vxfs_inode.c
+++ b/fs/freevxfs/vxfs_inode.c
@@ -110,10 +110,10 @@ static inline void dip2vip_cpy(struct vxfs_sb_info *sbi,
inode->i_size = vip->vii_size;
inode->i_atime.tv_sec = vip->vii_atime;
- inode->i_ctime.tv_sec = vip->vii_ctime;
+ inode_ctime_set_sec(inode, vip->vii_ctime);
inode->i_mtime.tv_sec = vip->vii_mtime;
inode->i_atime.tv_nsec = 0;
- inode->i_ctime.tv_nsec = 0;
+ inode_ctime_set_nsec(inode, 0);
inode->i_mtime.tv_nsec = 0;
inode->i_blocks = vip->vii_blocks;
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 33/79] fuse: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (30 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Miklos Szeredi
Cc: Al Viro, Jan Kara, linux-fsdevel, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/fuse/control.c | 2 +-
fs/fuse/dir.c | 8 ++++----
fs/fuse/inode.c | 18 ++++++++++--------
3 files changed, 15 insertions(+), 13 deletions(-)
diff --git a/fs/fuse/control.c b/fs/fuse/control.c
index 247ef4f76761..b5df89d0c95a 100644
--- a/fs/fuse/control.c
+++ b/fs/fuse/control.c
@@ -235,7 +235,7 @@ static struct dentry *fuse_ctl_add_dentry(struct dentry *parent,
inode->i_mode = mode;
inode->i_uid = fc->user_id;
inode->i_gid = fc->group_id;
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
/* setting ->i_op to NULL is not allowed */
if (iop)
inode->i_op = iop;
diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index 5a4a7155cf1c..03fbb83b7127 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -933,7 +933,7 @@ void fuse_flush_time_update(struct inode *inode)
static void fuse_update_ctime_in_cache(struct inode *inode)
{
if (!IS_NOCMTIME(inode)) {
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
mark_inode_dirty_sync(inode);
fuse_flush_time_update(inode);
}
@@ -1715,8 +1715,8 @@ int fuse_flush_times(struct inode *inode, struct fuse_file *ff)
inarg.mtimensec = inode->i_mtime.tv_nsec;
if (fm->fc->minor >= 23) {
inarg.valid |= FATTR_CTIME;
- inarg.ctime = inode->i_ctime.tv_sec;
- inarg.ctimensec = inode->i_ctime.tv_nsec;
+ inarg.ctime = inode_ctime_peek(inode).tv_sec;
+ inarg.ctimensec = inode_ctime_peek(inode).tv_nsec;
}
if (ff) {
inarg.valid |= FATTR_FH;
@@ -1857,7 +1857,7 @@ int fuse_do_setattr(struct dentry *dentry, struct iattr *attr,
if (attr->ia_valid & ATTR_MTIME)
inode->i_mtime = attr->ia_mtime;
if (attr->ia_valid & ATTR_CTIME)
- inode->i_ctime = attr->ia_ctime;
+ inode_ctime_set(inode, attr->ia_ctime);
/* FIXME: clear I_DIRTY_SYNC? */
}
diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index 660be31aaabc..54e06d3874e7 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -194,8 +194,8 @@ void fuse_change_attributes_common(struct inode *inode, struct fuse_attr *attr,
inode->i_mtime.tv_nsec = attr->mtimensec;
}
if (!(cache_mask & STATX_CTIME)) {
- inode->i_ctime.tv_sec = attr->ctime;
- inode->i_ctime.tv_nsec = attr->ctimensec;
+ inode_ctime_set_sec(inode, attr->ctime);
+ inode_ctime_set_nsec(inode, attr->ctimensec);
}
if (attr->blksize != 0)
@@ -259,8 +259,8 @@ void fuse_change_attributes(struct inode *inode, struct fuse_attr *attr,
attr->mtimensec = inode->i_mtime.tv_nsec;
}
if (cache_mask & STATX_CTIME) {
- attr->ctime = inode->i_ctime.tv_sec;
- attr->ctimensec = inode->i_ctime.tv_nsec;
+ attr->ctime = inode_ctime_peek(inode).tv_sec;
+ attr->ctimensec = inode_ctime_peek(inode).tv_nsec;
}
if ((attr_version != 0 && fi->attr_version > attr_version) ||
@@ -318,8 +318,8 @@ static void fuse_init_inode(struct inode *inode, struct fuse_attr *attr,
inode->i_size = attr->size;
inode->i_mtime.tv_sec = attr->mtime;
inode->i_mtime.tv_nsec = attr->mtimensec;
- inode->i_ctime.tv_sec = attr->ctime;
- inode->i_ctime.tv_nsec = attr->ctimensec;
+ inode_ctime_set_sec(inode, attr->ctime);
+ inode_ctime_set_nsec(inode, attr->ctimensec);
if (S_ISREG(inode->i_mode)) {
fuse_init_common(inode);
fuse_init_file_inode(inode, attr->flags);
@@ -1398,16 +1398,18 @@ EXPORT_SYMBOL_GPL(fuse_dev_free);
static void fuse_fill_attr_from_inode(struct fuse_attr *attr,
const struct fuse_inode *fi)
{
+ struct timespec64 ctime = inode_ctime_peek(&fi->inode);
+
*attr = (struct fuse_attr){
.ino = fi->inode.i_ino,
.size = fi->inode.i_size,
.blocks = fi->inode.i_blocks,
.atime = fi->inode.i_atime.tv_sec,
.mtime = fi->inode.i_mtime.tv_sec,
- .ctime = fi->inode.i_ctime.tv_sec,
+ .ctime = ctime.tv_sec,
.atimensec = fi->inode.i_atime.tv_nsec,
.mtimensec = fi->inode.i_mtime.tv_nsec,
- .ctimensec = fi->inode.i_ctime.tv_nsec,
+ .ctimensec = ctime.tv_nsec,
.mode = fi->inode.i_mode,
.nlink = fi->inode.i_nlink,
.uid = fi->inode.i_uid.val,
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 34/79] gfs2: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Bob Peterson, Andreas Gruenbacher
Cc: Al Viro, Jan Kara, cluster-devel, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/gfs2/acl.c | 2 +-
fs/gfs2/bmap.c | 11 +++++------
fs/gfs2/dir.c | 15 ++++++++-------
fs/gfs2/file.c | 2 +-
fs/gfs2/glops.c | 4 ++--
fs/gfs2/inode.c | 8 ++++----
fs/gfs2/super.c | 4 ++--
fs/gfs2/xattr.c | 8 ++++----
8 files changed, 27 insertions(+), 27 deletions(-)
diff --git a/fs/gfs2/acl.c b/fs/gfs2/acl.c
index a392aa0f041d..b267dae0dc63 100644
--- a/fs/gfs2/acl.c
+++ b/fs/gfs2/acl.c
@@ -142,7 +142,7 @@ int gfs2_set_acl(struct mnt_idmap *idmap, struct dentry *dentry,
ret = __gfs2_set_acl(inode, acl, type);
if (!ret && mode != inode->i_mode) {
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
inode->i_mode = mode;
mark_inode_dirty(inode);
}
diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c
index 8d611fbcf0bd..743b09a0b196 100644
--- a/fs/gfs2/bmap.c
+++ b/fs/gfs2/bmap.c
@@ -1386,7 +1386,7 @@ static int trunc_start(struct inode *inode, u64 newsize)
ip->i_diskflags |= GFS2_DIF_TRUNC_IN_PROG;
i_size_write(inode, newsize);
- ip->i_inode.i_mtime = ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
gfs2_dinode_out(ip, dibh->b_data);
if (journaled)
@@ -1583,8 +1583,7 @@ static int sweep_bh_for_rgrps(struct gfs2_inode *ip, struct gfs2_holder *rd_gh,
/* Every transaction boundary, we rewrite the dinode
to keep its di_blocks current in case of failure. */
- ip->i_inode.i_mtime = ip->i_inode.i_ctime =
- current_time(&ip->i_inode);
+ ip->i_inode.i_mtime = inode_ctime_set_current(&ip->i_inode);
gfs2_trans_add_meta(ip->i_gl, dibh);
gfs2_dinode_out(ip, dibh->b_data);
brelse(dibh);
@@ -1950,7 +1949,7 @@ static int punch_hole(struct gfs2_inode *ip, u64 offset, u64 length)
gfs2_statfs_change(sdp, 0, +btotal, 0);
gfs2_quota_change(ip, -(s64)btotal, ip->i_inode.i_uid,
ip->i_inode.i_gid);
- ip->i_inode.i_mtime = ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ ip->i_inode.i_mtime = inode_ctime_set_current(&ip->i_inode);
gfs2_trans_add_meta(ip->i_gl, dibh);
gfs2_dinode_out(ip, dibh->b_data);
up_write(&ip->i_rw_mutex);
@@ -1993,7 +1992,7 @@ static int trunc_end(struct gfs2_inode *ip)
gfs2_buffer_clear_tail(dibh, sizeof(struct gfs2_dinode));
gfs2_ordered_del_inode(ip);
}
- ip->i_inode.i_mtime = ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ ip->i_inode.i_mtime = inode_ctime_set_current(&ip->i_inode);
ip->i_diskflags &= ~GFS2_DIF_TRUNC_IN_PROG;
gfs2_trans_add_meta(ip->i_gl, dibh);
@@ -2094,7 +2093,7 @@ static int do_grow(struct inode *inode, u64 size)
goto do_end_trans;
truncate_setsize(inode, size);
- ip->i_inode.i_mtime = ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ ip->i_inode.i_mtime = inode_ctime_set_current(&ip->i_inode);
gfs2_trans_add_meta(ip->i_gl, dibh);
gfs2_dinode_out(ip, dibh->b_data);
brelse(dibh);
diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c
index 54a6d17b8c25..c07cb9883ea1 100644
--- a/fs/gfs2/dir.c
+++ b/fs/gfs2/dir.c
@@ -130,7 +130,7 @@ static int gfs2_dir_write_stuffed(struct gfs2_inode *ip, const char *buf,
memcpy(dibh->b_data + offset + sizeof(struct gfs2_dinode), buf, size);
if (ip->i_inode.i_size < offset + size)
i_size_write(&ip->i_inode, offset + size);
- ip->i_inode.i_mtime = ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ ip->i_inode.i_mtime = inode_ctime_set_current(&ip->i_inode);
gfs2_dinode_out(ip, dibh->b_data);
brelse(dibh);
@@ -227,7 +227,7 @@ static int gfs2_dir_write_data(struct gfs2_inode *ip, const char *buf,
if (ip->i_inode.i_size < offset + copied)
i_size_write(&ip->i_inode, offset + copied);
- ip->i_inode.i_mtime = ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ ip->i_inode.i_mtime = inode_ctime_set_current(&ip->i_inode);
gfs2_trans_add_meta(ip->i_gl, dibh);
gfs2_dinode_out(ip, dibh->b_data);
@@ -1814,7 +1814,7 @@ int gfs2_dir_add(struct inode *inode, const struct qstr *name,
gfs2_inum_out(nip, dent);
dent->de_type = cpu_to_be16(IF2DT(nip->i_inode.i_mode));
dent->de_rahead = cpu_to_be16(gfs2_inode_ra_len(nip));
- tv = current_time(&ip->i_inode);
+ tv = inode_ctime_set_current(&ip->i_inode);
if (ip->i_diskflags & GFS2_DIF_EXHASH) {
leaf = (struct gfs2_leaf *)bh->b_data;
be16_add_cpu(&leaf->lf_entries, 1);
@@ -1825,7 +1825,7 @@ int gfs2_dir_add(struct inode *inode, const struct qstr *name,
da->bh = NULL;
brelse(bh);
ip->i_entries++;
- ip->i_inode.i_mtime = ip->i_inode.i_ctime = tv;
+ ip->i_inode.i_mtime = tv;
if (S_ISDIR(nip->i_inode.i_mode))
inc_nlink(&ip->i_inode);
mark_inode_dirty(inode);
@@ -1876,7 +1876,7 @@ int gfs2_dir_del(struct gfs2_inode *dip, const struct dentry *dentry)
const struct qstr *name = &dentry->d_name;
struct gfs2_dirent *dent, *prev = NULL;
struct buffer_head *bh;
- struct timespec64 tv = current_time(&dip->i_inode);
+ struct timespec64 tv;
/* Returns _either_ the entry (if its first in block) or the
previous entry otherwise */
@@ -1895,6 +1895,7 @@ int gfs2_dir_del(struct gfs2_inode *dip, const struct dentry *dentry)
dent = (struct gfs2_dirent *)((char *)dent + be16_to_cpu(prev->de_rec_len));
}
+ tv = inode_ctime_set_current(&dip->i_inode);
dirent_del(dip, bh, prev, dent);
if (dip->i_diskflags & GFS2_DIF_EXHASH) {
struct gfs2_leaf *leaf = (struct gfs2_leaf *)bh->b_data;
@@ -1910,7 +1911,7 @@ int gfs2_dir_del(struct gfs2_inode *dip, const struct dentry *dentry)
if (!dip->i_entries)
gfs2_consist_inode(dip);
dip->i_entries--;
- dip->i_inode.i_mtime = dip->i_inode.i_ctime = tv;
+ dip->i_inode.i_mtime = tv;
if (d_is_dir(dentry))
drop_nlink(&dip->i_inode);
mark_inode_dirty(&dip->i_inode);
@@ -1951,7 +1952,7 @@ int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename,
dent->de_type = cpu_to_be16(new_type);
brelse(bh);
- dip->i_inode.i_mtime = dip->i_inode.i_ctime = current_time(&dip->i_inode);
+ dip->i_inode.i_mtime = inode_ctime_set_current(&dip->i_inode);
mark_inode_dirty_sync(&dip->i_inode);
return 0;
}
diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c
index 1bf3c4453516..cb754c5f1d2d 100644
--- a/fs/gfs2/file.c
+++ b/fs/gfs2/file.c
@@ -260,7 +260,7 @@ static int do_gfs2_set_flags(struct inode *inode, u32 reqflags, u32 mask)
error = gfs2_meta_inode_buffer(ip, &bh);
if (error)
goto out_trans_end;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
gfs2_trans_add_meta(ip->i_gl, bh);
ip->i_diskflags = new_flags;
gfs2_dinode_out(ip, bh->b_data);
diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c
index 7c48c7afd6a4..2aba6f82194f 100644
--- a/fs/gfs2/glops.c
+++ b/fs/gfs2/glops.c
@@ -437,8 +437,8 @@ static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf)
inode->i_atime = atime;
inode->i_mtime.tv_sec = be64_to_cpu(str->di_mtime);
inode->i_mtime.tv_nsec = be32_to_cpu(str->di_mtime_nsec);
- inode->i_ctime.tv_sec = be64_to_cpu(str->di_ctime);
- inode->i_ctime.tv_nsec = be32_to_cpu(str->di_ctime_nsec);
+ inode_ctime_set_sec(inode, be64_to_cpu(str->di_ctime));
+ inode_ctime_set_nsec(inode, be32_to_cpu(str->di_ctime_nsec));
ip->i_goal = be64_to_cpu(str->di_goal_meta);
ip->i_generation = be64_to_cpu(str->di_generation);
diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c
index 17c994a0c0d0..3a9c9b6ea456 100644
--- a/fs/gfs2/inode.c
+++ b/fs/gfs2/inode.c
@@ -690,7 +690,7 @@ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry,
set_nlink(inode, S_ISDIR(mode) ? 2 : 1);
inode->i_rdev = dev;
inode->i_size = size;
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
munge_mode_uid_gid(dip, inode);
check_and_update_goal(dip);
ip->i_goal = dip->i_goal;
@@ -1029,7 +1029,7 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir,
gfs2_trans_add_meta(ip->i_gl, dibh);
inc_nlink(&ip->i_inode);
- ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ inode_ctime_set_current(&ip->i_inode);
ihold(inode);
d_instantiate(dentry, inode);
mark_inode_dirty(inode);
@@ -1114,7 +1114,7 @@ static int gfs2_unlink_inode(struct gfs2_inode *dip,
return error;
ip->i_entries = 0;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
if (S_ISDIR(inode->i_mode))
clear_nlink(inode);
else
@@ -1371,7 +1371,7 @@ static int update_moved_ino(struct gfs2_inode *ip, struct gfs2_inode *ndip,
if (dir_rename)
return gfs2_dir_mvino(ip, &gfs2_qdotdot, ndip, DT_DIR);
- ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ inode_ctime_set_current(&ip->i_inode);
mark_inode_dirty_sync(&ip->i_inode);
return 0;
}
diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c
index 9be72d5aafea..fc495df15e71 100644
--- a/fs/gfs2/super.c
+++ b/fs/gfs2/super.c
@@ -412,7 +412,7 @@ void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf)
str->di_blocks = cpu_to_be64(gfs2_get_inode_blocks(inode));
str->di_atime = cpu_to_be64(inode->i_atime.tv_sec);
str->di_mtime = cpu_to_be64(inode->i_mtime.tv_sec);
- str->di_ctime = cpu_to_be64(inode->i_ctime.tv_sec);
+ str->di_ctime = cpu_to_be64(inode_ctime_peek(inode).tv_sec);
str->di_goal_meta = cpu_to_be64(ip->i_goal);
str->di_goal_data = cpu_to_be64(ip->i_goal);
@@ -429,7 +429,7 @@ void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf)
str->di_eattr = cpu_to_be64(ip->i_eattr);
str->di_atime_nsec = cpu_to_be32(inode->i_atime.tv_nsec);
str->di_mtime_nsec = cpu_to_be32(inode->i_mtime.tv_nsec);
- str->di_ctime_nsec = cpu_to_be32(inode->i_ctime.tv_nsec);
+ str->di_ctime_nsec = cpu_to_be32(inode_ctime_peek(inode).tv_nsec);
}
/**
diff --git a/fs/gfs2/xattr.c b/fs/gfs2/xattr.c
index 93b36d026bb4..8f807d18ec52 100644
--- a/fs/gfs2/xattr.c
+++ b/fs/gfs2/xattr.c
@@ -311,7 +311,7 @@ static int ea_dealloc_unstuffed(struct gfs2_inode *ip, struct buffer_head *bh,
ea->ea_num_ptrs = 0;
}
- ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ inode_ctime_set_current(&ip->i_inode);
__mark_inode_dirty(&ip->i_inode, I_DIRTY_DATASYNC);
gfs2_trans_end(sdp);
@@ -763,7 +763,7 @@ static int ea_alloc_skeleton(struct gfs2_inode *ip, struct gfs2_ea_request *er,
if (error)
goto out_end_trans;
- ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ inode_ctime_set_current(&ip->i_inode);
__mark_inode_dirty(&ip->i_inode, I_DIRTY_DATASYNC);
out_end_trans:
@@ -888,7 +888,7 @@ static int ea_set_simple_noalloc(struct gfs2_inode *ip, struct buffer_head *bh,
if (es->es_el)
ea_set_remove_stuffed(ip, es->es_el);
- ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ inode_ctime_set_current(&ip->i_inode);
__mark_inode_dirty(&ip->i_inode, I_DIRTY_DATASYNC);
gfs2_trans_end(GFS2_SB(&ip->i_inode));
@@ -1106,7 +1106,7 @@ static int ea_remove_stuffed(struct gfs2_inode *ip, struct gfs2_ea_location *el)
ea->ea_type = GFS2_EATYPE_UNUSED;
}
- ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ inode_ctime_set_current(&ip->i_inode);
__mark_inode_dirty(&ip->i_inode, I_DIRTY_DATASYNC);
gfs2_trans_end(GFS2_SB(&ip->i_inode));
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [Cluster-devel] [PATCH 34/79] gfs2: switch to new ctime accessors
@ 2023-06-21 14:45 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: cluster-devel.redhat.com
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/gfs2/acl.c | 2 +-
fs/gfs2/bmap.c | 11 +++++------
fs/gfs2/dir.c | 15 ++++++++-------
fs/gfs2/file.c | 2 +-
fs/gfs2/glops.c | 4 ++--
fs/gfs2/inode.c | 8 ++++----
fs/gfs2/super.c | 4 ++--
fs/gfs2/xattr.c | 8 ++++----
8 files changed, 27 insertions(+), 27 deletions(-)
diff --git a/fs/gfs2/acl.c b/fs/gfs2/acl.c
index a392aa0f041d..b267dae0dc63 100644
--- a/fs/gfs2/acl.c
+++ b/fs/gfs2/acl.c
@@ -142,7 +142,7 @@ int gfs2_set_acl(struct mnt_idmap *idmap, struct dentry *dentry,
ret = __gfs2_set_acl(inode, acl, type);
if (!ret && mode != inode->i_mode) {
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
inode->i_mode = mode;
mark_inode_dirty(inode);
}
diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c
index 8d611fbcf0bd..743b09a0b196 100644
--- a/fs/gfs2/bmap.c
+++ b/fs/gfs2/bmap.c
@@ -1386,7 +1386,7 @@ static int trunc_start(struct inode *inode, u64 newsize)
ip->i_diskflags |= GFS2_DIF_TRUNC_IN_PROG;
i_size_write(inode, newsize);
- ip->i_inode.i_mtime = ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
gfs2_dinode_out(ip, dibh->b_data);
if (journaled)
@@ -1583,8 +1583,7 @@ static int sweep_bh_for_rgrps(struct gfs2_inode *ip, struct gfs2_holder *rd_gh,
/* Every transaction boundary, we rewrite the dinode
to keep its di_blocks current in case of failure. */
- ip->i_inode.i_mtime = ip->i_inode.i_ctime =
- current_time(&ip->i_inode);
+ ip->i_inode.i_mtime = inode_ctime_set_current(&ip->i_inode);
gfs2_trans_add_meta(ip->i_gl, dibh);
gfs2_dinode_out(ip, dibh->b_data);
brelse(dibh);
@@ -1950,7 +1949,7 @@ static int punch_hole(struct gfs2_inode *ip, u64 offset, u64 length)
gfs2_statfs_change(sdp, 0, +btotal, 0);
gfs2_quota_change(ip, -(s64)btotal, ip->i_inode.i_uid,
ip->i_inode.i_gid);
- ip->i_inode.i_mtime = ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ ip->i_inode.i_mtime = inode_ctime_set_current(&ip->i_inode);
gfs2_trans_add_meta(ip->i_gl, dibh);
gfs2_dinode_out(ip, dibh->b_data);
up_write(&ip->i_rw_mutex);
@@ -1993,7 +1992,7 @@ static int trunc_end(struct gfs2_inode *ip)
gfs2_buffer_clear_tail(dibh, sizeof(struct gfs2_dinode));
gfs2_ordered_del_inode(ip);
}
- ip->i_inode.i_mtime = ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ ip->i_inode.i_mtime = inode_ctime_set_current(&ip->i_inode);
ip->i_diskflags &= ~GFS2_DIF_TRUNC_IN_PROG;
gfs2_trans_add_meta(ip->i_gl, dibh);
@@ -2094,7 +2093,7 @@ static int do_grow(struct inode *inode, u64 size)
goto do_end_trans;
truncate_setsize(inode, size);
- ip->i_inode.i_mtime = ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ ip->i_inode.i_mtime = inode_ctime_set_current(&ip->i_inode);
gfs2_trans_add_meta(ip->i_gl, dibh);
gfs2_dinode_out(ip, dibh->b_data);
brelse(dibh);
diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c
index 54a6d17b8c25..c07cb9883ea1 100644
--- a/fs/gfs2/dir.c
+++ b/fs/gfs2/dir.c
@@ -130,7 +130,7 @@ static int gfs2_dir_write_stuffed(struct gfs2_inode *ip, const char *buf,
memcpy(dibh->b_data + offset + sizeof(struct gfs2_dinode), buf, size);
if (ip->i_inode.i_size < offset + size)
i_size_write(&ip->i_inode, offset + size);
- ip->i_inode.i_mtime = ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ ip->i_inode.i_mtime = inode_ctime_set_current(&ip->i_inode);
gfs2_dinode_out(ip, dibh->b_data);
brelse(dibh);
@@ -227,7 +227,7 @@ static int gfs2_dir_write_data(struct gfs2_inode *ip, const char *buf,
if (ip->i_inode.i_size < offset + copied)
i_size_write(&ip->i_inode, offset + copied);
- ip->i_inode.i_mtime = ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ ip->i_inode.i_mtime = inode_ctime_set_current(&ip->i_inode);
gfs2_trans_add_meta(ip->i_gl, dibh);
gfs2_dinode_out(ip, dibh->b_data);
@@ -1814,7 +1814,7 @@ int gfs2_dir_add(struct inode *inode, const struct qstr *name,
gfs2_inum_out(nip, dent);
dent->de_type = cpu_to_be16(IF2DT(nip->i_inode.i_mode));
dent->de_rahead = cpu_to_be16(gfs2_inode_ra_len(nip));
- tv = current_time(&ip->i_inode);
+ tv = inode_ctime_set_current(&ip->i_inode);
if (ip->i_diskflags & GFS2_DIF_EXHASH) {
leaf = (struct gfs2_leaf *)bh->b_data;
be16_add_cpu(&leaf->lf_entries, 1);
@@ -1825,7 +1825,7 @@ int gfs2_dir_add(struct inode *inode, const struct qstr *name,
da->bh = NULL;
brelse(bh);
ip->i_entries++;
- ip->i_inode.i_mtime = ip->i_inode.i_ctime = tv;
+ ip->i_inode.i_mtime = tv;
if (S_ISDIR(nip->i_inode.i_mode))
inc_nlink(&ip->i_inode);
mark_inode_dirty(inode);
@@ -1876,7 +1876,7 @@ int gfs2_dir_del(struct gfs2_inode *dip, const struct dentry *dentry)
const struct qstr *name = &dentry->d_name;
struct gfs2_dirent *dent, *prev = NULL;
struct buffer_head *bh;
- struct timespec64 tv = current_time(&dip->i_inode);
+ struct timespec64 tv;
/* Returns _either_ the entry (if its first in block) or the
previous entry otherwise */
@@ -1895,6 +1895,7 @@ int gfs2_dir_del(struct gfs2_inode *dip, const struct dentry *dentry)
dent = (struct gfs2_dirent *)((char *)dent + be16_to_cpu(prev->de_rec_len));
}
+ tv = inode_ctime_set_current(&dip->i_inode);
dirent_del(dip, bh, prev, dent);
if (dip->i_diskflags & GFS2_DIF_EXHASH) {
struct gfs2_leaf *leaf = (struct gfs2_leaf *)bh->b_data;
@@ -1910,7 +1911,7 @@ int gfs2_dir_del(struct gfs2_inode *dip, const struct dentry *dentry)
if (!dip->i_entries)
gfs2_consist_inode(dip);
dip->i_entries--;
- dip->i_inode.i_mtime = dip->i_inode.i_ctime = tv;
+ dip->i_inode.i_mtime = tv;
if (d_is_dir(dentry))
drop_nlink(&dip->i_inode);
mark_inode_dirty(&dip->i_inode);
@@ -1951,7 +1952,7 @@ int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename,
dent->de_type = cpu_to_be16(new_type);
brelse(bh);
- dip->i_inode.i_mtime = dip->i_inode.i_ctime = current_time(&dip->i_inode);
+ dip->i_inode.i_mtime = inode_ctime_set_current(&dip->i_inode);
mark_inode_dirty_sync(&dip->i_inode);
return 0;
}
diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c
index 1bf3c4453516..cb754c5f1d2d 100644
--- a/fs/gfs2/file.c
+++ b/fs/gfs2/file.c
@@ -260,7 +260,7 @@ static int do_gfs2_set_flags(struct inode *inode, u32 reqflags, u32 mask)
error = gfs2_meta_inode_buffer(ip, &bh);
if (error)
goto out_trans_end;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
gfs2_trans_add_meta(ip->i_gl, bh);
ip->i_diskflags = new_flags;
gfs2_dinode_out(ip, bh->b_data);
diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c
index 7c48c7afd6a4..2aba6f82194f 100644
--- a/fs/gfs2/glops.c
+++ b/fs/gfs2/glops.c
@@ -437,8 +437,8 @@ static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf)
inode->i_atime = atime;
inode->i_mtime.tv_sec = be64_to_cpu(str->di_mtime);
inode->i_mtime.tv_nsec = be32_to_cpu(str->di_mtime_nsec);
- inode->i_ctime.tv_sec = be64_to_cpu(str->di_ctime);
- inode->i_ctime.tv_nsec = be32_to_cpu(str->di_ctime_nsec);
+ inode_ctime_set_sec(inode, be64_to_cpu(str->di_ctime));
+ inode_ctime_set_nsec(inode, be32_to_cpu(str->di_ctime_nsec));
ip->i_goal = be64_to_cpu(str->di_goal_meta);
ip->i_generation = be64_to_cpu(str->di_generation);
diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c
index 17c994a0c0d0..3a9c9b6ea456 100644
--- a/fs/gfs2/inode.c
+++ b/fs/gfs2/inode.c
@@ -690,7 +690,7 @@ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry,
set_nlink(inode, S_ISDIR(mode) ? 2 : 1);
inode->i_rdev = dev;
inode->i_size = size;
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
munge_mode_uid_gid(dip, inode);
check_and_update_goal(dip);
ip->i_goal = dip->i_goal;
@@ -1029,7 +1029,7 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir,
gfs2_trans_add_meta(ip->i_gl, dibh);
inc_nlink(&ip->i_inode);
- ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ inode_ctime_set_current(&ip->i_inode);
ihold(inode);
d_instantiate(dentry, inode);
mark_inode_dirty(inode);
@@ -1114,7 +1114,7 @@ static int gfs2_unlink_inode(struct gfs2_inode *dip,
return error;
ip->i_entries = 0;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
if (S_ISDIR(inode->i_mode))
clear_nlink(inode);
else
@@ -1371,7 +1371,7 @@ static int update_moved_ino(struct gfs2_inode *ip, struct gfs2_inode *ndip,
if (dir_rename)
return gfs2_dir_mvino(ip, &gfs2_qdotdot, ndip, DT_DIR);
- ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ inode_ctime_set_current(&ip->i_inode);
mark_inode_dirty_sync(&ip->i_inode);
return 0;
}
diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c
index 9be72d5aafea..fc495df15e71 100644
--- a/fs/gfs2/super.c
+++ b/fs/gfs2/super.c
@@ -412,7 +412,7 @@ void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf)
str->di_blocks = cpu_to_be64(gfs2_get_inode_blocks(inode));
str->di_atime = cpu_to_be64(inode->i_atime.tv_sec);
str->di_mtime = cpu_to_be64(inode->i_mtime.tv_sec);
- str->di_ctime = cpu_to_be64(inode->i_ctime.tv_sec);
+ str->di_ctime = cpu_to_be64(inode_ctime_peek(inode).tv_sec);
str->di_goal_meta = cpu_to_be64(ip->i_goal);
str->di_goal_data = cpu_to_be64(ip->i_goal);
@@ -429,7 +429,7 @@ void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf)
str->di_eattr = cpu_to_be64(ip->i_eattr);
str->di_atime_nsec = cpu_to_be32(inode->i_atime.tv_nsec);
str->di_mtime_nsec = cpu_to_be32(inode->i_mtime.tv_nsec);
- str->di_ctime_nsec = cpu_to_be32(inode->i_ctime.tv_nsec);
+ str->di_ctime_nsec = cpu_to_be32(inode_ctime_peek(inode).tv_nsec);
}
/**
diff --git a/fs/gfs2/xattr.c b/fs/gfs2/xattr.c
index 93b36d026bb4..8f807d18ec52 100644
--- a/fs/gfs2/xattr.c
+++ b/fs/gfs2/xattr.c
@@ -311,7 +311,7 @@ static int ea_dealloc_unstuffed(struct gfs2_inode *ip, struct buffer_head *bh,
ea->ea_num_ptrs = 0;
}
- ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ inode_ctime_set_current(&ip->i_inode);
__mark_inode_dirty(&ip->i_inode, I_DIRTY_DATASYNC);
gfs2_trans_end(sdp);
@@ -763,7 +763,7 @@ static int ea_alloc_skeleton(struct gfs2_inode *ip, struct gfs2_ea_request *er,
if (error)
goto out_end_trans;
- ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ inode_ctime_set_current(&ip->i_inode);
__mark_inode_dirty(&ip->i_inode, I_DIRTY_DATASYNC);
out_end_trans:
@@ -888,7 +888,7 @@ static int ea_set_simple_noalloc(struct gfs2_inode *ip, struct buffer_head *bh,
if (es->es_el)
ea_set_remove_stuffed(ip, es->es_el);
- ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ inode_ctime_set_current(&ip->i_inode);
__mark_inode_dirty(&ip->i_inode, I_DIRTY_DATASYNC);
gfs2_trans_end(GFS2_SB(&ip->i_inode));
@@ -1106,7 +1106,7 @@ static int ea_remove_stuffed(struct gfs2_inode *ip, struct gfs2_ea_location *el)
ea->ea_type = GFS2_EATYPE_UNUSED;
}
- ip->i_inode.i_ctime = current_time(&ip->i_inode);
+ inode_ctime_set_current(&ip->i_inode);
__mark_inode_dirty(&ip->i_inode, I_DIRTY_DATASYNC);
gfs2_trans_end(GFS2_SB(&ip->i_inode));
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 35/79] hfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (32 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner; +Cc: Al Viro, Jan Kara, linux-fsdevel, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/hfs/catalog.c | 8 ++++----
fs/hfs/dir.c | 2 +-
fs/hfs/inode.c | 13 ++++++-------
fs/hfs/sysdep.c | 2 +-
4 files changed, 12 insertions(+), 13 deletions(-)
diff --git a/fs/hfs/catalog.c b/fs/hfs/catalog.c
index d365bf0b8c77..6fd2ae856178 100644
--- a/fs/hfs/catalog.c
+++ b/fs/hfs/catalog.c
@@ -133,7 +133,7 @@ int hfs_cat_create(u32 cnid, struct inode *dir, const struct qstr *str, struct i
goto err1;
dir->i_size++;
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
mark_inode_dirty(dir);
hfs_find_exit(&fd);
return 0;
@@ -269,7 +269,7 @@ int hfs_cat_delete(u32 cnid, struct inode *dir, const struct qstr *str)
}
dir->i_size--;
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
mark_inode_dirty(dir);
res = 0;
out:
@@ -337,7 +337,7 @@ int hfs_cat_move(u32 cnid, struct inode *src_dir, const struct qstr *src_name,
if (err)
goto out;
dst_dir->i_size++;
- dst_dir->i_mtime = dst_dir->i_ctime = current_time(dst_dir);
+ dst_dir->i_mtime = inode_ctime_set_current(dst_dir);
mark_inode_dirty(dst_dir);
/* finally remove the old entry */
@@ -349,7 +349,7 @@ int hfs_cat_move(u32 cnid, struct inode *src_dir, const struct qstr *src_name,
if (err)
goto out;
src_dir->i_size--;
- src_dir->i_mtime = src_dir->i_ctime = current_time(src_dir);
+ src_dir->i_mtime = inode_ctime_set_current(src_dir);
mark_inode_dirty(src_dir);
type = entry.type;
diff --git a/fs/hfs/dir.c b/fs/hfs/dir.c
index 3e1e3dcf0b48..bb9e651c1008 100644
--- a/fs/hfs/dir.c
+++ b/fs/hfs/dir.c
@@ -263,7 +263,7 @@ static int hfs_remove(struct inode *dir, struct dentry *dentry)
if (res)
return res;
clear_nlink(inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
hfs_delete_inode(inode);
mark_inode_dirty(inode);
return 0;
diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c
index 441d7fc952e3..d8008d926a19 100644
--- a/fs/hfs/inode.c
+++ b/fs/hfs/inode.c
@@ -200,7 +200,7 @@ struct inode *hfs_new_inode(struct inode *dir, const struct qstr *name, umode_t
inode->i_uid = current_fsuid();
inode->i_gid = current_fsgid();
set_nlink(inode, 1);
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
HFS_I(inode)->flags = 0;
HFS_I(inode)->rsrc_inode = NULL;
HFS_I(inode)->fs_blocks = 0;
@@ -355,8 +355,8 @@ static int hfs_read_inode(struct inode *inode, void *data)
inode->i_mode |= S_IWUGO;
inode->i_mode &= ~hsb->s_file_umask;
inode->i_mode |= S_IFREG;
- inode->i_ctime = inode->i_atime = inode->i_mtime =
- hfs_m_to_utime(rec->file.MdDat);
+ inode->i_atime = inode->i_mtime = hfs_m_to_utime(rec->file.MdDat);
+ inode_ctime_set(inode, inode->i_mtime);
inode->i_op = &hfs_file_inode_operations;
inode->i_fop = &hfs_file_operations;
inode->i_mapping->a_ops = &hfs_aops;
@@ -366,8 +366,8 @@ static int hfs_read_inode(struct inode *inode, void *data)
inode->i_size = be16_to_cpu(rec->dir.Val) + 2;
HFS_I(inode)->fs_blocks = 0;
inode->i_mode = S_IFDIR | (S_IRWXUGO & ~hsb->s_dir_umask);
- inode->i_ctime = inode->i_atime = inode->i_mtime =
- hfs_m_to_utime(rec->dir.MdDat);
+ inode->i_atime = inode->i_mtime = hfs_m_to_utime(rec->dir.MdDat);
+ inode_ctime_set(inode, inode->i_mtime);
inode->i_op = &hfs_dir_inode_operations;
inode->i_fop = &hfs_dir_operations;
break;
@@ -654,8 +654,7 @@ int hfs_inode_setattr(struct mnt_idmap *idmap, struct dentry *dentry,
truncate_setsize(inode, attr->ia_size);
hfs_file_truncate(inode);
- inode->i_atime = inode->i_mtime = inode->i_ctime =
- current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
}
setattr_copy(&nop_mnt_idmap, inode, attr);
diff --git a/fs/hfs/sysdep.c b/fs/hfs/sysdep.c
index 2875961fdc10..e2949390fadc 100644
--- a/fs/hfs/sysdep.c
+++ b/fs/hfs/sysdep.c
@@ -28,7 +28,7 @@ static int hfs_revalidate_dentry(struct dentry *dentry, unsigned int flags)
/* fix up inode on a timezone change */
diff = sys_tz.tz_minuteswest * 60 - HFS_I(inode)->tz_secondswest;
if (diff) {
- inode->i_ctime.tv_sec += diff;
+ inode_ctime_set_sec(inode, inode_ctime_peek(inode).tv_sec + diff);
inode->i_atime.tv_sec += diff;
inode->i_mtime.tv_sec += diff;
HFS_I(inode)->tz_secondswest += diff;
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 36/79] hfsplus: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (33 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner; +Cc: Al Viro, Jan Kara, linux-fsdevel, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/hfsplus/catalog.c | 8 ++++----
fs/hfsplus/dir.c | 6 +++---
fs/hfsplus/inode.c | 14 +++++++-------
3 files changed, 14 insertions(+), 14 deletions(-)
diff --git a/fs/hfsplus/catalog.c b/fs/hfsplus/catalog.c
index 35472cba750e..0e1938729669 100644
--- a/fs/hfsplus/catalog.c
+++ b/fs/hfsplus/catalog.c
@@ -312,7 +312,7 @@ int hfsplus_create_cat(u32 cnid, struct inode *dir,
dir->i_size++;
if (S_ISDIR(inode->i_mode))
hfsplus_subfolders_inc(dir);
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
hfsplus_mark_inode_dirty(dir, HFSPLUS_I_CAT_DIRTY);
hfs_find_exit(&fd);
@@ -417,7 +417,7 @@ int hfsplus_delete_cat(u32 cnid, struct inode *dir, const struct qstr *str)
dir->i_size--;
if (type == HFSPLUS_FOLDER)
hfsplus_subfolders_dec(dir);
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
hfsplus_mark_inode_dirty(dir, HFSPLUS_I_CAT_DIRTY);
if (type == HFSPLUS_FILE || type == HFSPLUS_FOLDER) {
@@ -494,7 +494,7 @@ int hfsplus_rename_cat(u32 cnid,
dst_dir->i_size++;
if (type == HFSPLUS_FOLDER)
hfsplus_subfolders_inc(dst_dir);
- dst_dir->i_mtime = dst_dir->i_ctime = current_time(dst_dir);
+ dst_dir->i_mtime = inode_ctime_set_current(dst_dir);
/* finally remove the old entry */
err = hfsplus_cat_build_key(sb, src_fd.search_key,
@@ -511,7 +511,7 @@ int hfsplus_rename_cat(u32 cnid,
src_dir->i_size--;
if (type == HFSPLUS_FOLDER)
hfsplus_subfolders_dec(src_dir);
- src_dir->i_mtime = src_dir->i_ctime = current_time(src_dir);
+ src_dir->i_mtime = inode_ctime_set_current(src_dir);
/* remove old thread entry */
hfsplus_cat_build_key_with_cnid(sb, src_fd.search_key, cnid);
diff --git a/fs/hfsplus/dir.c b/fs/hfsplus/dir.c
index 56fb5f1312e7..e7b6de12ecef 100644
--- a/fs/hfsplus/dir.c
+++ b/fs/hfsplus/dir.c
@@ -346,7 +346,7 @@ static int hfsplus_link(struct dentry *src_dentry, struct inode *dst_dir,
inc_nlink(inode);
hfsplus_instantiate(dst_dentry, inode, cnid);
ihold(inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
mark_inode_dirty(inode);
sbi->file_count++;
hfsplus_mark_mdb_dirty(dst_dir->i_sb);
@@ -405,7 +405,7 @@ static int hfsplus_unlink(struct inode *dir, struct dentry *dentry)
hfsplus_delete_inode(inode);
} else
sbi->file_count--;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
mark_inode_dirty(inode);
out:
mutex_unlock(&sbi->vh_mutex);
@@ -426,7 +426,7 @@ static int hfsplus_rmdir(struct inode *dir, struct dentry *dentry)
if (res)
goto out;
clear_nlink(inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
hfsplus_delete_inode(inode);
mark_inode_dirty(inode);
out:
diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c
index 7d1a675e037d..b9c02df839c8 100644
--- a/fs/hfsplus/inode.c
+++ b/fs/hfsplus/inode.c
@@ -267,7 +267,7 @@ static int hfsplus_setattr(struct mnt_idmap *idmap,
}
truncate_setsize(inode, attr->ia_size);
hfsplus_file_truncate(inode);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
}
setattr_copy(&nop_mnt_idmap, inode, attr);
@@ -392,7 +392,7 @@ struct inode *hfsplus_new_inode(struct super_block *sb, struct inode *dir,
inode->i_ino = sbi->next_cnid++;
inode_init_owner(&nop_mnt_idmap, inode, dir, mode);
set_nlink(inode, 1);
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
hip = HFSPLUS_I(inode);
INIT_LIST_HEAD(&hip->open_dir_list);
@@ -523,7 +523,7 @@ int hfsplus_cat_read_inode(struct inode *inode, struct hfs_find_data *fd)
inode->i_size = 2 + be32_to_cpu(folder->valence);
inode->i_atime = hfsp_mt2ut(folder->access_date);
inode->i_mtime = hfsp_mt2ut(folder->content_mod_date);
- inode->i_ctime = hfsp_mt2ut(folder->attribute_mod_date);
+ inode_ctime_set(inode, hfsp_mt2ut(folder->attribute_mod_date));
HFSPLUS_I(inode)->create_date = folder->create_date;
HFSPLUS_I(inode)->fs_blocks = 0;
if (folder->flags & cpu_to_be16(HFSPLUS_HAS_FOLDER_COUNT)) {
@@ -564,7 +564,7 @@ int hfsplus_cat_read_inode(struct inode *inode, struct hfs_find_data *fd)
}
inode->i_atime = hfsp_mt2ut(file->access_date);
inode->i_mtime = hfsp_mt2ut(file->content_mod_date);
- inode->i_ctime = hfsp_mt2ut(file->attribute_mod_date);
+ inode_ctime_set(inode, hfsp_mt2ut(file->attribute_mod_date));
HFSPLUS_I(inode)->create_date = file->create_date;
} else {
pr_err("bad catalog entry used to create inode\n");
@@ -609,7 +609,7 @@ int hfsplus_cat_write_inode(struct inode *inode)
hfsplus_cat_set_perms(inode, &folder->permissions);
folder->access_date = hfsp_ut2mt(inode->i_atime);
folder->content_mod_date = hfsp_ut2mt(inode->i_mtime);
- folder->attribute_mod_date = hfsp_ut2mt(inode->i_ctime);
+ folder->attribute_mod_date = hfsp_ut2mt(inode_ctime_peek(inode));
folder->valence = cpu_to_be32(inode->i_size - 2);
if (folder->flags & cpu_to_be16(HFSPLUS_HAS_FOLDER_COUNT)) {
folder->subfolders =
@@ -644,7 +644,7 @@ int hfsplus_cat_write_inode(struct inode *inode)
file->flags &= cpu_to_be16(~HFSPLUS_FILE_LOCKED);
file->access_date = hfsp_ut2mt(inode->i_atime);
file->content_mod_date = hfsp_ut2mt(inode->i_mtime);
- file->attribute_mod_date = hfsp_ut2mt(inode->i_ctime);
+ file->attribute_mod_date = hfsp_ut2mt(inode_ctime_peek(inode));
hfs_bnode_write(fd.bnode, &entry, fd.entryoffset,
sizeof(struct hfsplus_cat_file));
}
@@ -700,7 +700,7 @@ int hfsplus_fileattr_set(struct mnt_idmap *idmap,
else
hip->userflags &= ~HFSPLUS_FLG_NODUMP;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
mark_inode_dirty(inode);
return 0;
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 37/79] hostfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Richard Weinberger, Anton Ivanov, Johannes Berg
Cc: Al Viro, Jan Kara, linux-um, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/hostfs/hostfs_kern.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c
index 46387090eb76..2248a2e599a7 100644
--- a/fs/hostfs/hostfs_kern.c
+++ b/fs/hostfs/hostfs_kern.c
@@ -517,8 +517,8 @@ static int hostfs_inode_update(struct inode *ino, const struct hostfs_stat *st)
(struct timespec64){ st->atime.tv_sec, st->atime.tv_nsec };
ino->i_mtime =
(struct timespec64){ st->mtime.tv_sec, st->mtime.tv_nsec };
- ino->i_ctime =
- (struct timespec64){ st->ctime.tv_sec, st->ctime.tv_nsec };
+ inode_ctime_set(ino,
+ (struct timespec64){ st->ctime.tv_sec, st->ctime.tv_nsec });
ino->i_size = st->size;
ino->i_blocks = st->blocks;
return 0;
--
2.41.0
_______________________________________________
linux-um mailing list
linux-um@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-um
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 37/79] hostfs: switch to new ctime accessors
@ 2023-06-21 14:45 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Richard Weinberger, Anton Ivanov, Johannes Berg
Cc: Al Viro, Jan Kara, linux-um, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/hostfs/hostfs_kern.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c
index 46387090eb76..2248a2e599a7 100644
--- a/fs/hostfs/hostfs_kern.c
+++ b/fs/hostfs/hostfs_kern.c
@@ -517,8 +517,8 @@ static int hostfs_inode_update(struct inode *ino, const struct hostfs_stat *st)
(struct timespec64){ st->atime.tv_sec, st->atime.tv_nsec };
ino->i_mtime =
(struct timespec64){ st->mtime.tv_sec, st->mtime.tv_nsec };
- ino->i_ctime =
- (struct timespec64){ st->ctime.tv_sec, st->ctime.tv_nsec };
+ inode_ctime_set(ino,
+ (struct timespec64){ st->ctime.tv_sec, st->ctime.tv_nsec });
ino->i_size = st->size;
ino->i_blocks = st->blocks;
return 0;
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 38/79] hpfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (35 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Mikulas Patocka; +Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/hpfs/dir.c | 8 ++++----
fs/hpfs/inode.c | 6 +++---
fs/hpfs/namei.c | 26 +++++++++++++++-----------
fs/hpfs/super.c | 5 +++--
4 files changed, 25 insertions(+), 20 deletions(-)
diff --git a/fs/hpfs/dir.c b/fs/hpfs/dir.c
index f32f15669996..fe808e6a3962 100644
--- a/fs/hpfs/dir.c
+++ b/fs/hpfs/dir.c
@@ -277,10 +277,10 @@ struct dentry *hpfs_lookup(struct inode *dir, struct dentry *dentry, unsigned in
* inode.
*/
- if (!result->i_ctime.tv_sec) {
- if (!(result->i_ctime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(de->creation_date))))
- result->i_ctime.tv_sec = 1;
- result->i_ctime.tv_nsec = 0;
+ if (!inode_ctime_peek(result).tv_sec) {
+ if (!(inode_ctime_set_sec(result, local_to_gmt(dir->i_sb, le32_to_cpu(de->creation_date)))))
+ inode_ctime_set_sec(result, 1);
+ inode_ctime_set_nsec(result, 0);
result->i_mtime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(de->write_date));
result->i_mtime.tv_nsec = 0;
result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(de->read_date));
diff --git a/fs/hpfs/inode.c b/fs/hpfs/inode.c
index e50e92a42432..48e71035d88c 100644
--- a/fs/hpfs/inode.c
+++ b/fs/hpfs/inode.c
@@ -36,7 +36,7 @@ void hpfs_init_inode(struct inode *i)
hpfs_inode->i_rddir_off = NULL;
hpfs_inode->i_dirty = 0;
- i->i_ctime.tv_sec = i->i_ctime.tv_nsec = 0;
+ inode_ctime_set_sec(i, inode_ctime_set_nsec(i, 0));
i->i_mtime.tv_sec = i->i_mtime.tv_nsec = 0;
i->i_atime.tv_sec = i->i_atime.tv_nsec = 0;
}
@@ -232,7 +232,7 @@ void hpfs_write_inode_nolock(struct inode *i)
if (de) {
de->write_date = cpu_to_le32(gmt_to_local(i->i_sb, i->i_mtime.tv_sec));
de->read_date = cpu_to_le32(gmt_to_local(i->i_sb, i->i_atime.tv_sec));
- de->creation_date = cpu_to_le32(gmt_to_local(i->i_sb, i->i_ctime.tv_sec));
+ de->creation_date = cpu_to_le32(gmt_to_local(i->i_sb, inode_ctime_peek(i).tv_sec));
de->read_only = !(i->i_mode & 0222);
de->ea_size = cpu_to_le32(hpfs_inode->i_ea_size);
hpfs_mark_4buffers_dirty(&qbh);
@@ -242,7 +242,7 @@ void hpfs_write_inode_nolock(struct inode *i)
if ((de = map_dirent(i, hpfs_inode->i_dno, "\001\001", 2, NULL, &qbh))) {
de->write_date = cpu_to_le32(gmt_to_local(i->i_sb, i->i_mtime.tv_sec));
de->read_date = cpu_to_le32(gmt_to_local(i->i_sb, i->i_atime.tv_sec));
- de->creation_date = cpu_to_le32(gmt_to_local(i->i_sb, i->i_ctime.tv_sec));
+ de->creation_date = cpu_to_le32(gmt_to_local(i->i_sb, inode_ctime_peek(i).tv_sec));
de->read_only = !(i->i_mode & 0222);
de->ea_size = cpu_to_le32(/*hpfs_inode->i_ea_size*/0);
de->file_size = cpu_to_le32(0);
diff --git a/fs/hpfs/namei.c b/fs/hpfs/namei.c
index 69fb40b2c99a..c065ef3fec52 100644
--- a/fs/hpfs/namei.c
+++ b/fs/hpfs/namei.c
@@ -13,10 +13,10 @@ static void hpfs_update_directory_times(struct inode *dir)
{
time64_t t = local_to_gmt(dir->i_sb, local_get_seconds(dir->i_sb));
if (t == dir->i_mtime.tv_sec &&
- t == dir->i_ctime.tv_sec)
+ t == inode_ctime_peek(dir).tv_sec)
return;
- dir->i_mtime.tv_sec = dir->i_ctime.tv_sec = t;
- dir->i_mtime.tv_nsec = dir->i_ctime.tv_nsec = 0;
+ dir->i_mtime.tv_sec = inode_ctime_set_sec(dir, t);
+ dir->i_mtime.tv_nsec = inode_ctime_set_nsec(dir, 0);
hpfs_write_inode_nolock(dir);
}
@@ -59,8 +59,9 @@ static int hpfs_mkdir(struct mnt_idmap *idmap, struct inode *dir,
result->i_ino = fno;
hpfs_i(result)->i_parent_dir = dir->i_ino;
hpfs_i(result)->i_dno = dno;
- result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date));
- result->i_ctime.tv_nsec = 0;
+ inode_ctime_set_sec(result,
+ result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date)));
+ inode_ctime_set_nsec(result, 0);
result->i_mtime.tv_nsec = 0;
result->i_atime.tv_nsec = 0;
hpfs_i(result)->i_ea_size = 0;
@@ -167,8 +168,9 @@ static int hpfs_create(struct mnt_idmap *idmap, struct inode *dir,
result->i_fop = &hpfs_file_ops;
set_nlink(result, 1);
hpfs_i(result)->i_parent_dir = dir->i_ino;
- result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date));
- result->i_ctime.tv_nsec = 0;
+ inode_ctime_set_sec(result,
+ result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date)));
+ inode_ctime_set_nsec(result, 0);
result->i_mtime.tv_nsec = 0;
result->i_atime.tv_nsec = 0;
hpfs_i(result)->i_ea_size = 0;
@@ -250,8 +252,9 @@ static int hpfs_mknod(struct mnt_idmap *idmap, struct inode *dir,
hpfs_init_inode(result);
result->i_ino = fno;
hpfs_i(result)->i_parent_dir = dir->i_ino;
- result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date));
- result->i_ctime.tv_nsec = 0;
+ inode_ctime_set_sec(result,
+ result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date)));
+ inode_ctime_set_nsec(result, 0);
result->i_mtime.tv_nsec = 0;
result->i_atime.tv_nsec = 0;
hpfs_i(result)->i_ea_size = 0;
@@ -326,8 +329,9 @@ static int hpfs_symlink(struct mnt_idmap *idmap, struct inode *dir,
result->i_ino = fno;
hpfs_init_inode(result);
hpfs_i(result)->i_parent_dir = dir->i_ino;
- result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date));
- result->i_ctime.tv_nsec = 0;
+ inode_ctime_set_sec(result,
+ result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date)));
+ inode_ctime_set_nsec(result, 0);
result->i_mtime.tv_nsec = 0;
result->i_atime.tv_nsec = 0;
hpfs_i(result)->i_ea_size = 0;
diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c
index 1cb89595b875..cc7602d5485e 100644
--- a/fs/hpfs/super.c
+++ b/fs/hpfs/super.c
@@ -729,8 +729,9 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent)
root->i_atime.tv_nsec = 0;
root->i_mtime.tv_sec = local_to_gmt(s, le32_to_cpu(de->write_date));
root->i_mtime.tv_nsec = 0;
- root->i_ctime.tv_sec = local_to_gmt(s, le32_to_cpu(de->creation_date));
- root->i_ctime.tv_nsec = 0;
+ inode_ctime_set_sec(root,
+ local_to_gmt(s, le32_to_cpu(de->creation_date)));
+ inode_ctime_set_nsec(root, 0);
hpfs_i(root)->i_ea_size = le32_to_cpu(de->ea_size);
hpfs_i(root)->i_parent_dir = root->i_ino;
if (root->i_size == -1)
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 39/79] hugetlbfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (36 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Mike Kravetz, Muchun Song
Cc: Al Viro, Jan Kara, linux-mm, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/hugetlbfs/inode.c | 12 ++++++------
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index 90361a922cec..7be5a8f5927f 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -889,7 +889,7 @@ static long hugetlbfs_fallocate(struct file *file, int mode, loff_t offset,
if (!(mode & FALLOC_FL_KEEP_SIZE) && offset + len > inode->i_size)
i_size_write(inode, offset + len);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
out:
inode_unlock(inode);
return error;
@@ -937,7 +937,7 @@ static struct inode *hugetlbfs_get_root(struct super_block *sb,
inode->i_mode = S_IFDIR | ctx->mode;
inode->i_uid = ctx->uid;
inode->i_gid = ctx->gid;
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
inode->i_op = &hugetlbfs_dir_inode_operations;
inode->i_fop = &simple_dir_operations;
/* directory inodes start off with i_nlink == 2 (for "." entry) */
@@ -981,7 +981,7 @@ static struct inode *hugetlbfs_get_inode(struct super_block *sb,
lockdep_set_class(&inode->i_mapping->i_mmap_rwsem,
&hugetlbfs_i_mmap_rwsem_key);
inode->i_mapping->a_ops = &hugetlbfs_aops;
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
inode->i_mapping->private_data = resv_map;
info->seals = F_SEAL_SEAL;
switch (mode & S_IFMT) {
@@ -1024,7 +1024,7 @@ static int hugetlbfs_mknod(struct mnt_idmap *idmap, struct inode *dir,
inode = hugetlbfs_get_inode(dir->i_sb, dir, mode, dev);
if (!inode)
return -ENOSPC;
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
d_instantiate(dentry, inode);
dget(dentry);/* Extra count - pin the dentry in core */
return 0;
@@ -1056,7 +1056,7 @@ static int hugetlbfs_tmpfile(struct mnt_idmap *idmap,
inode = hugetlbfs_get_inode(dir->i_sb, dir, mode | S_IFREG, 0);
if (!inode)
return -ENOSPC;
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
d_tmpfile(file, inode);
return finish_open_simple(file, 0);
}
@@ -1078,7 +1078,7 @@ static int hugetlbfs_symlink(struct mnt_idmap *idmap,
} else
iput(inode);
}
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
return error;
}
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 40/79] isofs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (37 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Jan Kara; +Cc: Al Viro, linux-fsdevel, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/isofs/inode.c | 4 ++--
fs/isofs/rock.c | 16 +++++++---------
2 files changed, 9 insertions(+), 11 deletions(-)
diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c
index df9d70588b60..035fa0271d6e 100644
--- a/fs/isofs/inode.c
+++ b/fs/isofs/inode.c
@@ -1425,10 +1425,10 @@ static int isofs_read_inode(struct inode *inode, int relocated)
inode->i_mtime.tv_sec =
inode->i_atime.tv_sec =
- inode->i_ctime.tv_sec = iso_date(de->date, high_sierra);
+ inode_ctime_set_sec(inode, iso_date(de->date, high_sierra));
inode->i_mtime.tv_nsec =
inode->i_atime.tv_nsec =
- inode->i_ctime.tv_nsec = 0;
+ inode_ctime_set_nsec(inode, 0);
ei->i_first_extent = (isonum_733(de->extent) +
isonum_711(de->ext_attr_length));
diff --git a/fs/isofs/rock.c b/fs/isofs/rock.c
index 48f58c6c9e69..6b7f2a62124d 100644
--- a/fs/isofs/rock.c
+++ b/fs/isofs/rock.c
@@ -421,10 +421,9 @@ parse_rock_ridge_inode_internal(struct iso_directory_record *de,
/* Rock ridge never appears on a High Sierra disk */
cnt = 0;
if (rr->u.TF.flags & TF_CREATE) {
- inode->i_ctime.tv_sec =
- iso_date(rr->u.TF.times[cnt++].time,
- 0);
- inode->i_ctime.tv_nsec = 0;
+ inode_ctime_set_sec(inode,
+ iso_date(rr->u.TF.times[cnt++].time, 0));
+ inode_ctime_set_nsec(inode, 0);
}
if (rr->u.TF.flags & TF_MODIFY) {
inode->i_mtime.tv_sec =
@@ -439,10 +438,9 @@ parse_rock_ridge_inode_internal(struct iso_directory_record *de,
inode->i_atime.tv_nsec = 0;
}
if (rr->u.TF.flags & TF_ATTRIBUTES) {
- inode->i_ctime.tv_sec =
- iso_date(rr->u.TF.times[cnt++].time,
- 0);
- inode->i_ctime.tv_nsec = 0;
+ inode_ctime_set_sec(inode,
+ iso_date(rr->u.TF.times[cnt++].time, 0));
+ inode_ctime_set_nsec(inode, 0);
}
break;
case SIG('S', 'L'):
@@ -534,7 +532,7 @@ parse_rock_ridge_inode_internal(struct iso_directory_record *de,
inode->i_size = reloc->i_size;
inode->i_blocks = reloc->i_blocks;
inode->i_atime = reloc->i_atime;
- inode->i_ctime = reloc->i_ctime;
+ inode_ctime_set(inode, inode_ctime_peek(reloc));
inode->i_mtime = reloc->i_mtime;
iput(reloc);
break;
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 41/79] jffs2: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, David Woodhouse, Richard Weinberger
Cc: Al Viro, Jan Kara, linux-mtd, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/jffs2/dir.c | 19 ++++++++++---------
fs/jffs2/file.c | 3 ++-
fs/jffs2/fs.c | 10 +++++-----
fs/jffs2/os-linux.h | 2 +-
4 files changed, 18 insertions(+), 16 deletions(-)
diff --git a/fs/jffs2/dir.c b/fs/jffs2/dir.c
index 5075a0a6d594..2ade386a2c02 100644
--- a/fs/jffs2/dir.c
+++ b/fs/jffs2/dir.c
@@ -204,7 +204,7 @@ static int jffs2_create(struct mnt_idmap *idmap, struct inode *dir_i,
if (ret)
goto fail;
- dir_i->i_mtime = dir_i->i_ctime = ITIME(je32_to_cpu(ri->ctime));
+ dir_i->i_mtime = inode_ctime_set(dir_i, ITIME(je32_to_cpu(ri->ctime)));
jffs2_free_raw_inode(ri);
@@ -237,7 +237,7 @@ static int jffs2_unlink(struct inode *dir_i, struct dentry *dentry)
if (dead_f->inocache)
set_nlink(d_inode(dentry), dead_f->inocache->pino_nlink);
if (!ret)
- dir_i->i_mtime = dir_i->i_ctime = ITIME(now);
+ dir_i->i_mtime = inode_ctime_set(dir_i, ITIME(now));
return ret;
}
/***********************************************************************/
@@ -271,7 +271,7 @@ static int jffs2_link (struct dentry *old_dentry, struct inode *dir_i, struct de
set_nlink(d_inode(old_dentry), ++f->inocache->pino_nlink);
mutex_unlock(&f->sem);
d_instantiate(dentry, d_inode(old_dentry));
- dir_i->i_mtime = dir_i->i_ctime = ITIME(now);
+ dir_i->i_mtime = inode_ctime_set(dir_i, ITIME(now));
ihold(d_inode(old_dentry));
}
return ret;
@@ -422,7 +422,7 @@ static int jffs2_symlink (struct mnt_idmap *idmap, struct inode *dir_i,
goto fail;
}
- dir_i->i_mtime = dir_i->i_ctime = ITIME(je32_to_cpu(rd->mctime));
+ dir_i->i_mtime = inode_ctime_set(dir_i, ITIME(je32_to_cpu(rd->mctime)));
jffs2_free_raw_dirent(rd);
@@ -566,7 +566,7 @@ static int jffs2_mkdir (struct mnt_idmap *idmap, struct inode *dir_i,
goto fail;
}
- dir_i->i_mtime = dir_i->i_ctime = ITIME(je32_to_cpu(rd->mctime));
+ dir_i->i_mtime = inode_ctime_set(dir_i, ITIME(je32_to_cpu(rd->mctime)));
inc_nlink(dir_i);
jffs2_free_raw_dirent(rd);
@@ -607,7 +607,7 @@ static int jffs2_rmdir (struct inode *dir_i, struct dentry *dentry)
ret = jffs2_do_unlink(c, dir_f, dentry->d_name.name,
dentry->d_name.len, f, now);
if (!ret) {
- dir_i->i_mtime = dir_i->i_ctime = ITIME(now);
+ dir_i->i_mtime = inode_ctime_set(dir_i, ITIME(now));
clear_nlink(d_inode(dentry));
drop_nlink(dir_i);
}
@@ -743,7 +743,7 @@ static int jffs2_mknod (struct mnt_idmap *idmap, struct inode *dir_i,
goto fail;
}
- dir_i->i_mtime = dir_i->i_ctime = ITIME(je32_to_cpu(rd->mctime));
+ dir_i->i_mtime = inode_ctime_set(dir_i, ITIME(je32_to_cpu(rd->mctime)));
jffs2_free_raw_dirent(rd);
@@ -864,14 +864,15 @@ static int jffs2_rename (struct mnt_idmap *idmap,
* caller won't do it on its own since we are returning an error.
*/
d_invalidate(new_dentry);
- new_dir_i->i_mtime = new_dir_i->i_ctime = ITIME(now);
+ new_dir_i->i_mtime = inode_ctime_set(new_dir_i, ITIME(now));
return ret;
}
if (d_is_dir(old_dentry))
drop_nlink(old_dir_i);
- new_dir_i->i_mtime = new_dir_i->i_ctime = old_dir_i->i_mtime = old_dir_i->i_ctime = ITIME(now);
+ new_dir_i->i_mtime = inode_ctime_set(new_dir_i, ITIME(now));
+ old_dir_i->i_mtime = inode_ctime_set(old_dir_i, ITIME(now));
return 0;
}
diff --git a/fs/jffs2/file.c b/fs/jffs2/file.c
index 2345ca3f09ee..1935e4eb9aa8 100644
--- a/fs/jffs2/file.c
+++ b/fs/jffs2/file.c
@@ -317,7 +317,8 @@ static int jffs2_write_end(struct file *filp, struct address_space *mapping,
inode->i_size = pos + writtenlen;
inode->i_blocks = (inode->i_size + 511) >> 9;
- inode->i_ctime = inode->i_mtime = ITIME(je32_to_cpu(ri->ctime));
+ inode->i_mtime = ITIME(je32_to_cpu(ri->ctime));
+ inode_ctime_set(inode, inode->i_mtime);
}
}
diff --git a/fs/jffs2/fs.c b/fs/jffs2/fs.c
index 038516bee1ab..1d77c4616ca5 100644
--- a/fs/jffs2/fs.c
+++ b/fs/jffs2/fs.c
@@ -115,7 +115,7 @@ int jffs2_do_setattr (struct inode *inode, struct iattr *iattr)
ri->isize = cpu_to_je32((ivalid & ATTR_SIZE)?iattr->ia_size:inode->i_size);
ri->atime = cpu_to_je32(I_SEC((ivalid & ATTR_ATIME)?iattr->ia_atime:inode->i_atime));
ri->mtime = cpu_to_je32(I_SEC((ivalid & ATTR_MTIME)?iattr->ia_mtime:inode->i_mtime));
- ri->ctime = cpu_to_je32(I_SEC((ivalid & ATTR_CTIME)?iattr->ia_ctime:inode->i_ctime));
+ ri->ctime = cpu_to_je32(I_SEC((ivalid & ATTR_CTIME)?iattr->ia_ctime:inode_ctime_peek(inode)));
ri->offset = cpu_to_je32(0);
ri->csize = ri->dsize = cpu_to_je32(mdatalen);
@@ -148,7 +148,7 @@ int jffs2_do_setattr (struct inode *inode, struct iattr *iattr)
}
/* It worked. Update the inode */
inode->i_atime = ITIME(je32_to_cpu(ri->atime));
- inode->i_ctime = ITIME(je32_to_cpu(ri->ctime));
+ inode_ctime_set(inode, ITIME(je32_to_cpu(ri->ctime)));
inode->i_mtime = ITIME(je32_to_cpu(ri->mtime));
inode->i_mode = jemode_to_cpu(ri->mode);
i_uid_write(inode, je16_to_cpu(ri->uid));
@@ -284,7 +284,7 @@ struct inode *jffs2_iget(struct super_block *sb, unsigned long ino)
inode->i_size = je32_to_cpu(latest_node.isize);
inode->i_atime = ITIME(je32_to_cpu(latest_node.atime));
inode->i_mtime = ITIME(je32_to_cpu(latest_node.mtime));
- inode->i_ctime = ITIME(je32_to_cpu(latest_node.ctime));
+ inode_ctime_set(inode, ITIME(je32_to_cpu(latest_node.ctime)));
set_nlink(inode, f->inocache->pino_nlink);
@@ -388,7 +388,7 @@ void jffs2_dirty_inode(struct inode *inode, int flags)
iattr.ia_gid = inode->i_gid;
iattr.ia_atime = inode->i_atime;
iattr.ia_mtime = inode->i_mtime;
- iattr.ia_ctime = inode->i_ctime;
+ iattr.ia_ctime = inode_ctime_peek(inode);
jffs2_do_setattr(inode, &iattr);
}
@@ -475,7 +475,7 @@ struct inode *jffs2_new_inode (struct inode *dir_i, umode_t mode, struct jffs2_r
inode->i_mode = jemode_to_cpu(ri->mode);
i_gid_write(inode, je16_to_cpu(ri->gid));
i_uid_write(inode, je16_to_cpu(ri->uid));
- inode->i_atime = inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
ri->atime = ri->mtime = ri->ctime = cpu_to_je32(I_SEC(inode->i_mtime));
inode->i_blocks = 0;
diff --git a/fs/jffs2/os-linux.h b/fs/jffs2/os-linux.h
index 8da19766c101..2c22fdd4762e 100644
--- a/fs/jffs2/os-linux.h
+++ b/fs/jffs2/os-linux.h
@@ -35,7 +35,7 @@ struct kvec;
#define ITIME(sec) ((struct timespec64){sec, 0})
#define JFFS2_NOW() JFFS2_CLAMP_TIME(ktime_get_real_seconds())
#define I_SEC(tv) JFFS2_CLAMP_TIME((tv).tv_sec)
-#define JFFS2_F_I_CTIME(f) I_SEC(OFNI_EDONI_2SFFJ(f)->i_ctime)
+#define JFFS2_F_I_CTIME(f) I_SEC(inode_ctime_peek(OFNI_EDONI_2SFFJ(f)))
#define JFFS2_F_I_MTIME(f) I_SEC(OFNI_EDONI_2SFFJ(f)->i_mtime)
#define JFFS2_F_I_ATIME(f) I_SEC(OFNI_EDONI_2SFFJ(f)->i_atime)
#define sleep_on_spinunlock(wq, s) \
--
2.41.0
______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 41/79] jffs2: switch to new ctime accessors
@ 2023-06-21 14:45 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, David Woodhouse, Richard Weinberger
Cc: Al Viro, Jan Kara, linux-mtd, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/jffs2/dir.c | 19 ++++++++++---------
fs/jffs2/file.c | 3 ++-
fs/jffs2/fs.c | 10 +++++-----
fs/jffs2/os-linux.h | 2 +-
4 files changed, 18 insertions(+), 16 deletions(-)
diff --git a/fs/jffs2/dir.c b/fs/jffs2/dir.c
index 5075a0a6d594..2ade386a2c02 100644
--- a/fs/jffs2/dir.c
+++ b/fs/jffs2/dir.c
@@ -204,7 +204,7 @@ static int jffs2_create(struct mnt_idmap *idmap, struct inode *dir_i,
if (ret)
goto fail;
- dir_i->i_mtime = dir_i->i_ctime = ITIME(je32_to_cpu(ri->ctime));
+ dir_i->i_mtime = inode_ctime_set(dir_i, ITIME(je32_to_cpu(ri->ctime)));
jffs2_free_raw_inode(ri);
@@ -237,7 +237,7 @@ static int jffs2_unlink(struct inode *dir_i, struct dentry *dentry)
if (dead_f->inocache)
set_nlink(d_inode(dentry), dead_f->inocache->pino_nlink);
if (!ret)
- dir_i->i_mtime = dir_i->i_ctime = ITIME(now);
+ dir_i->i_mtime = inode_ctime_set(dir_i, ITIME(now));
return ret;
}
/***********************************************************************/
@@ -271,7 +271,7 @@ static int jffs2_link (struct dentry *old_dentry, struct inode *dir_i, struct de
set_nlink(d_inode(old_dentry), ++f->inocache->pino_nlink);
mutex_unlock(&f->sem);
d_instantiate(dentry, d_inode(old_dentry));
- dir_i->i_mtime = dir_i->i_ctime = ITIME(now);
+ dir_i->i_mtime = inode_ctime_set(dir_i, ITIME(now));
ihold(d_inode(old_dentry));
}
return ret;
@@ -422,7 +422,7 @@ static int jffs2_symlink (struct mnt_idmap *idmap, struct inode *dir_i,
goto fail;
}
- dir_i->i_mtime = dir_i->i_ctime = ITIME(je32_to_cpu(rd->mctime));
+ dir_i->i_mtime = inode_ctime_set(dir_i, ITIME(je32_to_cpu(rd->mctime)));
jffs2_free_raw_dirent(rd);
@@ -566,7 +566,7 @@ static int jffs2_mkdir (struct mnt_idmap *idmap, struct inode *dir_i,
goto fail;
}
- dir_i->i_mtime = dir_i->i_ctime = ITIME(je32_to_cpu(rd->mctime));
+ dir_i->i_mtime = inode_ctime_set(dir_i, ITIME(je32_to_cpu(rd->mctime)));
inc_nlink(dir_i);
jffs2_free_raw_dirent(rd);
@@ -607,7 +607,7 @@ static int jffs2_rmdir (struct inode *dir_i, struct dentry *dentry)
ret = jffs2_do_unlink(c, dir_f, dentry->d_name.name,
dentry->d_name.len, f, now);
if (!ret) {
- dir_i->i_mtime = dir_i->i_ctime = ITIME(now);
+ dir_i->i_mtime = inode_ctime_set(dir_i, ITIME(now));
clear_nlink(d_inode(dentry));
drop_nlink(dir_i);
}
@@ -743,7 +743,7 @@ static int jffs2_mknod (struct mnt_idmap *idmap, struct inode *dir_i,
goto fail;
}
- dir_i->i_mtime = dir_i->i_ctime = ITIME(je32_to_cpu(rd->mctime));
+ dir_i->i_mtime = inode_ctime_set(dir_i, ITIME(je32_to_cpu(rd->mctime)));
jffs2_free_raw_dirent(rd);
@@ -864,14 +864,15 @@ static int jffs2_rename (struct mnt_idmap *idmap,
* caller won't do it on its own since we are returning an error.
*/
d_invalidate(new_dentry);
- new_dir_i->i_mtime = new_dir_i->i_ctime = ITIME(now);
+ new_dir_i->i_mtime = inode_ctime_set(new_dir_i, ITIME(now));
return ret;
}
if (d_is_dir(old_dentry))
drop_nlink(old_dir_i);
- new_dir_i->i_mtime = new_dir_i->i_ctime = old_dir_i->i_mtime = old_dir_i->i_ctime = ITIME(now);
+ new_dir_i->i_mtime = inode_ctime_set(new_dir_i, ITIME(now));
+ old_dir_i->i_mtime = inode_ctime_set(old_dir_i, ITIME(now));
return 0;
}
diff --git a/fs/jffs2/file.c b/fs/jffs2/file.c
index 2345ca3f09ee..1935e4eb9aa8 100644
--- a/fs/jffs2/file.c
+++ b/fs/jffs2/file.c
@@ -317,7 +317,8 @@ static int jffs2_write_end(struct file *filp, struct address_space *mapping,
inode->i_size = pos + writtenlen;
inode->i_blocks = (inode->i_size + 511) >> 9;
- inode->i_ctime = inode->i_mtime = ITIME(je32_to_cpu(ri->ctime));
+ inode->i_mtime = ITIME(je32_to_cpu(ri->ctime));
+ inode_ctime_set(inode, inode->i_mtime);
}
}
diff --git a/fs/jffs2/fs.c b/fs/jffs2/fs.c
index 038516bee1ab..1d77c4616ca5 100644
--- a/fs/jffs2/fs.c
+++ b/fs/jffs2/fs.c
@@ -115,7 +115,7 @@ int jffs2_do_setattr (struct inode *inode, struct iattr *iattr)
ri->isize = cpu_to_je32((ivalid & ATTR_SIZE)?iattr->ia_size:inode->i_size);
ri->atime = cpu_to_je32(I_SEC((ivalid & ATTR_ATIME)?iattr->ia_atime:inode->i_atime));
ri->mtime = cpu_to_je32(I_SEC((ivalid & ATTR_MTIME)?iattr->ia_mtime:inode->i_mtime));
- ri->ctime = cpu_to_je32(I_SEC((ivalid & ATTR_CTIME)?iattr->ia_ctime:inode->i_ctime));
+ ri->ctime = cpu_to_je32(I_SEC((ivalid & ATTR_CTIME)?iattr->ia_ctime:inode_ctime_peek(inode)));
ri->offset = cpu_to_je32(0);
ri->csize = ri->dsize = cpu_to_je32(mdatalen);
@@ -148,7 +148,7 @@ int jffs2_do_setattr (struct inode *inode, struct iattr *iattr)
}
/* It worked. Update the inode */
inode->i_atime = ITIME(je32_to_cpu(ri->atime));
- inode->i_ctime = ITIME(je32_to_cpu(ri->ctime));
+ inode_ctime_set(inode, ITIME(je32_to_cpu(ri->ctime)));
inode->i_mtime = ITIME(je32_to_cpu(ri->mtime));
inode->i_mode = jemode_to_cpu(ri->mode);
i_uid_write(inode, je16_to_cpu(ri->uid));
@@ -284,7 +284,7 @@ struct inode *jffs2_iget(struct super_block *sb, unsigned long ino)
inode->i_size = je32_to_cpu(latest_node.isize);
inode->i_atime = ITIME(je32_to_cpu(latest_node.atime));
inode->i_mtime = ITIME(je32_to_cpu(latest_node.mtime));
- inode->i_ctime = ITIME(je32_to_cpu(latest_node.ctime));
+ inode_ctime_set(inode, ITIME(je32_to_cpu(latest_node.ctime)));
set_nlink(inode, f->inocache->pino_nlink);
@@ -388,7 +388,7 @@ void jffs2_dirty_inode(struct inode *inode, int flags)
iattr.ia_gid = inode->i_gid;
iattr.ia_atime = inode->i_atime;
iattr.ia_mtime = inode->i_mtime;
- iattr.ia_ctime = inode->i_ctime;
+ iattr.ia_ctime = inode_ctime_peek(inode);
jffs2_do_setattr(inode, &iattr);
}
@@ -475,7 +475,7 @@ struct inode *jffs2_new_inode (struct inode *dir_i, umode_t mode, struct jffs2_r
inode->i_mode = jemode_to_cpu(ri->mode);
i_gid_write(inode, je16_to_cpu(ri->gid));
i_uid_write(inode, je16_to_cpu(ri->uid));
- inode->i_atime = inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
ri->atime = ri->mtime = ri->ctime = cpu_to_je32(I_SEC(inode->i_mtime));
inode->i_blocks = 0;
diff --git a/fs/jffs2/os-linux.h b/fs/jffs2/os-linux.h
index 8da19766c101..2c22fdd4762e 100644
--- a/fs/jffs2/os-linux.h
+++ b/fs/jffs2/os-linux.h
@@ -35,7 +35,7 @@ struct kvec;
#define ITIME(sec) ((struct timespec64){sec, 0})
#define JFFS2_NOW() JFFS2_CLAMP_TIME(ktime_get_real_seconds())
#define I_SEC(tv) JFFS2_CLAMP_TIME((tv).tv_sec)
-#define JFFS2_F_I_CTIME(f) I_SEC(OFNI_EDONI_2SFFJ(f)->i_ctime)
+#define JFFS2_F_I_CTIME(f) I_SEC(inode_ctime_peek(OFNI_EDONI_2SFFJ(f)))
#define JFFS2_F_I_MTIME(f) I_SEC(OFNI_EDONI_2SFFJ(f)->i_mtime)
#define JFFS2_F_I_ATIME(f) I_SEC(OFNI_EDONI_2SFFJ(f)->i_atime)
#define sleep_on_spinunlock(wq, s) \
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 42/79] jfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (39 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
2023-06-21 17:47 ` Dave Kleikamp
-1 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Dave Kleikamp
Cc: Al Viro, Jan Kara, jfs-discussion, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/jfs/acl.c | 2 +-
fs/jfs/inode.c | 2 +-
fs/jfs/ioctl.c | 2 +-
fs/jfs/jfs_imap.c | 8 ++++----
fs/jfs/jfs_inode.c | 4 ++--
fs/jfs/namei.c | 25 +++++++++++++------------
fs/jfs/super.c | 2 +-
fs/jfs/xattr.c | 2 +-
8 files changed, 24 insertions(+), 23 deletions(-)
diff --git a/fs/jfs/acl.c b/fs/jfs/acl.c
index fb96f872d207..3cd6eb00d881 100644
--- a/fs/jfs/acl.c
+++ b/fs/jfs/acl.c
@@ -116,7 +116,7 @@ int jfs_set_acl(struct mnt_idmap *idmap, struct dentry *dentry,
if (!rc) {
if (update_mode) {
inode->i_mode = mode;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
mark_inode_dirty(inode);
}
rc = txCommit(tid, 1, &inode, 0);
diff --git a/fs/jfs/inode.c b/fs/jfs/inode.c
index 8ac10e396050..d9f708fc6439 100644
--- a/fs/jfs/inode.c
+++ b/fs/jfs/inode.c
@@ -393,7 +393,7 @@ void jfs_truncate_nolock(struct inode *ip, loff_t length)
break;
}
- ip->i_mtime = ip->i_ctime = current_time(ip);
+ ip->i_mtime = inode_ctime_set_current(ip);
mark_inode_dirty(ip);
txCommit(tid, 1, &ip, 0);
diff --git a/fs/jfs/ioctl.c b/fs/jfs/ioctl.c
index ed7989bc2db1..db22929938a2 100644
--- a/fs/jfs/ioctl.c
+++ b/fs/jfs/ioctl.c
@@ -96,7 +96,7 @@ int jfs_fileattr_set(struct mnt_idmap *idmap,
jfs_inode->mode2 = flags;
jfs_set_inode_flags(inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
mark_inode_dirty(inode);
return 0;
diff --git a/fs/jfs/jfs_imap.c b/fs/jfs/jfs_imap.c
index 390cbfce391f..f32fb967d360 100644
--- a/fs/jfs/jfs_imap.c
+++ b/fs/jfs/jfs_imap.c
@@ -3064,8 +3064,8 @@ static int copy_from_dinode(struct dinode * dip, struct inode *ip)
ip->i_atime.tv_nsec = le32_to_cpu(dip->di_atime.tv_nsec);
ip->i_mtime.tv_sec = le32_to_cpu(dip->di_mtime.tv_sec);
ip->i_mtime.tv_nsec = le32_to_cpu(dip->di_mtime.tv_nsec);
- ip->i_ctime.tv_sec = le32_to_cpu(dip->di_ctime.tv_sec);
- ip->i_ctime.tv_nsec = le32_to_cpu(dip->di_ctime.tv_nsec);
+ inode_ctime_set_sec(ip, le32_to_cpu(dip->di_ctime.tv_sec));
+ inode_ctime_set_nsec(ip, le32_to_cpu(dip->di_ctime.tv_nsec));
ip->i_blocks = LBLK2PBLK(ip->i_sb, le64_to_cpu(dip->di_nblocks));
ip->i_generation = le32_to_cpu(dip->di_gen);
@@ -3139,8 +3139,8 @@ static void copy_to_dinode(struct dinode * dip, struct inode *ip)
dip->di_atime.tv_sec = cpu_to_le32(ip->i_atime.tv_sec);
dip->di_atime.tv_nsec = cpu_to_le32(ip->i_atime.tv_nsec);
- dip->di_ctime.tv_sec = cpu_to_le32(ip->i_ctime.tv_sec);
- dip->di_ctime.tv_nsec = cpu_to_le32(ip->i_ctime.tv_nsec);
+ dip->di_ctime.tv_sec = cpu_to_le32(inode_ctime_peek(ip).tv_sec);
+ dip->di_ctime.tv_nsec = cpu_to_le32(inode_ctime_peek(ip).tv_nsec);
dip->di_mtime.tv_sec = cpu_to_le32(ip->i_mtime.tv_sec);
dip->di_mtime.tv_nsec = cpu_to_le32(ip->i_mtime.tv_nsec);
dip->di_ixpxd = jfs_ip->ixpxd; /* in-memory pxd's are little-endian */
diff --git a/fs/jfs/jfs_inode.c b/fs/jfs/jfs_inode.c
index 9e1f02767201..5bbae1ff5129 100644
--- a/fs/jfs/jfs_inode.c
+++ b/fs/jfs/jfs_inode.c
@@ -97,8 +97,8 @@ struct inode *ialloc(struct inode *parent, umode_t mode)
jfs_inode->mode2 |= inode->i_mode;
inode->i_blocks = 0;
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
- jfs_inode->otime = inode->i_ctime.tv_sec;
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
+ jfs_inode->otime = inode_ctime_peek(inode).tv_sec;
inode->i_generation = JFS_SBI(sb)->gengen++;
jfs_inode->cflag = 0;
diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c
index 494b9f4043cf..405b4790b916 100644
--- a/fs/jfs/namei.c
+++ b/fs/jfs/namei.c
@@ -149,7 +149,7 @@ static int jfs_create(struct mnt_idmap *idmap, struct inode *dip,
mark_inode_dirty(ip);
- dip->i_ctime = dip->i_mtime = current_time(dip);
+ dip->i_mtime = inode_ctime_set_current(dip);
mark_inode_dirty(dip);
@@ -284,7 +284,7 @@ static int jfs_mkdir(struct mnt_idmap *idmap, struct inode *dip,
/* update parent directory inode */
inc_nlink(dip); /* for '..' from child directory */
- dip->i_ctime = dip->i_mtime = current_time(dip);
+ dip->i_mtime = inode_ctime_set_current(dip);
mark_inode_dirty(dip);
rc = txCommit(tid, 2, &iplist[0], 0);
@@ -390,7 +390,7 @@ static int jfs_rmdir(struct inode *dip, struct dentry *dentry)
/* update parent directory's link count corresponding
* to ".." entry of the target directory deleted
*/
- dip->i_ctime = dip->i_mtime = current_time(dip);
+ dip->i_mtime = inode_ctime_set_current(dip);
inode_dec_link_count(dip);
/*
@@ -512,7 +512,8 @@ static int jfs_unlink(struct inode *dip, struct dentry *dentry)
ASSERT(ip->i_nlink);
- ip->i_ctime = dip->i_ctime = dip->i_mtime = current_time(ip);
+ dip->i_mtime = inode_ctime_set_current(ip);
+ inode_ctime_set(dip, dip->i_mtime);
mark_inode_dirty(dip);
/* update target's inode */
@@ -822,8 +823,8 @@ static int jfs_link(struct dentry *old_dentry,
/* update object inode */
inc_nlink(ip); /* for new link */
- ip->i_ctime = current_time(ip);
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ inode_ctime_set_current(ip);
+ dir->i_mtime = inode_ctime_set_current(dir);
mark_inode_dirty(dir);
ihold(ip);
@@ -1023,7 +1024,7 @@ static int jfs_symlink(struct mnt_idmap *idmap, struct inode *dip,
mark_inode_dirty(ip);
- dip->i_ctime = dip->i_mtime = current_time(dip);
+ dip->i_mtime = inode_ctime_set_current(dip);
mark_inode_dirty(dip);
/*
* commit update of parent directory and link object
@@ -1200,7 +1201,7 @@ static int jfs_rename(struct mnt_idmap *idmap, struct inode *old_dir,
tblk->xflag |= COMMIT_DELETE;
tblk->u.ip = new_ip;
} else {
- new_ip->i_ctime = current_time(new_ip);
+ inode_ctime_set_current(new_ip);
mark_inode_dirty(new_ip);
}
} else {
@@ -1263,10 +1264,10 @@ static int jfs_rename(struct mnt_idmap *idmap, struct inode *old_dir,
/*
* Update ctime on changed/moved inodes & mark dirty
*/
- old_ip->i_ctime = current_time(old_ip);
+ inode_ctime_set_current(old_ip);
mark_inode_dirty(old_ip);
- new_dir->i_ctime = new_dir->i_mtime = current_time(new_dir);
+ new_dir->i_mtime = inode_ctime_set_current(new_dir);
mark_inode_dirty(new_dir);
/* Build list of inodes modified by this transaction */
@@ -1278,7 +1279,7 @@ static int jfs_rename(struct mnt_idmap *idmap, struct inode *old_dir,
if (old_dir != new_dir) {
iplist[ipcount++] = new_dir;
- old_dir->i_ctime = old_dir->i_mtime = current_time(old_dir);
+ old_dir->i_mtime = inode_ctime_set_current(old_dir);
mark_inode_dirty(old_dir);
}
@@ -1411,7 +1412,7 @@ static int jfs_mknod(struct mnt_idmap *idmap, struct inode *dir,
mark_inode_dirty(ip);
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
mark_inode_dirty(dir);
diff --git a/fs/jfs/super.c b/fs/jfs/super.c
index d2f82cb7db1b..c810b98254dd 100644
--- a/fs/jfs/super.c
+++ b/fs/jfs/super.c
@@ -818,7 +818,7 @@ static ssize_t jfs_quota_write(struct super_block *sb, int type,
}
if (inode->i_size < off+len-towrite)
i_size_write(inode, off+len-towrite);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
mark_inode_dirty(inode);
inode_unlock(inode);
return len - towrite;
diff --git a/fs/jfs/xattr.c b/fs/jfs/xattr.c
index 931e50018f88..71f005bbe486 100644
--- a/fs/jfs/xattr.c
+++ b/fs/jfs/xattr.c
@@ -647,7 +647,7 @@ static int ea_put(tid_t tid, struct inode *inode, struct ea_buffer *ea_buf,
if (old_blocks)
dquot_free_block(inode, old_blocks);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
return 0;
}
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 43/79] kernfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (40 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
2023-06-21 15:30 ` Greg Kroah-Hartman
-1 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Greg Kroah-Hartman, Tejun Heo
Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/kernfs/inode.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/kernfs/inode.c b/fs/kernfs/inode.c
index b22b74d1a115..781093c8acdd 100644
--- a/fs/kernfs/inode.c
+++ b/fs/kernfs/inode.c
@@ -152,7 +152,7 @@ static inline void set_default_inode_attr(struct inode *inode, umode_t mode)
{
inode->i_mode = mode;
inode->i_atime = inode->i_mtime =
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
}
static inline void set_inode_attr(struct inode *inode,
@@ -162,7 +162,7 @@ static inline void set_inode_attr(struct inode *inode,
inode->i_gid = attrs->ia_gid;
inode->i_atime = attrs->ia_atime;
inode->i_mtime = attrs->ia_mtime;
- inode->i_ctime = attrs->ia_ctime;
+ inode_ctime_set(inode, attrs->ia_ctime);
}
static void kernfs_refresh_inode(struct kernfs_node *kn, struct inode *inode)
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 44/79] minix: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (41 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner; +Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/minix/bitmap.c | 2 +-
fs/minix/dir.c | 6 +++---
fs/minix/inode.c | 11 ++++++-----
fs/minix/itree_common.c | 4 ++--
fs/minix/namei.c | 6 +++---
5 files changed, 15 insertions(+), 14 deletions(-)
diff --git a/fs/minix/bitmap.c b/fs/minix/bitmap.c
index 870207ba23f1..9d41dce3b8db 100644
--- a/fs/minix/bitmap.c
+++ b/fs/minix/bitmap.c
@@ -251,7 +251,7 @@ struct inode *minix_new_inode(const struct inode *dir, umode_t mode)
}
inode_init_owner(&nop_mnt_idmap, inode, dir, mode);
inode->i_ino = j;
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
inode->i_blocks = 0;
memset(&minix_i(inode)->u, 0, sizeof(minix_i(inode)->u));
insert_inode_hash(inode);
diff --git a/fs/minix/dir.c b/fs/minix/dir.c
index bf9858f76b6a..03d556635ffc 100644
--- a/fs/minix/dir.c
+++ b/fs/minix/dir.c
@@ -281,7 +281,7 @@ int minix_add_link(struct dentry *dentry, struct inode *inode)
de->inode = inode->i_ino;
}
dir_commit_chunk(page, pos, sbi->s_dirsize);
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
mark_inode_dirty(dir);
err = minix_handle_dirsync(dir);
out_put:
@@ -313,7 +313,7 @@ int minix_delete_entry(struct minix_dir_entry *de, struct page *page)
else
de->inode = 0;
dir_commit_chunk(page, pos, len);
- inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
mark_inode_dirty(inode);
return minix_handle_dirsync(inode);
}
@@ -436,7 +436,7 @@ int minix_set_link(struct minix_dir_entry *de, struct page *page,
else
de->inode = inode->i_ino;
dir_commit_chunk(page, pos, sbi->s_dirsize);
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
mark_inode_dirty(dir);
return minix_handle_dirsync(dir);
}
diff --git a/fs/minix/inode.c b/fs/minix/inode.c
index e9fbb5303a22..e9c45ecf6ec2 100644
--- a/fs/minix/inode.c
+++ b/fs/minix/inode.c
@@ -501,10 +501,11 @@ static struct inode *V1_minix_iget(struct inode *inode)
i_gid_write(inode, raw_inode->i_gid);
set_nlink(inode, raw_inode->i_nlinks);
inode->i_size = raw_inode->i_size;
- inode->i_mtime.tv_sec = inode->i_atime.tv_sec = inode->i_ctime.tv_sec = raw_inode->i_time;
+ inode->i_mtime.tv_sec = inode->i_atime.tv_sec = inode_ctime_set_sec(inode,
+ raw_inode->i_time);
inode->i_mtime.tv_nsec = 0;
inode->i_atime.tv_nsec = 0;
- inode->i_ctime.tv_nsec = 0;
+ inode_ctime_set_nsec(inode, 0);
inode->i_blocks = 0;
for (i = 0; i < 9; i++)
minix_inode->u.i1_data[i] = raw_inode->i_zone[i];
@@ -543,10 +544,10 @@ static struct inode *V2_minix_iget(struct inode *inode)
inode->i_size = raw_inode->i_size;
inode->i_mtime.tv_sec = raw_inode->i_mtime;
inode->i_atime.tv_sec = raw_inode->i_atime;
- inode->i_ctime.tv_sec = raw_inode->i_ctime;
+ inode_ctime_set_sec(inode, raw_inode->i_ctime);
inode->i_mtime.tv_nsec = 0;
inode->i_atime.tv_nsec = 0;
- inode->i_ctime.tv_nsec = 0;
+ inode_ctime_set_nsec(inode, 0);
inode->i_blocks = 0;
for (i = 0; i < 10; i++)
minix_inode->u.i2_data[i] = raw_inode->i_zone[i];
@@ -622,7 +623,7 @@ static struct buffer_head * V2_minix_update_inode(struct inode * inode)
raw_inode->i_size = inode->i_size;
raw_inode->i_mtime = inode->i_mtime.tv_sec;
raw_inode->i_atime = inode->i_atime.tv_sec;
- raw_inode->i_ctime = inode->i_ctime.tv_sec;
+ raw_inode->i_ctime = inode_ctime_peek(inode).tv_sec;
if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode))
raw_inode->i_zone[0] = old_encode_dev(inode->i_rdev);
else for (i = 0; i < 10; i++)
diff --git a/fs/minix/itree_common.c b/fs/minix/itree_common.c
index 446148792f41..78e48a6c78a2 100644
--- a/fs/minix/itree_common.c
+++ b/fs/minix/itree_common.c
@@ -131,7 +131,7 @@ static inline int splice_branch(struct inode *inode,
/* We are done with atomic stuff, now do the rest of housekeeping */
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
/* had we spliced it onto indirect block? */
if (where->bh)
@@ -350,7 +350,7 @@ static inline void truncate (struct inode * inode)
}
first_whole++;
}
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
mark_inode_dirty(inode);
}
diff --git a/fs/minix/namei.c b/fs/minix/namei.c
index 956d5183828d..f55ea33486c9 100644
--- a/fs/minix/namei.c
+++ b/fs/minix/namei.c
@@ -98,7 +98,7 @@ static int minix_link(struct dentry * old_dentry, struct inode * dir,
{
struct inode *inode = d_inode(old_dentry);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
inode_inc_link_count(inode);
ihold(inode);
return add_nondir(dentry, inode);
@@ -154,7 +154,7 @@ static int minix_unlink(struct inode * dir, struct dentry *dentry)
if (err)
return err;
- inode->i_ctime = dir->i_ctime;
+ inode_ctime_set(inode, inode_ctime_peek(dir));
inode_dec_link_count(inode);
return 0;
}
@@ -218,7 +218,7 @@ static int minix_rename(struct mnt_idmap *idmap,
put_page(new_page);
if (err)
goto out_dir;
- new_inode->i_ctime = current_time(new_inode);
+ inode_ctime_set_current(new_inode);
if (dir_de)
drop_nlink(new_inode);
inode_dec_link_count(new_inode);
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 45/79] nfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (42 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Trond Myklebust, Anna Schumaker
Cc: Al Viro, Jan Kara, linux-nfs, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/nfs/callback_proc.c | 2 +-
fs/nfs/fscache.h | 4 ++--
fs/nfs/inode.c | 21 +++++++++++----------
3 files changed, 14 insertions(+), 13 deletions(-)
diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c
index c1eda73254e1..995654050815 100644
--- a/fs/nfs/callback_proc.c
+++ b/fs/nfs/callback_proc.c
@@ -59,7 +59,7 @@ __be32 nfs4_callback_getattr(void *argp, void *resp,
res->change_attr = delegation->change_attr;
if (nfs_have_writebacks(inode))
res->change_attr++;
- res->ctime = inode->i_ctime;
+ res->ctime = inode_ctime_peek(inode);
res->mtime = inode->i_mtime;
res->bitmap[0] = (FATTR4_WORD0_CHANGE|FATTR4_WORD0_SIZE) &
args->bitmap[0];
diff --git a/fs/nfs/fscache.h b/fs/nfs/fscache.h
index e1706e736c64..98be316c64a0 100644
--- a/fs/nfs/fscache.h
+++ b/fs/nfs/fscache.h
@@ -116,8 +116,8 @@ static inline void nfs_fscache_update_auxdata(struct nfs_fscache_inode_auxdata *
memset(auxdata, 0, sizeof(*auxdata));
auxdata->mtime_sec = inode->i_mtime.tv_sec;
auxdata->mtime_nsec = inode->i_mtime.tv_nsec;
- auxdata->ctime_sec = inode->i_ctime.tv_sec;
- auxdata->ctime_nsec = inode->i_ctime.tv_nsec;
+ auxdata->ctime_sec = inode_ctime_peek(inode).tv_sec;
+ auxdata->ctime_nsec = inode_ctime_peek(inode).tv_nsec;
if (NFS_SERVER(inode)->nfs_client->rpc_ops->version == 4)
auxdata->change_attr = inode_peek_iversion_raw(inode);
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index a910b9a638c5..bc4cac08bb24 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -514,7 +514,8 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr)
memset(&inode->i_atime, 0, sizeof(inode->i_atime));
memset(&inode->i_mtime, 0, sizeof(inode->i_mtime));
- memset(&inode->i_ctime, 0, sizeof(inode->i_ctime));
+ inode_ctime_set_sec(inode, 0);
+ inode_ctime_set_nsec(inode, 0);
inode_set_iversion_raw(inode, 0);
inode->i_size = 0;
clear_nlink(inode);
@@ -535,7 +536,7 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr)
else if (fattr_supported & NFS_ATTR_FATTR_MTIME)
nfs_set_cache_invalid(inode, NFS_INO_INVALID_MTIME);
if (fattr->valid & NFS_ATTR_FATTR_CTIME)
- inode->i_ctime = fattr->ctime;
+ inode_ctime_set(inode, fattr->ctime);
else if (fattr_supported & NFS_ATTR_FATTR_CTIME)
nfs_set_cache_invalid(inode, NFS_INO_INVALID_CTIME);
if (fattr->valid & NFS_ATTR_FATTR_CHANGE)
@@ -731,7 +732,7 @@ void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr,
if ((attr->ia_valid & ATTR_GID) != 0)
inode->i_gid = attr->ia_gid;
if (fattr->valid & NFS_ATTR_FATTR_CTIME)
- inode->i_ctime = fattr->ctime;
+ inode_ctime_set(inode, fattr->ctime);
else
nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE
| NFS_INO_INVALID_CTIME);
@@ -749,7 +750,7 @@ void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr,
nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATIME);
if (fattr->valid & NFS_ATTR_FATTR_CTIME)
- inode->i_ctime = fattr->ctime;
+ inode_ctime_set(inode, fattr->ctime);
else
nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE
| NFS_INO_INVALID_CTIME);
@@ -765,7 +766,7 @@ void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr,
nfs_set_cache_invalid(inode, NFS_INO_INVALID_MTIME);
if (fattr->valid & NFS_ATTR_FATTR_CTIME)
- inode->i_ctime = fattr->ctime;
+ inode_ctime_set(inode, fattr->ctime);
else
nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE
| NFS_INO_INVALID_CTIME);
@@ -1444,11 +1445,11 @@ static void nfs_wcc_update_inode(struct inode *inode, struct nfs_fattr *fattr)
nfs_set_cache_invalid(inode, NFS_INO_INVALID_XATTR);
}
/* If we have atomic WCC data, we may update some attributes */
- ts = inode->i_ctime;
+ ts = inode_ctime_peek(inode);
if ((fattr->valid & NFS_ATTR_FATTR_PRECTIME)
&& (fattr->valid & NFS_ATTR_FATTR_CTIME)
&& timespec64_equal(&ts, &fattr->pre_ctime)) {
- inode->i_ctime = fattr->ctime;
+ inode_ctime_set(inode, fattr->ctime);
}
ts = inode->i_mtime;
@@ -1510,7 +1511,7 @@ static int nfs_check_inode_attributes(struct inode *inode, struct nfs_fattr *fat
if ((fattr->valid & NFS_ATTR_FATTR_MTIME) && !timespec64_equal(&ts, &fattr->mtime))
invalid |= NFS_INO_INVALID_MTIME;
- ts = inode->i_ctime;
+ ts = inode_ctime_peek(inode);
if ((fattr->valid & NFS_ATTR_FATTR_CTIME) && !timespec64_equal(&ts, &fattr->ctime))
invalid |= NFS_INO_INVALID_CTIME;
@@ -1997,7 +1998,7 @@ int nfs_post_op_update_inode_force_wcc_locked(struct inode *inode, struct nfs_fa
}
if ((fattr->valid & NFS_ATTR_FATTR_CTIME) != 0 &&
(fattr->valid & NFS_ATTR_FATTR_PRECTIME) == 0) {
- fattr->pre_ctime = inode->i_ctime;
+ fattr->pre_ctime = inode_ctime_peek(inode);
fattr->valid |= NFS_ATTR_FATTR_PRECTIME;
}
if ((fattr->valid & NFS_ATTR_FATTR_MTIME) != 0 &&
@@ -2190,7 +2191,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
save_cache_validity & NFS_INO_INVALID_MTIME;
if (fattr->valid & NFS_ATTR_FATTR_CTIME)
- inode->i_ctime = fattr->ctime;
+ inode_ctime_set(inode, fattr->ctime);
else if (fattr_supported & NFS_ATTR_FATTR_CTIME)
nfsi->cache_validity |=
save_cache_validity & NFS_INO_INVALID_CTIME;
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 46/79] nfsd: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (43 preceding siblings ...)
(?)
@ 2023-06-21 14:45 ` Jeff Layton
2023-06-21 17:36 ` Chuck Lever
-1 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:45 UTC (permalink / raw)
To: Christian Brauner, Chuck Lever, Neil Brown, Olga Kornievskaia,
Dai Ngo, Tom Talpey
Cc: Al Viro, Jan Kara, linux-nfs, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/nfsd/nfsctl.c | 2 +-
fs/nfsd/vfs.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index 1b8b1aab9a15..bebb65d3ad6b 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -1131,7 +1131,7 @@ static struct inode *nfsd_get_inode(struct super_block *sb, umode_t mode)
/* Following advice from simple_fill_super documentation: */
inode->i_ino = iunique(sb, NFSD_MaxReserved);
inode->i_mode = mode;
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
switch (mode & S_IFMT) {
case S_IFDIR:
inode->i_fop = &simple_dir_operations;
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index 59b7d60ae33e..febdfdf16ec1 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -520,7 +520,7 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp,
nfsd_sanitize_attrs(inode, iap);
- if (check_guard && guardtime != inode->i_ctime.tv_sec)
+ if (check_guard && guardtime != inode_ctime_peek(inode).tv_sec)
return nfserr_notsync;
/*
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 47/79] nilfs2: switch to new ctime accessors
@ 2023-06-21 14:46 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Ryusuke Konishi
Cc: Al Viro, Jan Kara, linux-nilfs, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/nilfs2/dir.c | 6 +++---
fs/nilfs2/inode.c | 12 ++++++------
fs/nilfs2/ioctl.c | 2 +-
fs/nilfs2/namei.c | 8 ++++----
4 files changed, 14 insertions(+), 14 deletions(-)
diff --git a/fs/nilfs2/dir.c b/fs/nilfs2/dir.c
index decd6471300b..a51d13a95651 100644
--- a/fs/nilfs2/dir.c
+++ b/fs/nilfs2/dir.c
@@ -429,7 +429,7 @@ void nilfs_set_link(struct inode *dir, struct nilfs_dir_entry *de,
nilfs_set_de_type(de, inode);
nilfs_commit_chunk(page, mapping, from, to);
nilfs_put_page(page);
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
}
/*
@@ -519,7 +519,7 @@ int nilfs_add_link(struct dentry *dentry, struct inode *inode)
de->inode = cpu_to_le64(inode->i_ino);
nilfs_set_de_type(de, inode);
nilfs_commit_chunk(page, page->mapping, from, to);
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
nilfs_mark_inode_dirty(dir);
/* OFFSET_CACHE */
out_put:
@@ -567,7 +567,7 @@ int nilfs_delete_entry(struct nilfs_dir_entry *dir, struct page *page)
pde->rec_len = nilfs_rec_len_to_disk(to - from);
dir->inode = 0;
nilfs_commit_chunk(page, mapping, from, to);
- inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
out:
nilfs_put_page(page);
return err;
diff --git a/fs/nilfs2/inode.c b/fs/nilfs2/inode.c
index a8ce522ac747..dee831f7426f 100644
--- a/fs/nilfs2/inode.c
+++ b/fs/nilfs2/inode.c
@@ -366,7 +366,7 @@ struct inode *nilfs_new_inode(struct inode *dir, umode_t mode)
atomic64_inc(&root->inodes_count);
inode_init_owner(&nop_mnt_idmap, inode, dir, mode);
inode->i_ino = ino;
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
if (S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode)) {
err = nilfs_bmap_read(ii->i_bmap, NULL);
@@ -450,10 +450,10 @@ int nilfs_read_inode_common(struct inode *inode,
set_nlink(inode, le16_to_cpu(raw_inode->i_links_count));
inode->i_size = le64_to_cpu(raw_inode->i_size);
inode->i_atime.tv_sec = le64_to_cpu(raw_inode->i_mtime);
- inode->i_ctime.tv_sec = le64_to_cpu(raw_inode->i_ctime);
+ inode_ctime_set_sec(inode, le64_to_cpu(raw_inode->i_ctime));
inode->i_mtime.tv_sec = le64_to_cpu(raw_inode->i_mtime);
inode->i_atime.tv_nsec = le32_to_cpu(raw_inode->i_mtime_nsec);
- inode->i_ctime.tv_nsec = le32_to_cpu(raw_inode->i_ctime_nsec);
+ inode_ctime_set_nsec(inode, le32_to_cpu(raw_inode->i_ctime_nsec));
inode->i_mtime.tv_nsec = le32_to_cpu(raw_inode->i_mtime_nsec);
if (nilfs_is_metadata_file_inode(inode) && !S_ISREG(inode->i_mode))
return -EIO; /* this inode is for metadata and corrupted */
@@ -768,9 +768,9 @@ void nilfs_write_inode_common(struct inode *inode,
raw_inode->i_gid = cpu_to_le32(i_gid_read(inode));
raw_inode->i_links_count = cpu_to_le16(inode->i_nlink);
raw_inode->i_size = cpu_to_le64(inode->i_size);
- raw_inode->i_ctime = cpu_to_le64(inode->i_ctime.tv_sec);
+ raw_inode->i_ctime = cpu_to_le64(inode_ctime_peek(inode).tv_sec);
raw_inode->i_mtime = cpu_to_le64(inode->i_mtime.tv_sec);
- raw_inode->i_ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
+ raw_inode->i_ctime_nsec = cpu_to_le32(inode_ctime_peek(inode).tv_nsec);
raw_inode->i_mtime_nsec = cpu_to_le32(inode->i_mtime.tv_nsec);
raw_inode->i_blocks = cpu_to_le64(inode->i_blocks);
@@ -875,7 +875,7 @@ void nilfs_truncate(struct inode *inode)
nilfs_truncate_bmap(ii, blkoff);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
if (IS_SYNC(inode))
nilfs_set_transaction_flag(NILFS_TI_SYNC);
diff --git a/fs/nilfs2/ioctl.c b/fs/nilfs2/ioctl.c
index 1dfbc0c34513..811fab46e277 100644
--- a/fs/nilfs2/ioctl.c
+++ b/fs/nilfs2/ioctl.c
@@ -149,7 +149,7 @@ int nilfs_fileattr_set(struct mnt_idmap *idmap,
NILFS_I(inode)->i_flags = oldflags | (flags & FS_FL_USER_MODIFIABLE);
nilfs_set_inode_flags(inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
if (IS_SYNC(inode))
nilfs_set_transaction_flag(NILFS_TI_SYNC);
diff --git a/fs/nilfs2/namei.c b/fs/nilfs2/namei.c
index c7024da8f1e2..251f6021c3db 100644
--- a/fs/nilfs2/namei.c
+++ b/fs/nilfs2/namei.c
@@ -185,7 +185,7 @@ static int nilfs_link(struct dentry *old_dentry, struct inode *dir,
if (err)
return err;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
inode_inc_link_count(inode);
ihold(inode);
@@ -283,7 +283,7 @@ static int nilfs_do_unlink(struct inode *dir, struct dentry *dentry)
if (err)
goto out;
- inode->i_ctime = dir->i_ctime;
+ inode_ctime_set(inode, inode_ctime_peek(dir));
drop_nlink(inode);
err = 0;
out:
@@ -387,7 +387,7 @@ static int nilfs_rename(struct mnt_idmap *idmap,
goto out_dir;
nilfs_set_link(new_dir, new_de, new_page, old_inode);
nilfs_mark_inode_dirty(new_dir);
- new_inode->i_ctime = current_time(new_inode);
+ inode_ctime_set_current(new_inode);
if (dir_de)
drop_nlink(new_inode);
drop_nlink(new_inode);
@@ -406,7 +406,7 @@ static int nilfs_rename(struct mnt_idmap *idmap,
* Like most other Unix systems, set the ctime for inodes on a
* rename.
*/
- old_inode->i_ctime = current_time(old_inode);
+ inode_ctime_set_current(old_inode);
nilfs_delete_entry(old_de, old_page);
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 47/79] nilfs2: switch to new ctime accessors
@ 2023-06-21 14:46 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Ryusuke Konishi
Cc: Al Viro, Jan Kara, linux-nilfs-u79uwXL29TY76Z2rM5mHXA,
linux-kernel-u79uwXL29TY76Z2rM5mHXA
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
---
fs/nilfs2/dir.c | 6 +++---
fs/nilfs2/inode.c | 12 ++++++------
fs/nilfs2/ioctl.c | 2 +-
fs/nilfs2/namei.c | 8 ++++----
4 files changed, 14 insertions(+), 14 deletions(-)
diff --git a/fs/nilfs2/dir.c b/fs/nilfs2/dir.c
index decd6471300b..a51d13a95651 100644
--- a/fs/nilfs2/dir.c
+++ b/fs/nilfs2/dir.c
@@ -429,7 +429,7 @@ void nilfs_set_link(struct inode *dir, struct nilfs_dir_entry *de,
nilfs_set_de_type(de, inode);
nilfs_commit_chunk(page, mapping, from, to);
nilfs_put_page(page);
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
}
/*
@@ -519,7 +519,7 @@ int nilfs_add_link(struct dentry *dentry, struct inode *inode)
de->inode = cpu_to_le64(inode->i_ino);
nilfs_set_de_type(de, inode);
nilfs_commit_chunk(page, page->mapping, from, to);
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
nilfs_mark_inode_dirty(dir);
/* OFFSET_CACHE */
out_put:
@@ -567,7 +567,7 @@ int nilfs_delete_entry(struct nilfs_dir_entry *dir, struct page *page)
pde->rec_len = nilfs_rec_len_to_disk(to - from);
dir->inode = 0;
nilfs_commit_chunk(page, mapping, from, to);
- inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
out:
nilfs_put_page(page);
return err;
diff --git a/fs/nilfs2/inode.c b/fs/nilfs2/inode.c
index a8ce522ac747..dee831f7426f 100644
--- a/fs/nilfs2/inode.c
+++ b/fs/nilfs2/inode.c
@@ -366,7 +366,7 @@ struct inode *nilfs_new_inode(struct inode *dir, umode_t mode)
atomic64_inc(&root->inodes_count);
inode_init_owner(&nop_mnt_idmap, inode, dir, mode);
inode->i_ino = ino;
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
if (S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode)) {
err = nilfs_bmap_read(ii->i_bmap, NULL);
@@ -450,10 +450,10 @@ int nilfs_read_inode_common(struct inode *inode,
set_nlink(inode, le16_to_cpu(raw_inode->i_links_count));
inode->i_size = le64_to_cpu(raw_inode->i_size);
inode->i_atime.tv_sec = le64_to_cpu(raw_inode->i_mtime);
- inode->i_ctime.tv_sec = le64_to_cpu(raw_inode->i_ctime);
+ inode_ctime_set_sec(inode, le64_to_cpu(raw_inode->i_ctime));
inode->i_mtime.tv_sec = le64_to_cpu(raw_inode->i_mtime);
inode->i_atime.tv_nsec = le32_to_cpu(raw_inode->i_mtime_nsec);
- inode->i_ctime.tv_nsec = le32_to_cpu(raw_inode->i_ctime_nsec);
+ inode_ctime_set_nsec(inode, le32_to_cpu(raw_inode->i_ctime_nsec));
inode->i_mtime.tv_nsec = le32_to_cpu(raw_inode->i_mtime_nsec);
if (nilfs_is_metadata_file_inode(inode) && !S_ISREG(inode->i_mode))
return -EIO; /* this inode is for metadata and corrupted */
@@ -768,9 +768,9 @@ void nilfs_write_inode_common(struct inode *inode,
raw_inode->i_gid = cpu_to_le32(i_gid_read(inode));
raw_inode->i_links_count = cpu_to_le16(inode->i_nlink);
raw_inode->i_size = cpu_to_le64(inode->i_size);
- raw_inode->i_ctime = cpu_to_le64(inode->i_ctime.tv_sec);
+ raw_inode->i_ctime = cpu_to_le64(inode_ctime_peek(inode).tv_sec);
raw_inode->i_mtime = cpu_to_le64(inode->i_mtime.tv_sec);
- raw_inode->i_ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
+ raw_inode->i_ctime_nsec = cpu_to_le32(inode_ctime_peek(inode).tv_nsec);
raw_inode->i_mtime_nsec = cpu_to_le32(inode->i_mtime.tv_nsec);
raw_inode->i_blocks = cpu_to_le64(inode->i_blocks);
@@ -875,7 +875,7 @@ void nilfs_truncate(struct inode *inode)
nilfs_truncate_bmap(ii, blkoff);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
if (IS_SYNC(inode))
nilfs_set_transaction_flag(NILFS_TI_SYNC);
diff --git a/fs/nilfs2/ioctl.c b/fs/nilfs2/ioctl.c
index 1dfbc0c34513..811fab46e277 100644
--- a/fs/nilfs2/ioctl.c
+++ b/fs/nilfs2/ioctl.c
@@ -149,7 +149,7 @@ int nilfs_fileattr_set(struct mnt_idmap *idmap,
NILFS_I(inode)->i_flags = oldflags | (flags & FS_FL_USER_MODIFIABLE);
nilfs_set_inode_flags(inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
if (IS_SYNC(inode))
nilfs_set_transaction_flag(NILFS_TI_SYNC);
diff --git a/fs/nilfs2/namei.c b/fs/nilfs2/namei.c
index c7024da8f1e2..251f6021c3db 100644
--- a/fs/nilfs2/namei.c
+++ b/fs/nilfs2/namei.c
@@ -185,7 +185,7 @@ static int nilfs_link(struct dentry *old_dentry, struct inode *dir,
if (err)
return err;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
inode_inc_link_count(inode);
ihold(inode);
@@ -283,7 +283,7 @@ static int nilfs_do_unlink(struct inode *dir, struct dentry *dentry)
if (err)
goto out;
- inode->i_ctime = dir->i_ctime;
+ inode_ctime_set(inode, inode_ctime_peek(dir));
drop_nlink(inode);
err = 0;
out:
@@ -387,7 +387,7 @@ static int nilfs_rename(struct mnt_idmap *idmap,
goto out_dir;
nilfs_set_link(new_dir, new_de, new_page, old_inode);
nilfs_mark_inode_dirty(new_dir);
- new_inode->i_ctime = current_time(new_inode);
+ inode_ctime_set_current(new_inode);
if (dir_de)
drop_nlink(new_inode);
drop_nlink(new_inode);
@@ -406,7 +406,7 @@ static int nilfs_rename(struct mnt_idmap *idmap,
* Like most other Unix systems, set the ctime for inodes on a
* rename.
*/
- old_inode->i_ctime = current_time(old_inode);
+ inode_ctime_set_current(old_inode);
nilfs_delete_entry(old_de, old_page);
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 48/79] ntfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (45 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Anton Altaparmakov, Namjae Jeon
Cc: Al Viro, Jan Kara, linux-ntfs-dev, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/ntfs/inode.c | 15 ++++++++-------
fs/ntfs/mft.c | 3 +--
2 files changed, 9 insertions(+), 9 deletions(-)
diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c
index 6c3f38d66579..285142c22275 100644
--- a/fs/ntfs/inode.c
+++ b/fs/ntfs/inode.c
@@ -654,7 +654,7 @@ static int ntfs_read_locked_inode(struct inode *vi)
* always changes, when mtime is changed. ctime can be changed on its
* own, mtime is then not changed, e.g. when a file is renamed.
*/
- vi->i_ctime = ntfs2utc(si->last_mft_change_time);
+ inode_ctime_set(vi, ntfs2utc(si->last_mft_change_time));
/*
* Last access to the data within the file. Not changed during a rename
* for example but changed whenever the file is written to.
@@ -1218,7 +1218,7 @@ static int ntfs_read_locked_attr_inode(struct inode *base_vi, struct inode *vi)
vi->i_gid = base_vi->i_gid;
set_nlink(vi, base_vi->i_nlink);
vi->i_mtime = base_vi->i_mtime;
- vi->i_ctime = base_vi->i_ctime;
+ inode_ctime_set(vi, inode_ctime_peek(base_vi));
vi->i_atime = base_vi->i_atime;
vi->i_generation = ni->seq_no = base_ni->seq_no;
@@ -1484,7 +1484,7 @@ static int ntfs_read_locked_index_inode(struct inode *base_vi, struct inode *vi)
vi->i_gid = base_vi->i_gid;
set_nlink(vi, base_vi->i_nlink);
vi->i_mtime = base_vi->i_mtime;
- vi->i_ctime = base_vi->i_ctime;
+ inode_ctime_set(vi, inode_ctime_peek(base_vi));
vi->i_atime = base_vi->i_atime;
vi->i_generation = ni->seq_no = base_ni->seq_no;
/* Set inode type to zero but preserve permissions. */
@@ -2804,13 +2804,14 @@ int ntfs_truncate(struct inode *vi)
*/
if (!IS_NOCMTIME(VFS_I(base_ni)) && !IS_RDONLY(VFS_I(base_ni))) {
struct timespec64 now = current_time(VFS_I(base_ni));
+ struct timespec64 ctime = inode_ctime_peek(VFS_I(base_ni));
int sync_it = 0;
if (!timespec64_equal(&VFS_I(base_ni)->i_mtime, &now) ||
- !timespec64_equal(&VFS_I(base_ni)->i_ctime, &now))
+ !timespec64_equal(&ctime, &now))
sync_it = 1;
VFS_I(base_ni)->i_mtime = now;
- VFS_I(base_ni)->i_ctime = now;
+ inode_ctime_set(VFS_I(base_ni), now);
if (sync_it)
mark_inode_dirty_sync(VFS_I(base_ni));
@@ -2928,7 +2929,7 @@ int ntfs_setattr(struct mnt_idmap *idmap, struct dentry *dentry,
if (ia_valid & ATTR_MTIME)
vi->i_mtime = attr->ia_mtime;
if (ia_valid & ATTR_CTIME)
- vi->i_ctime = attr->ia_ctime;
+ inode_ctime_set(vi, attr->ia_ctime);
mark_inode_dirty(vi);
out:
return err;
@@ -3004,7 +3005,7 @@ int __ntfs_write_inode(struct inode *vi, int sync)
si->last_data_change_time = nt;
modified = true;
}
- nt = utc2ntfs(vi->i_ctime);
+ nt = utc2ntfs(inode_ctime_peek(vi));
if (si->last_mft_change_time != nt) {
ntfs_debug("Updating ctime for inode 0x%lx: old = 0x%llx, "
"new = 0x%llx", vi->i_ino, (long long)
diff --git a/fs/ntfs/mft.c b/fs/ntfs/mft.c
index 0155f106ec34..68821e312ed2 100644
--- a/fs/ntfs/mft.c
+++ b/fs/ntfs/mft.c
@@ -2682,8 +2682,7 @@ ntfs_inode *ntfs_mft_record_alloc(ntfs_volume *vol, const int mode,
vi->i_mode &= ~S_IWUGO;
/* Set the inode times to the current time. */
- vi->i_atime = vi->i_mtime = vi->i_ctime =
- current_time(vi);
+ vi->i_atime = vi->i_mtime = inode_ctime_set_current(vi);
/*
* Set the file size to 0, the ntfs inode sizes are set to 0 by
* the call to ntfs_init_big_inode() below.
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 49/79] ntfs3: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (46 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Konstantin Komarov
Cc: Al Viro, Jan Kara, ntfs3, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/ntfs3/file.c | 6 +++---
fs/ntfs3/frecord.c | 4 +++-
fs/ntfs3/inode.c | 14 ++++++++------
fs/ntfs3/namei.c | 10 +++++-----
fs/ntfs3/xattr.c | 4 ++--
5 files changed, 21 insertions(+), 17 deletions(-)
diff --git a/fs/ntfs3/file.c b/fs/ntfs3/file.c
index 1d6c824246c4..032452b3e242 100644
--- a/fs/ntfs3/file.c
+++ b/fs/ntfs3/file.c
@@ -342,7 +342,7 @@ static int ntfs_extend(struct inode *inode, loff_t pos, size_t count,
err = 0;
}
- inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
mark_inode_dirty(inode);
if (IS_SYNC(inode)) {
@@ -400,7 +400,7 @@ static int ntfs_truncate(struct inode *inode, loff_t new_size)
ni_unlock(ni);
ni->std_fa |= FILE_ATTRIBUTE_ARCHIVE;
- inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
if (!IS_DIRSYNC(inode)) {
dirty = 1;
} else {
@@ -642,7 +642,7 @@ static long ntfs_fallocate(struct file *file, int mode, loff_t vbo, loff_t len)
filemap_invalidate_unlock(mapping);
if (!err) {
- inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
mark_inode_dirty(inode);
}
diff --git a/fs/ntfs3/frecord.c b/fs/ntfs3/frecord.c
index 16bd9faa2d28..4712474d74a7 100644
--- a/fs/ntfs3/frecord.c
+++ b/fs/ntfs3/frecord.c
@@ -3264,6 +3264,7 @@ int ni_write_inode(struct inode *inode, int sync, const char *hint)
if (is_rec_inuse(ni->mi.mrec) &&
!(sbi->flags & NTFS_FLAGS_LOG_REPLAYING) && inode->i_nlink) {
+ struct timespec64 ctime;
bool modified = false;
/* Update times in standard attribute. */
@@ -3280,7 +3281,8 @@ int ni_write_inode(struct inode *inode, int sync, const char *hint)
modified = true;
}
- dup.c_time = kernel2nt(&inode->i_ctime);
+ ctime = inode_ctime_peek(inode);
+ dup.c_time = kernel2nt(&ctime);
if (std->c_time != dup.c_time) {
std->c_time = dup.c_time;
modified = true;
diff --git a/fs/ntfs3/inode.c b/fs/ntfs3/inode.c
index dc7e7ab701c6..ac1726fc4e16 100644
--- a/fs/ntfs3/inode.c
+++ b/fs/ntfs3/inode.c
@@ -44,6 +44,7 @@ static struct inode *ntfs_read_mft(struct inode *inode,
u64 t64;
struct MFT_REC *rec;
struct runs_tree *run;
+ struct timespec64 ctime;
inode->i_op = NULL;
/* Setup 'uid' and 'gid' */
@@ -169,7 +170,8 @@ static struct inode *ntfs_read_mft(struct inode *inode,
nt2kernel(std5->cr_time, &ni->i_crtime);
#endif
nt2kernel(std5->a_time, &inode->i_atime);
- nt2kernel(std5->c_time, &inode->i_ctime);
+ ctime = inode_ctime_peek(inode);
+ nt2kernel(std5->c_time, &ctime);
nt2kernel(std5->m_time, &inode->i_mtime);
ni->std_fa = std5->fa;
@@ -958,7 +960,7 @@ int ntfs_write_end(struct file *file, struct address_space *mapping, loff_t pos,
if (err >= 0) {
if (!(ni->std_fa & FILE_ATTRIBUTE_ARCHIVE)) {
- inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
ni->std_fa |= FILE_ATTRIBUTE_ARCHIVE;
dirty = true;
}
@@ -1658,8 +1660,8 @@ struct inode *ntfs_create_inode(struct mnt_idmap *idmap, struct inode *dir,
d_instantiate(dentry, inode);
/* Set original time. inode times (i_ctime) may be changed in ntfs_init_acl. */
- inode->i_atime = inode->i_mtime = inode->i_ctime = dir->i_mtime =
- dir->i_ctime = ni->i_crtime;
+ inode->i_atime = inode->i_mtime = inode_ctime_set(inode, ni->i_crtime);
+ dir->i_mtime = inode_ctime_set(dir, ni->i_crtime);
mark_inode_dirty(dir);
mark_inode_dirty(inode);
@@ -1765,9 +1767,9 @@ int ntfs_unlink_inode(struct inode *dir, const struct dentry *dentry)
if (!err) {
drop_nlink(inode);
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
mark_inode_dirty(dir);
- inode->i_ctime = dir->i_ctime;
+ inode_ctime_set(inode, inode_ctime_peek(dir));
if (inode->i_nlink)
mark_inode_dirty(inode);
} else if (!ni_remove_name_undo(dir_ni, ni, de, de2, undo_remove)) {
diff --git a/fs/ntfs3/namei.c b/fs/ntfs3/namei.c
index 70f8c859e0ad..30e2341533fa 100644
--- a/fs/ntfs3/namei.c
+++ b/fs/ntfs3/namei.c
@@ -156,8 +156,8 @@ static int ntfs_link(struct dentry *ode, struct inode *dir, struct dentry *de)
err = ntfs_link_inode(inode, de);
if (!err) {
- dir->i_ctime = dir->i_mtime = inode->i_ctime =
- current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
+ inode_ctime_set_current(inode);
mark_inode_dirty(inode);
mark_inode_dirty(dir);
d_instantiate(de, inode);
@@ -324,12 +324,12 @@ static int ntfs_rename(struct mnt_idmap *idmap, struct inode *dir,
/* Restore after failed rename failed too. */
_ntfs_bad_inode(inode);
} else if (!err) {
- inode->i_ctime = dir->i_ctime = dir->i_mtime =
- current_time(dir);
+ inode_ctime_set_current(inode);
+ dir->i_mtime = inode_ctime_set_current(dir);
mark_inode_dirty(inode);
mark_inode_dirty(dir);
if (dir != new_dir) {
- new_dir->i_mtime = new_dir->i_ctime = dir->i_ctime;
+ new_dir->i_mtime = inode_ctime_set_current(new_dir);
mark_inode_dirty(new_dir);
}
diff --git a/fs/ntfs3/xattr.c b/fs/ntfs3/xattr.c
index 023f314e8950..fd05965e2a06 100644
--- a/fs/ntfs3/xattr.c
+++ b/fs/ntfs3/xattr.c
@@ -637,7 +637,7 @@ static noinline int ntfs_set_acl_ex(struct mnt_idmap *idmap,
if (!err) {
set_cached_acl(inode, type, acl);
inode->i_mode = mode;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
mark_inode_dirty(inode);
}
@@ -924,7 +924,7 @@ static noinline int ntfs_setxattr(const struct xattr_handler *handler,
NULL);
out:
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
mark_inode_dirty(inode);
return err;
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [Ocfs2-devel] [PATCH 50/79] ocfs2: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
@ 2023-06-21 14:46 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton via Ocfs2-devel @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Mark Fasheh, Joel Becker, Joseph Qi
Cc: Jan Kara, Al Viro, ocfs2-devel, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/ocfs2/acl.c | 6 +++---
fs/ocfs2/alloc.c | 6 +++---
fs/ocfs2/aops.c | 2 +-
fs/ocfs2/dir.c | 8 ++++----
fs/ocfs2/dlmfs/dlmfs.c | 4 ++--
fs/ocfs2/dlmglue.c | 10 ++++++----
fs/ocfs2/file.c | 16 +++++++++-------
fs/ocfs2/inode.c | 14 ++++++++------
fs/ocfs2/move_extents.c | 6 +++---
fs/ocfs2/namei.c | 22 ++++++++++++----------
fs/ocfs2/refcounttree.c | 14 +++++++-------
fs/ocfs2/xattr.c | 6 +++---
12 files changed, 61 insertions(+), 53 deletions(-)
diff --git a/fs/ocfs2/acl.c b/fs/ocfs2/acl.c
index 9fd03eaf15f8..ad03021412b4 100644
--- a/fs/ocfs2/acl.c
+++ b/fs/ocfs2/acl.c
@@ -191,10 +191,10 @@ static int ocfs2_acl_set_mode(struct inode *inode, struct buffer_head *di_bh,
}
inode->i_mode = new_mode;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
di->i_mode = cpu_to_le16(inode->i_mode);
- di->i_ctime = cpu_to_le64(inode->i_ctime.tv_sec);
- di->i_ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
+ di->i_ctime = cpu_to_le64(inode_ctime_peek(inode).tv_sec);
+ di->i_ctime_nsec = cpu_to_le32(inode_ctime_peek(inode).tv_nsec);
ocfs2_update_inode_fsync_trans(handle, inode, 0);
ocfs2_journal_dirty(handle, di_bh);
diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c
index 51c93929a146..0455b5e0d659 100644
--- a/fs/ocfs2/alloc.c
+++ b/fs/ocfs2/alloc.c
@@ -7436,10 +7436,10 @@ int ocfs2_truncate_inline(struct inode *inode, struct buffer_head *di_bh,
}
inode->i_blocks = ocfs2_inode_sector_count(inode);
- inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
- di->i_ctime = di->i_mtime = cpu_to_le64(inode->i_ctime.tv_sec);
- di->i_ctime_nsec = di->i_mtime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
+ di->i_ctime = di->i_mtime = cpu_to_le64(inode_ctime_peek(inode).tv_sec);
+ di->i_ctime_nsec = di->i_mtime_nsec = cpu_to_le32(inode_ctime_peek(inode).tv_nsec);
ocfs2_update_inode_fsync_trans(handle, inode, 1);
ocfs2_journal_dirty(handle, di_bh);
diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
index 8dfc284e85f0..873fad14046d 100644
--- a/fs/ocfs2/aops.c
+++ b/fs/ocfs2/aops.c
@@ -2048,7 +2048,7 @@ int ocfs2_write_end_nolock(struct address_space *mapping,
}
inode->i_blocks = ocfs2_inode_sector_count(inode);
di->i_size = cpu_to_le64((u64)i_size_read(inode));
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
di->i_mtime = di->i_ctime = cpu_to_le64(inode->i_mtime.tv_sec);
di->i_mtime_nsec = di->i_ctime_nsec = cpu_to_le32(inode->i_mtime.tv_nsec);
if (handle)
diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c
index 694471fc46b8..06fffb5c0fc7 100644
--- a/fs/ocfs2/dir.c
+++ b/fs/ocfs2/dir.c
@@ -1658,7 +1658,7 @@ int __ocfs2_add_entry(handle_t *handle,
offset, ocfs2_dir_trailer_blk_off(dir->i_sb));
if (ocfs2_dirent_would_fit(de, rec_len)) {
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
retval = ocfs2_mark_inode_dirty(handle, dir, parent_fe_bh);
if (retval < 0) {
mlog_errno(retval);
@@ -2962,11 +2962,11 @@ static int ocfs2_expand_inline_dir(struct inode *dir, struct buffer_head *di_bh,
ocfs2_dinode_new_extent_list(dir, di);
i_size_write(dir, sb->s_blocksize);
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
di->i_size = cpu_to_le64(sb->s_blocksize);
- di->i_ctime = di->i_mtime = cpu_to_le64(dir->i_ctime.tv_sec);
- di->i_ctime_nsec = di->i_mtime_nsec = cpu_to_le32(dir->i_ctime.tv_nsec);
+ di->i_ctime = di->i_mtime = cpu_to_le64(inode_ctime_peek(dir).tv_sec);
+ di->i_ctime_nsec = di->i_mtime_nsec = cpu_to_le32(inode_ctime_peek(dir).tv_nsec);
ocfs2_update_inode_fsync_trans(handle, dir, 1);
/*
diff --git a/fs/ocfs2/dlmfs/dlmfs.c b/fs/ocfs2/dlmfs/dlmfs.c
index ba26c5567cff..8b5a161d827c 100644
--- a/fs/ocfs2/dlmfs/dlmfs.c
+++ b/fs/ocfs2/dlmfs/dlmfs.c
@@ -337,7 +337,7 @@ static struct inode *dlmfs_get_root_inode(struct super_block *sb)
if (inode) {
inode->i_ino = get_next_ino();
inode_init_owner(&nop_mnt_idmap, inode, NULL, mode);
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
inc_nlink(inode);
inode->i_fop = &simple_dir_operations;
@@ -360,7 +360,7 @@ static struct inode *dlmfs_get_inode(struct inode *parent,
inode->i_ino = get_next_ino();
inode_init_owner(&nop_mnt_idmap, inode, parent, mode);
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
ip = DLMFS_I(inode);
ip->ip_conn = DLMFS_I(parent)->ip_conn;
diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c
index c28bc983a7b1..0d5679955657 100644
--- a/fs/ocfs2/dlmglue.c
+++ b/fs/ocfs2/dlmglue.c
@@ -2162,6 +2162,7 @@ static void __ocfs2_stuff_meta_lvb(struct inode *inode)
struct ocfs2_inode_info *oi = OCFS2_I(inode);
struct ocfs2_lock_res *lockres = &oi->ip_inode_lockres;
struct ocfs2_meta_lvb *lvb;
+ struct timespec64 ctime;
lvb = ocfs2_dlm_lvb(&lockres->l_lksb);
@@ -2184,8 +2185,8 @@ static void __ocfs2_stuff_meta_lvb(struct inode *inode)
lvb->lvb_inlink = cpu_to_be16(inode->i_nlink);
lvb->lvb_iatime_packed =
cpu_to_be64(ocfs2_pack_timespec(&inode->i_atime));
- lvb->lvb_ictime_packed =
- cpu_to_be64(ocfs2_pack_timespec(&inode->i_ctime));
+ ctime = inode_ctime_peek(inode);
+ lvb->lvb_ictime_packed = cpu_to_be64(ocfs2_pack_timespec(&ctime));
lvb->lvb_imtime_packed =
cpu_to_be64(ocfs2_pack_timespec(&inode->i_mtime));
lvb->lvb_iattr = cpu_to_be32(oi->ip_attr);
@@ -2208,6 +2209,7 @@ static int ocfs2_refresh_inode_from_lvb(struct inode *inode)
struct ocfs2_inode_info *oi = OCFS2_I(inode);
struct ocfs2_lock_res *lockres = &oi->ip_inode_lockres;
struct ocfs2_meta_lvb *lvb;
+ struct timespec64 ctime;
mlog_meta_lvb(0, lockres);
@@ -2238,8 +2240,8 @@ static int ocfs2_refresh_inode_from_lvb(struct inode *inode)
be64_to_cpu(lvb->lvb_iatime_packed));
ocfs2_unpack_timespec(&inode->i_mtime,
be64_to_cpu(lvb->lvb_imtime_packed));
- ocfs2_unpack_timespec(&inode->i_ctime,
- be64_to_cpu(lvb->lvb_ictime_packed));
+ ctime = inode_ctime_peek(inode);
+ ocfs2_unpack_timespec(&ctime, be64_to_cpu(lvb->lvb_ictime_packed));
spin_unlock(&oi->ip_lock);
return 0;
}
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index 91a194596552..52f126ebd0c2 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -232,8 +232,10 @@ int ocfs2_should_update_atime(struct inode *inode,
return 0;
if (vfsmnt->mnt_flags & MNT_RELATIME) {
+ struct timespec64 ctime = inode_ctime_peek(inode);
+
if ((timespec64_compare(&inode->i_atime, &inode->i_mtime) <= 0) ||
- (timespec64_compare(&inode->i_atime, &inode->i_ctime) <= 0))
+ (timespec64_compare(&inode->i_atime, &ctime) <= 0))
return 1;
return 0;
@@ -294,7 +296,7 @@ int ocfs2_set_inode_size(handle_t *handle,
i_size_write(inode, new_i_size);
inode->i_blocks = ocfs2_inode_sector_count(inode);
- inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
status = ocfs2_mark_inode_dirty(handle, inode, fe_bh);
if (status < 0) {
@@ -415,12 +417,12 @@ static int ocfs2_orphan_for_truncate(struct ocfs2_super *osb,
}
i_size_write(inode, new_i_size);
- inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
di = (struct ocfs2_dinode *) fe_bh->b_data;
di->i_size = cpu_to_le64(new_i_size);
- di->i_ctime = di->i_mtime = cpu_to_le64(inode->i_ctime.tv_sec);
- di->i_ctime_nsec = di->i_mtime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
+ di->i_ctime = di->i_mtime = cpu_to_le64(inode_ctime_peek(inode).tv_sec);
+ di->i_ctime_nsec = di->i_mtime_nsec = cpu_to_le32(inode_ctime_peek(inode).tv_nsec);
ocfs2_update_inode_fsync_trans(handle, inode, 0);
ocfs2_journal_dirty(handle, fe_bh);
@@ -824,7 +826,7 @@ static int ocfs2_write_zero_page(struct inode *inode, u64 abs_from,
i_size_write(inode, abs_to);
inode->i_blocks = ocfs2_inode_sector_count(inode);
di->i_size = cpu_to_le64((u64)i_size_read(inode));
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
di->i_mtime = di->i_ctime = cpu_to_le64(inode->i_mtime.tv_sec);
di->i_ctime_nsec = cpu_to_le32(inode->i_mtime.tv_nsec);
di->i_mtime_nsec = di->i_ctime_nsec;
@@ -2043,7 +2045,7 @@ static int __ocfs2_change_file_space(struct file *file, struct inode *inode,
goto out_inode_unlock;
}
- inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
ret = ocfs2_mark_inode_dirty(handle, inode, di_bh);
if (ret < 0)
mlog_errno(ret);
diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c
index bb116c39b581..1627948574fc 100644
--- a/fs/ocfs2/inode.c
+++ b/fs/ocfs2/inode.c
@@ -263,6 +263,7 @@ void ocfs2_populate_inode(struct inode *inode, struct ocfs2_dinode *fe,
struct super_block *sb;
struct ocfs2_super *osb;
int use_plocks = 1;
+ struct timespec64 ts;
sb = inode->i_sb;
osb = OCFS2_SB(sb);
@@ -306,8 +307,9 @@ void ocfs2_populate_inode(struct inode *inode, struct ocfs2_dinode *fe,
inode->i_atime.tv_nsec = le32_to_cpu(fe->i_atime_nsec);
inode->i_mtime.tv_sec = le64_to_cpu(fe->i_mtime);
inode->i_mtime.tv_nsec = le32_to_cpu(fe->i_mtime_nsec);
- inode->i_ctime.tv_sec = le64_to_cpu(fe->i_ctime);
- inode->i_ctime.tv_nsec = le32_to_cpu(fe->i_ctime_nsec);
+ ts.tv_sec = le64_to_cpu(fe->i_ctime);
+ ts.tv_nsec = le32_to_cpu(fe->i_ctime_nsec);
+ inode_ctime_set(inode, ts);
if (OCFS2_I(inode)->ip_blkno != le64_to_cpu(fe->i_blkno))
mlog(ML_ERROR,
@@ -1314,8 +1316,8 @@ int ocfs2_mark_inode_dirty(handle_t *handle,
fe->i_mode = cpu_to_le16(inode->i_mode);
fe->i_atime = cpu_to_le64(inode->i_atime.tv_sec);
fe->i_atime_nsec = cpu_to_le32(inode->i_atime.tv_nsec);
- fe->i_ctime = cpu_to_le64(inode->i_ctime.tv_sec);
- fe->i_ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
+ fe->i_ctime = cpu_to_le64(inode_ctime_peek(inode).tv_sec);
+ fe->i_ctime_nsec = cpu_to_le32(inode_ctime_peek(inode).tv_nsec);
fe->i_mtime = cpu_to_le64(inode->i_mtime.tv_sec);
fe->i_mtime_nsec = cpu_to_le32(inode->i_mtime.tv_nsec);
@@ -1352,8 +1354,8 @@ void ocfs2_refresh_inode(struct inode *inode,
inode->i_atime.tv_nsec = le32_to_cpu(fe->i_atime_nsec);
inode->i_mtime.tv_sec = le64_to_cpu(fe->i_mtime);
inode->i_mtime.tv_nsec = le32_to_cpu(fe->i_mtime_nsec);
- inode->i_ctime.tv_sec = le64_to_cpu(fe->i_ctime);
- inode->i_ctime.tv_nsec = le32_to_cpu(fe->i_ctime_nsec);
+ inode_ctime_set_sec(inode, le64_to_cpu(fe->i_ctime));
+ inode_ctime_set_nsec(inode, le32_to_cpu(fe->i_ctime_nsec));
spin_unlock(&OCFS2_I(inode)->ip_lock);
}
diff --git a/fs/ocfs2/move_extents.c b/fs/ocfs2/move_extents.c
index b1e32ec4a9d4..f8543a3509ee 100644
--- a/fs/ocfs2/move_extents.c
+++ b/fs/ocfs2/move_extents.c
@@ -950,9 +950,9 @@ static int ocfs2_move_extents(struct ocfs2_move_extents_context *context)
}
di = (struct ocfs2_dinode *)di_bh->b_data;
- inode->i_ctime = current_time(inode);
- di->i_ctime = cpu_to_le64(inode->i_ctime.tv_sec);
- di->i_ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
+ inode_ctime_set_current(inode);
+ di->i_ctime = cpu_to_le64(inode_ctime_peek(inode).tv_sec);
+ di->i_ctime_nsec = cpu_to_le32(inode_ctime_peek(inode).tv_nsec);
ocfs2_update_inode_fsync_trans(handle, inode, 0);
ocfs2_journal_dirty(handle, di_bh);
diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c
index 17c52225b87d..3e0c6e2c2669 100644
--- a/fs/ocfs2/namei.c
+++ b/fs/ocfs2/namei.c
@@ -690,6 +690,7 @@ static int ocfs2_link(struct dentry *old_dentry,
struct ocfs2_dinode *fe = NULL;
struct ocfs2_super *osb = OCFS2_SB(dir->i_sb);
struct ocfs2_dir_lookup_result lookup = { NULL, };
+ struct timespec64 ts;
sigset_t oldset;
u64 old_de_ino;
@@ -793,10 +794,10 @@ static int ocfs2_link(struct dentry *old_dentry,
}
inc_nlink(inode);
- inode->i_ctime = current_time(inode);
+ ts = inode_ctime_set_current(inode);
ocfs2_set_links_count(fe, inode->i_nlink);
- fe->i_ctime = cpu_to_le64(inode->i_ctime.tv_sec);
- fe->i_ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
+ fe->i_ctime = cpu_to_le64(ts.tv_sec);
+ fe->i_ctime_nsec = cpu_to_le32(ts.tv_nsec);
ocfs2_journal_dirty(handle, fe_bh);
err = ocfs2_add_entry(handle, dentry, inode,
@@ -995,7 +996,7 @@ static int ocfs2_unlink(struct inode *dir,
ocfs2_set_links_count(fe, inode->i_nlink);
ocfs2_journal_dirty(handle, fe_bh);
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
if (S_ISDIR(inode->i_mode))
drop_nlink(dir);
@@ -1537,7 +1538,7 @@ static int ocfs2_rename(struct mnt_idmap *idmap,
new_dir_bh, &target_insert);
}
- old_inode->i_ctime = current_time(old_inode);
+ inode_ctime_set_current(old_inode);
mark_inode_dirty(old_inode);
status = ocfs2_journal_access_di(handle, INODE_CACHE(old_inode),
@@ -1546,8 +1547,8 @@ static int ocfs2_rename(struct mnt_idmap *idmap,
if (status >= 0) {
old_di = (struct ocfs2_dinode *) old_inode_bh->b_data;
- old_di->i_ctime = cpu_to_le64(old_inode->i_ctime.tv_sec);
- old_di->i_ctime_nsec = cpu_to_le32(old_inode->i_ctime.tv_nsec);
+ old_di->i_ctime = cpu_to_le64(inode_ctime_peek(old_inode).tv_sec);
+ old_di->i_ctime_nsec = cpu_to_le32(inode_ctime_peek(old_inode).tv_nsec);
ocfs2_journal_dirty(handle, old_inode_bh);
} else
mlog_errno(status);
@@ -1586,9 +1587,9 @@ static int ocfs2_rename(struct mnt_idmap *idmap,
if (new_inode) {
drop_nlink(new_inode);
- new_inode->i_ctime = current_time(new_inode);
+ inode_ctime_set_current(new_inode);
}
- old_dir->i_ctime = old_dir->i_mtime = current_time(old_dir);
+ old_dir->i_mtime = inode_ctime_set_current(old_dir);
if (update_dot_dot) {
status = ocfs2_update_entry(old_inode, handle,
@@ -1610,7 +1611,8 @@ static int ocfs2_rename(struct mnt_idmap *idmap,
if (old_dir != new_dir) {
/* Keep the same times on both directories.*/
- new_dir->i_ctime = new_dir->i_mtime = old_dir->i_ctime;
+ new_dir->i_mtime = inode_ctime_peek(old_dir);
+ inode_ctime_set(new_dir, new_dir->i_mtime);
/*
* This will also pick up the i_nlink change from the
diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c
index 564ab48d03ef..8416c624782f 100644
--- a/fs/ocfs2/refcounttree.c
+++ b/fs/ocfs2/refcounttree.c
@@ -3750,9 +3750,9 @@ static int ocfs2_change_ctime(struct inode *inode,
goto out_commit;
}
- inode->i_ctime = current_time(inode);
- di->i_ctime = cpu_to_le64(inode->i_ctime.tv_sec);
- di->i_ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
+ inode_ctime_set_current(inode);
+ di->i_ctime = cpu_to_le64(inode_ctime_peek(inode).tv_sec);
+ di->i_ctime_nsec = cpu_to_le32(inode_ctime_peek(inode).tv_nsec);
ocfs2_journal_dirty(handle, di_bh);
@@ -4073,10 +4073,10 @@ static int ocfs2_complete_reflink(struct inode *s_inode,
* we want mtime to appear identical to the source and
* update ctime.
*/
- t_inode->i_ctime = current_time(t_inode);
+ inode_ctime_set_current(t_inode);
- di->i_ctime = cpu_to_le64(t_inode->i_ctime.tv_sec);
- di->i_ctime_nsec = cpu_to_le32(t_inode->i_ctime.tv_nsec);
+ di->i_ctime = cpu_to_le64(inode_ctime_peek(t_inode).tv_sec);
+ di->i_ctime_nsec = cpu_to_le32(inode_ctime_peek(t_inode).tv_nsec);
t_inode->i_mtime = s_inode->i_mtime;
di->i_mtime = s_di->i_mtime;
@@ -4456,7 +4456,7 @@ int ocfs2_reflink_update_dest(struct inode *dest,
if (newlen > i_size_read(dest))
i_size_write(dest, newlen);
spin_unlock(&OCFS2_I(dest)->ip_lock);
- dest->i_ctime = dest->i_mtime = current_time(dest);
+ dest->i_mtime = inode_ctime_set_current(dest);
ret = ocfs2_mark_inode_dirty(handle, dest, d_bh);
if (ret) {
diff --git a/fs/ocfs2/xattr.c b/fs/ocfs2/xattr.c
index 4ac77ff6e676..a137e9e05fa4 100644
--- a/fs/ocfs2/xattr.c
+++ b/fs/ocfs2/xattr.c
@@ -3421,9 +3421,9 @@ static int __ocfs2_xattr_set_handle(struct inode *inode,
goto out;
}
- inode->i_ctime = current_time(inode);
- di->i_ctime = cpu_to_le64(inode->i_ctime.tv_sec);
- di->i_ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
+ inode_ctime_set_current(inode);
+ di->i_ctime = cpu_to_le64(inode_ctime_peek(inode).tv_sec);
+ di->i_ctime_nsec = cpu_to_le32(inode_ctime_peek(inode).tv_nsec);
ocfs2_journal_dirty(ctxt->handle, xis->inode_bh);
}
out:
--
2.41.0
_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 50/79] ocfs2: switch to new ctime accessors
@ 2023-06-21 14:46 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Mark Fasheh, Joel Becker, Joseph Qi
Cc: Al Viro, Jan Kara, ocfs2-devel, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/ocfs2/acl.c | 6 +++---
fs/ocfs2/alloc.c | 6 +++---
fs/ocfs2/aops.c | 2 +-
fs/ocfs2/dir.c | 8 ++++----
fs/ocfs2/dlmfs/dlmfs.c | 4 ++--
fs/ocfs2/dlmglue.c | 10 ++++++----
fs/ocfs2/file.c | 16 +++++++++-------
fs/ocfs2/inode.c | 14 ++++++++------
fs/ocfs2/move_extents.c | 6 +++---
fs/ocfs2/namei.c | 22 ++++++++++++----------
fs/ocfs2/refcounttree.c | 14 +++++++-------
fs/ocfs2/xattr.c | 6 +++---
12 files changed, 61 insertions(+), 53 deletions(-)
diff --git a/fs/ocfs2/acl.c b/fs/ocfs2/acl.c
index 9fd03eaf15f8..ad03021412b4 100644
--- a/fs/ocfs2/acl.c
+++ b/fs/ocfs2/acl.c
@@ -191,10 +191,10 @@ static int ocfs2_acl_set_mode(struct inode *inode, struct buffer_head *di_bh,
}
inode->i_mode = new_mode;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
di->i_mode = cpu_to_le16(inode->i_mode);
- di->i_ctime = cpu_to_le64(inode->i_ctime.tv_sec);
- di->i_ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
+ di->i_ctime = cpu_to_le64(inode_ctime_peek(inode).tv_sec);
+ di->i_ctime_nsec = cpu_to_le32(inode_ctime_peek(inode).tv_nsec);
ocfs2_update_inode_fsync_trans(handle, inode, 0);
ocfs2_journal_dirty(handle, di_bh);
diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c
index 51c93929a146..0455b5e0d659 100644
--- a/fs/ocfs2/alloc.c
+++ b/fs/ocfs2/alloc.c
@@ -7436,10 +7436,10 @@ int ocfs2_truncate_inline(struct inode *inode, struct buffer_head *di_bh,
}
inode->i_blocks = ocfs2_inode_sector_count(inode);
- inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
- di->i_ctime = di->i_mtime = cpu_to_le64(inode->i_ctime.tv_sec);
- di->i_ctime_nsec = di->i_mtime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
+ di->i_ctime = di->i_mtime = cpu_to_le64(inode_ctime_peek(inode).tv_sec);
+ di->i_ctime_nsec = di->i_mtime_nsec = cpu_to_le32(inode_ctime_peek(inode).tv_nsec);
ocfs2_update_inode_fsync_trans(handle, inode, 1);
ocfs2_journal_dirty(handle, di_bh);
diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
index 8dfc284e85f0..873fad14046d 100644
--- a/fs/ocfs2/aops.c
+++ b/fs/ocfs2/aops.c
@@ -2048,7 +2048,7 @@ int ocfs2_write_end_nolock(struct address_space *mapping,
}
inode->i_blocks = ocfs2_inode_sector_count(inode);
di->i_size = cpu_to_le64((u64)i_size_read(inode));
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
di->i_mtime = di->i_ctime = cpu_to_le64(inode->i_mtime.tv_sec);
di->i_mtime_nsec = di->i_ctime_nsec = cpu_to_le32(inode->i_mtime.tv_nsec);
if (handle)
diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c
index 694471fc46b8..06fffb5c0fc7 100644
--- a/fs/ocfs2/dir.c
+++ b/fs/ocfs2/dir.c
@@ -1658,7 +1658,7 @@ int __ocfs2_add_entry(handle_t *handle,
offset, ocfs2_dir_trailer_blk_off(dir->i_sb));
if (ocfs2_dirent_would_fit(de, rec_len)) {
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
retval = ocfs2_mark_inode_dirty(handle, dir, parent_fe_bh);
if (retval < 0) {
mlog_errno(retval);
@@ -2962,11 +2962,11 @@ static int ocfs2_expand_inline_dir(struct inode *dir, struct buffer_head *di_bh,
ocfs2_dinode_new_extent_list(dir, di);
i_size_write(dir, sb->s_blocksize);
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
di->i_size = cpu_to_le64(sb->s_blocksize);
- di->i_ctime = di->i_mtime = cpu_to_le64(dir->i_ctime.tv_sec);
- di->i_ctime_nsec = di->i_mtime_nsec = cpu_to_le32(dir->i_ctime.tv_nsec);
+ di->i_ctime = di->i_mtime = cpu_to_le64(inode_ctime_peek(dir).tv_sec);
+ di->i_ctime_nsec = di->i_mtime_nsec = cpu_to_le32(inode_ctime_peek(dir).tv_nsec);
ocfs2_update_inode_fsync_trans(handle, dir, 1);
/*
diff --git a/fs/ocfs2/dlmfs/dlmfs.c b/fs/ocfs2/dlmfs/dlmfs.c
index ba26c5567cff..8b5a161d827c 100644
--- a/fs/ocfs2/dlmfs/dlmfs.c
+++ b/fs/ocfs2/dlmfs/dlmfs.c
@@ -337,7 +337,7 @@ static struct inode *dlmfs_get_root_inode(struct super_block *sb)
if (inode) {
inode->i_ino = get_next_ino();
inode_init_owner(&nop_mnt_idmap, inode, NULL, mode);
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
inc_nlink(inode);
inode->i_fop = &simple_dir_operations;
@@ -360,7 +360,7 @@ static struct inode *dlmfs_get_inode(struct inode *parent,
inode->i_ino = get_next_ino();
inode_init_owner(&nop_mnt_idmap, inode, parent, mode);
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
ip = DLMFS_I(inode);
ip->ip_conn = DLMFS_I(parent)->ip_conn;
diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c
index c28bc983a7b1..0d5679955657 100644
--- a/fs/ocfs2/dlmglue.c
+++ b/fs/ocfs2/dlmglue.c
@@ -2162,6 +2162,7 @@ static void __ocfs2_stuff_meta_lvb(struct inode *inode)
struct ocfs2_inode_info *oi = OCFS2_I(inode);
struct ocfs2_lock_res *lockres = &oi->ip_inode_lockres;
struct ocfs2_meta_lvb *lvb;
+ struct timespec64 ctime;
lvb = ocfs2_dlm_lvb(&lockres->l_lksb);
@@ -2184,8 +2185,8 @@ static void __ocfs2_stuff_meta_lvb(struct inode *inode)
lvb->lvb_inlink = cpu_to_be16(inode->i_nlink);
lvb->lvb_iatime_packed =
cpu_to_be64(ocfs2_pack_timespec(&inode->i_atime));
- lvb->lvb_ictime_packed =
- cpu_to_be64(ocfs2_pack_timespec(&inode->i_ctime));
+ ctime = inode_ctime_peek(inode);
+ lvb->lvb_ictime_packed = cpu_to_be64(ocfs2_pack_timespec(&ctime));
lvb->lvb_imtime_packed =
cpu_to_be64(ocfs2_pack_timespec(&inode->i_mtime));
lvb->lvb_iattr = cpu_to_be32(oi->ip_attr);
@@ -2208,6 +2209,7 @@ static int ocfs2_refresh_inode_from_lvb(struct inode *inode)
struct ocfs2_inode_info *oi = OCFS2_I(inode);
struct ocfs2_lock_res *lockres = &oi->ip_inode_lockres;
struct ocfs2_meta_lvb *lvb;
+ struct timespec64 ctime;
mlog_meta_lvb(0, lockres);
@@ -2238,8 +2240,8 @@ static int ocfs2_refresh_inode_from_lvb(struct inode *inode)
be64_to_cpu(lvb->lvb_iatime_packed));
ocfs2_unpack_timespec(&inode->i_mtime,
be64_to_cpu(lvb->lvb_imtime_packed));
- ocfs2_unpack_timespec(&inode->i_ctime,
- be64_to_cpu(lvb->lvb_ictime_packed));
+ ctime = inode_ctime_peek(inode);
+ ocfs2_unpack_timespec(&ctime, be64_to_cpu(lvb->lvb_ictime_packed));
spin_unlock(&oi->ip_lock);
return 0;
}
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index 91a194596552..52f126ebd0c2 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -232,8 +232,10 @@ int ocfs2_should_update_atime(struct inode *inode,
return 0;
if (vfsmnt->mnt_flags & MNT_RELATIME) {
+ struct timespec64 ctime = inode_ctime_peek(inode);
+
if ((timespec64_compare(&inode->i_atime, &inode->i_mtime) <= 0) ||
- (timespec64_compare(&inode->i_atime, &inode->i_ctime) <= 0))
+ (timespec64_compare(&inode->i_atime, &ctime) <= 0))
return 1;
return 0;
@@ -294,7 +296,7 @@ int ocfs2_set_inode_size(handle_t *handle,
i_size_write(inode, new_i_size);
inode->i_blocks = ocfs2_inode_sector_count(inode);
- inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
status = ocfs2_mark_inode_dirty(handle, inode, fe_bh);
if (status < 0) {
@@ -415,12 +417,12 @@ static int ocfs2_orphan_for_truncate(struct ocfs2_super *osb,
}
i_size_write(inode, new_i_size);
- inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
di = (struct ocfs2_dinode *) fe_bh->b_data;
di->i_size = cpu_to_le64(new_i_size);
- di->i_ctime = di->i_mtime = cpu_to_le64(inode->i_ctime.tv_sec);
- di->i_ctime_nsec = di->i_mtime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
+ di->i_ctime = di->i_mtime = cpu_to_le64(inode_ctime_peek(inode).tv_sec);
+ di->i_ctime_nsec = di->i_mtime_nsec = cpu_to_le32(inode_ctime_peek(inode).tv_nsec);
ocfs2_update_inode_fsync_trans(handle, inode, 0);
ocfs2_journal_dirty(handle, fe_bh);
@@ -824,7 +826,7 @@ static int ocfs2_write_zero_page(struct inode *inode, u64 abs_from,
i_size_write(inode, abs_to);
inode->i_blocks = ocfs2_inode_sector_count(inode);
di->i_size = cpu_to_le64((u64)i_size_read(inode));
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
di->i_mtime = di->i_ctime = cpu_to_le64(inode->i_mtime.tv_sec);
di->i_ctime_nsec = cpu_to_le32(inode->i_mtime.tv_nsec);
di->i_mtime_nsec = di->i_ctime_nsec;
@@ -2043,7 +2045,7 @@ static int __ocfs2_change_file_space(struct file *file, struct inode *inode,
goto out_inode_unlock;
}
- inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
ret = ocfs2_mark_inode_dirty(handle, inode, di_bh);
if (ret < 0)
mlog_errno(ret);
diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c
index bb116c39b581..1627948574fc 100644
--- a/fs/ocfs2/inode.c
+++ b/fs/ocfs2/inode.c
@@ -263,6 +263,7 @@ void ocfs2_populate_inode(struct inode *inode, struct ocfs2_dinode *fe,
struct super_block *sb;
struct ocfs2_super *osb;
int use_plocks = 1;
+ struct timespec64 ts;
sb = inode->i_sb;
osb = OCFS2_SB(sb);
@@ -306,8 +307,9 @@ void ocfs2_populate_inode(struct inode *inode, struct ocfs2_dinode *fe,
inode->i_atime.tv_nsec = le32_to_cpu(fe->i_atime_nsec);
inode->i_mtime.tv_sec = le64_to_cpu(fe->i_mtime);
inode->i_mtime.tv_nsec = le32_to_cpu(fe->i_mtime_nsec);
- inode->i_ctime.tv_sec = le64_to_cpu(fe->i_ctime);
- inode->i_ctime.tv_nsec = le32_to_cpu(fe->i_ctime_nsec);
+ ts.tv_sec = le64_to_cpu(fe->i_ctime);
+ ts.tv_nsec = le32_to_cpu(fe->i_ctime_nsec);
+ inode_ctime_set(inode, ts);
if (OCFS2_I(inode)->ip_blkno != le64_to_cpu(fe->i_blkno))
mlog(ML_ERROR,
@@ -1314,8 +1316,8 @@ int ocfs2_mark_inode_dirty(handle_t *handle,
fe->i_mode = cpu_to_le16(inode->i_mode);
fe->i_atime = cpu_to_le64(inode->i_atime.tv_sec);
fe->i_atime_nsec = cpu_to_le32(inode->i_atime.tv_nsec);
- fe->i_ctime = cpu_to_le64(inode->i_ctime.tv_sec);
- fe->i_ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
+ fe->i_ctime = cpu_to_le64(inode_ctime_peek(inode).tv_sec);
+ fe->i_ctime_nsec = cpu_to_le32(inode_ctime_peek(inode).tv_nsec);
fe->i_mtime = cpu_to_le64(inode->i_mtime.tv_sec);
fe->i_mtime_nsec = cpu_to_le32(inode->i_mtime.tv_nsec);
@@ -1352,8 +1354,8 @@ void ocfs2_refresh_inode(struct inode *inode,
inode->i_atime.tv_nsec = le32_to_cpu(fe->i_atime_nsec);
inode->i_mtime.tv_sec = le64_to_cpu(fe->i_mtime);
inode->i_mtime.tv_nsec = le32_to_cpu(fe->i_mtime_nsec);
- inode->i_ctime.tv_sec = le64_to_cpu(fe->i_ctime);
- inode->i_ctime.tv_nsec = le32_to_cpu(fe->i_ctime_nsec);
+ inode_ctime_set_sec(inode, le64_to_cpu(fe->i_ctime));
+ inode_ctime_set_nsec(inode, le32_to_cpu(fe->i_ctime_nsec));
spin_unlock(&OCFS2_I(inode)->ip_lock);
}
diff --git a/fs/ocfs2/move_extents.c b/fs/ocfs2/move_extents.c
index b1e32ec4a9d4..f8543a3509ee 100644
--- a/fs/ocfs2/move_extents.c
+++ b/fs/ocfs2/move_extents.c
@@ -950,9 +950,9 @@ static int ocfs2_move_extents(struct ocfs2_move_extents_context *context)
}
di = (struct ocfs2_dinode *)di_bh->b_data;
- inode->i_ctime = current_time(inode);
- di->i_ctime = cpu_to_le64(inode->i_ctime.tv_sec);
- di->i_ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
+ inode_ctime_set_current(inode);
+ di->i_ctime = cpu_to_le64(inode_ctime_peek(inode).tv_sec);
+ di->i_ctime_nsec = cpu_to_le32(inode_ctime_peek(inode).tv_nsec);
ocfs2_update_inode_fsync_trans(handle, inode, 0);
ocfs2_journal_dirty(handle, di_bh);
diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c
index 17c52225b87d..3e0c6e2c2669 100644
--- a/fs/ocfs2/namei.c
+++ b/fs/ocfs2/namei.c
@@ -690,6 +690,7 @@ static int ocfs2_link(struct dentry *old_dentry,
struct ocfs2_dinode *fe = NULL;
struct ocfs2_super *osb = OCFS2_SB(dir->i_sb);
struct ocfs2_dir_lookup_result lookup = { NULL, };
+ struct timespec64 ts;
sigset_t oldset;
u64 old_de_ino;
@@ -793,10 +794,10 @@ static int ocfs2_link(struct dentry *old_dentry,
}
inc_nlink(inode);
- inode->i_ctime = current_time(inode);
+ ts = inode_ctime_set_current(inode);
ocfs2_set_links_count(fe, inode->i_nlink);
- fe->i_ctime = cpu_to_le64(inode->i_ctime.tv_sec);
- fe->i_ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
+ fe->i_ctime = cpu_to_le64(ts.tv_sec);
+ fe->i_ctime_nsec = cpu_to_le32(ts.tv_nsec);
ocfs2_journal_dirty(handle, fe_bh);
err = ocfs2_add_entry(handle, dentry, inode,
@@ -995,7 +996,7 @@ static int ocfs2_unlink(struct inode *dir,
ocfs2_set_links_count(fe, inode->i_nlink);
ocfs2_journal_dirty(handle, fe_bh);
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
if (S_ISDIR(inode->i_mode))
drop_nlink(dir);
@@ -1537,7 +1538,7 @@ static int ocfs2_rename(struct mnt_idmap *idmap,
new_dir_bh, &target_insert);
}
- old_inode->i_ctime = current_time(old_inode);
+ inode_ctime_set_current(old_inode);
mark_inode_dirty(old_inode);
status = ocfs2_journal_access_di(handle, INODE_CACHE(old_inode),
@@ -1546,8 +1547,8 @@ static int ocfs2_rename(struct mnt_idmap *idmap,
if (status >= 0) {
old_di = (struct ocfs2_dinode *) old_inode_bh->b_data;
- old_di->i_ctime = cpu_to_le64(old_inode->i_ctime.tv_sec);
- old_di->i_ctime_nsec = cpu_to_le32(old_inode->i_ctime.tv_nsec);
+ old_di->i_ctime = cpu_to_le64(inode_ctime_peek(old_inode).tv_sec);
+ old_di->i_ctime_nsec = cpu_to_le32(inode_ctime_peek(old_inode).tv_nsec);
ocfs2_journal_dirty(handle, old_inode_bh);
} else
mlog_errno(status);
@@ -1586,9 +1587,9 @@ static int ocfs2_rename(struct mnt_idmap *idmap,
if (new_inode) {
drop_nlink(new_inode);
- new_inode->i_ctime = current_time(new_inode);
+ inode_ctime_set_current(new_inode);
}
- old_dir->i_ctime = old_dir->i_mtime = current_time(old_dir);
+ old_dir->i_mtime = inode_ctime_set_current(old_dir);
if (update_dot_dot) {
status = ocfs2_update_entry(old_inode, handle,
@@ -1610,7 +1611,8 @@ static int ocfs2_rename(struct mnt_idmap *idmap,
if (old_dir != new_dir) {
/* Keep the same times on both directories.*/
- new_dir->i_ctime = new_dir->i_mtime = old_dir->i_ctime;
+ new_dir->i_mtime = inode_ctime_peek(old_dir);
+ inode_ctime_set(new_dir, new_dir->i_mtime);
/*
* This will also pick up the i_nlink change from the
diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c
index 564ab48d03ef..8416c624782f 100644
--- a/fs/ocfs2/refcounttree.c
+++ b/fs/ocfs2/refcounttree.c
@@ -3750,9 +3750,9 @@ static int ocfs2_change_ctime(struct inode *inode,
goto out_commit;
}
- inode->i_ctime = current_time(inode);
- di->i_ctime = cpu_to_le64(inode->i_ctime.tv_sec);
- di->i_ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
+ inode_ctime_set_current(inode);
+ di->i_ctime = cpu_to_le64(inode_ctime_peek(inode).tv_sec);
+ di->i_ctime_nsec = cpu_to_le32(inode_ctime_peek(inode).tv_nsec);
ocfs2_journal_dirty(handle, di_bh);
@@ -4073,10 +4073,10 @@ static int ocfs2_complete_reflink(struct inode *s_inode,
* we want mtime to appear identical to the source and
* update ctime.
*/
- t_inode->i_ctime = current_time(t_inode);
+ inode_ctime_set_current(t_inode);
- di->i_ctime = cpu_to_le64(t_inode->i_ctime.tv_sec);
- di->i_ctime_nsec = cpu_to_le32(t_inode->i_ctime.tv_nsec);
+ di->i_ctime = cpu_to_le64(inode_ctime_peek(t_inode).tv_sec);
+ di->i_ctime_nsec = cpu_to_le32(inode_ctime_peek(t_inode).tv_nsec);
t_inode->i_mtime = s_inode->i_mtime;
di->i_mtime = s_di->i_mtime;
@@ -4456,7 +4456,7 @@ int ocfs2_reflink_update_dest(struct inode *dest,
if (newlen > i_size_read(dest))
i_size_write(dest, newlen);
spin_unlock(&OCFS2_I(dest)->ip_lock);
- dest->i_ctime = dest->i_mtime = current_time(dest);
+ dest->i_mtime = inode_ctime_set_current(dest);
ret = ocfs2_mark_inode_dirty(handle, dest, d_bh);
if (ret) {
diff --git a/fs/ocfs2/xattr.c b/fs/ocfs2/xattr.c
index 4ac77ff6e676..a137e9e05fa4 100644
--- a/fs/ocfs2/xattr.c
+++ b/fs/ocfs2/xattr.c
@@ -3421,9 +3421,9 @@ static int __ocfs2_xattr_set_handle(struct inode *inode,
goto out;
}
- inode->i_ctime = current_time(inode);
- di->i_ctime = cpu_to_le64(inode->i_ctime.tv_sec);
- di->i_ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
+ inode_ctime_set_current(inode);
+ di->i_ctime = cpu_to_le64(inode_ctime_peek(inode).tv_sec);
+ di->i_ctime_nsec = cpu_to_le32(inode_ctime_peek(inode).tv_nsec);
ocfs2_journal_dirty(ctxt->handle, xis->inode_bh);
}
out:
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 51/79] omfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (48 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
2023-06-23 12:12 ` Bob Copeland
-1 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Bob Copeland
Cc: Al Viro, Jan Kara, linux-karma-devel, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/omfs/dir.c | 4 ++--
fs/omfs/inode.c | 10 +++++-----
2 files changed, 7 insertions(+), 7 deletions(-)
diff --git a/fs/omfs/dir.c b/fs/omfs/dir.c
index 82cf7e9a665f..9b302b859335 100644
--- a/fs/omfs/dir.c
+++ b/fs/omfs/dir.c
@@ -143,7 +143,7 @@ static int omfs_add_link(struct dentry *dentry, struct inode *inode)
mark_buffer_dirty(bh);
brelse(bh);
- dir->i_ctime = current_time(dir);
+ inode_ctime_set_current(dir);
/* mark affected inodes dirty to rebuild checksums */
mark_inode_dirty(dir);
@@ -399,7 +399,7 @@ static int omfs_rename(struct mnt_idmap *idmap, struct inode *old_dir,
if (err)
goto out;
- old_inode->i_ctime = current_time(old_inode);
+ inode_ctime_set_current(old_inode);
mark_inode_dirty(old_inode);
out:
return err;
diff --git a/fs/omfs/inode.c b/fs/omfs/inode.c
index c4c79e07efc7..d420e5a53de3 100644
--- a/fs/omfs/inode.c
+++ b/fs/omfs/inode.c
@@ -51,7 +51,7 @@ struct inode *omfs_new_inode(struct inode *dir, umode_t mode)
inode_init_owner(&nop_mnt_idmap, inode, NULL, mode);
inode->i_mapping->a_ops = &omfs_aops;
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
switch (mode & S_IFMT) {
case S_IFDIR:
inode->i_op = &omfs_dir_inops;
@@ -134,8 +134,8 @@ static int __omfs_write_inode(struct inode *inode, int wait)
oi->i_head.h_magic = OMFS_IMAGIC;
oi->i_size = cpu_to_be64(inode->i_size);
- ctime = inode->i_ctime.tv_sec * 1000LL +
- ((inode->i_ctime.tv_nsec + 999)/1000);
+ ctime = inode_ctime_peek(inode).tv_sec * 1000LL +
+ ((inode_ctime_peek(inode).tv_nsec + 999)/1000);
oi->i_ctime = cpu_to_be64(ctime);
omfs_update_checksums(oi);
@@ -232,10 +232,10 @@ struct inode *omfs_iget(struct super_block *sb, ino_t ino)
inode->i_atime.tv_sec = ctime;
inode->i_mtime.tv_sec = ctime;
- inode->i_ctime.tv_sec = ctime;
+ inode_ctime_set_sec(inode, ctime);
inode->i_atime.tv_nsec = nsecs;
inode->i_mtime.tv_nsec = nsecs;
- inode->i_ctime.tv_nsec = nsecs;
+ inode_ctime_set_nsec(inode, nsecs);
inode->i_mapping->a_ops = &omfs_aops;
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 52/79] openpromfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (49 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner; +Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/openpromfs/inode.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c
index f0b7f4d51a17..5d5066078638 100644
--- a/fs/openpromfs/inode.c
+++ b/fs/openpromfs/inode.c
@@ -237,7 +237,7 @@ static struct dentry *openpromfs_lookup(struct inode *dir, struct dentry *dentry
if (IS_ERR(inode))
return ERR_CAST(inode);
if (inode->i_state & I_NEW) {
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
ent_oi = OP_I(inode);
ent_oi->type = ent_type;
ent_oi->u = ent_data;
@@ -388,7 +388,7 @@ static int openprom_fill_super(struct super_block *s, struct fs_context *fc)
}
root_inode->i_mtime = root_inode->i_atime =
- root_inode->i_ctime = current_time(root_inode);
+ inode_ctime_set_current(root_inode);
root_inode->i_op = &openprom_inode_operations;
root_inode->i_fop = &openprom_operations;
root_inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO;
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 53/79] orangefs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (50 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Mike Marshall, Martin Brandenburg
Cc: Al Viro, Jan Kara, devel, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/orangefs/namei.c | 2 +-
fs/orangefs/orangefs-utils.c | 6 +++---
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/fs/orangefs/namei.c b/fs/orangefs/namei.c
index 77518e248cf7..14085537ba1c 100644
--- a/fs/orangefs/namei.c
+++ b/fs/orangefs/namei.c
@@ -421,7 +421,7 @@ static int orangefs_rename(struct mnt_idmap *idmap,
ret);
if (new_dentry->d_inode)
- new_dentry->d_inode->i_ctime = current_time(new_dentry->d_inode);
+ inode_ctime_set_current(new_dentry->d_inode);
op_release(new_op);
return ret;
diff --git a/fs/orangefs/orangefs-utils.c b/fs/orangefs/orangefs-utils.c
index 46b7dcff18ac..4d12efb2ee85 100644
--- a/fs/orangefs/orangefs-utils.c
+++ b/fs/orangefs/orangefs-utils.c
@@ -361,11 +361,11 @@ int orangefs_inode_getattr(struct inode *inode, int flags)
downcall.resp.getattr.attributes.atime;
inode->i_mtime.tv_sec = (time64_t)new_op->
downcall.resp.getattr.attributes.mtime;
- inode->i_ctime.tv_sec = (time64_t)new_op->
- downcall.resp.getattr.attributes.ctime;
+ inode_ctime_set_sec(inode,
+ (time64_t)new_op->downcall.resp.getattr.attributes.ctime);
inode->i_atime.tv_nsec = 0;
inode->i_mtime.tv_nsec = 0;
- inode->i_ctime.tv_nsec = 0;
+ inode_ctime_set_nsec(inode, 0);
/* special case: mark the root inode as sticky */
inode->i_mode = type | (is_root_handle(inode) ? S_ISVTX : 0) |
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 54/79] overlayfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (51 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
2023-06-22 6:44 ` Amir Goldstein
-1 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Miklos Szeredi
Cc: Al Viro, Jan Kara, linux-unionfs, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/overlayfs/file.c | 7 +++++--
fs/overlayfs/util.c | 2 +-
2 files changed, 6 insertions(+), 3 deletions(-)
diff --git a/fs/overlayfs/file.c b/fs/overlayfs/file.c
index 21245b00722a..b47013d4bd4e 100644
--- a/fs/overlayfs/file.c
+++ b/fs/overlayfs/file.c
@@ -239,6 +239,7 @@ static loff_t ovl_llseek(struct file *file, loff_t offset, int whence)
static void ovl_file_accessed(struct file *file)
{
struct inode *inode, *upperinode;
+ struct timespec64 ct, uct;
if (file->f_flags & O_NOATIME)
return;
@@ -249,10 +250,12 @@ static void ovl_file_accessed(struct file *file)
if (!upperinode)
return;
+ ct = inode_ctime_peek(inode);
+ uct = inode_ctime_peek(upperinode);
if ((!timespec64_equal(&inode->i_mtime, &upperinode->i_mtime) ||
- !timespec64_equal(&inode->i_ctime, &upperinode->i_ctime))) {
+ !timespec64_equal(&ct, &uct))) {
inode->i_mtime = upperinode->i_mtime;
- inode->i_ctime = upperinode->i_ctime;
+ inode_ctime_set(inode, uct);
}
touch_atime(&file->f_path);
diff --git a/fs/overlayfs/util.c b/fs/overlayfs/util.c
index 7ef9e13c404a..e3746f9a202f 100644
--- a/fs/overlayfs/util.c
+++ b/fs/overlayfs/util.c
@@ -1202,6 +1202,6 @@ void ovl_copyattr(struct inode *inode)
inode->i_mode = realinode->i_mode;
inode->i_atime = realinode->i_atime;
inode->i_mtime = realinode->i_mtime;
- inode->i_ctime = realinode->i_ctime;
+ inode_ctime_set(inode, inode_ctime_peek(realinode));
i_size_write(inode, i_size_read(realinode));
}
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 55/79] proc: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (52 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
2023-06-30 22:13 ` Luis Chamberlain
-1 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Luis Chamberlain, Kees Cook, Iurii Zaikin
Cc: Al Viro, Jan Kara, linux-kernel, linux-fsdevel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/proc/base.c | 2 +-
fs/proc/inode.c | 2 +-
fs/proc/proc_sysctl.c | 2 +-
fs/proc/self.c | 2 +-
fs/proc/thread_self.c | 2 +-
5 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/fs/proc/base.c b/fs/proc/base.c
index 05452c3b9872..354f7f9ad05b 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -1902,7 +1902,7 @@ struct inode *proc_pid_make_inode(struct super_block *sb,
ei = PROC_I(inode);
inode->i_mode = mode;
inode->i_ino = get_next_ino();
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
inode->i_op = &proc_def_inode_operations;
/*
diff --git a/fs/proc/inode.c b/fs/proc/inode.c
index 67b09a1d9433..61f4fc0a6261 100644
--- a/fs/proc/inode.c
+++ b/fs/proc/inode.c
@@ -660,7 +660,7 @@ struct inode *proc_get_inode(struct super_block *sb, struct proc_dir_entry *de)
inode->i_private = de->data;
inode->i_ino = de->low_ino;
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
PROC_I(inode)->pde = de;
if (is_empty_pde(de)) {
make_empty_dir_inode(inode);
diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c
index 4e5488975415..abb7e524b28b 100644
--- a/fs/proc/proc_sysctl.c
+++ b/fs/proc/proc_sysctl.c
@@ -463,7 +463,7 @@ static struct inode *proc_sys_make_inode(struct super_block *sb,
head->count++;
spin_unlock(&sysctl_lock);
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
inode->i_mode = table->mode;
if (!S_ISDIR(table->mode)) {
inode->i_mode |= S_IFREG;
diff --git a/fs/proc/self.c b/fs/proc/self.c
index 72cd69bcaf4a..bb5dbdd30627 100644
--- a/fs/proc/self.c
+++ b/fs/proc/self.c
@@ -46,7 +46,7 @@ int proc_setup_self(struct super_block *s)
struct inode *inode = new_inode(s);
if (inode) {
inode->i_ino = self_inum;
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
inode->i_mode = S_IFLNK | S_IRWXUGO;
inode->i_uid = GLOBAL_ROOT_UID;
inode->i_gid = GLOBAL_ROOT_GID;
diff --git a/fs/proc/thread_self.c b/fs/proc/thread_self.c
index a553273fbd41..373680bd0635 100644
--- a/fs/proc/thread_self.c
+++ b/fs/proc/thread_self.c
@@ -46,7 +46,7 @@ int proc_setup_thread_self(struct super_block *s)
struct inode *inode = new_inode(s);
if (inode) {
inode->i_ino = thread_self_inum;
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
inode->i_mode = S_IFLNK | S_IRWXUGO;
inode->i_uid = GLOBAL_ROOT_UID;
inode->i_gid = GLOBAL_ROOT_GID;
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 56/79] pstore: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (53 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
2023-06-21 17:50 ` Kees Cook
-1 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Kees Cook, Tony Luck, Guilherme G. Piccoli
Cc: Al Viro, Jan Kara, linux-hardening, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/pstore/inode.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/pstore/inode.c b/fs/pstore/inode.c
index ffbadb8b3032..20624d6006e6 100644
--- a/fs/pstore/inode.c
+++ b/fs/pstore/inode.c
@@ -223,7 +223,7 @@ static struct inode *pstore_get_inode(struct super_block *sb)
struct inode *inode = new_inode(sb);
if (inode) {
inode->i_ino = get_next_ino();
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
}
return inode;
}
@@ -390,7 +390,7 @@ int pstore_mkfile(struct dentry *root, struct pstore_record *record)
inode->i_private = private;
if (record->time.tv_sec)
- inode->i_mtime = inode->i_ctime = record->time;
+ inode->i_mtime = inode_ctime_set(inode, record->time);
d_add(dentry, inode);
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 57/79] qnx4: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (54 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
2023-06-23 14:26 ` Anders Larsen
-1 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Anders Larsen; +Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/qnx4/inode.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c
index 391ea402920d..3fde90b3f99b 100644
--- a/fs/qnx4/inode.c
+++ b/fs/qnx4/inode.c
@@ -305,8 +305,8 @@ struct inode *qnx4_iget(struct super_block *sb, unsigned long ino)
inode->i_mtime.tv_nsec = 0;
inode->i_atime.tv_sec = le32_to_cpu(raw_inode->di_atime);
inode->i_atime.tv_nsec = 0;
- inode->i_ctime.tv_sec = le32_to_cpu(raw_inode->di_ctime);
- inode->i_ctime.tv_nsec = 0;
+ inode_ctime_set_sec(inode, le32_to_cpu(raw_inode->di_ctime));
+ inode_ctime_set_nsec(inode, 0);
inode->i_blocks = le32_to_cpu(raw_inode->di_first_xtnt.xtnt_size);
memcpy(qnx4_inode, raw_inode, QNX4_DIR_ENTRY_SIZE);
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 58/79] qnx6: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (55 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner; +Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/qnx6/inode.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/qnx6/inode.c b/fs/qnx6/inode.c
index 85b2fa3b211c..ff86c7100aaa 100644
--- a/fs/qnx6/inode.c
+++ b/fs/qnx6/inode.c
@@ -562,8 +562,8 @@ struct inode *qnx6_iget(struct super_block *sb, unsigned ino)
inode->i_mtime.tv_nsec = 0;
inode->i_atime.tv_sec = fs32_to_cpu(sbi, raw_inode->di_atime);
inode->i_atime.tv_nsec = 0;
- inode->i_ctime.tv_sec = fs32_to_cpu(sbi, raw_inode->di_ctime);
- inode->i_ctime.tv_nsec = 0;
+ inode_ctime_set_sec(inode, fs32_to_cpu(sbi, raw_inode->di_ctime));
+ inode_ctime_set_nsec(inode, 0);
/* calc blocks based on 512 byte blocksize */
inode->i_blocks = (inode->i_size + 511) >> 9;
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 59/79] ramfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (56 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner; +Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/ramfs/inode.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/fs/ramfs/inode.c b/fs/ramfs/inode.c
index fef477c78107..87f59dc5f6f2 100644
--- a/fs/ramfs/inode.c
+++ b/fs/ramfs/inode.c
@@ -65,7 +65,7 @@ struct inode *ramfs_get_inode(struct super_block *sb,
inode->i_mapping->a_ops = &ram_aops;
mapping_set_gfp_mask(inode->i_mapping, GFP_HIGHUSER);
mapping_set_unevictable(inode->i_mapping);
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
switch (mode & S_IFMT) {
default:
init_special_inode(inode, mode, dev);
@@ -105,7 +105,7 @@ ramfs_mknod(struct mnt_idmap *idmap, struct inode *dir,
d_instantiate(dentry, inode);
dget(dentry); /* Extra count - pin the dentry in core */
error = 0;
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
}
return error;
}
@@ -138,7 +138,7 @@ static int ramfs_symlink(struct mnt_idmap *idmap, struct inode *dir,
if (!error) {
d_instantiate(dentry, inode);
dget(dentry);
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
} else
iput(inode);
}
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 60/79] reiserfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (57 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner; +Cc: Al Viro, Jan Kara, reiserfs-devel, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/reiserfs/inode.c | 14 +++++++-------
fs/reiserfs/ioctl.c | 4 ++--
fs/reiserfs/namei.c | 21 ++++++++++-----------
fs/reiserfs/stree.c | 4 ++--
fs/reiserfs/super.c | 2 +-
fs/reiserfs/xattr.c | 5 +++--
fs/reiserfs/xattr_acl.c | 2 +-
7 files changed, 26 insertions(+), 26 deletions(-)
diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c
index 77bd3b27059f..3d4e18af8dac 100644
--- a/fs/reiserfs/inode.c
+++ b/fs/reiserfs/inode.c
@@ -1259,9 +1259,9 @@ static void init_inode(struct inode *inode, struct treepath *path)
inode->i_size = sd_v1_size(sd);
inode->i_atime.tv_sec = sd_v1_atime(sd);
inode->i_mtime.tv_sec = sd_v1_mtime(sd);
- inode->i_ctime.tv_sec = sd_v1_ctime(sd);
+ inode_ctime_set_sec(inode, sd_v1_ctime(sd));
inode->i_atime.tv_nsec = 0;
- inode->i_ctime.tv_nsec = 0;
+ inode_ctime_set_nsec(inode, 0);
inode->i_mtime.tv_nsec = 0;
inode->i_blocks = sd_v1_blocks(sd);
@@ -1314,8 +1314,8 @@ static void init_inode(struct inode *inode, struct treepath *path)
i_gid_write(inode, sd_v2_gid(sd));
inode->i_mtime.tv_sec = sd_v2_mtime(sd);
inode->i_atime.tv_sec = sd_v2_atime(sd);
- inode->i_ctime.tv_sec = sd_v2_ctime(sd);
- inode->i_ctime.tv_nsec = 0;
+ inode_ctime_set_sec(inode, sd_v2_ctime(sd));
+ inode_ctime_set_nsec(inode, 0);
inode->i_mtime.tv_nsec = 0;
inode->i_atime.tv_nsec = 0;
inode->i_blocks = sd_v2_blocks(sd);
@@ -1374,7 +1374,7 @@ static void inode2sd(void *sd, struct inode *inode, loff_t size)
set_sd_v2_gid(sd_v2, i_gid_read(inode));
set_sd_v2_mtime(sd_v2, inode->i_mtime.tv_sec);
set_sd_v2_atime(sd_v2, inode->i_atime.tv_sec);
- set_sd_v2_ctime(sd_v2, inode->i_ctime.tv_sec);
+ set_sd_v2_ctime(sd_v2, inode_ctime_peek(inode).tv_sec);
set_sd_v2_blocks(sd_v2, to_fake_used_blocks(inode, SD_V2_SIZE));
if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode))
set_sd_v2_rdev(sd_v2, new_encode_dev(inode->i_rdev));
@@ -1394,7 +1394,7 @@ static void inode2sd_v1(void *sd, struct inode *inode, loff_t size)
set_sd_v1_nlink(sd_v1, inode->i_nlink);
set_sd_v1_size(sd_v1, size);
set_sd_v1_atime(sd_v1, inode->i_atime.tv_sec);
- set_sd_v1_ctime(sd_v1, inode->i_ctime.tv_sec);
+ set_sd_v1_ctime(sd_v1, inode_ctime_peek(inode).tv_sec);
set_sd_v1_mtime(sd_v1, inode->i_mtime.tv_sec);
if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode))
@@ -1986,7 +1986,7 @@ int reiserfs_new_inode(struct reiserfs_transaction_handle *th,
/* uid and gid must already be set by the caller for quota init */
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
inode->i_size = i_size;
inode->i_blocks = 0;
inode->i_bytes = 0;
diff --git a/fs/reiserfs/ioctl.c b/fs/reiserfs/ioctl.c
index 6bf9b54e58ca..5d129a0b3ce7 100644
--- a/fs/reiserfs/ioctl.c
+++ b/fs/reiserfs/ioctl.c
@@ -55,7 +55,7 @@ int reiserfs_fileattr_set(struct mnt_idmap *idmap,
}
sd_attrs_to_i_attrs(flags, inode);
REISERFS_I(inode)->i_attrs = flags;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
mark_inode_dirty(inode);
err = 0;
unlock:
@@ -107,7 +107,7 @@ long reiserfs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
err = -EFAULT;
goto setversion_out;
}
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
mark_inode_dirty(inode);
setversion_out:
mnt_drop_write_file(filp);
diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c
index 52240cc891cf..fa4d11b21279 100644
--- a/fs/reiserfs/namei.c
+++ b/fs/reiserfs/namei.c
@@ -572,7 +572,7 @@ static int reiserfs_add_entry(struct reiserfs_transaction_handle *th,
}
dir->i_size += paste_size;
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
if (!S_ISDIR(inode->i_mode) && visible)
/* reiserfs_mkdir or reiserfs_rename will do that by itself */
reiserfs_update_sd(th, dir);
@@ -966,7 +966,8 @@ static int reiserfs_rmdir(struct inode *dir, struct dentry *dentry)
inode->i_nlink);
clear_nlink(inode);
- inode->i_ctime = dir->i_ctime = dir->i_mtime = current_time(dir);
+ inode_ctime_set_current(inode);
+ dir->i_mtime = inode_ctime_set_current(dir);
reiserfs_update_sd(&th, inode);
DEC_DIR_INODE_NLINK(dir)
@@ -1070,11 +1071,11 @@ static int reiserfs_unlink(struct inode *dir, struct dentry *dentry)
inc_nlink(inode);
goto end_unlink;
}
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
reiserfs_update_sd(&th, inode);
dir->i_size -= (de.de_entrylen + DEH_SIZE);
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
reiserfs_update_sd(&th, dir);
if (!savelink)
@@ -1250,7 +1251,7 @@ static int reiserfs_link(struct dentry *old_dentry, struct inode *dir,
return err ? err : retval;
}
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
reiserfs_update_sd(&th, inode);
ihold(inode);
@@ -1325,7 +1326,6 @@ static int reiserfs_rename(struct mnt_idmap *idmap,
int jbegin_count;
umode_t old_inode_mode;
unsigned long savelink = 1;
- struct timespec64 ctime;
if (flags & ~RENAME_NOREPLACE)
return -EINVAL;
@@ -1576,14 +1576,13 @@ static int reiserfs_rename(struct mnt_idmap *idmap,
mark_de_hidden(old_de.de_deh + old_de.de_entry_num);
journal_mark_dirty(&th, old_de.de_bh);
- ctime = current_time(old_dir);
- old_dir->i_ctime = old_dir->i_mtime = ctime;
- new_dir->i_ctime = new_dir->i_mtime = ctime;
+ old_dir->i_mtime = inode_ctime_set_current(old_dir);
+ new_dir->i_mtime = inode_ctime_set_current(new_dir);
/*
* thanks to Alex Adriaanse <alex_a@caltech.edu> for patch
* which adds ctime update of renamed object
*/
- old_inode->i_ctime = ctime;
+ inode_ctime_set_current(old_inode);
if (new_dentry_inode) {
/* adjust link number of the victim */
@@ -1592,7 +1591,7 @@ static int reiserfs_rename(struct mnt_idmap *idmap,
} else {
drop_nlink(new_dentry_inode);
}
- new_dentry_inode->i_ctime = ctime;
+ inode_ctime_set_current(new_dentry_inode);
savelink = new_dentry_inode->i_nlink;
}
diff --git a/fs/reiserfs/stree.c b/fs/reiserfs/stree.c
index ce5003986789..c7f0aa0e9a2e 100644
--- a/fs/reiserfs/stree.c
+++ b/fs/reiserfs/stree.c
@@ -2004,7 +2004,7 @@ int reiserfs_do_truncate(struct reiserfs_transaction_handle *th,
if (update_timestamps) {
inode->i_mtime = current_time(inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
}
reiserfs_update_sd(th, inode);
@@ -2029,7 +2029,7 @@ int reiserfs_do_truncate(struct reiserfs_transaction_handle *th,
if (update_timestamps) {
/* this is truncate, not file closing */
inode->i_mtime = current_time(inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
}
reiserfs_update_sd(th, inode);
diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index 929acce6e731..0ff67c74aac2 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -2587,7 +2587,7 @@ static ssize_t reiserfs_quota_write(struct super_block *sb, int type,
return err;
if (inode->i_size < off + len - towrite)
i_size_write(inode, off + len - towrite);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
mark_inode_dirty(inode);
return len - towrite;
}
diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c
index 651027967159..469d700d97a9 100644
--- a/fs/reiserfs/xattr.c
+++ b/fs/reiserfs/xattr.c
@@ -466,12 +466,13 @@ int reiserfs_commit_write(struct file *f, struct page *page,
static void update_ctime(struct inode *inode)
{
struct timespec64 now = current_time(inode);
+ struct timespec64 ctime = inode_ctime_peek(inode);
if (inode_unhashed(inode) || !inode->i_nlink ||
- timespec64_equal(&inode->i_ctime, &now))
+ timespec64_equal(&ctime, &now))
return;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
mark_inode_dirty(inode);
}
diff --git a/fs/reiserfs/xattr_acl.c b/fs/reiserfs/xattr_acl.c
index 138060452678..07ef471aa436 100644
--- a/fs/reiserfs/xattr_acl.c
+++ b/fs/reiserfs/xattr_acl.c
@@ -285,7 +285,7 @@ __reiserfs_set_acl(struct reiserfs_transaction_handle *th, struct inode *inode,
if (error == -ENODATA) {
error = 0;
if (type == ACL_TYPE_ACCESS) {
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
mark_inode_dirty(inode);
}
}
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 61/79] romfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (58 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner; +Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/romfs/super.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/romfs/super.c b/fs/romfs/super.c
index c59b230d55b4..1adea5c249a6 100644
--- a/fs/romfs/super.c
+++ b/fs/romfs/super.c
@@ -322,8 +322,8 @@ static struct inode *romfs_iget(struct super_block *sb, unsigned long pos)
set_nlink(i, 1); /* Hard to decide.. */
i->i_size = be32_to_cpu(ri.size);
- i->i_mtime.tv_sec = i->i_atime.tv_sec = i->i_ctime.tv_sec = 0;
- i->i_mtime.tv_nsec = i->i_atime.tv_nsec = i->i_ctime.tv_nsec = 0;
+ i->i_mtime.tv_sec = i->i_atime.tv_sec = inode_ctime_set_sec(i, 0);
+ i->i_mtime.tv_nsec = i->i_atime.tv_nsec = inode_ctime_set_nsec(i, 0);
/* set up mode and ops */
mode = romfs_modemap[nextfh & ROMFH_TYPE];
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 62/79] smb: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (59 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
2023-06-21 17:45 ` Tom Talpey
2023-06-23 5:03 ` Sergey Senozhatsky
-1 siblings, 2 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Steve French, Paulo Alcantara,
Ronnie Sahlberg, Shyam Prasad N, Tom Talpey, Namjae Jeon,
Sergey Senozhatsky
Cc: Al Viro, Jan Kara, linux-cifs, samba-technical, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/smb/client/file.c | 4 ++--
fs/smb/client/fscache.h | 5 +++--
fs/smb/client/inode.c | 15 +++++++--------
fs/smb/client/smb2ops.c | 2 +-
fs/smb/server/smb2pdu.c | 8 ++++----
5 files changed, 17 insertions(+), 17 deletions(-)
diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c
index 0a5fe8d5314b..07fed94b7501 100644
--- a/fs/smb/client/file.c
+++ b/fs/smb/client/file.c
@@ -1085,7 +1085,7 @@ int cifs_close(struct inode *inode, struct file *file)
!test_bit(CIFS_INO_CLOSE_ON_LOCK, &cinode->flags) &&
dclose) {
if (test_and_clear_bit(CIFS_INO_MODIFIED_ATTR, &cinode->flags)) {
- inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
}
spin_lock(&cinode->deferred_lock);
cifs_add_deferred_close(cfile, dclose);
@@ -2596,7 +2596,7 @@ static int cifs_partialpagewrite(struct page *page, unsigned from, unsigned to)
write_data, to - from, &offset);
cifsFileInfo_put(open_file);
/* Does mm or vfs already set times? */
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
if ((bytes_written > 0) && (offset))
rc = 0;
else if (bytes_written < 0)
diff --git a/fs/smb/client/fscache.h b/fs/smb/client/fscache.h
index 173999610997..1c50d286c3a4 100644
--- a/fs/smb/client/fscache.h
+++ b/fs/smb/client/fscache.h
@@ -50,12 +50,13 @@ void cifs_fscache_fill_coherency(struct inode *inode,
struct cifs_fscache_inode_coherency_data *cd)
{
struct cifsInodeInfo *cifsi = CIFS_I(inode);
+ struct timespec64 ctime = inode_ctime_peek(&cifsi->netfs.inode);
memset(cd, 0, sizeof(*cd));
cd->last_write_time_sec = cpu_to_le64(cifsi->netfs.inode.i_mtime.tv_sec);
cd->last_write_time_nsec = cpu_to_le32(cifsi->netfs.inode.i_mtime.tv_nsec);
- cd->last_change_time_sec = cpu_to_le64(cifsi->netfs.inode.i_ctime.tv_sec);
- cd->last_change_time_nsec = cpu_to_le32(cifsi->netfs.inode.i_ctime.tv_nsec);
+ cd->last_change_time_sec = cpu_to_le64(ctime.tv_sec);
+ cd->last_change_time_nsec = cpu_to_le32(ctime.tv_nsec);
}
diff --git a/fs/smb/client/inode.c b/fs/smb/client/inode.c
index 1087ac6104a9..d986b280b6c3 100644
--- a/fs/smb/client/inode.c
+++ b/fs/smb/client/inode.c
@@ -172,7 +172,7 @@ cifs_fattr_to_inode(struct inode *inode, struct cifs_fattr *fattr)
else
inode->i_atime = fattr->cf_atime;
inode->i_mtime = fattr->cf_mtime;
- inode->i_ctime = fattr->cf_ctime;
+ inode_ctime_set(inode, fattr->cf_ctime);
inode->i_rdev = fattr->cf_rdev;
cifs_nlink_fattr_to_inode(inode, fattr);
inode->i_uid = fattr->cf_uid;
@@ -1744,9 +1744,9 @@ int cifs_unlink(struct inode *dir, struct dentry *dentry)
cifs_inode = CIFS_I(inode);
cifs_inode->time = 0; /* will force revalidate to get info
when needed */
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
}
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
cifs_inode = CIFS_I(dir);
CIFS_I(dir)->time = 0; /* force revalidate of dir as well */
unlink_out:
@@ -2060,8 +2060,8 @@ int cifs_rmdir(struct inode *inode, struct dentry *direntry)
*/
cifsInode->time = 0;
- d_inode(direntry)->i_ctime = inode->i_ctime = inode->i_mtime =
- current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
+ inode_ctime_set(d_inode(direntry), inode->i_mtime);
rmdir_exit:
free_dentry_path(page);
@@ -2267,9 +2267,8 @@ cifs_rename2(struct mnt_idmap *idmap, struct inode *source_dir,
/* force revalidate to go get info when needed */
CIFS_I(source_dir)->time = CIFS_I(target_dir)->time = 0;
- source_dir->i_ctime = source_dir->i_mtime = target_dir->i_ctime =
- target_dir->i_mtime = current_time(source_dir);
-
+ source_dir->i_mtime = inode_ctime_set_current(source_dir);
+ target_dir->i_mtime = inode_ctime_set_current(target_dir);
cifs_rename_exit:
kfree(info_buf_source);
free_dentry_path(page2);
diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c
index 5639d8c48570..52564a5dd2b1 100644
--- a/fs/smb/client/smb2ops.c
+++ b/fs/smb/client/smb2ops.c
@@ -1391,7 +1391,7 @@ smb2_close_getattr(const unsigned int xid, struct cifs_tcon *tcon,
if (file_inf.LastWriteTime)
inode->i_mtime = cifs_NTtimeToUnix(file_inf.LastWriteTime);
if (file_inf.ChangeTime)
- inode->i_ctime = cifs_NTtimeToUnix(file_inf.ChangeTime);
+ inode_ctime_set(inode, cifs_NTtimeToUnix(file_inf.ChangeTime));
if (file_inf.LastAccessTime)
inode->i_atime = cifs_NTtimeToUnix(file_inf.LastAccessTime);
diff --git a/fs/smb/server/smb2pdu.c b/fs/smb/server/smb2pdu.c
index d31926194ebf..7a741ab6835c 100644
--- a/fs/smb/server/smb2pdu.c
+++ b/fs/smb/server/smb2pdu.c
@@ -4768,7 +4768,7 @@ static int find_file_posix_info(struct smb2_query_info_rsp *rsp,
file_info->LastAccessTime = cpu_to_le64(time);
time = ksmbd_UnixTimeToNT(inode->i_mtime);
file_info->LastWriteTime = cpu_to_le64(time);
- time = ksmbd_UnixTimeToNT(inode->i_ctime);
+ time = ksmbd_UnixTimeToNT(inode_ctime_peek(inode));
file_info->ChangeTime = cpu_to_le64(time);
file_info->DosAttributes = fp->f_ci->m_fattr;
file_info->Inode = cpu_to_le64(inode->i_ino);
@@ -5409,7 +5409,7 @@ int smb2_close(struct ksmbd_work *work)
rsp->LastAccessTime = cpu_to_le64(time);
time = ksmbd_UnixTimeToNT(inode->i_mtime);
rsp->LastWriteTime = cpu_to_le64(time);
- time = ksmbd_UnixTimeToNT(inode->i_ctime);
+ time = ksmbd_UnixTimeToNT(inode_ctime_peek(inode));
rsp->ChangeTime = cpu_to_le64(time);
ksmbd_fd_put(work, fp);
} else {
@@ -5628,7 +5628,7 @@ static int set_file_basic_info(struct ksmbd_file *fp,
if (file_info->ChangeTime)
attrs.ia_ctime = ksmbd_NTtimeToUnix(file_info->ChangeTime);
else
- attrs.ia_ctime = inode->i_ctime;
+ attrs.ia_ctime = inode_ctime_peek(inode);
if (file_info->LastWriteTime) {
attrs.ia_mtime = ksmbd_NTtimeToUnix(file_info->LastWriteTime);
@@ -5673,7 +5673,7 @@ static int set_file_basic_info(struct ksmbd_file *fp,
return -EACCES;
inode_lock(inode);
- inode->i_ctime = attrs.ia_ctime;
+ inode_ctime_set(inode, attrs.ia_ctime);
attrs.ia_valid &= ~ATTR_CTIME;
rc = notify_change(idmap, dentry, &attrs, NULL);
inode_unlock(inode);
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 63/79] squashfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (60 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Phillip Lougher; +Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/squashfs/inode.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/fs/squashfs/inode.c b/fs/squashfs/inode.c
index 24463145b351..3d619b294c52 100644
--- a/fs/squashfs/inode.c
+++ b/fs/squashfs/inode.c
@@ -61,7 +61,7 @@ static int squashfs_new_inode(struct super_block *sb, struct inode *inode,
inode->i_ino = le32_to_cpu(sqsh_ino->inode_number);
inode->i_mtime.tv_sec = le32_to_cpu(sqsh_ino->mtime);
inode->i_atime.tv_sec = inode->i_mtime.tv_sec;
- inode->i_ctime.tv_sec = inode->i_mtime.tv_sec;
+ inode_ctime_set_sec(inode, inode->i_mtime.tv_sec);
inode->i_mode = le16_to_cpu(sqsh_ino->mode);
inode->i_size = 0;
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 64/79] sysv: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (61 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner; +Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/sysv/dir.c | 6 +++---
fs/sysv/ialloc.c | 2 +-
fs/sysv/inode.c | 6 +++---
fs/sysv/itree.c | 4 ++--
fs/sysv/namei.c | 6 +++---
5 files changed, 12 insertions(+), 12 deletions(-)
diff --git a/fs/sysv/dir.c b/fs/sysv/dir.c
index 0140010aa0c3..7c145f2e8b97 100644
--- a/fs/sysv/dir.c
+++ b/fs/sysv/dir.c
@@ -224,7 +224,7 @@ int sysv_add_link(struct dentry *dentry, struct inode *inode)
memset (de->name + namelen, 0, SYSV_DIRSIZE - namelen - 2);
de->inode = cpu_to_fs16(SYSV_SB(inode->i_sb), inode->i_ino);
dir_commit_chunk(page, pos, SYSV_DIRSIZE);
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
mark_inode_dirty(dir);
err = sysv_handle_dirsync(dir);
out_page:
@@ -249,7 +249,7 @@ int sysv_delete_entry(struct sysv_dir_entry *de, struct page *page)
}
de->inode = 0;
dir_commit_chunk(page, pos, SYSV_DIRSIZE);
- inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
mark_inode_dirty(inode);
return sysv_handle_dirsync(inode);
}
@@ -346,7 +346,7 @@ int sysv_set_link(struct sysv_dir_entry *de, struct page *page,
}
de->inode = cpu_to_fs16(SYSV_SB(inode->i_sb), inode->i_ino);
dir_commit_chunk(page, pos, SYSV_DIRSIZE);
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
mark_inode_dirty(dir);
return sysv_handle_dirsync(inode);
}
diff --git a/fs/sysv/ialloc.c b/fs/sysv/ialloc.c
index e732879036ab..ade57ef69e3f 100644
--- a/fs/sysv/ialloc.c
+++ b/fs/sysv/ialloc.c
@@ -165,7 +165,7 @@ struct inode * sysv_new_inode(const struct inode * dir, umode_t mode)
dirty_sb(sb);
inode_init_owner(&nop_mnt_idmap, inode, dir, mode);
inode->i_ino = fs16_to_cpu(sbi, ino);
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
inode->i_blocks = 0;
memset(SYSV_I(inode)->i_data, 0, sizeof(SYSV_I(inode)->i_data));
SYSV_I(inode)->i_dir_start_lookup = 0;
diff --git a/fs/sysv/inode.c b/fs/sysv/inode.c
index 9e8d4a6fb2f3..53143b9e125a 100644
--- a/fs/sysv/inode.c
+++ b/fs/sysv/inode.c
@@ -202,8 +202,8 @@ struct inode *sysv_iget(struct super_block *sb, unsigned int ino)
inode->i_size = fs32_to_cpu(sbi, raw_inode->i_size);
inode->i_atime.tv_sec = fs32_to_cpu(sbi, raw_inode->i_atime);
inode->i_mtime.tv_sec = fs32_to_cpu(sbi, raw_inode->i_mtime);
- inode->i_ctime.tv_sec = fs32_to_cpu(sbi, raw_inode->i_ctime);
- inode->i_ctime.tv_nsec = 0;
+ inode_ctime_set_sec(inode, fs32_to_cpu(sbi, raw_inode->i_ctime));
+ inode_ctime_set_nsec(inode, 0);
inode->i_atime.tv_nsec = 0;
inode->i_mtime.tv_nsec = 0;
inode->i_blocks = 0;
@@ -256,7 +256,7 @@ static int __sysv_write_inode(struct inode *inode, int wait)
raw_inode->i_size = cpu_to_fs32(sbi, inode->i_size);
raw_inode->i_atime = cpu_to_fs32(sbi, inode->i_atime.tv_sec);
raw_inode->i_mtime = cpu_to_fs32(sbi, inode->i_mtime.tv_sec);
- raw_inode->i_ctime = cpu_to_fs32(sbi, inode->i_ctime.tv_sec);
+ raw_inode->i_ctime = cpu_to_fs32(sbi, inode_ctime_peek(inode).tv_sec);
si = SYSV_I(inode);
if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode))
diff --git a/fs/sysv/itree.c b/fs/sysv/itree.c
index 58d7f43a1371..5b3bc071d270 100644
--- a/fs/sysv/itree.c
+++ b/fs/sysv/itree.c
@@ -183,7 +183,7 @@ static inline int splice_branch(struct inode *inode,
*where->p = where->key;
write_unlock(&pointers_lock);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
/* had we spliced it onto indirect block? */
if (where->bh)
@@ -423,7 +423,7 @@ void sysv_truncate (struct inode * inode)
}
n++;
}
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
if (IS_SYNC(inode))
sysv_sync_inode (inode);
else
diff --git a/fs/sysv/namei.c b/fs/sysv/namei.c
index fcf163fea3ad..d5aa9468c362 100644
--- a/fs/sysv/namei.c
+++ b/fs/sysv/namei.c
@@ -103,7 +103,7 @@ static int sysv_link(struct dentry * old_dentry, struct inode * dir,
{
struct inode *inode = d_inode(old_dentry);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
inode_inc_link_count(inode);
ihold(inode);
@@ -161,7 +161,7 @@ static int sysv_unlink(struct inode * dir, struct dentry * dentry)
err = sysv_delete_entry(de, page);
if (!err) {
- inode->i_ctime = dir->i_ctime;
+ inode_ctime_set(inode, inode_ctime_peek(dir));
inode_dec_link_count(inode);
}
unmap_and_put_page(page, de);
@@ -230,7 +230,7 @@ static int sysv_rename(struct mnt_idmap *idmap, struct inode *old_dir,
unmap_and_put_page(new_page, new_de);
if (err)
goto out_dir;
- new_inode->i_ctime = current_time(new_inode);
+ inode_ctime_set_current(new_inode);
if (dir_de)
drop_nlink(new_inode);
inode_dec_link_count(new_inode);
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 65/79] tracefs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (62 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Steven Rostedt, Masami Hiramatsu
Cc: Al Viro, Jan Kara, linux-kernel, linux-trace-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/tracefs/inode.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/fs/tracefs/inode.c b/fs/tracefs/inode.c
index 57ac8aa4a724..aae4b5645efd 100644
--- a/fs/tracefs/inode.c
+++ b/fs/tracefs/inode.c
@@ -132,7 +132,7 @@ static struct inode *tracefs_get_inode(struct super_block *sb)
struct inode *inode = new_inode(sb);
if (inode) {
inode->i_ino = get_next_ino();
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
}
return inode;
}
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 66/79] ubifs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
@ 2023-06-21 14:46 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Richard Weinberger
Cc: Al Viro, Jan Kara, linux-mtd, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/ubifs/debug.c | 4 ++--
fs/ubifs/dir.c | 39 +++++++++++++++++++--------------------
fs/ubifs/file.c | 16 +++++++++-------
fs/ubifs/ioctl.c | 2 +-
fs/ubifs/journal.c | 4 ++--
fs/ubifs/super.c | 4 ++--
fs/ubifs/xattr.c | 6 +++---
7 files changed, 38 insertions(+), 37 deletions(-)
diff --git a/fs/ubifs/debug.c b/fs/ubifs/debug.c
index 9c9d3f0e36a4..114c2b00aa29 100644
--- a/fs/ubifs/debug.c
+++ b/fs/ubifs/debug.c
@@ -243,8 +243,8 @@ void ubifs_dump_inode(struct ubifs_info *c, const struct inode *inode)
(unsigned int)inode->i_mtime.tv_sec,
(unsigned int)inode->i_mtime.tv_nsec);
pr_err("\tctime %u.%u\n",
- (unsigned int)inode->i_ctime.tv_sec,
- (unsigned int)inode->i_ctime.tv_nsec);
+ (unsigned int) inode_ctime_peek(inode).tv_sec,
+ (unsigned int) inode_ctime_peek(inode).tv_nsec);
pr_err("\tcreat_sqnum %llu\n", ui->creat_sqnum);
pr_err("\txattr_size %u\n", ui->xattr_size);
pr_err("\txattr_cnt %u\n", ui->xattr_cnt);
diff --git a/fs/ubifs/dir.c b/fs/ubifs/dir.c
index ef0499edc248..39588e3e3a8b 100644
--- a/fs/ubifs/dir.c
+++ b/fs/ubifs/dir.c
@@ -96,8 +96,7 @@ struct inode *ubifs_new_inode(struct ubifs_info *c, struct inode *dir,
inode->i_flags |= S_NOCMTIME;
inode_init_owner(&nop_mnt_idmap, inode, dir, mode);
- inode->i_mtime = inode->i_atime = inode->i_ctime =
- current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
inode->i_mapping->nrpages = 0;
if (!is_xattr) {
@@ -325,7 +324,7 @@ static int ubifs_create(struct mnt_idmap *idmap, struct inode *dir,
mutex_lock(&dir_ui->ui_mutex);
dir->i_size += sz_change;
dir_ui->ui_size = dir->i_size;
- dir->i_mtime = dir->i_ctime = inode->i_ctime;
+ dir->i_mtime = inode_ctime_set(inode, inode_ctime_peek(inode));
err = ubifs_jnl_update(c, dir, &nm, inode, 0, 0);
if (err)
goto out_cancel;
@@ -765,10 +764,10 @@ static int ubifs_link(struct dentry *old_dentry, struct inode *dir,
inc_nlink(inode);
ihold(inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
dir->i_size += sz_change;
dir_ui->ui_size = dir->i_size;
- dir->i_mtime = dir->i_ctime = inode->i_ctime;
+ dir->i_mtime = inode_ctime_set(inode, inode_ctime_peek(inode));
err = ubifs_jnl_update(c, dir, &nm, inode, 0, 0);
if (err)
goto out_cancel;
@@ -838,11 +837,11 @@ static int ubifs_unlink(struct inode *dir, struct dentry *dentry)
}
lock_2_inodes(dir, inode);
- inode->i_ctime = current_time(dir);
+ inode_ctime_set(inode, current_time(dir));
drop_nlink(inode);
dir->i_size -= sz_change;
dir_ui->ui_size = dir->i_size;
- dir->i_mtime = dir->i_ctime = inode->i_ctime;
+ dir->i_mtime = inode_ctime_set(inode, inode_ctime_peek(inode));
err = ubifs_jnl_update(c, dir, &nm, inode, 1, 0);
if (err)
goto out_cancel;
@@ -940,12 +939,12 @@ static int ubifs_rmdir(struct inode *dir, struct dentry *dentry)
}
lock_2_inodes(dir, inode);
- inode->i_ctime = current_time(dir);
+ inode_ctime_set(inode, current_time(dir));
clear_nlink(inode);
drop_nlink(dir);
dir->i_size -= sz_change;
dir_ui->ui_size = dir->i_size;
- dir->i_mtime = dir->i_ctime = inode->i_ctime;
+ dir->i_mtime = inode_ctime_set(inode, inode_ctime_peek(inode));
err = ubifs_jnl_update(c, dir, &nm, inode, 1, 0);
if (err)
goto out_cancel;
@@ -1019,7 +1018,7 @@ static int ubifs_mkdir(struct mnt_idmap *idmap, struct inode *dir,
inc_nlink(dir);
dir->i_size += sz_change;
dir_ui->ui_size = dir->i_size;
- dir->i_mtime = dir->i_ctime = inode->i_ctime;
+ dir->i_mtime = inode_ctime_set(inode, inode_ctime_peek(inode));
err = ubifs_jnl_update(c, dir, &nm, inode, 0, 0);
if (err) {
ubifs_err(c, "cannot create directory, error %d", err);
@@ -1110,7 +1109,7 @@ static int ubifs_mknod(struct mnt_idmap *idmap, struct inode *dir,
mutex_lock(&dir_ui->ui_mutex);
dir->i_size += sz_change;
dir_ui->ui_size = dir->i_size;
- dir->i_mtime = dir->i_ctime = inode->i_ctime;
+ dir->i_mtime = inode_ctime_set(inode, inode_ctime_peek(inode));
err = ubifs_jnl_update(c, dir, &nm, inode, 0, 0);
if (err)
goto out_cancel;
@@ -1210,7 +1209,7 @@ static int ubifs_symlink(struct mnt_idmap *idmap, struct inode *dir,
mutex_lock(&dir_ui->ui_mutex);
dir->i_size += sz_change;
dir_ui->ui_size = dir->i_size;
- dir->i_mtime = dir->i_ctime = inode->i_ctime;
+ dir->i_mtime = inode_ctime_set(inode, inode_ctime_peek(inode));
err = ubifs_jnl_update(c, dir, &nm, inode, 0, 0);
if (err)
goto out_cancel;
@@ -1415,7 +1414,7 @@ static int do_rename(struct inode *old_dir, struct dentry *old_dentry,
* rename.
*/
time = current_time(old_dir);
- old_inode->i_ctime = time;
+ inode_ctime_set(old_inode, time);
/* We must adjust parent link count when renaming directories */
if (is_dir) {
@@ -1444,8 +1443,8 @@ static int do_rename(struct inode *old_dir, struct dentry *old_dentry,
old_dir->i_size -= old_sz;
ubifs_inode(old_dir)->ui_size = old_dir->i_size;
- old_dir->i_mtime = old_dir->i_ctime = time;
- new_dir->i_mtime = new_dir->i_ctime = time;
+ old_dir->i_mtime = inode_ctime_set(old_dir, time);
+ new_dir->i_mtime = inode_ctime_set(new_dir, time);
/*
* And finally, if we unlinked a direntry which happened to have the
@@ -1462,7 +1461,7 @@ static int do_rename(struct inode *old_dir, struct dentry *old_dentry,
clear_nlink(new_inode);
else
drop_nlink(new_inode);
- new_inode->i_ctime = time;
+ inode_ctime_set(new_inode, time);
} else {
new_dir->i_size += new_sz;
ubifs_inode(new_dir)->ui_size = new_dir->i_size;
@@ -1589,10 +1588,10 @@ static int ubifs_xrename(struct inode *old_dir, struct dentry *old_dentry,
lock_4_inodes(old_dir, new_dir, NULL, NULL);
time = current_time(old_dir);
- fst_inode->i_ctime = time;
- snd_inode->i_ctime = time;
- old_dir->i_mtime = old_dir->i_ctime = time;
- new_dir->i_mtime = new_dir->i_ctime = time;
+ inode_ctime_set(fst_inode, time);
+ inode_ctime_set(snd_inode, time);
+ old_dir->i_mtime = inode_ctime_set(old_dir, time);
+ new_dir->i_mtime = inode_ctime_set(new_dir, time);
if (old_dir != new_dir) {
if (S_ISDIR(fst_inode->i_mode) && !S_ISDIR(snd_inode->i_mode)) {
diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c
index 6738fe43040b..f9ba69f6fbab 100644
--- a/fs/ubifs/file.c
+++ b/fs/ubifs/file.c
@@ -1092,7 +1092,7 @@ static void do_attr_changes(struct inode *inode, const struct iattr *attr)
if (attr->ia_valid & ATTR_MTIME)
inode->i_mtime = attr->ia_mtime;
if (attr->ia_valid & ATTR_CTIME)
- inode->i_ctime = attr->ia_ctime;
+ inode_ctime_set(inode, attr->ia_ctime);
if (attr->ia_valid & ATTR_MODE) {
umode_t mode = attr->ia_mode;
@@ -1192,7 +1192,7 @@ static int do_truncation(struct ubifs_info *c, struct inode *inode,
mutex_lock(&ui->ui_mutex);
ui->ui_size = inode->i_size;
/* Truncation changes inode [mc]time */
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
/* Other attributes may be changed at the same time as well */
do_attr_changes(inode, attr);
err = ubifs_jnl_truncate(c, inode, old_size, new_size);
@@ -1239,7 +1239,7 @@ static int do_setattr(struct ubifs_info *c, struct inode *inode,
mutex_lock(&ui->ui_mutex);
if (attr->ia_valid & ATTR_SIZE) {
/* Truncation changes inode [mc]time */
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
/* 'truncate_setsize()' changed @i_size, update @ui_size */
ui->ui_size = inode->i_size;
}
@@ -1364,8 +1364,10 @@ int ubifs_fsync(struct file *file, loff_t start, loff_t end, int datasync)
static inline int mctime_update_needed(const struct inode *inode,
const struct timespec64 *now)
{
+ struct timespec64 ctime = inode_ctime_peek(inode);
+
if (!timespec64_equal(&inode->i_mtime, now) ||
- !timespec64_equal(&inode->i_ctime, now))
+ !timespec64_equal(&ctime, now))
return 1;
return 0;
}
@@ -1396,7 +1398,7 @@ int ubifs_update_time(struct inode *inode, struct timespec64 *time,
if (flags & S_ATIME)
inode->i_atime = *time;
if (flags & S_CTIME)
- inode->i_ctime = *time;
+ inode_ctime_set(inode, *time);
if (flags & S_MTIME)
inode->i_mtime = *time;
@@ -1432,7 +1434,7 @@ static int update_mctime(struct inode *inode)
return err;
mutex_lock(&ui->ui_mutex);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
release = ui->dirty;
mark_inode_dirty_sync(inode);
mutex_unlock(&ui->ui_mutex);
@@ -1570,7 +1572,7 @@ static vm_fault_t ubifs_vm_page_mkwrite(struct vm_fault *vmf)
struct ubifs_inode *ui = ubifs_inode(inode);
mutex_lock(&ui->ui_mutex);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
release = ui->dirty;
mark_inode_dirty_sync(inode);
mutex_unlock(&ui->ui_mutex);
diff --git a/fs/ubifs/ioctl.c b/fs/ubifs/ioctl.c
index 67c5108abd89..a4da17f4cceb 100644
--- a/fs/ubifs/ioctl.c
+++ b/fs/ubifs/ioctl.c
@@ -118,7 +118,7 @@ static int setflags(struct inode *inode, int flags)
ui->flags &= ~ioctl2ubifs(UBIFS_SETTABLE_IOCTL_FLAGS);
ui->flags |= ioctl2ubifs(flags);
ubifs_set_inode_flags(inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
release = ui->dirty;
mark_inode_dirty_sync(inode);
mutex_unlock(&ui->ui_mutex);
diff --git a/fs/ubifs/journal.c b/fs/ubifs/journal.c
index dc52ac0f4a34..86f5d73cd613 100644
--- a/fs/ubifs/journal.c
+++ b/fs/ubifs/journal.c
@@ -454,8 +454,8 @@ static void pack_inode(struct ubifs_info *c, struct ubifs_ino_node *ino,
ino->creat_sqnum = cpu_to_le64(ui->creat_sqnum);
ino->atime_sec = cpu_to_le64(inode->i_atime.tv_sec);
ino->atime_nsec = cpu_to_le32(inode->i_atime.tv_nsec);
- ino->ctime_sec = cpu_to_le64(inode->i_ctime.tv_sec);
- ino->ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
+ ino->ctime_sec = cpu_to_le64(inode_ctime_peek(inode).tv_sec);
+ ino->ctime_nsec = cpu_to_le32(inode_ctime_peek(inode).tv_nsec);
ino->mtime_sec = cpu_to_le64(inode->i_mtime.tv_sec);
ino->mtime_nsec = cpu_to_le32(inode->i_mtime.tv_nsec);
ino->uid = cpu_to_le32(i_uid_read(inode));
diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c
index 32cb14759796..29c2f40acac9 100644
--- a/fs/ubifs/super.c
+++ b/fs/ubifs/super.c
@@ -146,8 +146,8 @@ struct inode *ubifs_iget(struct super_block *sb, unsigned long inum)
inode->i_atime.tv_nsec = le32_to_cpu(ino->atime_nsec);
inode->i_mtime.tv_sec = (int64_t)le64_to_cpu(ino->mtime_sec);
inode->i_mtime.tv_nsec = le32_to_cpu(ino->mtime_nsec);
- inode->i_ctime.tv_sec = (int64_t)le64_to_cpu(ino->ctime_sec);
- inode->i_ctime.tv_nsec = le32_to_cpu(ino->ctime_nsec);
+ inode_ctime_set_sec(inode, (int64_t)le64_to_cpu(ino->ctime_sec));
+ inode_ctime_set_nsec(inode, le32_to_cpu(ino->ctime_nsec));
inode->i_mode = le32_to_cpu(ino->mode);
inode->i_size = le64_to_cpu(ino->size);
diff --git a/fs/ubifs/xattr.c b/fs/ubifs/xattr.c
index 349228dd1191..d5955e1fbb7d 100644
--- a/fs/ubifs/xattr.c
+++ b/fs/ubifs/xattr.c
@@ -134,7 +134,7 @@ static int create_xattr(struct ubifs_info *c, struct inode *host,
ui->data_len = size;
mutex_lock(&host_ui->ui_mutex);
- host->i_ctime = current_time(host);
+ inode_ctime_set_current(host);
host_ui->xattr_cnt += 1;
host_ui->xattr_size += CALC_DENT_SIZE(fname_len(nm));
host_ui->xattr_size += CALC_XATTR_BYTES(size);
@@ -215,7 +215,7 @@ static int change_xattr(struct ubifs_info *c, struct inode *host,
ui->data_len = size;
mutex_lock(&host_ui->ui_mutex);
- host->i_ctime = current_time(host);
+ inode_ctime_set_current(host);
host_ui->xattr_size -= CALC_XATTR_BYTES(old_size);
host_ui->xattr_size += CALC_XATTR_BYTES(size);
@@ -474,7 +474,7 @@ static int remove_xattr(struct ubifs_info *c, struct inode *host,
return err;
mutex_lock(&host_ui->ui_mutex);
- host->i_ctime = current_time(host);
+ inode_ctime_set_current(host);
host_ui->xattr_cnt -= 1;
host_ui->xattr_size -= CALC_DENT_SIZE(fname_len(nm));
host_ui->xattr_size -= CALC_XATTR_BYTES(ui->data_len);
--
2.41.0
______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 66/79] ubifs: switch to new ctime accessors
@ 2023-06-21 14:46 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Richard Weinberger
Cc: Al Viro, Jan Kara, linux-mtd, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/ubifs/debug.c | 4 ++--
fs/ubifs/dir.c | 39 +++++++++++++++++++--------------------
fs/ubifs/file.c | 16 +++++++++-------
fs/ubifs/ioctl.c | 2 +-
fs/ubifs/journal.c | 4 ++--
fs/ubifs/super.c | 4 ++--
fs/ubifs/xattr.c | 6 +++---
7 files changed, 38 insertions(+), 37 deletions(-)
diff --git a/fs/ubifs/debug.c b/fs/ubifs/debug.c
index 9c9d3f0e36a4..114c2b00aa29 100644
--- a/fs/ubifs/debug.c
+++ b/fs/ubifs/debug.c
@@ -243,8 +243,8 @@ void ubifs_dump_inode(struct ubifs_info *c, const struct inode *inode)
(unsigned int)inode->i_mtime.tv_sec,
(unsigned int)inode->i_mtime.tv_nsec);
pr_err("\tctime %u.%u\n",
- (unsigned int)inode->i_ctime.tv_sec,
- (unsigned int)inode->i_ctime.tv_nsec);
+ (unsigned int) inode_ctime_peek(inode).tv_sec,
+ (unsigned int) inode_ctime_peek(inode).tv_nsec);
pr_err("\tcreat_sqnum %llu\n", ui->creat_sqnum);
pr_err("\txattr_size %u\n", ui->xattr_size);
pr_err("\txattr_cnt %u\n", ui->xattr_cnt);
diff --git a/fs/ubifs/dir.c b/fs/ubifs/dir.c
index ef0499edc248..39588e3e3a8b 100644
--- a/fs/ubifs/dir.c
+++ b/fs/ubifs/dir.c
@@ -96,8 +96,7 @@ struct inode *ubifs_new_inode(struct ubifs_info *c, struct inode *dir,
inode->i_flags |= S_NOCMTIME;
inode_init_owner(&nop_mnt_idmap, inode, dir, mode);
- inode->i_mtime = inode->i_atime = inode->i_ctime =
- current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
inode->i_mapping->nrpages = 0;
if (!is_xattr) {
@@ -325,7 +324,7 @@ static int ubifs_create(struct mnt_idmap *idmap, struct inode *dir,
mutex_lock(&dir_ui->ui_mutex);
dir->i_size += sz_change;
dir_ui->ui_size = dir->i_size;
- dir->i_mtime = dir->i_ctime = inode->i_ctime;
+ dir->i_mtime = inode_ctime_set(inode, inode_ctime_peek(inode));
err = ubifs_jnl_update(c, dir, &nm, inode, 0, 0);
if (err)
goto out_cancel;
@@ -765,10 +764,10 @@ static int ubifs_link(struct dentry *old_dentry, struct inode *dir,
inc_nlink(inode);
ihold(inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
dir->i_size += sz_change;
dir_ui->ui_size = dir->i_size;
- dir->i_mtime = dir->i_ctime = inode->i_ctime;
+ dir->i_mtime = inode_ctime_set(inode, inode_ctime_peek(inode));
err = ubifs_jnl_update(c, dir, &nm, inode, 0, 0);
if (err)
goto out_cancel;
@@ -838,11 +837,11 @@ static int ubifs_unlink(struct inode *dir, struct dentry *dentry)
}
lock_2_inodes(dir, inode);
- inode->i_ctime = current_time(dir);
+ inode_ctime_set(inode, current_time(dir));
drop_nlink(inode);
dir->i_size -= sz_change;
dir_ui->ui_size = dir->i_size;
- dir->i_mtime = dir->i_ctime = inode->i_ctime;
+ dir->i_mtime = inode_ctime_set(inode, inode_ctime_peek(inode));
err = ubifs_jnl_update(c, dir, &nm, inode, 1, 0);
if (err)
goto out_cancel;
@@ -940,12 +939,12 @@ static int ubifs_rmdir(struct inode *dir, struct dentry *dentry)
}
lock_2_inodes(dir, inode);
- inode->i_ctime = current_time(dir);
+ inode_ctime_set(inode, current_time(dir));
clear_nlink(inode);
drop_nlink(dir);
dir->i_size -= sz_change;
dir_ui->ui_size = dir->i_size;
- dir->i_mtime = dir->i_ctime = inode->i_ctime;
+ dir->i_mtime = inode_ctime_set(inode, inode_ctime_peek(inode));
err = ubifs_jnl_update(c, dir, &nm, inode, 1, 0);
if (err)
goto out_cancel;
@@ -1019,7 +1018,7 @@ static int ubifs_mkdir(struct mnt_idmap *idmap, struct inode *dir,
inc_nlink(dir);
dir->i_size += sz_change;
dir_ui->ui_size = dir->i_size;
- dir->i_mtime = dir->i_ctime = inode->i_ctime;
+ dir->i_mtime = inode_ctime_set(inode, inode_ctime_peek(inode));
err = ubifs_jnl_update(c, dir, &nm, inode, 0, 0);
if (err) {
ubifs_err(c, "cannot create directory, error %d", err);
@@ -1110,7 +1109,7 @@ static int ubifs_mknod(struct mnt_idmap *idmap, struct inode *dir,
mutex_lock(&dir_ui->ui_mutex);
dir->i_size += sz_change;
dir_ui->ui_size = dir->i_size;
- dir->i_mtime = dir->i_ctime = inode->i_ctime;
+ dir->i_mtime = inode_ctime_set(inode, inode_ctime_peek(inode));
err = ubifs_jnl_update(c, dir, &nm, inode, 0, 0);
if (err)
goto out_cancel;
@@ -1210,7 +1209,7 @@ static int ubifs_symlink(struct mnt_idmap *idmap, struct inode *dir,
mutex_lock(&dir_ui->ui_mutex);
dir->i_size += sz_change;
dir_ui->ui_size = dir->i_size;
- dir->i_mtime = dir->i_ctime = inode->i_ctime;
+ dir->i_mtime = inode_ctime_set(inode, inode_ctime_peek(inode));
err = ubifs_jnl_update(c, dir, &nm, inode, 0, 0);
if (err)
goto out_cancel;
@@ -1415,7 +1414,7 @@ static int do_rename(struct inode *old_dir, struct dentry *old_dentry,
* rename.
*/
time = current_time(old_dir);
- old_inode->i_ctime = time;
+ inode_ctime_set(old_inode, time);
/* We must adjust parent link count when renaming directories */
if (is_dir) {
@@ -1444,8 +1443,8 @@ static int do_rename(struct inode *old_dir, struct dentry *old_dentry,
old_dir->i_size -= old_sz;
ubifs_inode(old_dir)->ui_size = old_dir->i_size;
- old_dir->i_mtime = old_dir->i_ctime = time;
- new_dir->i_mtime = new_dir->i_ctime = time;
+ old_dir->i_mtime = inode_ctime_set(old_dir, time);
+ new_dir->i_mtime = inode_ctime_set(new_dir, time);
/*
* And finally, if we unlinked a direntry which happened to have the
@@ -1462,7 +1461,7 @@ static int do_rename(struct inode *old_dir, struct dentry *old_dentry,
clear_nlink(new_inode);
else
drop_nlink(new_inode);
- new_inode->i_ctime = time;
+ inode_ctime_set(new_inode, time);
} else {
new_dir->i_size += new_sz;
ubifs_inode(new_dir)->ui_size = new_dir->i_size;
@@ -1589,10 +1588,10 @@ static int ubifs_xrename(struct inode *old_dir, struct dentry *old_dentry,
lock_4_inodes(old_dir, new_dir, NULL, NULL);
time = current_time(old_dir);
- fst_inode->i_ctime = time;
- snd_inode->i_ctime = time;
- old_dir->i_mtime = old_dir->i_ctime = time;
- new_dir->i_mtime = new_dir->i_ctime = time;
+ inode_ctime_set(fst_inode, time);
+ inode_ctime_set(snd_inode, time);
+ old_dir->i_mtime = inode_ctime_set(old_dir, time);
+ new_dir->i_mtime = inode_ctime_set(new_dir, time);
if (old_dir != new_dir) {
if (S_ISDIR(fst_inode->i_mode) && !S_ISDIR(snd_inode->i_mode)) {
diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c
index 6738fe43040b..f9ba69f6fbab 100644
--- a/fs/ubifs/file.c
+++ b/fs/ubifs/file.c
@@ -1092,7 +1092,7 @@ static void do_attr_changes(struct inode *inode, const struct iattr *attr)
if (attr->ia_valid & ATTR_MTIME)
inode->i_mtime = attr->ia_mtime;
if (attr->ia_valid & ATTR_CTIME)
- inode->i_ctime = attr->ia_ctime;
+ inode_ctime_set(inode, attr->ia_ctime);
if (attr->ia_valid & ATTR_MODE) {
umode_t mode = attr->ia_mode;
@@ -1192,7 +1192,7 @@ static int do_truncation(struct ubifs_info *c, struct inode *inode,
mutex_lock(&ui->ui_mutex);
ui->ui_size = inode->i_size;
/* Truncation changes inode [mc]time */
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
/* Other attributes may be changed at the same time as well */
do_attr_changes(inode, attr);
err = ubifs_jnl_truncate(c, inode, old_size, new_size);
@@ -1239,7 +1239,7 @@ static int do_setattr(struct ubifs_info *c, struct inode *inode,
mutex_lock(&ui->ui_mutex);
if (attr->ia_valid & ATTR_SIZE) {
/* Truncation changes inode [mc]time */
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
/* 'truncate_setsize()' changed @i_size, update @ui_size */
ui->ui_size = inode->i_size;
}
@@ -1364,8 +1364,10 @@ int ubifs_fsync(struct file *file, loff_t start, loff_t end, int datasync)
static inline int mctime_update_needed(const struct inode *inode,
const struct timespec64 *now)
{
+ struct timespec64 ctime = inode_ctime_peek(inode);
+
if (!timespec64_equal(&inode->i_mtime, now) ||
- !timespec64_equal(&inode->i_ctime, now))
+ !timespec64_equal(&ctime, now))
return 1;
return 0;
}
@@ -1396,7 +1398,7 @@ int ubifs_update_time(struct inode *inode, struct timespec64 *time,
if (flags & S_ATIME)
inode->i_atime = *time;
if (flags & S_CTIME)
- inode->i_ctime = *time;
+ inode_ctime_set(inode, *time);
if (flags & S_MTIME)
inode->i_mtime = *time;
@@ -1432,7 +1434,7 @@ static int update_mctime(struct inode *inode)
return err;
mutex_lock(&ui->ui_mutex);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
release = ui->dirty;
mark_inode_dirty_sync(inode);
mutex_unlock(&ui->ui_mutex);
@@ -1570,7 +1572,7 @@ static vm_fault_t ubifs_vm_page_mkwrite(struct vm_fault *vmf)
struct ubifs_inode *ui = ubifs_inode(inode);
mutex_lock(&ui->ui_mutex);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
release = ui->dirty;
mark_inode_dirty_sync(inode);
mutex_unlock(&ui->ui_mutex);
diff --git a/fs/ubifs/ioctl.c b/fs/ubifs/ioctl.c
index 67c5108abd89..a4da17f4cceb 100644
--- a/fs/ubifs/ioctl.c
+++ b/fs/ubifs/ioctl.c
@@ -118,7 +118,7 @@ static int setflags(struct inode *inode, int flags)
ui->flags &= ~ioctl2ubifs(UBIFS_SETTABLE_IOCTL_FLAGS);
ui->flags |= ioctl2ubifs(flags);
ubifs_set_inode_flags(inode);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
release = ui->dirty;
mark_inode_dirty_sync(inode);
mutex_unlock(&ui->ui_mutex);
diff --git a/fs/ubifs/journal.c b/fs/ubifs/journal.c
index dc52ac0f4a34..86f5d73cd613 100644
--- a/fs/ubifs/journal.c
+++ b/fs/ubifs/journal.c
@@ -454,8 +454,8 @@ static void pack_inode(struct ubifs_info *c, struct ubifs_ino_node *ino,
ino->creat_sqnum = cpu_to_le64(ui->creat_sqnum);
ino->atime_sec = cpu_to_le64(inode->i_atime.tv_sec);
ino->atime_nsec = cpu_to_le32(inode->i_atime.tv_nsec);
- ino->ctime_sec = cpu_to_le64(inode->i_ctime.tv_sec);
- ino->ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);
+ ino->ctime_sec = cpu_to_le64(inode_ctime_peek(inode).tv_sec);
+ ino->ctime_nsec = cpu_to_le32(inode_ctime_peek(inode).tv_nsec);
ino->mtime_sec = cpu_to_le64(inode->i_mtime.tv_sec);
ino->mtime_nsec = cpu_to_le32(inode->i_mtime.tv_nsec);
ino->uid = cpu_to_le32(i_uid_read(inode));
diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c
index 32cb14759796..29c2f40acac9 100644
--- a/fs/ubifs/super.c
+++ b/fs/ubifs/super.c
@@ -146,8 +146,8 @@ struct inode *ubifs_iget(struct super_block *sb, unsigned long inum)
inode->i_atime.tv_nsec = le32_to_cpu(ino->atime_nsec);
inode->i_mtime.tv_sec = (int64_t)le64_to_cpu(ino->mtime_sec);
inode->i_mtime.tv_nsec = le32_to_cpu(ino->mtime_nsec);
- inode->i_ctime.tv_sec = (int64_t)le64_to_cpu(ino->ctime_sec);
- inode->i_ctime.tv_nsec = le32_to_cpu(ino->ctime_nsec);
+ inode_ctime_set_sec(inode, (int64_t)le64_to_cpu(ino->ctime_sec));
+ inode_ctime_set_nsec(inode, le32_to_cpu(ino->ctime_nsec));
inode->i_mode = le32_to_cpu(ino->mode);
inode->i_size = le64_to_cpu(ino->size);
diff --git a/fs/ubifs/xattr.c b/fs/ubifs/xattr.c
index 349228dd1191..d5955e1fbb7d 100644
--- a/fs/ubifs/xattr.c
+++ b/fs/ubifs/xattr.c
@@ -134,7 +134,7 @@ static int create_xattr(struct ubifs_info *c, struct inode *host,
ui->data_len = size;
mutex_lock(&host_ui->ui_mutex);
- host->i_ctime = current_time(host);
+ inode_ctime_set_current(host);
host_ui->xattr_cnt += 1;
host_ui->xattr_size += CALC_DENT_SIZE(fname_len(nm));
host_ui->xattr_size += CALC_XATTR_BYTES(size);
@@ -215,7 +215,7 @@ static int change_xattr(struct ubifs_info *c, struct inode *host,
ui->data_len = size;
mutex_lock(&host_ui->ui_mutex);
- host->i_ctime = current_time(host);
+ inode_ctime_set_current(host);
host_ui->xattr_size -= CALC_XATTR_BYTES(old_size);
host_ui->xattr_size += CALC_XATTR_BYTES(size);
@@ -474,7 +474,7 @@ static int remove_xattr(struct ubifs_info *c, struct inode *host,
return err;
mutex_lock(&host_ui->ui_mutex);
- host->i_ctime = current_time(host);
+ inode_ctime_set_current(host);
host_ui->xattr_cnt -= 1;
host_ui->xattr_size -= CALC_DENT_SIZE(fname_len(nm));
host_ui->xattr_size -= CALC_XATTR_BYTES(ui->data_len);
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 67/79] udf: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (64 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Jan Kara; +Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/udf/ialloc.c | 2 +-
fs/udf/inode.c | 17 ++++++++++-------
fs/udf/namei.c | 24 ++++++++++++------------
3 files changed, 23 insertions(+), 20 deletions(-)
diff --git a/fs/udf/ialloc.c b/fs/udf/ialloc.c
index 5f7ac8c84798..6320b15113e4 100644
--- a/fs/udf/ialloc.c
+++ b/fs/udf/ialloc.c
@@ -100,7 +100,7 @@ struct inode *udf_new_inode(struct inode *dir, umode_t mode)
iinfo->i_alloc_type = ICBTAG_FLAG_AD_SHORT;
else
iinfo->i_alloc_type = ICBTAG_FLAG_AD_LONG;
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
iinfo->i_crtime = inode->i_mtime;
if (unlikely(insert_inode_locked(inode) < 0)) {
make_bad_inode(inode);
diff --git a/fs/udf/inode.c b/fs/udf/inode.c
index 28cdfc57d946..4c31a51e0d18 100644
--- a/fs/udf/inode.c
+++ b/fs/udf/inode.c
@@ -910,7 +910,7 @@ static int inode_getblk(struct inode *inode, struct udf_map_rq *map)
map->oflags = UDF_BLK_NEW | UDF_BLK_MAPPED;
iinfo->i_next_alloc_block = map->lblk + 1;
iinfo->i_next_alloc_goal = newblocknum + 1;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
if (IS_SYNC(inode))
udf_sync_inode(inode);
@@ -1298,7 +1298,7 @@ int udf_setsize(struct inode *inode, loff_t newsize)
goto out_unlock;
}
update_time:
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
if (IS_SYNC(inode))
udf_sync_inode(inode);
else
@@ -1329,6 +1329,7 @@ static int udf_read_inode(struct inode *inode, bool hidden_inode)
int bs = inode->i_sb->s_blocksize;
int ret = -EIO;
uint32_t uid, gid;
+ struct timespec64 ctime;
reread:
if (iloc->partitionReferenceNum >= sbi->s_partitions) {
@@ -1501,13 +1502,14 @@ static int udf_read_inode(struct inode *inode, bool hidden_inode)
inode->i_size = le64_to_cpu(fe->informationLength);
iinfo->i_lenExtents = inode->i_size;
+ ctime = inode_ctime_peek(inode);
if (iinfo->i_efe == 0) {
inode->i_blocks = le64_to_cpu(fe->logicalBlocksRecorded) <<
(inode->i_sb->s_blocksize_bits - 9);
udf_disk_stamp_to_time(&inode->i_atime, fe->accessTime);
udf_disk_stamp_to_time(&inode->i_mtime, fe->modificationTime);
- udf_disk_stamp_to_time(&inode->i_ctime, fe->attrTime);
+ udf_disk_stamp_to_time(&ctime, fe->attrTime);
iinfo->i_unique = le64_to_cpu(fe->uniqueID);
iinfo->i_lenEAttr = le32_to_cpu(fe->lengthExtendedAttr);
@@ -1522,7 +1524,7 @@ static int udf_read_inode(struct inode *inode, bool hidden_inode)
udf_disk_stamp_to_time(&inode->i_atime, efe->accessTime);
udf_disk_stamp_to_time(&inode->i_mtime, efe->modificationTime);
udf_disk_stamp_to_time(&iinfo->i_crtime, efe->createTime);
- udf_disk_stamp_to_time(&inode->i_ctime, efe->attrTime);
+ udf_disk_stamp_to_time(&ctime, efe->attrTime);
iinfo->i_unique = le64_to_cpu(efe->uniqueID);
iinfo->i_lenEAttr = le32_to_cpu(efe->lengthExtendedAttr);
@@ -1799,7 +1801,7 @@ static int udf_update_inode(struct inode *inode, int do_sync)
udf_time_to_disk_stamp(&fe->accessTime, inode->i_atime);
udf_time_to_disk_stamp(&fe->modificationTime, inode->i_mtime);
- udf_time_to_disk_stamp(&fe->attrTime, inode->i_ctime);
+ udf_time_to_disk_stamp(&fe->attrTime, inode_ctime_peek(inode));
memset(&(fe->impIdent), 0, sizeof(struct regid));
strcpy(fe->impIdent.ident, UDF_ID_DEVELOPER);
fe->impIdent.identSuffix[0] = UDF_OS_CLASS_UNIX;
@@ -1830,12 +1832,13 @@ static int udf_update_inode(struct inode *inode, int do_sync)
udf_adjust_time(iinfo, inode->i_atime);
udf_adjust_time(iinfo, inode->i_mtime);
- udf_adjust_time(iinfo, inode->i_ctime);
+ udf_adjust_time(iinfo, inode_ctime_peek(inode));
udf_time_to_disk_stamp(&efe->accessTime, inode->i_atime);
udf_time_to_disk_stamp(&efe->modificationTime, inode->i_mtime);
udf_time_to_disk_stamp(&efe->createTime, iinfo->i_crtime);
- udf_time_to_disk_stamp(&efe->attrTime, inode->i_ctime);
+ udf_time_to_disk_stamp(&efe->attrTime,
+ inode_ctime_peek(inode));
memset(&(efe->impIdent), 0, sizeof(efe->impIdent));
strcpy(efe->impIdent.ident, UDF_ID_DEVELOPER);
diff --git a/fs/udf/namei.c b/fs/udf/namei.c
index a95579b043ab..50ee82a7d910 100644
--- a/fs/udf/namei.c
+++ b/fs/udf/namei.c
@@ -365,7 +365,7 @@ static int udf_add_nondir(struct dentry *dentry, struct inode *inode)
*(__le32 *)((struct allocDescImpUse *)iter.fi.icb.impUse)->impUse =
cpu_to_le32(iinfo->i_unique & 0x00000000FFFFFFFFUL);
udf_fiiter_write_fi(&iter, NULL);
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
mark_inode_dirty(dir);
udf_fiiter_release(&iter);
udf_add_fid_counter(dir->i_sb, false, 1);
@@ -471,7 +471,7 @@ static int udf_mkdir(struct mnt_idmap *idmap, struct inode *dir,
udf_fiiter_release(&iter);
udf_add_fid_counter(dir->i_sb, true, 1);
inc_nlink(dir);
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
mark_inode_dirty(dir);
d_instantiate_new(dentry, inode);
@@ -523,8 +523,8 @@ static int udf_rmdir(struct inode *dir, struct dentry *dentry)
inode->i_size = 0;
inode_dec_link_count(dir);
udf_add_fid_counter(dir->i_sb, true, -1);
- inode->i_ctime = dir->i_ctime = dir->i_mtime =
- current_time(inode);
+ inode_ctime_set_current(inode);
+ dir->i_mtime = inode_ctime_set_current(dir);
mark_inode_dirty(dir);
ret = 0;
end_rmdir:
@@ -555,11 +555,11 @@ static int udf_unlink(struct inode *dir, struct dentry *dentry)
set_nlink(inode, 1);
}
udf_fiiter_delete_entry(&iter);
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
mark_inode_dirty(dir);
inode_dec_link_count(inode);
udf_add_fid_counter(dir->i_sb, false, -1);
- inode->i_ctime = dir->i_ctime;
+ inode_ctime_set(inode, inode_ctime_peek(dir));
ret = 0;
end_unlink:
udf_fiiter_release(&iter);
@@ -746,9 +746,9 @@ static int udf_link(struct dentry *old_dentry, struct inode *dir,
inc_nlink(inode);
udf_add_fid_counter(dir->i_sb, false, 1);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
mark_inode_dirty(inode);
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
mark_inode_dirty(dir);
ihold(inode);
d_instantiate(dentry, inode);
@@ -833,7 +833,7 @@ static int udf_rename(struct mnt_idmap *idmap, struct inode *old_dir,
* Like most other Unix systems, set the ctime for inodes on a
* rename.
*/
- old_inode->i_ctime = current_time(old_inode);
+ inode_ctime_set_current(old_inode);
mark_inode_dirty(old_inode);
/*
@@ -861,13 +861,13 @@ static int udf_rename(struct mnt_idmap *idmap, struct inode *old_dir,
}
if (new_inode) {
- new_inode->i_ctime = current_time(new_inode);
+ inode_ctime_set_current(new_inode);
inode_dec_link_count(new_inode);
udf_add_fid_counter(old_dir->i_sb, S_ISDIR(new_inode->i_mode),
-1);
}
- old_dir->i_ctime = old_dir->i_mtime = current_time(old_dir);
- new_dir->i_ctime = new_dir->i_mtime = current_time(new_dir);
+ old_dir->i_mtime = inode_ctime_set_current(old_dir);
+ new_dir->i_mtime = inode_ctime_set_current(new_dir);
mark_inode_dirty(old_dir);
mark_inode_dirty(new_dir);
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 68/79] ufs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (65 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Evgeniy Dushistov; +Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/ufs/dir.c | 6 +++---
fs/ufs/ialloc.c | 2 +-
fs/ufs/inode.c | 23 +++++++++++++----------
fs/ufs/namei.c | 8 ++++----
4 files changed, 21 insertions(+), 18 deletions(-)
diff --git a/fs/ufs/dir.c b/fs/ufs/dir.c
index 379d75796a5c..0ccd32ab7ffc 100644
--- a/fs/ufs/dir.c
+++ b/fs/ufs/dir.c
@@ -107,7 +107,7 @@ void ufs_set_link(struct inode *dir, struct ufs_dir_entry *de,
ufs_commit_chunk(page, pos, len);
ufs_put_page(page);
if (update_times)
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
mark_inode_dirty(dir);
ufs_handle_dirsync(dir);
}
@@ -397,7 +397,7 @@ int ufs_add_link(struct dentry *dentry, struct inode *inode)
ufs_set_de_type(sb, de, inode->i_mode);
ufs_commit_chunk(page, pos, rec_len);
- dir->i_mtime = dir->i_ctime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
mark_inode_dirty(dir);
err = ufs_handle_dirsync(dir);
@@ -539,7 +539,7 @@ int ufs_delete_entry(struct inode *inode, struct ufs_dir_entry *dir,
pde->d_reclen = cpu_to_fs16(sb, to - from);
dir->d_ino = 0;
ufs_commit_chunk(page, pos, to - from);
- inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
mark_inode_dirty(inode);
err = ufs_handle_dirsync(inode);
out:
diff --git a/fs/ufs/ialloc.c b/fs/ufs/ialloc.c
index 06bd84d555bd..3d64f7b3aba1 100644
--- a/fs/ufs/ialloc.c
+++ b/fs/ufs/ialloc.c
@@ -292,7 +292,7 @@ struct inode *ufs_new_inode(struct inode *dir, umode_t mode)
inode_init_owner(&nop_mnt_idmap, inode, dir, mode);
inode->i_blocks = 0;
inode->i_generation = 0;
- inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
ufsi->i_flags = UFS_I(dir)->i_flags;
ufsi->i_lastfrag = 0;
ufsi->i_shadow = 0;
diff --git a/fs/ufs/inode.c b/fs/ufs/inode.c
index a4246c83a8cd..64242e1378d6 100644
--- a/fs/ufs/inode.c
+++ b/fs/ufs/inode.c
@@ -296,7 +296,7 @@ ufs_inode_getfrag(struct inode *inode, unsigned index,
if (new)
*new = 1;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
if (IS_SYNC(inode))
ufs_sync_inode (inode);
mark_inode_dirty(inode);
@@ -378,7 +378,7 @@ ufs_inode_getblock(struct inode *inode, u64 ind_block,
mark_buffer_dirty(bh);
if (IS_SYNC(inode))
sync_dirty_buffer(bh);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
mark_inode_dirty(inode);
out:
brelse (bh);
@@ -580,11 +580,12 @@ static int ufs1_read_inode(struct inode *inode, struct ufs_inode *ufs_inode)
inode->i_size = fs64_to_cpu(sb, ufs_inode->ui_size);
inode->i_atime.tv_sec = (signed)fs32_to_cpu(sb, ufs_inode->ui_atime.tv_sec);
- inode->i_ctime.tv_sec = (signed)fs32_to_cpu(sb, ufs_inode->ui_ctime.tv_sec);
+ inode_ctime_set_sec(inode,
+ (signed)fs32_to_cpu(sb, ufs_inode->ui_ctime.tv_sec));
inode->i_mtime.tv_sec = (signed)fs32_to_cpu(sb, ufs_inode->ui_mtime.tv_sec);
inode->i_mtime.tv_nsec = 0;
inode->i_atime.tv_nsec = 0;
- inode->i_ctime.tv_nsec = 0;
+ inode_ctime_set_nsec(inode, 0);
inode->i_blocks = fs32_to_cpu(sb, ufs_inode->ui_blocks);
inode->i_generation = fs32_to_cpu(sb, ufs_inode->ui_gen);
ufsi->i_flags = fs32_to_cpu(sb, ufs_inode->ui_flags);
@@ -626,10 +627,10 @@ static int ufs2_read_inode(struct inode *inode, struct ufs2_inode *ufs2_inode)
inode->i_size = fs64_to_cpu(sb, ufs2_inode->ui_size);
inode->i_atime.tv_sec = fs64_to_cpu(sb, ufs2_inode->ui_atime);
- inode->i_ctime.tv_sec = fs64_to_cpu(sb, ufs2_inode->ui_ctime);
+ inode_ctime_set_sec(inode, fs64_to_cpu(sb, ufs2_inode->ui_ctime));
inode->i_mtime.tv_sec = fs64_to_cpu(sb, ufs2_inode->ui_mtime);
inode->i_atime.tv_nsec = fs32_to_cpu(sb, ufs2_inode->ui_atimensec);
- inode->i_ctime.tv_nsec = fs32_to_cpu(sb, ufs2_inode->ui_ctimensec);
+ inode_ctime_set_nsec(inode, fs32_to_cpu(sb, ufs2_inode->ui_ctimensec));
inode->i_mtime.tv_nsec = fs32_to_cpu(sb, ufs2_inode->ui_mtimensec);
inode->i_blocks = fs64_to_cpu(sb, ufs2_inode->ui_blocks);
inode->i_generation = fs32_to_cpu(sb, ufs2_inode->ui_gen);
@@ -726,7 +727,8 @@ static void ufs1_update_inode(struct inode *inode, struct ufs_inode *ufs_inode)
ufs_inode->ui_size = cpu_to_fs64(sb, inode->i_size);
ufs_inode->ui_atime.tv_sec = cpu_to_fs32(sb, inode->i_atime.tv_sec);
ufs_inode->ui_atime.tv_usec = 0;
- ufs_inode->ui_ctime.tv_sec = cpu_to_fs32(sb, inode->i_ctime.tv_sec);
+ ufs_inode->ui_ctime.tv_sec = cpu_to_fs32(sb,
+ inode_ctime_peek(inode).tv_sec);
ufs_inode->ui_ctime.tv_usec = 0;
ufs_inode->ui_mtime.tv_sec = cpu_to_fs32(sb, inode->i_mtime.tv_sec);
ufs_inode->ui_mtime.tv_usec = 0;
@@ -770,8 +772,9 @@ static void ufs2_update_inode(struct inode *inode, struct ufs2_inode *ufs_inode)
ufs_inode->ui_size = cpu_to_fs64(sb, inode->i_size);
ufs_inode->ui_atime = cpu_to_fs64(sb, inode->i_atime.tv_sec);
ufs_inode->ui_atimensec = cpu_to_fs32(sb, inode->i_atime.tv_nsec);
- ufs_inode->ui_ctime = cpu_to_fs64(sb, inode->i_ctime.tv_sec);
- ufs_inode->ui_ctimensec = cpu_to_fs32(sb, inode->i_ctime.tv_nsec);
+ ufs_inode->ui_ctime = cpu_to_fs64(sb, inode_ctime_peek(inode).tv_sec);
+ ufs_inode->ui_ctimensec = cpu_to_fs32(sb,
+ inode_ctime_peek(inode).tv_nsec);
ufs_inode->ui_mtime = cpu_to_fs64(sb, inode->i_mtime.tv_sec);
ufs_inode->ui_mtimensec = cpu_to_fs32(sb, inode->i_mtime.tv_nsec);
@@ -1205,7 +1208,7 @@ static int ufs_truncate(struct inode *inode, loff_t size)
truncate_setsize(inode, size);
ufs_truncate_blocks(inode);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
mark_inode_dirty(inode);
out:
UFSD("EXIT: err %d\n", err);
diff --git a/fs/ufs/namei.c b/fs/ufs/namei.c
index 36154b5aca6d..d3f69e3c6965 100644
--- a/fs/ufs/namei.c
+++ b/fs/ufs/namei.c
@@ -153,7 +153,7 @@ static int ufs_link (struct dentry * old_dentry, struct inode * dir,
struct inode *inode = d_inode(old_dentry);
int error;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
inode_inc_link_count(inode);
ihold(inode);
@@ -220,7 +220,7 @@ static int ufs_unlink(struct inode *dir, struct dentry *dentry)
if (err)
goto out;
- inode->i_ctime = dir->i_ctime;
+ inode_ctime_set(inode, inode_ctime_peek(dir));
inode_dec_link_count(inode);
err = 0;
out:
@@ -282,7 +282,7 @@ static int ufs_rename(struct mnt_idmap *idmap, struct inode *old_dir,
if (!new_de)
goto out_dir;
ufs_set_link(new_dir, new_de, new_page, old_inode, 1);
- new_inode->i_ctime = current_time(new_inode);
+ inode_ctime_set_current(new_inode);
if (dir_de)
drop_nlink(new_inode);
inode_dec_link_count(new_inode);
@@ -298,7 +298,7 @@ static int ufs_rename(struct mnt_idmap *idmap, struct inode *old_dir,
* Like most other Unix systems, set the ctime for inodes on a
* rename.
*/
- old_inode->i_ctime = current_time(old_inode);
+ inode_ctime_set_current(old_inode);
ufs_delete_entry(old_dir, old_de, old_page);
mark_inode_dirty(old_inode);
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 69/79] vboxsf: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (66 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Hans de Goede
Cc: Al Viro, Jan Kara, linux-fsdevel, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/vboxsf/utils.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/vboxsf/utils.c b/fs/vboxsf/utils.c
index dd0ae1188e87..a76dc1ec10f8 100644
--- a/fs/vboxsf/utils.c
+++ b/fs/vboxsf/utils.c
@@ -128,8 +128,8 @@ int vboxsf_init_inode(struct vboxsf_sbi *sbi, struct inode *inode,
inode->i_atime = ns_to_timespec64(
info->access_time.ns_relative_to_unix_epoch);
- inode->i_ctime = ns_to_timespec64(
- info->change_time.ns_relative_to_unix_epoch);
+ inode_ctime_set(inode,
+ ns_to_timespec64(info->change_time.ns_relative_to_unix_epoch));
inode->i_mtime = ns_to_timespec64(
info->modification_time.ns_relative_to_unix_epoch);
return 0;
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 70/79] xfs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (67 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Darrick J. Wong
Cc: Al Viro, Jan Kara, linux-xfs, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/xfs/libxfs/xfs_inode_buf.c | 4 ++--
fs/xfs/libxfs/xfs_trans_inode.c | 2 +-
fs/xfs/xfs_acl.c | 2 +-
fs/xfs/xfs_bmap_util.c | 6 ++++--
fs/xfs/xfs_inode.c | 2 +-
fs/xfs/xfs_inode_item.c | 2 +-
fs/xfs/xfs_iops.c | 4 ++--
fs/xfs/xfs_itable.c | 4 ++--
8 files changed, 14 insertions(+), 12 deletions(-)
diff --git a/fs/xfs/libxfs/xfs_inode_buf.c b/fs/xfs/libxfs/xfs_inode_buf.c
index 758aacd8166b..d5c1d3c86257 100644
--- a/fs/xfs/libxfs/xfs_inode_buf.c
+++ b/fs/xfs/libxfs/xfs_inode_buf.c
@@ -222,7 +222,7 @@ xfs_inode_from_disk(
*/
inode->i_atime = xfs_inode_from_disk_ts(from, from->di_atime);
inode->i_mtime = xfs_inode_from_disk_ts(from, from->di_mtime);
- inode->i_ctime = xfs_inode_from_disk_ts(from, from->di_ctime);
+ inode_ctime_set(inode, xfs_inode_from_disk_ts(from, from->di_ctime));
ip->i_disk_size = be64_to_cpu(from->di_size);
ip->i_nblocks = be64_to_cpu(from->di_nblocks);
@@ -316,7 +316,7 @@ xfs_inode_to_disk(
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_ctime = xfs_inode_to_disk_ts(ip, inode_ctime_peek(inode));
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);
diff --git a/fs/xfs/libxfs/xfs_trans_inode.c b/fs/xfs/libxfs/xfs_trans_inode.c
index cb4796b6e693..f924e81530be 100644
--- a/fs/xfs/libxfs/xfs_trans_inode.c
+++ b/fs/xfs/libxfs/xfs_trans_inode.c
@@ -67,7 +67,7 @@ xfs_trans_ichgtime(
if (flags & XFS_ICHGTIME_MOD)
inode->i_mtime = tv;
if (flags & XFS_ICHGTIME_CHG)
- inode->i_ctime = tv;
+ inode_ctime_set(inode, tv);
if (flags & XFS_ICHGTIME_CREATE)
ip->i_crtime = tv;
}
diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c
index 791db7d9c849..75f17d2bb97c 100644
--- a/fs/xfs/xfs_acl.c
+++ b/fs/xfs/xfs_acl.c
@@ -233,7 +233,7 @@ xfs_acl_set_mode(
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
inode->i_mode = mode;
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
if (xfs_has_wsync(mp))
diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c
index fbb675563208..0d4c141c4413 100644
--- a/fs/xfs/xfs_bmap_util.c
+++ b/fs/xfs/xfs_bmap_util.c
@@ -1644,6 +1644,7 @@ xfs_swap_extents(
uint64_t f;
int resblks = 0;
unsigned int flags = 0;
+ struct timespec64 ctime;
/*
* Lock the inodes against other IO, page faults and truncate to
@@ -1756,8 +1757,9 @@ xfs_swap_extents(
* process that the file was not changed out from
* under it.
*/
- if ((sbp->bs_ctime.tv_sec != VFS_I(ip)->i_ctime.tv_sec) ||
- (sbp->bs_ctime.tv_nsec != VFS_I(ip)->i_ctime.tv_nsec) ||
+ ctime = inode_ctime_peek(VFS_I(ip));
+ if ((sbp->bs_ctime.tv_sec != ctime.tv_sec) ||
+ (sbp->bs_ctime.tv_nsec != ctime.tv_nsec) ||
(sbp->bs_mtime.tv_sec != VFS_I(ip)->i_mtime.tv_sec) ||
(sbp->bs_mtime.tv_nsec != VFS_I(ip)->i_mtime.tv_nsec)) {
error = -EBUSY;
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index 9e62cc500140..763d6c95c56d 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -846,7 +846,7 @@ xfs_init_new_inode(
tv = current_time(inode);
inode->i_mtime = tv;
inode->i_atime = tv;
- inode->i_ctime = tv;
+ inode_ctime_set(inode, tv);
ip->i_extsize = 0;
ip->i_diflags = 0;
diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c
index 91c847a84e10..116038163098 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -528,7 +528,7 @@ xfs_inode_to_log_dinode(
memset(to->di_pad3, 0, sizeof(to->di_pad3));
to->di_atime = xfs_inode_to_log_dinode_ts(ip, inode->i_atime);
to->di_mtime = xfs_inode_to_log_dinode_ts(ip, inode->i_mtime);
- to->di_ctime = xfs_inode_to_log_dinode_ts(ip, inode->i_ctime);
+ to->di_ctime = xfs_inode_to_log_dinode_ts(ip, inode_ctime_peek(inode));
to->di_nlink = inode->i_nlink;
to->di_gen = inode->i_generation;
to->di_mode = inode->i_mode;
diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c
index 24718adb3c16..4096c4f86ef7 100644
--- a/fs/xfs/xfs_iops.c
+++ b/fs/xfs/xfs_iops.c
@@ -574,7 +574,7 @@ xfs_vn_getattr(
stat->ino = ip->i_ino;
stat->atime = inode->i_atime;
stat->mtime = inode->i_mtime;
- stat->ctime = inode->i_ctime;
+ stat->ctime = inode_ctime_peek(inode);
stat->blocks = XFS_FSB_TO_BB(mp, ip->i_nblocks + ip->i_delayed_blks);
if (xfs_has_v3inodes(mp)) {
@@ -1055,7 +1055,7 @@ xfs_vn_update_time(
xfs_ilock(ip, XFS_ILOCK_EXCL);
if (flags & S_CTIME)
- inode->i_ctime = *now;
+ inode_ctime_set(inode, *now);
if (flags & S_MTIME)
inode->i_mtime = *now;
if (flags & S_ATIME)
diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c
index f225413a993c..75b07eed0602 100644
--- a/fs/xfs/xfs_itable.c
+++ b/fs/xfs/xfs_itable.c
@@ -100,8 +100,8 @@ xfs_bulkstat_one_int(
buf->bs_atime_nsec = inode->i_atime.tv_nsec;
buf->bs_mtime = inode->i_mtime.tv_sec;
buf->bs_mtime_nsec = inode->i_mtime.tv_nsec;
- buf->bs_ctime = inode->i_ctime.tv_sec;
- buf->bs_ctime_nsec = inode->i_ctime.tv_nsec;
+ buf->bs_ctime = inode_ctime_peek(inode).tv_sec;
+ buf->bs_ctime_nsec = inode_ctime_peek(inode).tv_nsec;
buf->bs_gen = inode->i_generation;
buf->bs_mode = inode->i_mode;
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 71/79] zonefs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (68 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
2023-06-22 0:48 ` Damien Le Moal
-1 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Damien Le Moal, Naohiro Aota, Johannes Thumshirn
Cc: Al Viro, Jan Kara, linux-fsdevel, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
fs/zonefs/super.c | 8 +++++---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/fs/zonefs/super.c b/fs/zonefs/super.c
index bbe44a26a8e5..75be0e039ccf 100644
--- a/fs/zonefs/super.c
+++ b/fs/zonefs/super.c
@@ -658,7 +658,8 @@ static struct inode *zonefs_get_file_inode(struct inode *dir,
inode->i_ino = ino;
inode->i_mode = z->z_mode;
- inode->i_ctime = inode->i_mtime = inode->i_atime = dir->i_ctime;
+ inode->i_mtime = inode->i_atime = inode_ctime_peek(dir);
+ inode_ctime_set(inode, inode->i_mtime);
inode->i_uid = z->z_uid;
inode->i_gid = z->z_gid;
inode->i_size = z->z_wpoffset;
@@ -694,7 +695,8 @@ static struct inode *zonefs_get_zgroup_inode(struct super_block *sb,
inode->i_ino = ino;
inode_init_owner(&nop_mnt_idmap, inode, root, S_IFDIR | 0555);
inode->i_size = sbi->s_zgroup[ztype].g_nr_zones;
- inode->i_ctime = inode->i_mtime = inode->i_atime = root->i_ctime;
+ inode->i_mtime = inode->i_atime = inode_ctime_peek(root);
+ inode_ctime_set(inode, inode->i_mtime);
inode->i_private = &sbi->s_zgroup[ztype];
set_nlink(inode, 2);
@@ -1317,7 +1319,7 @@ static int zonefs_fill_super(struct super_block *sb, void *data, int silent)
inode->i_ino = bdev_nr_zones(sb->s_bdev);
inode->i_mode = S_IFDIR | 0555;
- inode->i_ctime = inode->i_mtime = inode->i_atime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
inode->i_op = &zonefs_dir_inode_operations;
inode->i_fop = &zonefs_dir_operations;
inode->i_size = 2;
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 72/79] mqueue: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (69 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner; +Cc: Al Viro, Jan Kara, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
ipc/mqueue.c | 20 +++++++++-----------
1 file changed, 9 insertions(+), 11 deletions(-)
diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index 71881bddad25..dd8564be3894 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -302,7 +302,7 @@ static struct inode *mqueue_get_inode(struct super_block *sb,
inode->i_mode = mode;
inode->i_uid = current_fsuid();
inode->i_gid = current_fsgid();
- inode->i_mtime = inode->i_ctime = inode->i_atime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
if (S_ISREG(mode)) {
struct mqueue_inode_info *info;
@@ -596,7 +596,7 @@ static int mqueue_create_attr(struct dentry *dentry, umode_t mode, void *arg)
put_ipc_ns(ipc_ns);
dir->i_size += DIRENT_SIZE;
- dir->i_ctime = dir->i_mtime = dir->i_atime = current_time(dir);
+ inode_ctime_set(dir, dir->i_mtime = dir->i_atime = current_time(dir));
d_instantiate(dentry, inode);
dget(dentry);
@@ -618,7 +618,7 @@ static int mqueue_unlink(struct inode *dir, struct dentry *dentry)
{
struct inode *inode = d_inode(dentry);
- dir->i_ctime = dir->i_mtime = dir->i_atime = current_time(dir);
+ inode_ctime_set(dir, dir->i_mtime = dir->i_atime = current_time(dir));
dir->i_size -= DIRENT_SIZE;
drop_nlink(inode);
dput(dentry);
@@ -656,7 +656,7 @@ static ssize_t mqueue_read_file(struct file *filp, char __user *u_data,
if (ret <= 0)
return ret;
- file_inode(filp)->i_atime = file_inode(filp)->i_ctime = current_time(file_inode(filp));
+ file_inode(filp)->i_atime = inode_ctime_set_current(file_inode(filp));
return ret;
}
@@ -1162,8 +1162,7 @@ static int do_mq_timedsend(mqd_t mqdes, const char __user *u_msg_ptr,
goto out_unlock;
__do_notify(info);
}
- inode->i_atime = inode->i_mtime = inode->i_ctime =
- current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
}
out_unlock:
spin_unlock(&info->lock);
@@ -1257,8 +1256,7 @@ static int do_mq_timedreceive(mqd_t mqdes, char __user *u_msg_ptr,
msg_ptr = msg_get(info);
- inode->i_atime = inode->i_mtime = inode->i_ctime =
- current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
/* There is now free space in queue. */
pipelined_receive(&wake_q, info);
@@ -1396,7 +1394,7 @@ static int do_mq_notify(mqd_t mqdes, const struct sigevent *notification)
if (notification == NULL) {
if (info->notify_owner == task_tgid(current)) {
remove_notification(info);
- inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode_ctime_set_current(inode);
}
} else if (info->notify_owner != NULL) {
ret = -EBUSY;
@@ -1422,7 +1420,7 @@ static int do_mq_notify(mqd_t mqdes, const struct sigevent *notification)
info->notify_owner = get_pid(task_tgid(current));
info->notify_user_ns = get_user_ns(current_user_ns());
- inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode_ctime_set_current(inode);
}
spin_unlock(&info->lock);
out_fput:
@@ -1485,7 +1483,7 @@ static int do_mq_getsetattr(int mqdes, struct mq_attr *new, struct mq_attr *old)
f.file->f_flags &= ~O_NONBLOCK;
spin_unlock(&f.file->f_lock);
- inode->i_atime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode_ctime_set_current(inode);
}
spin_unlock(&info->lock);
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 73/79] bpf: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (70 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa
Cc: Al Viro, Jan Kara, bpf, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
kernel/bpf/inode.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/kernel/bpf/inode.c b/kernel/bpf/inode.c
index 4174f76133df..d4489bb761df 100644
--- a/kernel/bpf/inode.c
+++ b/kernel/bpf/inode.c
@@ -120,7 +120,7 @@ static struct inode *bpf_get_inode(struct super_block *sb,
inode->i_ino = get_next_ino();
inode->i_atime = current_time(inode);
inode->i_mtime = inode->i_atime;
- inode->i_ctime = inode->i_atime;
+ inode_ctime_set(inode, inode->i_atime);
inode_init_owner(&nop_mnt_idmap, inode, dir, mode);
@@ -149,7 +149,7 @@ static void bpf_dentry_finalize(struct dentry *dentry, struct inode *inode,
dget(dentry);
dir->i_mtime = current_time(dir);
- dir->i_ctime = dir->i_mtime;
+ inode_ctime_set(dir, dir->i_mtime);
}
static int bpf_mkdir(struct mnt_idmap *idmap, struct inode *dir,
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 74/79] shmem: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (71 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Hugh Dickins, Andrew Morton
Cc: Al Viro, Jan Kara, linux-mm, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
mm/shmem.c | 28 +++++++++++++++-------------
1 file changed, 15 insertions(+), 13 deletions(-)
diff --git a/mm/shmem.c b/mm/shmem.c
index 4752084720b2..4979cb3e37e5 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -1064,7 +1064,7 @@ static void shmem_undo_range(struct inode *inode, loff_t lstart, loff_t lend,
void shmem_truncate_range(struct inode *inode, loff_t lstart, loff_t lend)
{
shmem_undo_range(inode, lstart, lend, false);
- inode->i_ctime = inode->i_mtime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
inode_inc_iversion(inode);
}
EXPORT_SYMBOL_GPL(shmem_truncate_range);
@@ -1161,9 +1161,9 @@ static int shmem_setattr(struct mnt_idmap *idmap,
if (attr->ia_valid & ATTR_MODE)
error = posix_acl_chmod(idmap, dentry, inode->i_mode);
if (!error && update_ctime) {
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
if (update_mtime)
- inode->i_mtime = inode->i_ctime;
+ inode->i_mtime = inode_ctime_peek(inode);
inode_inc_iversion(inode);
}
return error;
@@ -2389,7 +2389,7 @@ static struct inode *shmem_get_inode(struct mnt_idmap *idmap, struct super_block
inode->i_ino = ino;
inode_init_owner(idmap, inode, dir, mode);
inode->i_blocks = 0;
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
inode->i_generation = get_random_u32();
info = SHMEM_I(inode);
memset(info, 0, (char *)inode - (char *)info);
@@ -3101,7 +3101,7 @@ shmem_mknod(struct mnt_idmap *idmap, struct inode *dir,
error = 0;
dir->i_size += BOGO_DIRENT_SIZE;
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
inode_inc_iversion(dir);
d_instantiate(dentry, inode);
dget(dentry); /* Extra count - pin the dentry in core */
@@ -3177,7 +3177,8 @@ static int shmem_link(struct dentry *old_dentry, struct inode *dir, struct dentr
}
dir->i_size += BOGO_DIRENT_SIZE;
- inode->i_ctime = dir->i_ctime = dir->i_mtime = current_time(inode);
+ dir->i_mtime = inode_ctime_set_current(inode);
+ inode_ctime_set(dir, dir->i_mtime);
inode_inc_iversion(dir);
inc_nlink(inode);
ihold(inode); /* New dentry reference */
@@ -3195,7 +3196,8 @@ static int shmem_unlink(struct inode *dir, struct dentry *dentry)
shmem_free_inode(inode->i_sb);
dir->i_size -= BOGO_DIRENT_SIZE;
- inode->i_ctime = dir->i_ctime = dir->i_mtime = current_time(inode);
+ dir->i_mtime = inode_ctime_set_current(inode);
+ inode_ctime_set(dir, dir->i_mtime);
inode_inc_iversion(dir);
drop_nlink(inode);
dput(dentry); /* Undo the count from "create" - this does all the work */
@@ -3283,9 +3285,9 @@ static int shmem_rename2(struct mnt_idmap *idmap,
old_dir->i_size -= BOGO_DIRENT_SIZE;
new_dir->i_size += BOGO_DIRENT_SIZE;
- old_dir->i_ctime = old_dir->i_mtime =
- new_dir->i_ctime = new_dir->i_mtime =
- inode->i_ctime = current_time(old_dir);
+ old_dir->i_mtime = inode_ctime_set_current(old_dir);
+ new_dir->i_mtime = inode_ctime_set_current(new_dir);
+ inode_ctime_set_current(inode);
inode_inc_iversion(old_dir);
inode_inc_iversion(new_dir);
return 0;
@@ -3339,7 +3341,7 @@ static int shmem_symlink(struct mnt_idmap *idmap, struct inode *dir,
folio_put(folio);
}
dir->i_size += BOGO_DIRENT_SIZE;
- dir->i_ctime = dir->i_mtime = current_time(dir);
+ dir->i_mtime = inode_ctime_set_current(dir);
inode_inc_iversion(dir);
d_instantiate(dentry, inode);
dget(dentry);
@@ -3411,7 +3413,7 @@ static int shmem_fileattr_set(struct mnt_idmap *idmap,
(fa->flags & SHMEM_FL_USER_MODIFIABLE);
shmem_set_inode_flags(inode, info->fsflags);
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
inode_inc_iversion(inode);
return 0;
}
@@ -3481,7 +3483,7 @@ static int shmem_xattr_handler_set(const struct xattr_handler *handler,
name = xattr_full_name(handler, name);
err = simple_xattr_set(&info->xattrs, name, value, size, flags, NULL);
if (!err) {
- inode->i_ctime = current_time(inode);
+ inode_ctime_set_current(inode);
inode_inc_iversion(inode);
}
return err;
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 75/79] rpc_pipefs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (72 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Trond Myklebust, Anna Schumaker, Chuck Lever,
Neil Brown, Olga Kornievskaia, Dai Ngo, Tom Talpey,
David S. Miller, Eric Dumazet, Jakub Kicinski, Paolo Abeni
Cc: Al Viro, Jan Kara, linux-nfs, netdev, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
net/sunrpc/rpc_pipe.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c
index 0b6034fab9ab..aa382e7ae982 100644
--- a/net/sunrpc/rpc_pipe.c
+++ b/net/sunrpc/rpc_pipe.c
@@ -472,7 +472,7 @@ rpc_get_inode(struct super_block *sb, umode_t mode)
return NULL;
inode->i_ino = get_next_ino();
inode->i_mode = mode;
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
switch (mode & S_IFMT) {
case S_IFDIR:
inode->i_fop = &simple_dir_operations;
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 76/79] apparmor: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (73 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
2023-06-22 9:04 ` John Johansen
-1 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, John Johansen, Paul Moore, James Morris,
Serge E. Hallyn
Cc: Al Viro, Jan Kara, apparmor, linux-security-module, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
security/apparmor/apparmorfs.c | 6 +++---
security/apparmor/policy_unpack.c | 4 ++--
2 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/security/apparmor/apparmorfs.c b/security/apparmor/apparmorfs.c
index 3d0d370d6ffd..8c72a94dd9e3 100644
--- a/security/apparmor/apparmorfs.c
+++ b/security/apparmor/apparmorfs.c
@@ -226,7 +226,7 @@ static int __aafs_setup_d_inode(struct inode *dir, struct dentry *dentry,
inode->i_ino = get_next_ino();
inode->i_mode = mode;
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
inode->i_private = data;
if (S_ISDIR(mode)) {
inode->i_op = iops ? iops : &simple_dir_inode_operations;
@@ -1557,7 +1557,7 @@ void __aafs_profile_migrate_dents(struct aa_profile *old,
if (new->dents[i]) {
struct inode *inode = d_inode(new->dents[i]);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
}
old->dents[i] = NULL;
}
@@ -2546,7 +2546,7 @@ static int aa_mk_null_file(struct dentry *parent)
inode->i_ino = get_next_ino();
inode->i_mode = S_IFCHR | S_IRUGO | S_IWUGO;
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
init_special_inode(inode, S_IFCHR | S_IRUGO | S_IWUGO,
MKDEV(MEM_MAJOR, 3));
d_instantiate(dentry, inode);
diff --git a/security/apparmor/policy_unpack.c b/security/apparmor/policy_unpack.c
index 4c188a44d65c..334d69b118a9 100644
--- a/security/apparmor/policy_unpack.c
+++ b/security/apparmor/policy_unpack.c
@@ -89,10 +89,10 @@ void __aa_loaddata_update(struct aa_loaddata *data, long revision)
struct inode *inode;
inode = d_inode(data->dents[AAFS_LOADDATA_DIR]);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
inode = d_inode(data->dents[AAFS_LOADDATA_REVISION]);
- inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_mtime = inode_ctime_set_current(inode);
}
}
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 77/79] security: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (74 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
2023-06-23 14:15 ` Paul Moore
-1 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Paul Moore, James Morris, Serge E. Hallyn
Cc: Al Viro, Jan Kara, linux-security-module, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
security/inode.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/security/inode.c b/security/inode.c
index 6c326939750d..086280390793 100644
--- a/security/inode.c
+++ b/security/inode.c
@@ -145,7 +145,7 @@ static struct dentry *securityfs_create_dentry(const char *name, umode_t mode,
inode->i_ino = get_next_ino();
inode->i_mode = mode;
- inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
+ inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
inode->i_private = data;
if (S_ISDIR(mode)) {
inode->i_op = &simple_dir_inode_operations;
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 78/79] selinux: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
` (75 preceding siblings ...)
(?)
@ 2023-06-21 14:46 ` Jeff Layton
2023-06-23 14:17 ` Paul Moore
-1 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:46 UTC (permalink / raw)
To: Christian Brauner, Paul Moore, Stephen Smalley, Eric Paris
Cc: Al Viro, Jan Kara, selinux, linux-kernel
In later patches, we're going to change how the ctime.tv_nsec field is
utilized. Switch to using accessor functions instead of raw accesses of
inode->i_ctime.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
security/selinux/selinuxfs.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/security/selinux/selinuxfs.c b/security/selinux/selinuxfs.c
index bad1f6b685fd..d3908baddb30 100644
--- a/security/selinux/selinuxfs.c
+++ b/security/selinux/selinuxfs.c
@@ -1197,7 +1197,7 @@ static struct inode *sel_make_inode(struct super_block *sb, int mode)
if (ret) {
ret->i_mode = mode;
- ret->i_atime = ret->i_mtime = ret->i_ctime = current_time(ret);
+ ret->i_atime = ret->i_mtime = inode_ctime_set_current(ret);
}
return ret;
}
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [PATCH 79/79] fs: rename i_ctime field to __i_ctime
2023-06-21 14:45 ` [f2fs-dev] " Jeff Layton
(?)
(?)
@ 2023-06-21 14:49 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:49 UTC (permalink / raw)
To: Jeremy Kerr, Arnd Bergmann, Michael Ellerman, Nicholas Piggin,
Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Tom Talpey, Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
Now that everything in-tree is converted to use the accessor functions,
rename the i_ctime field in the inode to make its accesses more
self-documenting.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
include/linux/fs.h | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 9afb30606373..2ca46c532b49 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -642,7 +642,7 @@ struct inode {
loff_t i_size;
struct timespec64 i_atime;
struct timespec64 i_mtime;
- struct timespec64 i_ctime;
+ struct timespec64 __i_ctime; /* use inode_ctime_* accessors! */
spinlock_t i_lock; /* i_blocks, i_bytes, maybe i_size */
unsigned short i_bytes;
u8 i_blkbits;
@@ -1485,7 +1485,7 @@ struct timespec64 inode_ctime_set_current(struct inode *inode);
*/
static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
{
- return inode->i_ctime;
+ return inode->__i_ctime;
}
/**
@@ -1497,7 +1497,7 @@ static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
*/
static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
{
- inode->i_ctime = ts;
+ inode->__i_ctime = ts;
return ts;
}
@@ -1510,7 +1510,7 @@ static inline struct timespec64 inode_ctime_set(struct inode *inode, struct time
*/
static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
{
- inode->i_ctime.tv_sec = sec;
+ inode->__i_ctime.tv_sec = sec;
return sec;
}
@@ -1523,7 +1523,7 @@ static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
*/
static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
{
- inode->i_ctime.tv_nsec = nsec;
+ inode->__i_ctime.tv_nsec = nsec;
return nsec;
}
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [f2fs-dev] [PATCH 79/79] fs: rename i_ctime field to __i_ctime
@ 2023-06-21 14:49 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:49 UTC (permalink / raw)
To: Jeremy Kerr, Arnd Bergmann, Michael Ellerman, Nicholas Piggin,
Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Tom Talpey, Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
Now that everything in-tree is converted to use the accessor functions,
rename the i_ctime field in the inode to make its accesses more
self-documenting.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
include/linux/fs.h | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 9afb30606373..2ca46c532b49 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -642,7 +642,7 @@ struct inode {
loff_t i_size;
struct timespec64 i_atime;
struct timespec64 i_mtime;
- struct timespec64 i_ctime;
+ struct timespec64 __i_ctime; /* use inode_ctime_* accessors! */
spinlock_t i_lock; /* i_blocks, i_bytes, maybe i_size */
unsigned short i_bytes;
u8 i_blkbits;
@@ -1485,7 +1485,7 @@ struct timespec64 inode_ctime_set_current(struct inode *inode);
*/
static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
{
- return inode->i_ctime;
+ return inode->__i_ctime;
}
/**
@@ -1497,7 +1497,7 @@ static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
*/
static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
{
- inode->i_ctime = ts;
+ inode->__i_ctime = ts;
return ts;
}
@@ -1510,7 +1510,7 @@ static inline struct timespec64 inode_ctime_set(struct inode *inode, struct time
*/
static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
{
- inode->i_ctime.tv_sec = sec;
+ inode->__i_ctime.tv_sec = sec;
return sec;
}
@@ -1523,7 +1523,7 @@ static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
*/
static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
{
- inode->i_ctime.tv_nsec = nsec;
+ inode->__i_ctime.tv_nsec = nsec;
return nsec;
}
--
2.41.0
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [Ocfs2-devel] [PATCH 79/79] fs: rename i_ctime field to __i_ctime
@ 2023-06-21 14:49 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton via Ocfs2-devel @ 2023-06-21 14:49 UTC (permalink / raw)
To: Jeremy Kerr, Arnd Bergmann, Michael Ellerman, Nicholas Piggin,
Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Tom Talpey, Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
Now that everything in-tree is converted to use the accessor functions,
rename the i_ctime field in the inode to make its accesses more
self-documenting.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
include/linux/fs.h | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 9afb30606373..2ca46c532b49 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -642,7 +642,7 @@ struct inode {
loff_t i_size;
struct timespec64 i_atime;
struct timespec64 i_mtime;
- struct timespec64 i_ctime;
+ struct timespec64 __i_ctime; /* use inode_ctime_* accessors! */
spinlock_t i_lock; /* i_blocks, i_bytes, maybe i_size */
unsigned short i_bytes;
u8 i_blkbits;
@@ -1485,7 +1485,7 @@ struct timespec64 inode_ctime_set_current(struct inode *inode);
*/
static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
{
- return inode->i_ctime;
+ return inode->__i_ctime;
}
/**
@@ -1497,7 +1497,7 @@ static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
*/
static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
{
- inode->i_ctime = ts;
+ inode->__i_ctime = ts;
return ts;
}
@@ -1510,7 +1510,7 @@ static inline struct timespec64 inode_ctime_set(struct inode *inode, struct time
*/
static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
{
- inode->i_ctime.tv_sec = sec;
+ inode->__i_ctime.tv_sec = sec;
return sec;
}
@@ -1523,7 +1523,7 @@ static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
*/
static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
{
- inode->i_ctime.tv_nsec = nsec;
+ inode->__i_ctime.tv_nsec = nsec;
return nsec;
}
--
2.41.0
_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel
^ permalink raw reply related [flat|nested] 229+ messages in thread
* [Cluster-devel] [PATCH 79/79] fs: rename i_ctime field to __i_ctime
@ 2023-06-21 14:49 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 14:49 UTC (permalink / raw)
To: cluster-devel.redhat.com
Now that everything in-tree is converted to use the accessor functions,
rename the i_ctime field in the inode to make its accesses more
self-documenting.
Signed-off-by: Jeff Layton <jlayton@kernel.org>
---
include/linux/fs.h | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 9afb30606373..2ca46c532b49 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -642,7 +642,7 @@ struct inode {
loff_t i_size;
struct timespec64 i_atime;
struct timespec64 i_mtime;
- struct timespec64 i_ctime;
+ struct timespec64 __i_ctime; /* use inode_ctime_* accessors! */
spinlock_t i_lock; /* i_blocks, i_bytes, maybe i_size */
unsigned short i_bytes;
u8 i_blkbits;
@@ -1485,7 +1485,7 @@ struct timespec64 inode_ctime_set_current(struct inode *inode);
*/
static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
{
- return inode->i_ctime;
+ return inode->__i_ctime;
}
/**
@@ -1497,7 +1497,7 @@ static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
*/
static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
{
- inode->i_ctime = ts;
+ inode->__i_ctime = ts;
return ts;
}
@@ -1510,7 +1510,7 @@ static inline struct timespec64 inode_ctime_set(struct inode *inode, struct time
*/
static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
{
- inode->i_ctime.tv_sec = sec;
+ inode->__i_ctime.tv_sec = sec;
return sec;
}
@@ -1523,7 +1523,7 @@ static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
*/
static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
{
- inode->i_ctime.tv_nsec = nsec;
+ inode->__i_ctime.tv_nsec = nsec;
return nsec;
}
--
2.41.0
^ permalink raw reply related [flat|nested] 229+ messages in thread
* Re: [PATCH 20/79] cramfs: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 20/79] cramfs: " Jeff Layton
@ 2023-06-21 15:29 ` Nicolas Pitre
0 siblings, 0 replies; 229+ messages in thread
From: Nicolas Pitre @ 2023-06-21 15:29 UTC (permalink / raw)
To: Jeff Layton; +Cc: Christian Brauner, Al Viro, Jan Kara, linux-kernel
On Wed, 21 Jun 2023, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Acked-by: Nicolas Pitre <nico@fluxnic.net
Yet all that's needed here is to simply set time entries to 0 as cramfs
doesn't time stamp its files. It could as well be done with memset().
> ---
> fs/cramfs/inode.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c
> index 27c6597aa1be..039ce0f8858b 100644
> --- a/fs/cramfs/inode.c
> +++ b/fs/cramfs/inode.c
> @@ -133,7 +133,7 @@ static struct inode *get_cramfs_inode(struct super_block *sb,
> }
>
> /* Struct copy intentional */
> - inode->i_mtime = inode->i_atime = inode->i_ctime = zerotime;
> + inode->i_mtime = inode->i_atime = inode_ctime_set(inode, zerotime);
> /* inode->i_nlink is left 1 - arguably wrong for directories,
> but it's the best we can do without reading the directory
> contents. 1 yields the right result in GNU find, even
> --
> 2.41.0
>
>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 04/79] binderfs: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 04/79] binderfs: " Jeff Layton
@ 2023-06-21 15:29 ` Greg Kroah-Hartman
2023-06-21 16:35 ` Jan Kara
1 sibling, 0 replies; 229+ messages in thread
From: Greg Kroah-Hartman @ 2023-06-21 15:29 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Carlos Llamas,
Suren Baghdasaryan, Al Viro, Jan Kara, linux-kernel
On Wed, Jun 21, 2023 at 10:45:17AM -0400, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
> ---
> drivers/android/binderfs.c | 8 ++++----
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 06/79] ibm: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 06/79] ibm: " Jeff Layton
@ 2023-06-21 15:29 ` Greg Kroah-Hartman
2023-06-21 16:35 ` Jan Kara
1 sibling, 0 replies; 229+ messages in thread
From: Greg Kroah-Hartman @ 2023-06-21 15:29 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Arnd Bergmann, Brad Warrum, Ritu Agarwal,
Al Viro, Jan Kara, linux-kernel
On Wed, Jun 21, 2023 at 10:45:19AM -0400, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
> ---
> drivers/misc/ibmasm/ibmasmfs.c | 2 +-
> drivers/misc/ibmvmc.c | 2 +-
> 2 files changed, 2 insertions(+), 2 deletions(-)
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 21/79] debugfs: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 21/79] debugfs: " Jeff Layton
@ 2023-06-21 15:29 ` Greg Kroah-Hartman
0 siblings, 0 replies; 229+ messages in thread
From: Greg Kroah-Hartman @ 2023-06-21 15:29 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Rafael J. Wysocki, Al Viro, Jan Kara, linux-kernel
On Wed, Jun 21, 2023 at 10:45:34AM -0400, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 07/79] usb: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 07/79] usb: " Jeff Layton
@ 2023-06-21 15:29 ` Greg Kroah-Hartman
2023-06-21 16:36 ` Jan Kara
1 sibling, 0 replies; 229+ messages in thread
From: Greg Kroah-Hartman @ 2023-06-21 15:29 UTC (permalink / raw)
To: Jeff Layton; +Cc: Christian Brauner, Al Viro, Jan Kara, linux-usb, linux-kernel
On Wed, Jun 21, 2023 at 10:45:20AM -0400, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
> ---
> drivers/usb/core/devio.c | 16 ++++++++--------
> drivers/usb/gadget/function/f_fs.c | 6 +-----
> drivers/usb/gadget/legacy/inode.c | 3 +--
> 3 files changed, 10 insertions(+), 15 deletions(-)
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 43/79] kernfs: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 43/79] kernfs: " Jeff Layton
@ 2023-06-21 15:30 ` Greg Kroah-Hartman
0 siblings, 0 replies; 229+ messages in thread
From: Greg Kroah-Hartman @ 2023-06-21 15:30 UTC (permalink / raw)
To: Jeff Layton; +Cc: Christian Brauner, Tejun Heo, Al Viro, Jan Kara, linux-kernel
On Wed, Jun 21, 2023 at 10:45:56AM -0400, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
2023-06-21 14:45 ` [f2fs-dev] " Jeff Layton
` (4 preceding siblings ...)
(?)
@ 2023-06-21 16:34 ` Jan Kara
-1 siblings, 0 replies; 229+ messages in thread
From: Jan Kara @ 2023-06-21 16:34 UTC (permalink / raw)
To: Jeff Layton
Cc: Jeremy Kerr, Arnd Bergmann, Michael Ellerman, Nicholas Piggin,
Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Tom Talpey, Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On Wed 21-06-23 10:45:06, Jeff Layton wrote:
> struct timespec64 has unused bits in the tv_nsec field that can be used
> for other purposes. In future patches, we're going to change how the
> inode->i_ctime is accessed in certain inodes in order to make use of
> them. In order to do that safely though, we'll need to eradicate raw
> accesses of the inode->i_ctime field from the kernel.
>
> Add new accessor functions for the ctime that we can use to replace them.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Looks good to me. Feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
Honza
> ---
> fs/inode.c | 16 ++++++++++++++
> include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
> 2 files changed, 68 insertions(+), 1 deletion(-)
>
> diff --git a/fs/inode.c b/fs/inode.c
> index d37fad91c8da..c005e7328fbb 100644
> --- a/fs/inode.c
> +++ b/fs/inode.c
> @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
> }
> EXPORT_SYMBOL(current_time);
>
> +/**
> + * inode_ctime_set_current - set the ctime to current_time
> + * @inode: inode
> + *
> + * Set the inode->i_ctime to the current value for the inode. Returns
> + * the current value that was assigned to i_ctime.
> + */
> +struct timespec64 inode_ctime_set_current(struct inode *inode)
> +{
> + struct timespec64 now = current_time(inode);
> +
> + inode_set_ctime(inode, now);
> + return now;
> +}
> +EXPORT_SYMBOL(inode_ctime_set_current);
> +
> /**
> * in_group_or_capable - check whether caller is CAP_FSETID privileged
> * @idmap: idmap of the mount @inode was found from
> diff --git a/include/linux/fs.h b/include/linux/fs.h
> index 6867512907d6..9afb30606373 100644
> --- a/include/linux/fs.h
> +++ b/include/linux/fs.h
> @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
> kgid_has_mapping(fs_userns, kgid);
> }
>
> -extern struct timespec64 current_time(struct inode *inode);
> +struct timespec64 current_time(struct inode *inode);
> +struct timespec64 inode_ctime_set_current(struct inode *inode);
> +
> +/**
> + * inode_ctime_peek - fetch the current ctime from the inode
> + * @inode: inode from which to fetch ctime
> + *
> + * Grab the current ctime from the inode and return it.
> + */
> +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
> +{
> + return inode->i_ctime;
> +}
> +
> +/**
> + * inode_ctime_set - set the ctime in the inode to the given value
> + * @inode: inode in which to set the ctime
> + * @ts: timespec value to set the ctime
> + *
> + * Set the ctime in @inode to @ts.
> + */
> +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
> +{
> + inode->i_ctime = ts;
> + return ts;
> +}
> +
> +/**
> + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
> + * @inode: inode in which to set the ctime
> + * @sec: value to set the tv_sec field
> + *
> + * Set the sec field in the ctime. Returns @sec.
> + */
> +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
> +{
> + inode->i_ctime.tv_sec = sec;
> + return sec;
> +}
> +
> +/**
> + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
> + * @inode: inode in which to set the ctime
> + * @nsec: value to set the tv_nsec field
> + *
> + * Set the nsec field in the ctime. Returns @nsec.
> + */
> +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
> +{
> + inode->i_ctime.tv_nsec = nsec;
> + return nsec;
> +}
>
> /*
> * Snapshotting support.
> --
> 2.41.0
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [f2fs-dev] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-21 16:34 ` Jan Kara
0 siblings, 0 replies; 229+ messages in thread
From: Jan Kara @ 2023-06-21 16:34 UTC (permalink / raw)
To: Jeff Layton
Cc: Latchesar Ionkov, Rafael J. Wysocki, Darrick J. Wong,
Anders Larsen, Carlos Llamas, Andrii Nakryiko, Hugh Dickins,
John Johansen, Seth Forshee, Alexander Gordeev,
Christoph Hellwig, Mike Marshall, Paulo Alcantara, linux-xfs,
Bart Van Assche, Michael Ellerman, John Keeping, Zhang Yi,
James Morris, Christophe Leroy, Tyler Hicks, Alan Stern,
Christian Borntraeger, devel, Shyam Prasad N, Jan Harkes,
linux-um, Nicholas Piggin, Alexander Viro, Eric Van Hensbergen,
Suren Baghdasaryan, Trond Myklebust, Anton Altaparmakov,
Christian Brauner, Wolfram Sang, Greg Kroah-Hartman,
Stephen Smalley, linux-usb, linux-kernel, Ronnie Sahlberg,
Sergey Senozhatsky, Luis Chamberlain, Chuck Lever, Sven Schnelle,
Jiri Olsa, Jan Kara, Tejun Heo, Andrew Morton,
linux-trace-kernel, linux-hardening, Dave Kleikamp,
Sandeep Dhavale, Tetsuo Handa, Mimi Zohar, linux-mm,
Joel Fernandes, Eric Dumazet, Stanislav Fomichev,
Andrzej Pietrasiewicz, Hangyu Hua, linux-s390, linux-nilfs,
Paul Moore, Leon Romanovsky, John Fastabend,
Arve Hjønnevåg, Minghao Chi, codalist, selinux,
ZhangPeng, Udipto Goswami, Yonghong Song, Iurii Zaikin,
Namjae Jeon, Masami Hiramatsu, ecryptfs, Todd Kjos,
Vasily Gorbik, Yu Zhe, linuxppc-dev, reiserfs-devel,
Miklos Szeredi, Yue Hu, Jaegeuk Kim, Aditya Garg, Martijn Coenen,
OGAWA Hirofumi, Hao Luo, Tony Luck, Theodore Ts'o,
Nicolas Pitre, linux-ntfs-dev, Muchun Song, Roberto Sassu,
linux-f2fs-devel, Guilherme G. Piccoli, Jozef Martiniak,
Eric Biederman, Anna Schumaker, xu xin, Brad Warrum,
Mike Kravetz, Jingyu Wang, linux-efi, Dan Carpenter,
Martin Brandenburg, Tom Rix, Alexei Starovoitov, Chris Mason,
linux-mtd, Matthew Wilcox (Oracle),
Marc Dionne, linux-afs, Ian Kent, Naohiro Aota, Daniel Borkmann,
Dennis Dalessandro, linux-rdma, Linyu Yuan, coda,
Viacheslav Dubeyko, Ilya Dryomov, Paolo Abeni, Alexey Dobriyan,
Serge E. Hallyn, Jens Axboe, Zeng Jingxiang, Kees Cook,
Arnd Bergmann, autofs, Steven Rostedt, Yifei Liu, Damien Le Moal,
Eric Paris, ceph-devel, Gao Xiang, Jiangshan Yi, David Howells,
linux-nfs, linux-ext4, Song Liu, samba-technical, Steve French,
Jeremy Kerr, netdev, Bob Peterson, linux-fsdevel, bpf, ntfs3,
linux-erofs, David S. Miller, ocfs2-devel, jfs-discussion,
Dominique Martinet, Christian Schoenebeck, Bob Copeland,
KP Singh, Oleg Kanatov, Konstantin Komarov, Joseph Qi,
Yuta Hayama, Andreas Dilger, Mikulas Patocka, Zhengchao Shao,
Chen Zhongjin, Ard Biesheuvel, Anton Ivanov, Laurent Pinchart,
Andreas Gruenbacher, Richard Weinberger, Mark Fasheh,
Stefan Roesch, cluster-devel, Jason Gunthorpe, Jakub Kicinski,
Rik van Riel, Salah Triki, Evgeniy Dushistov, linux-cifs,
Heiko Carstens, apparmor, Josef Bacik, Liam R. Howlett,
Tom Talpey, Hans de Goede, Tigran A. Aivazian, Dave Chinner,
David Sterba, Xiubo Li, Ryusuke Konishi, Juergen Gross,
Johannes Thumshirn, Ritu Agarwal, Luis de Bethencourt,
Martin KaFai Lau, v9fs, Fabio M. De Francesco, linux-unionfs,
Ruihan Li, linux-security-module, Erez Zadok, Jeffle Xu,
Dr. David Alan Gilbert, Johannes Weiner, Phillip Lougher,
Johannes Berg, Sungjong Seo, David Woodhouse, linux-karma-devel,
linux-btrfs, Joel Becker
On Wed 21-06-23 10:45:06, Jeff Layton wrote:
> struct timespec64 has unused bits in the tv_nsec field that can be used
> for other purposes. In future patches, we're going to change how the
> inode->i_ctime is accessed in certain inodes in order to make use of
> them. In order to do that safely though, we'll need to eradicate raw
> accesses of the inode->i_ctime field from the kernel.
>
> Add new accessor functions for the ctime that we can use to replace them.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Looks good to me. Feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
Honza
> ---
> fs/inode.c | 16 ++++++++++++++
> include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
> 2 files changed, 68 insertions(+), 1 deletion(-)
>
> diff --git a/fs/inode.c b/fs/inode.c
> index d37fad91c8da..c005e7328fbb 100644
> --- a/fs/inode.c
> +++ b/fs/inode.c
> @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
> }
> EXPORT_SYMBOL(current_time);
>
> +/**
> + * inode_ctime_set_current - set the ctime to current_time
> + * @inode: inode
> + *
> + * Set the inode->i_ctime to the current value for the inode. Returns
> + * the current value that was assigned to i_ctime.
> + */
> +struct timespec64 inode_ctime_set_current(struct inode *inode)
> +{
> + struct timespec64 now = current_time(inode);
> +
> + inode_set_ctime(inode, now);
> + return now;
> +}
> +EXPORT_SYMBOL(inode_ctime_set_current);
> +
> /**
> * in_group_or_capable - check whether caller is CAP_FSETID privileged
> * @idmap: idmap of the mount @inode was found from
> diff --git a/include/linux/fs.h b/include/linux/fs.h
> index 6867512907d6..9afb30606373 100644
> --- a/include/linux/fs.h
> +++ b/include/linux/fs.h
> @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
> kgid_has_mapping(fs_userns, kgid);
> }
>
> -extern struct timespec64 current_time(struct inode *inode);
> +struct timespec64 current_time(struct inode *inode);
> +struct timespec64 inode_ctime_set_current(struct inode *inode);
> +
> +/**
> + * inode_ctime_peek - fetch the current ctime from the inode
> + * @inode: inode from which to fetch ctime
> + *
> + * Grab the current ctime from the inode and return it.
> + */
> +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
> +{
> + return inode->i_ctime;
> +}
> +
> +/**
> + * inode_ctime_set - set the ctime in the inode to the given value
> + * @inode: inode in which to set the ctime
> + * @ts: timespec value to set the ctime
> + *
> + * Set the ctime in @inode to @ts.
> + */
> +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
> +{
> + inode->i_ctime = ts;
> + return ts;
> +}
> +
> +/**
> + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
> + * @inode: inode in which to set the ctime
> + * @sec: value to set the tv_sec field
> + *
> + * Set the sec field in the ctime. Returns @sec.
> + */
> +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
> +{
> + inode->i_ctime.tv_sec = sec;
> + return sec;
> +}
> +
> +/**
> + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
> + * @inode: inode in which to set the ctime
> + * @nsec: value to set the tv_nsec field
> + *
> + * Set the nsec field in the ctime. Returns @nsec.
> + */
> +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
> +{
> + inode->i_ctime.tv_nsec = nsec;
> + return nsec;
> +}
>
> /*
> * Snapshotting support.
> --
> 2.41.0
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [Ocfs2-devel] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-21 16:34 ` Jan Kara
0 siblings, 0 replies; 229+ messages in thread
From: Jan Kara via Ocfs2-devel @ 2023-06-21 16:34 UTC (permalink / raw)
To: Jeff Layton
Cc: Latchesar Ionkov, Rafael J. Wysocki, Anders Larsen,
Carlos Llamas, Andrii Nakryiko, Hugh Dickins, John Johansen,
Seth Forshee, Alexander Gordeev, Mike Marshall, Paulo Alcantara,
linux-xfs, Bart Van Assche, Michael Ellerman, John Keeping,
Zhang Yi, James Morris, Christophe Leroy, Tyler Hicks,
Alan Stern, Christian Borntraeger, devel, Shyam Prasad N,
Jan Harkes, linux-um, Nicholas Piggin, Alexander Viro,
Eric Van Hensbergen, Suren Baghdasaryan, Trond Myklebust,
Anton Altaparmakov, Christian Brauner, Wolfram Sang,
Greg Kroah-Hartman, Stephen Smalley, linux-usb, linux-kernel,
Ronnie Sahlberg, Sergey Senozhatsky, Luis Chamberlain,
Chuck Lever, Sven Schnelle, Jiri Olsa, Jan Kara, Tejun Heo,
linux-trace-kernel, linux-hardening, Dave Kleikamp,
Sandeep Dhavale, Tetsuo Handa, Mimi Zohar, linux-mm,
Joel Fernandes, Eric Dumazet, Stanislav Fomichev,
Andrzej Pietrasiewicz, Hangyu Hua, linux-s390, linux-nilfs,
Paul Moore, Leon Romanovsky, John Fastabend,
Arve Hjønnevåg, Minghao Chi, codalist, selinux,
ZhangPeng, Udipto Goswami, Yonghong Song, Iurii Zaikin,
Namjae Jeon, Masami Hiramatsu, ecryptfs, Todd Kjos,
Vasily Gorbik, Yu Zhe, linuxppc-dev, reiserfs-devel,
Miklos Szeredi, Yue Hu, Jaegeuk Kim, Aditya Garg, Martijn Coenen,
OGAWA Hirofumi, Hao Luo, Tony Luck, Theodore Ts'o,
Nicolas Pitre, linux-ntfs-dev, Muchun Song, Roberto Sassu,
linux-f2fs-devel, Guilherme G. Piccoli, Jozef Martiniak,
Eric Biederman, Anna Schumaker, xu xin, Brad Warrum,
Mike Kravetz, Jingyu Wang, linux-efi, Dan Carpenter,
Martin Brandenburg, Tom Rix, Alexei Starovoitov, Chris Mason,
linux-mtd, Marc Dionne, linux-afs, Ian Kent, Naohiro Aota,
Daniel Borkmann, Dennis Dalessandro, linux-rdma, Linyu Yuan,
coda, Viacheslav Dubeyko, Ilya Dryomov, Paolo Abeni,
Alexey Dobriyan, Serge E. Hallyn, Zhihao Cheng, Jens Axboe,
Zeng Jingxiang, Kees Cook, Arnd Bergmann, autofs, Steven Rostedt,
Yifei Liu, Damien Le Moal, Eric Paris, ceph-devel, Gao Xiang,
Jiangshan Yi, David Howells, linux-nfs, linux-ext4, Song Liu,
samba-technical, Steve French, Jeremy Kerr, netdev, Bob Peterson,
linux-fsdevel, bpf, ntfs3, linux-erofs, David S. Miller,
ocfs2-devel, jfs-discussion, Dominique Martinet,
Christian Schoenebeck, Bob Copeland, KP Singh, Oleg Kanatov,
Konstantin Komarov, Yuta Hayama, Andreas Dilger, Mikulas Patocka,
Zhengchao Shao, Chen Zhongjin, Ard Biesheuvel, Anton Ivanov,
Laurent Pinchart, Andreas Gruenbacher, Richard Weinberger,
Stefan Roesch, cluster-devel, Jason Gunthorpe, Jakub Kicinski,
Rik van Riel, Salah Triki, Evgeniy Dushistov, linux-cifs,
Heiko Carstens, Chao Yu, apparmor, Josef Bacik, Liam R. Howlett,
Tom Talpey, Hans de Goede, Tigran A. Aivazian, Dave Chinner,
David Sterba, Xiubo Li, Ryusuke Konishi, Juergen Gross,
Johannes Thumshirn, Ritu Agarwal, Luis de Bethencourt,
Martin KaFai Lau, v9fs, Fabio M. De Francesco, linux-unionfs,
Ruihan Li, linux-security-module, Erez Zadok, Jeffle Xu,
Dr. David Alan Gilbert, Johannes Weiner, Phillip Lougher,
Johannes Berg, Sungjong Seo, David Woodhouse, linux-karma-devel,
linux-btrfs
On Wed 21-06-23 10:45:06, Jeff Layton wrote:
> struct timespec64 has unused bits in the tv_nsec field that can be used
> for other purposes. In future patches, we're going to change how the
> inode->i_ctime is accessed in certain inodes in order to make use of
> them. In order to do that safely though, we'll need to eradicate raw
> accesses of the inode->i_ctime field from the kernel.
>
> Add new accessor functions for the ctime that we can use to replace them.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Looks good to me. Feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
Honza
> ---
> fs/inode.c | 16 ++++++++++++++
> include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
> 2 files changed, 68 insertions(+), 1 deletion(-)
>
> diff --git a/fs/inode.c b/fs/inode.c
> index d37fad91c8da..c005e7328fbb 100644
> --- a/fs/inode.c
> +++ b/fs/inode.c
> @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
> }
> EXPORT_SYMBOL(current_time);
>
> +/**
> + * inode_ctime_set_current - set the ctime to current_time
> + * @inode: inode
> + *
> + * Set the inode->i_ctime to the current value for the inode. Returns
> + * the current value that was assigned to i_ctime.
> + */
> +struct timespec64 inode_ctime_set_current(struct inode *inode)
> +{
> + struct timespec64 now = current_time(inode);
> +
> + inode_set_ctime(inode, now);
> + return now;
> +}
> +EXPORT_SYMBOL(inode_ctime_set_current);
> +
> /**
> * in_group_or_capable - check whether caller is CAP_FSETID privileged
> * @idmap: idmap of the mount @inode was found from
> diff --git a/include/linux/fs.h b/include/linux/fs.h
> index 6867512907d6..9afb30606373 100644
> --- a/include/linux/fs.h
> +++ b/include/linux/fs.h
> @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
> kgid_has_mapping(fs_userns, kgid);
> }
>
> -extern struct timespec64 current_time(struct inode *inode);
> +struct timespec64 current_time(struct inode *inode);
> +struct timespec64 inode_ctime_set_current(struct inode *inode);
> +
> +/**
> + * inode_ctime_peek - fetch the current ctime from the inode
> + * @inode: inode from which to fetch ctime
> + *
> + * Grab the current ctime from the inode and return it.
> + */
> +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
> +{
> + return inode->i_ctime;
> +}
> +
> +/**
> + * inode_ctime_set - set the ctime in the inode to the given value
> + * @inode: inode in which to set the ctime
> + * @ts: timespec value to set the ctime
> + *
> + * Set the ctime in @inode to @ts.
> + */
> +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
> +{
> + inode->i_ctime = ts;
> + return ts;
> +}
> +
> +/**
> + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
> + * @inode: inode in which to set the ctime
> + * @sec: value to set the tv_sec field
> + *
> + * Set the sec field in the ctime. Returns @sec.
> + */
> +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
> +{
> + inode->i_ctime.tv_sec = sec;
> + return sec;
> +}
> +
> +/**
> + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
> + * @inode: inode in which to set the ctime
> + * @nsec: value to set the tv_nsec field
> + *
> + * Set the nsec field in the ctime. Returns @nsec.
> + */
> +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
> +{
> + inode->i_ctime.tv_nsec = nsec;
> + return nsec;
> +}
>
> /*
> * Snapshotting support.
> --
> 2.41.0
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-21 16:34 ` Jan Kara
0 siblings, 0 replies; 229+ messages in thread
From: Jan Kara @ 2023-06-21 16:34 UTC (permalink / raw)
To: Jeff Layton
Cc: Latchesar Ionkov, Rafael J. Wysocki, Darrick J. Wong,
Anders Larsen, Carlos Llamas, Andrii Nakryiko, Hugh Dickins,
John Johansen, Seth Forshee, Alexander Gordeev,
Christoph Hellwig, Mike Marshall, Paulo Alcantara, linux-xfs,
Bart Van Assche, John Keeping, Zhang Yi, James Morris,
Tyler Hicks, Alan Stern, Christian Borntraeger, devel,
Shyam Prasad N, Jan Harkes, linux-um, Nicholas Piggin,
Alexander Viro, Eric Van Hensbergen, Suren Baghdasaryan,
Trond Myklebust, Anton Altaparmakov, Christian Brauner,
Wolfram Sang, Greg Kroah-Hartman, Stephen Smalley, linux-usb,
linux-kernel, Ronnie Sahlberg, Sergey Senozhatsky,
Luis Chamberlain, Chuck Lever, Sven Schnelle, Jiri Olsa,
Jan Kara, Tejun Heo, Andrew Morton, linux-trace-kernel,
linux-hardening, Dave Kleikamp, Sandeep Dhavale, Tetsuo Handa,
Mimi Zohar, linux-mm, Joel Fernandes, Eric Dumazet,
Stanislav Fomichev, Andrzej Pietrasiewicz, Hangyu Hua,
linux-s390, linux-nilfs, Paul Moore, Leon Romanovsky,
John Fastabend, Arve Hjønnevåg, Minghao Chi, codalist,
selinux, ZhangPeng, Udipto Goswami, Yonghong Song, Iurii Zaikin,
Namjae Jeon, Masami Hiramatsu, ecryptfs, Todd Kjos,
Vasily Gorbik, Yu Zhe, linuxppc-dev, reiserfs-devel,
Miklos Szeredi, Yue Hu, Jaegeuk Kim, Aditya Garg, Martijn Coenen,
OGAWA Hirofumi, Hao Luo, Tony Luck, Theodore Ts'o,
Nicolas Pitre, linux-ntfs-dev, Muchun Song, Roberto Sassu,
linux-f2fs-devel, Guilherme G. Piccoli, Jozef Martiniak,
Eric Biederman, Anna Schumaker, xu xin, Brad Warrum,
Mike Kravetz, Jingyu Wang, linux-efi, Dan Carpenter,
Martin Brandenburg, Tom Rix, Alexei Starovoitov, Chris Mason,
linux-mtd, Matthew Wilcox (Oracle),
Marc Dionne, linux-afs, Ian Kent, Naohiro Aota, Daniel Borkmann,
Dennis Dalessandro, linux-rdma, Linyu Yuan, coda,
Viacheslav Dubeyko, Ilya Dryomov, Paolo Abeni, Alexey Dobriyan,
Serge E. Hallyn, Zhihao Cheng, Jens Axboe, Zeng Jingxiang,
Kees Cook, Arnd Bergmann, autofs, Steven Rostedt, Yifei Liu,
Damien Le Moal, Eric Paris, ceph-devel, Gao Xiang, Jiangshan Yi,
David Howells, linux-nfs, linux-ext4, Song Liu, samba-technical,
Steve French, Jeremy Kerr, netdev, Bob Peterson, linux-fsdevel,
bpf, ntfs3, linux-erofs, David S. Miller, ocfs2-devel,
jfs-discussion, Dominique Martinet, Christian Schoenebeck,
Bob Copeland, KP Singh, Oleg Kanatov, Konstantin Komarov,
Joseph Qi, Yuta Hayama, Andreas Dilger, Mikulas Patocka,
Zhengchao Shao, Chen Zhongjin, Ard Biesheuvel, Anton Ivanov,
Laurent Pinchart, Andreas Gruenbacher
On Wed 21-06-23 10:45:06, Jeff Layton wrote:
> struct timespec64 has unused bits in the tv_nsec field that can be used
> for other purposes. In future patches, we're going to change how the
> inode->i_ctime is accessed in certain inodes in order to make use of
> them. In order to do that safely though, we'll need to eradicate raw
> accesses of the inode->i_ctime field from the kernel.
>
> Add new accessor functions for the ctime that we can use to replace them.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Looks good to me. Feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
Honza
> ---
> fs/inode.c | 16 ++++++++++++++
> include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
> 2 files changed, 68 insertions(+), 1 deletion(-)
>
> diff --git a/fs/inode.c b/fs/inode.c
> index d37fad91c8da..c005e7328fbb 100644
> --- a/fs/inode.c
> +++ b/fs/inode.c
> @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
> }
> EXPORT_SYMBOL(current_time);
>
> +/**
> + * inode_ctime_set_current - set the ctime to current_time
> + * @inode: inode
> + *
> + * Set the inode->i_ctime to the current value for the inode. Returns
> + * the current value that was assigned to i_ctime.
> + */
> +struct timespec64 inode_ctime_set_current(struct inode *inode)
> +{
> + struct timespec64 now = current_time(inode);
> +
> + inode_set_ctime(inode, now);
> + return now;
> +}
> +EXPORT_SYMBOL(inode_ctime_set_current);
> +
> /**
> * in_group_or_capable - check whether caller is CAP_FSETID privileged
> * @idmap: idmap of the mount @inode was found from
> diff --git a/include/linux/fs.h b/include/linux/fs.h
> index 6867512907d6..9afb30606373 100644
> --- a/include/linux/fs.h
> +++ b/include/linux/fs.h
> @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
> kgid_has_mapping(fs_userns, kgid);
> }
>
> -extern struct timespec64 current_time(struct inode *inode);
> +struct timespec64 current_time(struct inode *inode);
> +struct timespec64 inode_ctime_set_current(struct inode *inode);
> +
> +/**
> + * inode_ctime_peek - fetch the current ctime from the inode
> + * @inode: inode from which to fetch ctime
> + *
> + * Grab the current ctime from the inode and return it.
> + */
> +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
> +{
> + return inode->i_ctime;
> +}
> +
> +/**
> + * inode_ctime_set - set the ctime in the inode to the given value
> + * @inode: inode in which to set the ctime
> + * @ts: timespec value to set the ctime
> + *
> + * Set the ctime in @inode to @ts.
> + */
> +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
> +{
> + inode->i_ctime = ts;
> + return ts;
> +}
> +
> +/**
> + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
> + * @inode: inode in which to set the ctime
> + * @sec: value to set the tv_sec field
> + *
> + * Set the sec field in the ctime. Returns @sec.
> + */
> +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
> +{
> + inode->i_ctime.tv_sec = sec;
> + return sec;
> +}
> +
> +/**
> + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
> + * @inode: inode in which to set the ctime
> + * @nsec: value to set the tv_nsec field
> + *
> + * Set the nsec field in the ctime. Returns @nsec.
> + */
> +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
> +{
> + inode->i_ctime.tv_nsec = nsec;
> + return nsec;
> +}
>
> /*
> * Snapshotting support.
> --
> 2.41.0
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-21 16:34 ` Jan Kara
0 siblings, 0 replies; 229+ messages in thread
From: Jan Kara @ 2023-06-21 16:34 UTC (permalink / raw)
To: Jeff Layton
Cc: Latchesar Ionkov, Rafael J. Wysocki, Darrick J. Wong,
Anders Larsen, Carlos Llamas, Andrii Nakryiko, Hugh Dickins,
John Johansen, Seth Forshee, Alexander Gordeev,
Christoph Hellwig, Mike Marshall, Paulo Alcantara, linux-xfs,
Bart Van Assche, Michael Ellerman, John Keeping, Zhang Yi,
James Morris, Christophe Leroy, Tyler Hicks, Alan Stern,
Christian Borntraeger, devel, Shyam Prasad N, Jan Harkes,
linux-um
On Wed 21-06-23 10:45:06, Jeff Layton wrote:
> struct timespec64 has unused bits in the tv_nsec field that can be used
> for other purposes. In future patches, we're going to change how the
> inode->i_ctime is accessed in certain inodes in order to make use of
> them. In order to do that safely though, we'll need to eradicate raw
> accesses of the inode->i_ctime field from the kernel.
>
> Add new accessor functions for the ctime that we can use to replace them.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Looks good to me. Feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
Honza
> ---
> fs/inode.c | 16 ++++++++++++++
> include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
> 2 files changed, 68 insertions(+), 1 deletion(-)
>
> diff --git a/fs/inode.c b/fs/inode.c
> index d37fad91c8da..c005e7328fbb 100644
> --- a/fs/inode.c
> +++ b/fs/inode.c
> @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
> }
> EXPORT_SYMBOL(current_time);
>
> +/**
> + * inode_ctime_set_current - set the ctime to current_time
> + * @inode: inode
> + *
> + * Set the inode->i_ctime to the current value for the inode. Returns
> + * the current value that was assigned to i_ctime.
> + */
> +struct timespec64 inode_ctime_set_current(struct inode *inode)
> +{
> + struct timespec64 now = current_time(inode);
> +
> + inode_set_ctime(inode, now);
> + return now;
> +}
> +EXPORT_SYMBOL(inode_ctime_set_current);
> +
> /**
> * in_group_or_capable - check whether caller is CAP_FSETID privileged
> * @idmap: idmap of the mount @inode was found from
> diff --git a/include/linux/fs.h b/include/linux/fs.h
> index 6867512907d6..9afb30606373 100644
> --- a/include/linux/fs.h
> +++ b/include/linux/fs.h
> @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
> kgid_has_mapping(fs_userns, kgid);
> }
>
> -extern struct timespec64 current_time(struct inode *inode);
> +struct timespec64 current_time(struct inode *inode);
> +struct timespec64 inode_ctime_set_current(struct inode *inode);
> +
> +/**
> + * inode_ctime_peek - fetch the current ctime from the inode
> + * @inode: inode from which to fetch ctime
> + *
> + * Grab the current ctime from the inode and return it.
> + */
> +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
> +{
> + return inode->i_ctime;
> +}
> +
> +/**
> + * inode_ctime_set - set the ctime in the inode to the given value
> + * @inode: inode in which to set the ctime
> + * @ts: timespec value to set the ctime
> + *
> + * Set the ctime in @inode to @ts.
> + */
> +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
> +{
> + inode->i_ctime = ts;
> + return ts;
> +}
> +
> +/**
> + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
> + * @inode: inode in which to set the ctime
> + * @sec: value to set the tv_sec field
> + *
> + * Set the sec field in the ctime. Returns @sec.
> + */
> +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
> +{
> + inode->i_ctime.tv_sec = sec;
> + return sec;
> +}
> +
> +/**
> + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
> + * @inode: inode in which to set the ctime
> + * @nsec: value to set the tv_nsec field
> + *
> + * Set the nsec field in the ctime. Returns @nsec.
> + */
> +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
> +{
> + inode->i_ctime.tv_nsec = nsec;
> + return nsec;
> +}
>
> /*
> * Snapshotting support.
> --
> 2.41.0
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
^ permalink raw reply [flat|nested] 229+ messages in thread
* [Cluster-devel] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-21 16:34 ` Jan Kara
0 siblings, 0 replies; 229+ messages in thread
From: Jan Kara @ 2023-06-21 16:34 UTC (permalink / raw)
To: cluster-devel.redhat.com
On Wed 21-06-23 10:45:06, Jeff Layton wrote:
> struct timespec64 has unused bits in the tv_nsec field that can be used
> for other purposes. In future patches, we're going to change how the
> inode->i_ctime is accessed in certain inodes in order to make use of
> them. In order to do that safely though, we'll need to eradicate raw
> accesses of the inode->i_ctime field from the kernel.
>
> Add new accessor functions for the ctime that we can use to replace them.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Looks good to me. Feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
Honza
> ---
> fs/inode.c | 16 ++++++++++++++
> include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
> 2 files changed, 68 insertions(+), 1 deletion(-)
>
> diff --git a/fs/inode.c b/fs/inode.c
> index d37fad91c8da..c005e7328fbb 100644
> --- a/fs/inode.c
> +++ b/fs/inode.c
> @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
> }
> EXPORT_SYMBOL(current_time);
>
> +/**
> + * inode_ctime_set_current - set the ctime to current_time
> + * @inode: inode
> + *
> + * Set the inode->i_ctime to the current value for the inode. Returns
> + * the current value that was assigned to i_ctime.
> + */
> +struct timespec64 inode_ctime_set_current(struct inode *inode)
> +{
> + struct timespec64 now = current_time(inode);
> +
> + inode_set_ctime(inode, now);
> + return now;
> +}
> +EXPORT_SYMBOL(inode_ctime_set_current);
> +
> /**
> * in_group_or_capable - check whether caller is CAP_FSETID privileged
> * @idmap: idmap of the mount @inode was found from
> diff --git a/include/linux/fs.h b/include/linux/fs.h
> index 6867512907d6..9afb30606373 100644
> --- a/include/linux/fs.h
> +++ b/include/linux/fs.h
> @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
> kgid_has_mapping(fs_userns, kgid);
> }
>
> -extern struct timespec64 current_time(struct inode *inode);
> +struct timespec64 current_time(struct inode *inode);
> +struct timespec64 inode_ctime_set_current(struct inode *inode);
> +
> +/**
> + * inode_ctime_peek - fetch the current ctime from the inode
> + * @inode: inode from which to fetch ctime
> + *
> + * Grab the current ctime from the inode and return it.
> + */
> +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
> +{
> + return inode->i_ctime;
> +}
> +
> +/**
> + * inode_ctime_set - set the ctime in the inode to the given value
> + * @inode: inode in which to set the ctime
> + * @ts: timespec value to set the ctime
> + *
> + * Set the ctime in @inode to @ts.
> + */
> +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
> +{
> + inode->i_ctime = ts;
> + return ts;
> +}
> +
> +/**
> + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
> + * @inode: inode in which to set the ctime
> + * @sec: value to set the tv_sec field
> + *
> + * Set the sec field in the ctime. Returns @sec.
> + */
> +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
> +{
> + inode->i_ctime.tv_sec = sec;
> + return sec;
> +}
> +
> +/**
> + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
> + * @inode: inode in which to set the ctime
> + * @nsec: value to set the tv_nsec field
> + *
> + * Set the nsec field in the ctime. Returns @nsec.
> + */
> +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
> +{
> + inode->i_ctime.tv_nsec = nsec;
> + return nsec;
> +}
>
> /*
> * Snapshotting support.
> --
> 2.41.0
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-21 16:34 ` Jan Kara
0 siblings, 0 replies; 229+ messages in thread
From: Jan Kara @ 2023-06-21 16:34 UTC (permalink / raw)
To: Jeff Layton
Cc: Latchesar Ionkov, Rafael J. Wysocki, Darrick J. Wong,
Anders Larsen, Carlos Llamas, Andrii Nakryiko, Hugh Dickins,
John Johansen, Seth Forshee, Alexander Gordeev,
Christoph Hellwig, Mike Marshall, Paulo Alcantara, linux-xfs,
Bart Van Assche, Michael Ellerman, John Keeping, Zhang Yi,
James Morris, Christophe Leroy, Tyler Hicks, Alan Stern,
Christian Borntraeger, devel
On Wed 21-06-23 10:45:06, Jeff Layton wrote:
> struct timespec64 has unused bits in the tv_nsec field that can be used
> for other purposes. In future patches, we're going to change how the
> inode->i_ctime is accessed in certain inodes in order to make use of
> them. In order to do that safely though, we'll need to eradicate raw
> accesses of the inode->i_ctime field from the kernel.
>
> Add new accessor functions for the ctime that we can use to replace them.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Looks good to me. Feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
Honza
> ---
> fs/inode.c | 16 ++++++++++++++
> include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
> 2 files changed, 68 insertions(+), 1 deletion(-)
>
> diff --git a/fs/inode.c b/fs/inode.c
> index d37fad91c8da..c005e7328fbb 100644
> --- a/fs/inode.c
> +++ b/fs/inode.c
> @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
> }
> EXPORT_SYMBOL(current_time);
>
> +/**
> + * inode_ctime_set_current - set the ctime to current_time
> + * @inode: inode
> + *
> + * Set the inode->i_ctime to the current value for the inode. Returns
> + * the current value that was assigned to i_ctime.
> + */
> +struct timespec64 inode_ctime_set_current(struct inode *inode)
> +{
> + struct timespec64 now = current_time(inode);
> +
> + inode_set_ctime(inode, now);
> + return now;
> +}
> +EXPORT_SYMBOL(inode_ctime_set_current);
> +
> /**
> * in_group_or_capable - check whether caller is CAP_FSETID privileged
> * @idmap: idmap of the mount @inode was found from
> diff --git a/include/linux/fs.h b/include/linux/fs.h
> index 6867512907d6..9afb30606373 100644
> --- a/include/linux/fs.h
> +++ b/include/linux/fs.h
> @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
> kgid_has_mapping(fs_userns, kgid);
> }
>
> -extern struct timespec64 current_time(struct inode *inode);
> +struct timespec64 current_time(struct inode *inode);
> +struct timespec64 inode_ctime_set_current(struct inode *inode);
> +
> +/**
> + * inode_ctime_peek - fetch the current ctime from the inode
> + * @inode: inode from which to fetch ctime
> + *
> + * Grab the current ctime from the inode and return it.
> + */
> +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
> +{
> + return inode->i_ctime;
> +}
> +
> +/**
> + * inode_ctime_set - set the ctime in the inode to the given value
> + * @inode: inode in which to set the ctime
> + * @ts: timespec value to set the ctime
> + *
> + * Set the ctime in @inode to @ts.
> + */
> +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
> +{
> + inode->i_ctime = ts;
> + return ts;
> +}
> +
> +/**
> + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
> + * @inode: inode in which to set the ctime
> + * @sec: value to set the tv_sec field
> + *
> + * Set the sec field in the ctime. Returns @sec.
> + */
> +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
> +{
> + inode->i_ctime.tv_sec = sec;
> + return sec;
> +}
> +
> +/**
> + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
> + * @inode: inode in which to set the ctime
> + * @nsec: value to set the tv_nsec field
> + *
> + * Set the nsec field in the ctime. Returns @nsec.
> + */
> +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
> +{
> + inode->i_ctime.tv_nsec = nsec;
> + return nsec;
> +}
>
> /*
> * Snapshotting support.
> --
> 2.41.0
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 02/79] spufs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
@ 2023-06-21 16:34 ` Jan Kara
-1 siblings, 0 replies; 229+ messages in thread
From: Jan Kara @ 2023-06-21 16:34 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Al Viro, Jan Kara,
linuxppc-dev, linux-kernel
On Wed 21-06-23 10:45:15, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Looks good to me. Feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
Honza
> ---
> arch/powerpc/platforms/cell/spufs/inode.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c
> index ea807aa0c31a..55418395bd9a 100644
> --- a/arch/powerpc/platforms/cell/spufs/inode.c
> +++ b/arch/powerpc/platforms/cell/spufs/inode.c
> @@ -86,7 +86,7 @@ spufs_new_inode(struct super_block *sb, umode_t mode)
> inode->i_mode = mode;
> inode->i_uid = current_fsuid();
> inode->i_gid = current_fsgid();
> - inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> out:
> return inode;
> }
> --
> 2.41.0
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 02/79] spufs: switch to new ctime accessors
@ 2023-06-21 16:34 ` Jan Kara
0 siblings, 0 replies; 229+ messages in thread
From: Jan Kara @ 2023-06-21 16:34 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Jan Kara, Arnd Bergmann, linux-kernel,
Al Viro, Nicholas Piggin, linuxppc-dev, Jeremy Kerr
On Wed 21-06-23 10:45:15, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Looks good to me. Feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
Honza
> ---
> arch/powerpc/platforms/cell/spufs/inode.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c
> index ea807aa0c31a..55418395bd9a 100644
> --- a/arch/powerpc/platforms/cell/spufs/inode.c
> +++ b/arch/powerpc/platforms/cell/spufs/inode.c
> @@ -86,7 +86,7 @@ spufs_new_inode(struct super_block *sb, umode_t mode)
> inode->i_mode = mode;
> inode->i_uid = current_fsuid();
> inode->i_gid = current_fsgid();
> - inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> out:
> return inode;
> }
> --
> 2.41.0
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 03/79] s390: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 03/79] s390: " Jeff Layton
@ 2023-06-21 16:34 ` Jan Kara
2023-06-22 17:35 ` Alexander Gordeev
1 sibling, 0 replies; 229+ messages in thread
From: Jan Kara @ 2023-06-21 16:34 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle, Al Viro,
Jan Kara, linux-s390, linux-kernel
On Wed 21-06-23 10:45:16, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Looks good to me. Feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
Honza
> ---
> arch/s390/hypfs/inode.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c
> index ee919bfc8186..30fa336ec63e 100644
> --- a/arch/s390/hypfs/inode.c
> +++ b/arch/s390/hypfs/inode.c
> @@ -53,7 +53,7 @@ static void hypfs_update_update(struct super_block *sb)
> struct inode *inode = d_inode(sb_info->update_file);
>
> sb_info->last_update = ktime_get_seconds();
> - inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> }
>
> /* directory tree removal functions */
> @@ -101,7 +101,7 @@ static struct inode *hypfs_make_inode(struct super_block *sb, umode_t mode)
> ret->i_mode = mode;
> ret->i_uid = hypfs_info->uid;
> ret->i_gid = hypfs_info->gid;
> - ret->i_atime = ret->i_mtime = ret->i_ctime = current_time(ret);
> + ret->i_atime = ret->i_mtime = inode_ctime_set_current(ret);
> if (S_ISDIR(mode))
> set_nlink(ret, 2);
> }
> --
> 2.41.0
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 04/79] binderfs: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 04/79] binderfs: " Jeff Layton
2023-06-21 15:29 ` Greg Kroah-Hartman
@ 2023-06-21 16:35 ` Jan Kara
1 sibling, 0 replies; 229+ messages in thread
From: Jan Kara @ 2023-06-21 16:35 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Greg Kroah-Hartman, Arve Hjønnevåg,
Todd Kjos, Martijn Coenen, Joel Fernandes, Carlos Llamas,
Suren Baghdasaryan, Al Viro, Jan Kara, linux-kernel
On Wed 21-06-23 10:45:17, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Looks good to me. Feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
Honza
> ---
> drivers/android/binderfs.c | 8 ++++----
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/android/binderfs.c b/drivers/android/binderfs.c
> index 76e7d6676657..b6b1220e1e55 100644
> --- a/drivers/android/binderfs.c
> +++ b/drivers/android/binderfs.c
> @@ -153,7 +153,7 @@ static int binderfs_binder_device_create(struct inode *ref_inode,
> goto err;
>
> inode->i_ino = minor + INODE_OFFSET;
> - inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
> + inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
> init_special_inode(inode, S_IFCHR | 0600,
> MKDEV(MAJOR(binderfs_dev), minor));
> inode->i_fop = &binder_fops;
> @@ -432,7 +432,7 @@ static int binderfs_binder_ctl_create(struct super_block *sb)
> }
>
> inode->i_ino = SECOND_INODE;
> - inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
> + inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
> init_special_inode(inode, S_IFCHR | 0600,
> MKDEV(MAJOR(binderfs_dev), minor));
> inode->i_fop = &binder_ctl_fops;
> @@ -474,7 +474,7 @@ static struct inode *binderfs_make_inode(struct super_block *sb, int mode)
> if (ret) {
> ret->i_ino = iunique(sb, BINDERFS_MAX_MINOR + INODE_OFFSET);
> ret->i_mode = mode;
> - ret->i_atime = ret->i_mtime = ret->i_ctime = current_time(ret);
> + ret->i_atime = ret->i_mtime = inode_ctime_set_current(ret);
> }
> return ret;
> }
> @@ -703,7 +703,7 @@ static int binderfs_fill_super(struct super_block *sb, struct fs_context *fc)
> inode->i_ino = FIRST_INODE;
> inode->i_fop = &simple_dir_operations;
> inode->i_mode = S_IFDIR | 0755;
> - inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
> + inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
> inode->i_op = &binderfs_dir_inode_operations;
> set_nlink(inode, 2);
>
> --
> 2.41.0
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 05/79] qib_fs: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 05/79] qib_fs: " Jeff Layton
@ 2023-06-21 16:35 ` Jan Kara
2023-06-21 19:37 ` Dennis Dalessandro
1 sibling, 0 replies; 229+ messages in thread
From: Jan Kara @ 2023-06-21 16:35 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Al Viro, Jan Kara, linux-rdma, linux-kernel
On Wed 21-06-23 10:45:18, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Looks good to me. Feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
Honza
> ---
> drivers/infiniband/hw/qib/qib_fs.c | 4 +---
> 1 file changed, 1 insertion(+), 3 deletions(-)
>
> diff --git a/drivers/infiniband/hw/qib/qib_fs.c b/drivers/infiniband/hw/qib/qib_fs.c
> index a973905afd13..db008f2884e4 100644
> --- a/drivers/infiniband/hw/qib/qib_fs.c
> +++ b/drivers/infiniband/hw/qib/qib_fs.c
> @@ -64,9 +64,7 @@ static int qibfs_mknod(struct inode *dir, struct dentry *dentry,
> inode->i_uid = GLOBAL_ROOT_UID;
> inode->i_gid = GLOBAL_ROOT_GID;
> inode->i_blocks = 0;
> - inode->i_atime = current_time(inode);
> - inode->i_mtime = inode->i_atime;
> - inode->i_ctime = inode->i_atime;
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> inode->i_private = data;
> if (S_ISDIR(mode)) {
> inode->i_op = &simple_dir_inode_operations;
> --
> 2.41.0
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 06/79] ibm: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 06/79] ibm: " Jeff Layton
2023-06-21 15:29 ` Greg Kroah-Hartman
@ 2023-06-21 16:35 ` Jan Kara
1 sibling, 0 replies; 229+ messages in thread
From: Jan Kara @ 2023-06-21 16:35 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Arnd Bergmann, Greg Kroah-Hartman,
Brad Warrum, Ritu Agarwal, Al Viro, Jan Kara, linux-kernel
On Wed 21-06-23 10:45:19, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Looks good to me. Feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
Honza
> ---
> drivers/misc/ibmasm/ibmasmfs.c | 2 +-
> drivers/misc/ibmvmc.c | 2 +-
> 2 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/misc/ibmasm/ibmasmfs.c b/drivers/misc/ibmasm/ibmasmfs.c
> index 35fec1bf1b3d..34d41b2a0d89 100644
> --- a/drivers/misc/ibmasm/ibmasmfs.c
> +++ b/drivers/misc/ibmasm/ibmasmfs.c
> @@ -139,7 +139,7 @@ static struct inode *ibmasmfs_make_inode(struct super_block *sb, int mode)
> if (ret) {
> ret->i_ino = get_next_ino();
> ret->i_mode = mode;
> - ret->i_atime = ret->i_mtime = ret->i_ctime = current_time(ret);
> + ret->i_atime = ret->i_mtime = inode_ctime_set_current(ret);
> }
> return ret;
> }
> diff --git a/drivers/misc/ibmvmc.c b/drivers/misc/ibmvmc.c
> index d7c7f0305257..b69c978c2590 100644
> --- a/drivers/misc/ibmvmc.c
> +++ b/drivers/misc/ibmvmc.c
> @@ -1124,7 +1124,7 @@ static ssize_t ibmvmc_write(struct file *file, const char *buffer,
> goto out;
>
> inode = file_inode(file);
> - inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_mtime = inode_ctime_set_current(inode);
> mark_inode_dirty(inode);
>
> dev_dbg(adapter->dev, "write: file = 0x%lx, count = 0x%lx\n",
> --
> 2.41.0
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 07/79] usb: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 07/79] usb: " Jeff Layton
2023-06-21 15:29 ` Greg Kroah-Hartman
@ 2023-06-21 16:36 ` Jan Kara
1 sibling, 0 replies; 229+ messages in thread
From: Jan Kara @ 2023-06-21 16:36 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Greg Kroah-Hartman, Al Viro, Jan Kara,
linux-usb, linux-kernel
On Wed 21-06-23 10:45:20, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Looks good to me. Feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
Honza
> ---
> drivers/usb/core/devio.c | 16 ++++++++--------
> drivers/usb/gadget/function/f_fs.c | 6 +-----
> drivers/usb/gadget/legacy/inode.c | 3 +--
> 3 files changed, 10 insertions(+), 15 deletions(-)
>
> diff --git a/drivers/usb/core/devio.c b/drivers/usb/core/devio.c
> index 1a16a8bdea60..02f718e0deaf 100644
> --- a/drivers/usb/core/devio.c
> +++ b/drivers/usb/core/devio.c
> @@ -2642,21 +2642,21 @@ static long usbdev_do_ioctl(struct file *file, unsigned int cmd,
> snoop(&dev->dev, "%s: CONTROL\n", __func__);
> ret = proc_control(ps, p);
> if (ret >= 0)
> - inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_mtime = inode_ctime_set_current(inode);
> break;
>
> case USBDEVFS_BULK:
> snoop(&dev->dev, "%s: BULK\n", __func__);
> ret = proc_bulk(ps, p);
> if (ret >= 0)
> - inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_mtime = inode_ctime_set_current(inode);
> break;
>
> case USBDEVFS_RESETEP:
> snoop(&dev->dev, "%s: RESETEP\n", __func__);
> ret = proc_resetep(ps, p);
> if (ret >= 0)
> - inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_mtime = inode_ctime_set_current(inode);
> break;
>
> case USBDEVFS_RESET:
> @@ -2668,7 +2668,7 @@ static long usbdev_do_ioctl(struct file *file, unsigned int cmd,
> snoop(&dev->dev, "%s: CLEAR_HALT\n", __func__);
> ret = proc_clearhalt(ps, p);
> if (ret >= 0)
> - inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_mtime = inode_ctime_set_current(inode);
> break;
>
> case USBDEVFS_GETDRIVER:
> @@ -2695,7 +2695,7 @@ static long usbdev_do_ioctl(struct file *file, unsigned int cmd,
> snoop(&dev->dev, "%s: SUBMITURB\n", __func__);
> ret = proc_submiturb(ps, p);
> if (ret >= 0)
> - inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_mtime = inode_ctime_set_current(inode);
> break;
>
> #ifdef CONFIG_COMPAT
> @@ -2703,14 +2703,14 @@ static long usbdev_do_ioctl(struct file *file, unsigned int cmd,
> snoop(&dev->dev, "%s: CONTROL32\n", __func__);
> ret = proc_control_compat(ps, p);
> if (ret >= 0)
> - inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_mtime = inode_ctime_set_current(inode);
> break;
>
> case USBDEVFS_BULK32:
> snoop(&dev->dev, "%s: BULK32\n", __func__);
> ret = proc_bulk_compat(ps, p);
> if (ret >= 0)
> - inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_mtime = inode_ctime_set_current(inode);
> break;
>
> case USBDEVFS_DISCSIGNAL32:
> @@ -2722,7 +2722,7 @@ static long usbdev_do_ioctl(struct file *file, unsigned int cmd,
> snoop(&dev->dev, "%s: SUBMITURB32\n", __func__);
> ret = proc_submiturb_compat(ps, p);
> if (ret >= 0)
> - inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_mtime = inode_ctime_set_current(inode);
> break;
>
> case USBDEVFS_IOCTL32:
> diff --git a/drivers/usb/gadget/function/f_fs.c b/drivers/usb/gadget/function/f_fs.c
> index f41a385a5c42..756c78043a04 100644
> --- a/drivers/usb/gadget/function/f_fs.c
> +++ b/drivers/usb/gadget/function/f_fs.c
> @@ -1377,16 +1377,12 @@ ffs_sb_make_inode(struct super_block *sb, void *data,
> inode = new_inode(sb);
>
> if (inode) {
> - struct timespec64 ts = current_time(inode);
> -
> inode->i_ino = get_next_ino();
> inode->i_mode = perms->mode;
> inode->i_uid = perms->uid;
> inode->i_gid = perms->gid;
> - inode->i_atime = ts;
> - inode->i_mtime = ts;
> - inode->i_ctime = ts;
> inode->i_private = data;
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> if (fops)
> inode->i_fop = fops;
> if (iops)
> diff --git a/drivers/usb/gadget/legacy/inode.c b/drivers/usb/gadget/legacy/inode.c
> index 28249d0bf062..b83a68feb316 100644
> --- a/drivers/usb/gadget/legacy/inode.c
> +++ b/drivers/usb/gadget/legacy/inode.c
> @@ -1969,8 +1969,7 @@ gadgetfs_make_inode (struct super_block *sb,
> inode->i_mode = mode;
> inode->i_uid = make_kuid(&init_user_ns, default_uid);
> inode->i_gid = make_kgid(&init_user_ns, default_gid);
> - inode->i_atime = inode->i_mtime = inode->i_ctime
> - = current_time(inode);
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> inode->i_private = data;
> inode->i_fop = fops;
> }
> --
> 2.41.0
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 08/79] 9p: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 08/79] 9p: " Jeff Layton
@ 2023-06-21 16:37 ` Jan Kara
0 siblings, 0 replies; 229+ messages in thread
From: Jan Kara @ 2023-06-21 16:37 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Eric Van Hensbergen, Latchesar Ionkov,
Dominique Martinet, Christian Schoenebeck, Al Viro, Jan Kara,
v9fs, linux-kernel
On Wed 21-06-23 10:45:21, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Looks good to me. Feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
Honza
> ---
> fs/9p/vfs_inode.c | 6 ++++--
> fs/9p/vfs_inode_dotl.c | 11 +++++------
> 2 files changed, 9 insertions(+), 8 deletions(-)
>
> diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c
> index 36b466e35887..098e4c7160a8 100644
> --- a/fs/9p/vfs_inode.c
> +++ b/fs/9p/vfs_inode.c
> @@ -261,7 +261,7 @@ int v9fs_init_inode(struct v9fs_session_info *v9ses,
> inode_init_owner(&nop_mnt_idmap, inode, NULL, mode);
> inode->i_blocks = 0;
> inode->i_rdev = rdev;
> - inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> inode->i_mapping->a_ops = &v9fs_addr_operations;
> inode->i_private = NULL;
>
> @@ -1153,12 +1153,14 @@ v9fs_stat2inode(struct p9_wstat *stat, struct inode *inode,
> umode_t mode;
> struct v9fs_session_info *v9ses = sb->s_fs_info;
> struct v9fs_inode *v9inode = V9FS_I(inode);
> + struct timespec64 ctime = { .tv_sec = stat->mtime,
> + .tv_nsec = 0 };
>
> set_nlink(inode, 1);
>
> inode->i_atime.tv_sec = stat->atime;
> inode->i_mtime.tv_sec = stat->mtime;
> - inode->i_ctime.tv_sec = stat->mtime;
> + inode_ctime_set(inode, ctime);
>
> inode->i_uid = v9ses->dfltuid;
> inode->i_gid = v9ses->dfltgid;
> diff --git a/fs/9p/vfs_inode_dotl.c b/fs/9p/vfs_inode_dotl.c
> index 5361cd2d7996..0041b5fc4407 100644
> --- a/fs/9p/vfs_inode_dotl.c
> +++ b/fs/9p/vfs_inode_dotl.c
> @@ -640,14 +640,15 @@ v9fs_stat2inode_dotl(struct p9_stat_dotl *stat, struct inode *inode,
> {
> umode_t mode;
> struct v9fs_inode *v9inode = V9FS_I(inode);
> + struct timespec64 ctime = { .tv_sec = stat->st_ctime_sec,
> + .tv_nsec = stat->st_ctime_nsec };
>
> if ((stat->st_result_mask & P9_STATS_BASIC) == P9_STATS_BASIC) {
> inode->i_atime.tv_sec = stat->st_atime_sec;
> inode->i_atime.tv_nsec = stat->st_atime_nsec;
> inode->i_mtime.tv_sec = stat->st_mtime_sec;
> inode->i_mtime.tv_nsec = stat->st_mtime_nsec;
> - inode->i_ctime.tv_sec = stat->st_ctime_sec;
> - inode->i_ctime.tv_nsec = stat->st_ctime_nsec;
> + inode_ctime_set(inode, ctime);
> inode->i_uid = stat->st_uid;
> inode->i_gid = stat->st_gid;
> set_nlink(inode, stat->st_nlink);
> @@ -668,10 +669,8 @@ v9fs_stat2inode_dotl(struct p9_stat_dotl *stat, struct inode *inode,
> inode->i_mtime.tv_sec = stat->st_mtime_sec;
> inode->i_mtime.tv_nsec = stat->st_mtime_nsec;
> }
> - if (stat->st_result_mask & P9_STATS_CTIME) {
> - inode->i_ctime.tv_sec = stat->st_ctime_sec;
> - inode->i_ctime.tv_nsec = stat->st_ctime_nsec;
> - }
> + if (stat->st_result_mask & P9_STATS_CTIME)
> + inode_ctime_set(inode, ctime);
> if (stat->st_result_mask & P9_STATS_UID)
> inode->i_uid = stat->st_uid;
> if (stat->st_result_mask & P9_STATS_GID)
> --
> 2.41.0
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 09/79] adfs: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 09/79] adfs: " Jeff Layton
@ 2023-06-21 16:38 ` Jan Kara
0 siblings, 0 replies; 229+ messages in thread
From: Jan Kara @ 2023-06-21 16:38 UTC (permalink / raw)
To: Jeff Layton; +Cc: Christian Brauner, Al Viro, Jan Kara, linux-kernel
On Wed 21-06-23 10:45:22, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Looks good to me. Feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
Honza
> ---
> fs/adfs/inode.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/fs/adfs/inode.c b/fs/adfs/inode.c
> index c3ac613d0975..266191e6b80c 100644
> --- a/fs/adfs/inode.c
> +++ b/fs/adfs/inode.c
> @@ -270,7 +270,7 @@ adfs_iget(struct super_block *sb, struct object_info *obj)
> inode->i_mode = adfs_atts2mode(sb, inode);
> adfs_adfs2unix_time(&inode->i_mtime, inode);
> inode->i_atime = inode->i_mtime;
> - inode->i_ctime = inode->i_mtime;
> + inode_ctime_set(inode, inode->i_mtime);
>
> if (S_ISDIR(inode->i_mode)) {
> inode->i_op = &adfs_dir_inode_operations;
> @@ -331,7 +331,7 @@ adfs_notify_change(struct mnt_idmap *idmap, struct dentry *dentry,
> if (ia_valid & ATTR_ATIME)
> inode->i_atime = attr->ia_atime;
> if (ia_valid & ATTR_CTIME)
> - inode->i_ctime = attr->ia_ctime;
> + inode_ctime_set(inode, attr->ia_ctime);
> if (ia_valid & ATTR_MODE) {
> ADFS_I(inode)->attr = adfs_mode2atts(sb, inode, attr->ia_mode);
> inode->i_mode = adfs_atts2mode(sb, inode);
> --
> 2.41.0
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 10/79] affs: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 10/79] affs: " Jeff Layton
@ 2023-06-21 16:39 ` Jan Kara
2023-06-22 11:37 ` David Sterba
1 sibling, 0 replies; 229+ messages in thread
From: Jan Kara @ 2023-06-21 16:39 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, David Sterba, Al Viro, Jan Kara,
linux-fsdevel, linux-kernel
On Wed 21-06-23 10:45:23, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Looks good to me. Feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
Honza
> ---
> fs/affs/amigaffs.c | 6 +++---
> fs/affs/inode.c | 17 +++++++++--------
> 2 files changed, 12 insertions(+), 11 deletions(-)
>
> diff --git a/fs/affs/amigaffs.c b/fs/affs/amigaffs.c
> index 29f11e10a7c7..2b508aa6707e 100644
> --- a/fs/affs/amigaffs.c
> +++ b/fs/affs/amigaffs.c
> @@ -60,7 +60,7 @@ affs_insert_hash(struct inode *dir, struct buffer_head *bh)
> mark_buffer_dirty_inode(dir_bh, dir);
> affs_brelse(dir_bh);
>
> - dir->i_mtime = dir->i_ctime = current_time(dir);
> + dir->i_mtime = inode_ctime_set_current(dir);
> inode_inc_iversion(dir);
> mark_inode_dirty(dir);
>
> @@ -114,7 +114,7 @@ affs_remove_hash(struct inode *dir, struct buffer_head *rem_bh)
>
> affs_brelse(bh);
>
> - dir->i_mtime = dir->i_ctime = current_time(dir);
> + dir->i_mtime = inode_ctime_set_current(dir);
> inode_inc_iversion(dir);
> mark_inode_dirty(dir);
>
> @@ -315,7 +315,7 @@ affs_remove_header(struct dentry *dentry)
> else
> clear_nlink(inode);
> affs_unlock_link(inode);
> - inode->i_ctime = current_time(inode);
> + inode_ctime_set_current(inode);
> mark_inode_dirty(inode);
>
> done:
> diff --git a/fs/affs/inode.c b/fs/affs/inode.c
> index 27f77a52c5c8..177bac4def5e 100644
> --- a/fs/affs/inode.c
> +++ b/fs/affs/inode.c
> @@ -19,6 +19,7 @@ struct inode *affs_iget(struct super_block *sb, unsigned long ino)
> {
> struct affs_sb_info *sbi = AFFS_SB(sb);
> struct buffer_head *bh;
> + struct timespec64 ctime;
> struct affs_tail *tail;
> struct inode *inode;
> u32 block;
> @@ -149,13 +150,13 @@ struct inode *affs_iget(struct super_block *sb, unsigned long ino)
> break;
> }
>
> - inode->i_mtime.tv_sec = inode->i_atime.tv_sec = inode->i_ctime.tv_sec
> - = (be32_to_cpu(tail->change.days) * 86400LL +
> - be32_to_cpu(tail->change.mins) * 60 +
> - be32_to_cpu(tail->change.ticks) / 50 +
> - AFFS_EPOCH_DELTA) +
> - sys_tz.tz_minuteswest * 60;
> - inode->i_mtime.tv_nsec = inode->i_ctime.tv_nsec = inode->i_atime.tv_nsec = 0;
> + ctime.tv_sec = (be32_to_cpu(tail->change.days) * 86400LL +
> + be32_to_cpu(tail->change.mins) * 60 +
> + be32_to_cpu(tail->change.ticks) / 50 +
> + AFFS_EPOCH_DELTA) +
> + sys_tz.tz_minuteswest * 60;
> + ctime.tv_nsec = 0;
> + inode->i_atime = inode->i_mtime = inode_ctime_set(inode, ctime);
> affs_brelse(bh);
> unlock_new_inode(inode);
> return inode;
> @@ -314,7 +315,7 @@ affs_new_inode(struct inode *dir)
> inode->i_gid = current_fsgid();
> inode->i_ino = block;
> set_nlink(inode, 1);
> - inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
> + inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
> atomic_set(&AFFS_I(inode)->i_opencnt, 0);
> AFFS_I(inode)->i_blkcnt = 0;
> AFFS_I(inode)->i_lc = NULL;
> --
> 2.41.0
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 12/79] fs: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 12/79] fs: " Jeff Layton
@ 2023-06-21 16:42 ` Jan Kara
0 siblings, 0 replies; 229+ messages in thread
From: Jan Kara @ 2023-06-21 16:42 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Alexander Viro, Eric Biederman, Kees Cook,
Jan Kara, linux-fsdevel, linux-kernel, linux-mm
On Wed 21-06-23 10:45:25, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Looks good to me. Feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
Honza
> ---
> fs/attr.c | 2 +-
> fs/bad_inode.c | 3 +--
> fs/binfmt_misc.c | 3 +--
> fs/inode.c | 12 ++++++++----
> fs/libfs.c | 32 +++++++++++++++++---------------
> fs/nsfs.c | 2 +-
> fs/pipe.c | 2 +-
> fs/posix_acl.c | 2 +-
> fs/stack.c | 2 +-
> fs/stat.c | 2 +-
> include/linux/fs_stack.h | 2 +-
> 11 files changed, 34 insertions(+), 30 deletions(-)
>
> diff --git a/fs/attr.c b/fs/attr.c
> index d60dc1edb526..2750e5f98dfb 100644
> --- a/fs/attr.c
> +++ b/fs/attr.c
> @@ -312,7 +312,7 @@ void setattr_copy(struct mnt_idmap *idmap, struct inode *inode,
> if (ia_valid & ATTR_MTIME)
> inode->i_mtime = attr->ia_mtime;
> if (ia_valid & ATTR_CTIME)
> - inode->i_ctime = attr->ia_ctime;
> + inode_ctime_set(inode, attr->ia_ctime);
> if (ia_valid & ATTR_MODE) {
> umode_t mode = attr->ia_mode;
> if (!in_group_or_capable(idmap, inode,
> diff --git a/fs/bad_inode.c b/fs/bad_inode.c
> index db649487d58c..bd3762e1b670 100644
> --- a/fs/bad_inode.c
> +++ b/fs/bad_inode.c
> @@ -209,8 +209,7 @@ void make_bad_inode(struct inode *inode)
> remove_inode_hash(inode);
>
> inode->i_mode = S_IFREG;
> - inode->i_atime = inode->i_mtime = inode->i_ctime =
> - current_time(inode);
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> inode->i_op = &bad_inode_ops;
> inode->i_opflags &= ~IOP_XATTR;
> inode->i_fop = &bad_file_ops;
> diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c
> index bb202ad369d5..6af92eb1b871 100644
> --- a/fs/binfmt_misc.c
> +++ b/fs/binfmt_misc.c
> @@ -547,8 +547,7 @@ static struct inode *bm_get_inode(struct super_block *sb, int mode)
> if (inode) {
> inode->i_ino = get_next_ino();
> inode->i_mode = mode;
> - inode->i_atime = inode->i_mtime = inode->i_ctime =
> - current_time(inode);
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> }
> return inode;
> }
> diff --git a/fs/inode.c b/fs/inode.c
> index c005e7328fbb..a7f484e9e7c1 100644
> --- a/fs/inode.c
> +++ b/fs/inode.c
> @@ -1851,6 +1851,7 @@ EXPORT_SYMBOL(bmap);
> static int relatime_need_update(struct vfsmount *mnt, struct inode *inode,
> struct timespec64 now)
> {
> + struct timespec64 ctime;
>
> if (!(mnt->mnt_flags & MNT_RELATIME))
> return 1;
> @@ -1862,7 +1863,8 @@ static int relatime_need_update(struct vfsmount *mnt, struct inode *inode,
> /*
> * Is ctime younger than or equal to atime? If yes, update atime:
> */
> - if (timespec64_compare(&inode->i_ctime, &inode->i_atime) >= 0)
> + ctime = inode_ctime_peek(inode);
> + if (timespec64_compare(&ctime, &inode->i_atime) >= 0)
> return 1;
>
> /*
> @@ -1885,7 +1887,7 @@ int generic_update_time(struct inode *inode, struct timespec64 *time, int flags)
> if (flags & S_ATIME)
> inode->i_atime = *time;
> if (flags & S_CTIME)
> - inode->i_ctime = *time;
> + inode_ctime_set(inode, *time);
> if (flags & S_MTIME)
> inode->i_mtime = *time;
>
> @@ -2071,6 +2073,7 @@ EXPORT_SYMBOL(file_remove_privs);
> static int inode_needs_update_time(struct inode *inode, struct timespec64 *now)
> {
> int sync_it = 0;
> + struct timespec64 ctime;
>
> /* First try to exhaust all avenues to not sync */
> if (IS_NOCMTIME(inode))
> @@ -2079,7 +2082,8 @@ static int inode_needs_update_time(struct inode *inode, struct timespec64 *now)
> if (!timespec64_equal(&inode->i_mtime, now))
> sync_it = S_MTIME;
>
> - if (!timespec64_equal(&inode->i_ctime, now))
> + ctime = inode_ctime_peek(inode);
> + if (!timespec64_equal(&ctime, now))
> sync_it |= S_CTIME;
>
> if (IS_I_VERSION(inode) && inode_iversion_need_inc(inode))
> @@ -2510,7 +2514,7 @@ struct timespec64 inode_ctime_set_current(struct inode *inode)
> {
> struct timespec64 now = current_time(inode);
>
> - inode_set_ctime(inode, now);
> + inode_ctime_set(inode, now);
> return now;
> }
> EXPORT_SYMBOL(inode_ctime_set_current);
> diff --git a/fs/libfs.c b/fs/libfs.c
> index 5b851315eeed..4a914f09fa87 100644
> --- a/fs/libfs.c
> +++ b/fs/libfs.c
> @@ -275,7 +275,7 @@ void simple_recursive_removal(struct dentry *dentry,
> while ((child = find_next_child(this, victim)) == NULL) {
> // kill and ascend
> // update metadata while it's still locked
> - inode->i_ctime = current_time(inode);
> + inode_ctime_set_current(inode);
> clear_nlink(inode);
> inode_unlock(inode);
> victim = this;
> @@ -293,8 +293,7 @@ void simple_recursive_removal(struct dentry *dentry,
> dput(victim); // unpin it
> }
> if (victim == dentry) {
> - inode->i_ctime = inode->i_mtime =
> - current_time(inode);
> + inode->i_mtime = inode_ctime_set_current(inode);
> if (d_is_dir(dentry))
> drop_nlink(inode);
> inode_unlock(inode);
> @@ -335,7 +334,7 @@ static int pseudo_fs_fill_super(struct super_block *s, struct fs_context *fc)
> */
> root->i_ino = 1;
> root->i_mode = S_IFDIR | S_IRUSR | S_IWUSR;
> - root->i_atime = root->i_mtime = root->i_ctime = current_time(root);
> + root->i_atime = root->i_mtime = inode_ctime_set_current(root);
> s->s_root = d_make_root(root);
> if (!s->s_root)
> return -ENOMEM;
> @@ -391,7 +390,8 @@ int simple_link(struct dentry *old_dentry, struct inode *dir, struct dentry *den
> {
> struct inode *inode = d_inode(old_dentry);
>
> - inode->i_ctime = dir->i_ctime = dir->i_mtime = current_time(inode);
> + inode_ctime_set_current(inode);
> + inode->i_mtime = inode_ctime_set_current(dir);
> inc_nlink(inode);
> ihold(inode);
> dget(dentry);
> @@ -425,7 +425,8 @@ int simple_unlink(struct inode *dir, struct dentry *dentry)
> {
> struct inode *inode = d_inode(dentry);
>
> - inode->i_ctime = dir->i_ctime = dir->i_mtime = current_time(inode);
> + inode_ctime_set_current(inode);
> + dir->i_mtime = inode_ctime_set_current(dir);
> drop_nlink(inode);
> dput(dentry);
> return 0;
> @@ -459,10 +460,10 @@ int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry,
> inc_nlink(old_dir);
> }
> }
> - old_dir->i_ctime = old_dir->i_mtime =
> - new_dir->i_ctime = new_dir->i_mtime =
> - d_inode(old_dentry)->i_ctime =
> - d_inode(new_dentry)->i_ctime = current_time(old_dir);
> + old_dir->i_mtime = inode_ctime_set_current(old_dir);
> + new_dir->i_mtime = inode_ctime_set_current(new_dir);
> + inode_ctime_set_current(d_inode(old_dentry));
> + inode_ctime_set_current(d_inode(new_dentry));
>
> return 0;
> }
> @@ -495,8 +496,9 @@ int simple_rename(struct mnt_idmap *idmap, struct inode *old_dir,
> inc_nlink(new_dir);
> }
>
> - old_dir->i_ctime = old_dir->i_mtime = new_dir->i_ctime =
> - new_dir->i_mtime = inode->i_ctime = current_time(old_dir);
> + old_dir->i_mtime = inode_ctime_set_current(old_dir);
> + new_dir->i_mtime = inode_ctime_set_current(new_dir);
> + inode_ctime_set_current(inode);
>
> return 0;
> }
> @@ -659,7 +661,7 @@ int simple_fill_super(struct super_block *s, unsigned long magic,
> */
> inode->i_ino = 1;
> inode->i_mode = S_IFDIR | 0755;
> - inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> inode->i_op = &simple_dir_inode_operations;
> inode->i_fop = &simple_dir_operations;
> set_nlink(inode, 2);
> @@ -685,7 +687,7 @@ int simple_fill_super(struct super_block *s, unsigned long magic,
> goto out;
> }
> inode->i_mode = S_IFREG | files->mode;
> - inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> inode->i_fop = files->ops;
> inode->i_ino = i;
> d_add(dentry, inode);
> @@ -1253,7 +1255,7 @@ struct inode *alloc_anon_inode(struct super_block *s)
> inode->i_uid = current_fsuid();
> inode->i_gid = current_fsgid();
> inode->i_flags |= S_PRIVATE;
> - inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> return inode;
> }
> EXPORT_SYMBOL(alloc_anon_inode);
> diff --git a/fs/nsfs.c b/fs/nsfs.c
> index f602a96a1afe..c052cc55eacd 100644
> --- a/fs/nsfs.c
> +++ b/fs/nsfs.c
> @@ -84,7 +84,7 @@ static int __ns_get_path(struct path *path, struct ns_common *ns)
> return -ENOMEM;
> }
> inode->i_ino = ns->inum;
> - inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
> + inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
> inode->i_flags |= S_IMMUTABLE;
> inode->i_mode = S_IFREG | S_IRUGO;
> inode->i_fop = &ns_file_operations;
> diff --git a/fs/pipe.c b/fs/pipe.c
> index 2d88f73f585a..bb90b6fc4a96 100644
> --- a/fs/pipe.c
> +++ b/fs/pipe.c
> @@ -899,7 +899,7 @@ static struct inode * get_pipe_inode(void)
> inode->i_mode = S_IFIFO | S_IRUSR | S_IWUSR;
> inode->i_uid = current_fsuid();
> inode->i_gid = current_fsgid();
> - inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
>
> return inode;
>
> diff --git a/fs/posix_acl.c b/fs/posix_acl.c
> index 7fa1b738bbab..cc9c390fd2af 100644
> --- a/fs/posix_acl.c
> +++ b/fs/posix_acl.c
> @@ -1027,7 +1027,7 @@ int simple_set_acl(struct mnt_idmap *idmap, struct dentry *dentry,
> return error;
> }
>
> - inode->i_ctime = current_time(inode);
> + inode_ctime_set_current(inode);
> if (IS_I_VERSION(inode))
> inode_inc_iversion(inode);
> set_cached_acl(inode, type, acl);
> diff --git a/fs/stack.c b/fs/stack.c
> index c9830924eb12..efd0de85bace 100644
> --- a/fs/stack.c
> +++ b/fs/stack.c
> @@ -68,7 +68,7 @@ void fsstack_copy_attr_all(struct inode *dest, const struct inode *src)
> dest->i_rdev = src->i_rdev;
> dest->i_atime = src->i_atime;
> dest->i_mtime = src->i_mtime;
> - dest->i_ctime = src->i_ctime;
> + inode_ctime_set(dest, inode_ctime_peek(src));
> dest->i_blkbits = src->i_blkbits;
> dest->i_flags = src->i_flags;
> set_nlink(dest, src->i_nlink);
> diff --git a/fs/stat.c b/fs/stat.c
> index 7c238da22ef0..5d87e34d6dd5 100644
> --- a/fs/stat.c
> +++ b/fs/stat.c
> @@ -58,7 +58,7 @@ void generic_fillattr(struct mnt_idmap *idmap, struct inode *inode,
> stat->size = i_size_read(inode);
> stat->atime = inode->i_atime;
> stat->mtime = inode->i_mtime;
> - stat->ctime = inode->i_ctime;
> + stat->ctime = inode_ctime_peek(inode);
> stat->blksize = i_blocksize(inode);
> stat->blocks = inode->i_blocks;
> }
> diff --git a/include/linux/fs_stack.h b/include/linux/fs_stack.h
> index 54210a42c30d..1488a118fe91 100644
> --- a/include/linux/fs_stack.h
> +++ b/include/linux/fs_stack.h
> @@ -24,7 +24,7 @@ static inline void fsstack_copy_attr_times(struct inode *dest,
> {
> dest->i_atime = src->i_atime;
> dest->i_mtime = src->i_mtime;
> - dest->i_ctime = src->i_ctime;
> + inode_ctime_set(dest, inode_ctime_peek(src));
> }
>
> #endif /* _LINUX_FS_STACK_H */
> --
> 2.41.0
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 11/79] afs: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 11/79] afs: " Jeff Layton
@ 2023-06-21 16:43 ` Jan Kara
0 siblings, 0 replies; 229+ messages in thread
From: Jan Kara @ 2023-06-21 16:43 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, David Howells, Marc Dionne, Al Viro, Jan Kara,
linux-afs, linux-kernel
On Wed 21-06-23 10:45:24, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Looks good to me. Feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
Honza
> ---
> fs/afs/dynroot.c | 2 +-
> fs/afs/inode.c | 6 +++---
> 2 files changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/fs/afs/dynroot.c b/fs/afs/dynroot.c
> index d7d9402ff718..ee36375de337 100644
> --- a/fs/afs/dynroot.c
> +++ b/fs/afs/dynroot.c
> @@ -88,9 +88,9 @@ struct inode *afs_iget_pseudo_dir(struct super_block *sb, bool root)
> set_nlink(inode, 2);
> inode->i_uid = GLOBAL_ROOT_UID;
> inode->i_gid = GLOBAL_ROOT_GID;
> - inode->i_ctime = inode->i_atime = inode->i_mtime = current_time(inode);
> inode->i_blocks = 0;
> inode->i_generation = 0;
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
>
> set_bit(AFS_VNODE_PSEUDODIR, &vnode->flags);
> if (!root) {
> diff --git a/fs/afs/inode.c b/fs/afs/inode.c
> index 866bab860a88..0b59903c2a43 100644
> --- a/fs/afs/inode.c
> +++ b/fs/afs/inode.c
> @@ -90,7 +90,7 @@ static int afs_inode_init_from_status(struct afs_operation *op,
> vnode->status = *status;
>
> t = status->mtime_client;
> - inode->i_ctime = t;
> + inode_ctime_set(inode, t);
> inode->i_mtime = t;
> inode->i_atime = t;
> inode->i_flags |= S_NOATIME;
> @@ -206,7 +206,7 @@ static void afs_apply_status(struct afs_operation *op,
> t = status->mtime_client;
> inode->i_mtime = t;
> if (vp->update_ctime)
> - inode->i_ctime = op->ctime;
> + inode_ctime_set(inode, op->ctime);
>
> if (vnode->status.data_version != status->data_version)
> data_changed = true;
> @@ -252,7 +252,7 @@ static void afs_apply_status(struct afs_operation *op,
> vnode->netfs.remote_i_size = status->size;
> if (change_size) {
> afs_set_i_size(vnode, status->size);
> - inode->i_ctime = t;
> + inode_ctime_set(inode, t);
> inode->i_atime = t;
> }
> }
> --
> 2.41.0
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 13/79] autofs: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 13/79] autofs: " Jeff Layton
@ 2023-06-21 16:43 ` Jan Kara
2023-06-27 1:48 ` Ian Kent
1 sibling, 0 replies; 229+ messages in thread
From: Jan Kara @ 2023-06-21 16:43 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Ian Kent, Al Viro, Jan Kara, autofs, linux-kernel
On Wed 21-06-23 10:45:26, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Looks good to me. Feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
Honza
> ---
> fs/autofs/inode.c | 2 +-
> fs/autofs/root.c | 6 +++---
> 2 files changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/fs/autofs/inode.c b/fs/autofs/inode.c
> index affa70360b1f..47e3054b29dc 100644
> --- a/fs/autofs/inode.c
> +++ b/fs/autofs/inode.c
> @@ -370,7 +370,7 @@ struct inode *autofs_get_inode(struct super_block *sb, umode_t mode)
> inode->i_uid = d_inode(sb->s_root)->i_uid;
> inode->i_gid = d_inode(sb->s_root)->i_gid;
> }
> - inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> inode->i_ino = get_next_ino();
>
> if (S_ISDIR(mode)) {
> diff --git a/fs/autofs/root.c b/fs/autofs/root.c
> index 93046c9dc461..4c0fc0f8d688 100644
> --- a/fs/autofs/root.c
> +++ b/fs/autofs/root.c
> @@ -600,7 +600,7 @@ static int autofs_dir_symlink(struct mnt_idmap *idmap,
> p_ino = autofs_dentry_ino(dentry->d_parent);
> p_ino->count++;
>
> - dir->i_mtime = dir->i_ctime = current_time(dir);
> + dir->i_mtime = inode_ctime_set_current(dir);
>
> return 0;
> }
> @@ -633,7 +633,7 @@ static int autofs_dir_unlink(struct inode *dir, struct dentry *dentry)
> d_inode(dentry)->i_size = 0;
> clear_nlink(d_inode(dentry));
>
> - dir->i_mtime = dir->i_ctime = current_time(dir);
> + dir->i_mtime = inode_ctime_set_current(dir);
>
> spin_lock(&sbi->lookup_lock);
> __autofs_add_expiring(dentry);
> @@ -749,7 +749,7 @@ static int autofs_dir_mkdir(struct mnt_idmap *idmap,
> p_ino = autofs_dentry_ino(dentry->d_parent);
> p_ino->count++;
> inc_nlink(dir);
> - dir->i_mtime = dir->i_ctime = current_time(dir);
> + dir->i_mtime = inode_ctime_set_current(dir);
>
> return 0;
> }
> --
> 2.41.0
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 14/79] befs: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 14/79] befs: " Jeff Layton
@ 2023-06-21 16:44 ` Jan Kara
0 siblings, 0 replies; 229+ messages in thread
From: Jan Kara @ 2023-06-21 16:44 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Luis de Bethencourt, Salah Triki, Al Viro,
Jan Kara, linux-kernel
On Wed 21-06-23 10:45:27, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Looks good to me. Feel free to add:
Reviewed-by: Jan Kara <jack@suse.cz>
Honza
> ---
> fs/befs/linuxvfs.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c
> index eee9237386e2..1eee7baa808b 100644
> --- a/fs/befs/linuxvfs.c
> +++ b/fs/befs/linuxvfs.c
> @@ -363,7 +363,7 @@ static struct inode *befs_iget(struct super_block *sb, unsigned long ino)
> inode->i_mtime.tv_sec =
> fs64_to_cpu(sb, raw_inode->last_modified_time) >> 16;
> inode->i_mtime.tv_nsec = 0; /* lower 16 bits are not a time */
> - inode->i_ctime = inode->i_mtime;
> + inode_ctime_set(inode, inode->i_mtime);
> inode->i_atime = inode->i_mtime;
>
> befs_ino->i_inode_num = fsrun_to_cpu(sb, raw_inode->inode_num);
> --
> 2.41.0
>
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 15/79] bfs: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 15/79] bfs: " Jeff Layton
@ 2023-06-21 16:48 ` Jan Kara
2023-06-21 16:57 ` Jeff Layton
0 siblings, 1 reply; 229+ messages in thread
From: Jan Kara @ 2023-06-21 16:48 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Tigran A. Aivazian, Al Viro, Jan Kara, linux-kernel
On Wed 21-06-23 10:45:28, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
...
> diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c
> index 1926bec2c850..c964316be32b 100644
> --- a/fs/bfs/inode.c
> +++ b/fs/bfs/inode.c
> @@ -82,10 +82,10 @@ struct inode *bfs_iget(struct super_block *sb, unsigned long ino)
> inode->i_blocks = BFS_FILEBLOCKS(di);
> inode->i_atime.tv_sec = le32_to_cpu(di->i_atime);
> inode->i_mtime.tv_sec = le32_to_cpu(di->i_mtime);
> - inode->i_ctime.tv_sec = le32_to_cpu(di->i_ctime);
> + inode_ctime_set_sec(inode, le32_to_cpu(di->i_ctime));
> inode->i_atime.tv_nsec = 0;
> inode->i_mtime.tv_nsec = 0;
> - inode->i_ctime.tv_nsec = 0;
> + inode_ctime_set_nsec(inode, 0);
So I'm somewhat wondering here - in other filesystem you construct
timespec64 and then use inode_ctime_set(). Here you use
inode_ctime_set_sec() + inode_ctime_set_nsec(). What's the benefit? It
seems these two functions are not used that much some maybe we could just
live with just inode_ctime_set() and constructing timespec64 when needed?
Honza
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 15/79] bfs: switch to new ctime accessors
2023-06-21 16:48 ` Jan Kara
@ 2023-06-21 16:57 ` Jeff Layton
2023-06-22 12:30 ` Jan Kara
0 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 16:57 UTC (permalink / raw)
To: Jan Kara; +Cc: Christian Brauner, Tigran A. Aivazian, Al Viro, linux-kernel
On Wed, 2023-06-21 at 18:48 +0200, Jan Kara wrote:
> On Wed 21-06-23 10:45:28, Jeff Layton wrote:
> > In later patches, we're going to change how the ctime.tv_nsec field is
> > utilized. Switch to using accessor functions instead of raw accesses of
> > inode->i_ctime.
> >
> > Signed-off-by: Jeff Layton <jlayton@kernel.org>
>
> ...
>
> > diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c
> > index 1926bec2c850..c964316be32b 100644
> > --- a/fs/bfs/inode.c
> > +++ b/fs/bfs/inode.c
> > @@ -82,10 +82,10 @@ struct inode *bfs_iget(struct super_block *sb, unsigned long ino)
> > inode->i_blocks = BFS_FILEBLOCKS(di);
> > inode->i_atime.tv_sec = le32_to_cpu(di->i_atime);
> > inode->i_mtime.tv_sec = le32_to_cpu(di->i_mtime);
> > - inode->i_ctime.tv_sec = le32_to_cpu(di->i_ctime);
> > + inode_ctime_set_sec(inode, le32_to_cpu(di->i_ctime));
> > inode->i_atime.tv_nsec = 0;
> > inode->i_mtime.tv_nsec = 0;
> > - inode->i_ctime.tv_nsec = 0;
> > + inode_ctime_set_nsec(inode, 0);
>
> So I'm somewhat wondering here - in other filesystem you construct
> timespec64 and then use inode_ctime_set(). Here you use
> inode_ctime_set_sec() + inode_ctime_set_nsec(). What's the benefit? It
> seems these two functions are not used that much some maybe we could just
> live with just inode_ctime_set() and constructing timespec64 when needed?
>
> Honza
The main advantage is that by using that, I didn't need to do quite so
much of this conversion by hand. My coccinelle skills are pretty
primitive. I went with whatever conversion was going to give minimal
changes, to the existing accesses for the most part.
We could certainly do it the way you suggest, it just means having to
re-touch a lot of this code by hand, or someone with better coccinelle
chops suggesting a way to declare a temporary variables in place.
--
Jeff Layton <jlayton@kernel.org>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
2023-06-21 14:45 ` [f2fs-dev] " Jeff Layton
(?)
(?)
@ 2023-06-21 17:29 ` Tom Talpey via Ocfs2-devel
-1 siblings, 0 replies; 229+ messages in thread
From: Tom Talpey @ 2023-06-21 17:29 UTC (permalink / raw)
To: Jeff Layton, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On 6/21/2023 10:45 AM, Jeff Layton wrote:
> struct timespec64 has unused bits in the tv_nsec field that can be used
> for other purposes. In future patches, we're going to change how the
> inode->i_ctime is accessed in certain inodes in order to make use of
> them. In order to do that safely though, we'll need to eradicate raw
> accesses of the inode->i_ctime field from the kernel.
>
> Add new accessor functions for the ctime that we can use to replace them.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
> ---
> fs/inode.c | 16 ++++++++++++++
> include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
> 2 files changed, 68 insertions(+), 1 deletion(-)
>
> diff --git a/fs/inode.c b/fs/inode.c
> index d37fad91c8da..c005e7328fbb 100644
> --- a/fs/inode.c
> +++ b/fs/inode.c
> @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
> }
> EXPORT_SYMBOL(current_time);
>
> +/**
> + * inode_ctime_set_current - set the ctime to current_time
> + * @inode: inode
> + *
> + * Set the inode->i_ctime to the current value for the inode. Returns
> + * the current value that was assigned to i_ctime.
> + */
> +struct timespec64 inode_ctime_set_current(struct inode *inode)
> +{
> + struct timespec64 now = current_time(inode);
> +
> + inode_set_ctime(inode, now);
> + return now;
> +}
> +EXPORT_SYMBOL(inode_ctime_set_current);
> +
> /**
> * in_group_or_capable - check whether caller is CAP_FSETID privileged
> * @idmap: idmap of the mount @inode was found from
> diff --git a/include/linux/fs.h b/include/linux/fs.h
> index 6867512907d6..9afb30606373 100644
> --- a/include/linux/fs.h
> +++ b/include/linux/fs.h
> @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
> kgid_has_mapping(fs_userns, kgid);
> }
>
> -extern struct timespec64 current_time(struct inode *inode);
> +struct timespec64 current_time(struct inode *inode);
> +struct timespec64 inode_ctime_set_current(struct inode *inode);
> +
> +/**
> + * inode_ctime_peek - fetch the current ctime from the inode
> + * @inode: inode from which to fetch ctime
> + *
> + * Grab the current ctime from the inode and return it.
> + */
> +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
> +{
> + return inode->i_ctime;
> +}
> +
> +/**
> + * inode_ctime_set - set the ctime in the inode to the given value
> + * @inode: inode in which to set the ctime
> + * @ts: timespec value to set the ctime
> + *
> + * Set the ctime in @inode to @ts.
> + */
> +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
> +{
> + inode->i_ctime = ts;
> + return ts;
> +}
> +
> +/**
> + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
I'm curious about why you choose to split the tv_sec and tv_nsec
set_ functions. Do any callers not set them both? Wouldn't a
single call enable a more atomic behavior someday?
inode_ctime_set_sec_nsec(struct inode *, time64_t, time64_t)
(or simply initialize a timespec64 and use inode_ctime_spec() )
Tom.
> + * @inode: inode in which to set the ctime
> + * @sec: value to set the tv_sec field
> + *
> + * Set the sec field in the ctime. Returns @sec.
> + */
> +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
> +{
> + inode->i_ctime.tv_sec = sec;
> + return sec;
> +}
> +
> +/**
> + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
> + * @inode: inode in which to set the ctime
> + * @nsec: value to set the tv_nsec field
> + *
> + * Set the nsec field in the ctime. Returns @nsec.
> + */
> +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
> +{
> + inode->i_ctime.tv_nsec = nsec;
> + return nsec;
> +}
>
> /*
> * Snapshotting support.
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [Ocfs2-devel] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-21 17:29 ` Tom Talpey via Ocfs2-devel
0 siblings, 0 replies; 229+ messages in thread
From: Tom Talpey via Ocfs2-devel @ 2023-06-21 17:29 UTC (permalink / raw)
To: Jeff Layton, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On 6/21/2023 10:45 AM, Jeff Layton wrote:
> struct timespec64 has unused bits in the tv_nsec field that can be used
> for other purposes. In future patches, we're going to change how the
> inode->i_ctime is accessed in certain inodes in order to make use of
> them. In order to do that safely though, we'll need to eradicate raw
> accesses of the inode->i_ctime field from the kernel.
>
> Add new accessor functions for the ctime that we can use to replace them.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
> ---
> fs/inode.c | 16 ++++++++++++++
> include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
> 2 files changed, 68 insertions(+), 1 deletion(-)
>
> diff --git a/fs/inode.c b/fs/inode.c
> index d37fad91c8da..c005e7328fbb 100644
> --- a/fs/inode.c
> +++ b/fs/inode.c
> @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
> }
> EXPORT_SYMBOL(current_time);
>
> +/**
> + * inode_ctime_set_current - set the ctime to current_time
> + * @inode: inode
> + *
> + * Set the inode->i_ctime to the current value for the inode. Returns
> + * the current value that was assigned to i_ctime.
> + */
> +struct timespec64 inode_ctime_set_current(struct inode *inode)
> +{
> + struct timespec64 now = current_time(inode);
> +
> + inode_set_ctime(inode, now);
> + return now;
> +}
> +EXPORT_SYMBOL(inode_ctime_set_current);
> +
> /**
> * in_group_or_capable - check whether caller is CAP_FSETID privileged
> * @idmap: idmap of the mount @inode was found from
> diff --git a/include/linux/fs.h b/include/linux/fs.h
> index 6867512907d6..9afb30606373 100644
> --- a/include/linux/fs.h
> +++ b/include/linux/fs.h
> @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
> kgid_has_mapping(fs_userns, kgid);
> }
>
> -extern struct timespec64 current_time(struct inode *inode);
> +struct timespec64 current_time(struct inode *inode);
> +struct timespec64 inode_ctime_set_current(struct inode *inode);
> +
> +/**
> + * inode_ctime_peek - fetch the current ctime from the inode
> + * @inode: inode from which to fetch ctime
> + *
> + * Grab the current ctime from the inode and return it.
> + */
> +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
> +{
> + return inode->i_ctime;
> +}
> +
> +/**
> + * inode_ctime_set - set the ctime in the inode to the given value
> + * @inode: inode in which to set the ctime
> + * @ts: timespec value to set the ctime
> + *
> + * Set the ctime in @inode to @ts.
> + */
> +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
> +{
> + inode->i_ctime = ts;
> + return ts;
> +}
> +
> +/**
> + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
I'm curious about why you choose to split the tv_sec and tv_nsec
set_ functions. Do any callers not set them both? Wouldn't a
single call enable a more atomic behavior someday?
inode_ctime_set_sec_nsec(struct inode *, time64_t, time64_t)
(or simply initialize a timespec64 and use inode_ctime_spec() )
Tom.
> + * @inode: inode in which to set the ctime
> + * @sec: value to set the tv_sec field
> + *
> + * Set the sec field in the ctime. Returns @sec.
> + */
> +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
> +{
> + inode->i_ctime.tv_sec = sec;
> + return sec;
> +}
> +
> +/**
> + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
> + * @inode: inode in which to set the ctime
> + * @nsec: value to set the tv_nsec field
> + *
> + * Set the nsec field in the ctime. Returns @nsec.
> + */
> +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
> +{
> + inode->i_ctime.tv_nsec = nsec;
> + return nsec;
> +}
>
> /*
> * Snapshotting support.
_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [f2fs-dev] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-21 17:29 ` Tom Talpey via Ocfs2-devel
0 siblings, 0 replies; 229+ messages in thread
From: Tom Talpey @ 2023-06-21 17:29 UTC (permalink / raw)
To: Jeff Layton, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On 6/21/2023 10:45 AM, Jeff Layton wrote:
> struct timespec64 has unused bits in the tv_nsec field that can be used
> for other purposes. In future patches, we're going to change how the
> inode->i_ctime is accessed in certain inodes in order to make use of
> them. In order to do that safely though, we'll need to eradicate raw
> accesses of the inode->i_ctime field from the kernel.
>
> Add new accessor functions for the ctime that we can use to replace them.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
> ---
> fs/inode.c | 16 ++++++++++++++
> include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
> 2 files changed, 68 insertions(+), 1 deletion(-)
>
> diff --git a/fs/inode.c b/fs/inode.c
> index d37fad91c8da..c005e7328fbb 100644
> --- a/fs/inode.c
> +++ b/fs/inode.c
> @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
> }
> EXPORT_SYMBOL(current_time);
>
> +/**
> + * inode_ctime_set_current - set the ctime to current_time
> + * @inode: inode
> + *
> + * Set the inode->i_ctime to the current value for the inode. Returns
> + * the current value that was assigned to i_ctime.
> + */
> +struct timespec64 inode_ctime_set_current(struct inode *inode)
> +{
> + struct timespec64 now = current_time(inode);
> +
> + inode_set_ctime(inode, now);
> + return now;
> +}
> +EXPORT_SYMBOL(inode_ctime_set_current);
> +
> /**
> * in_group_or_capable - check whether caller is CAP_FSETID privileged
> * @idmap: idmap of the mount @inode was found from
> diff --git a/include/linux/fs.h b/include/linux/fs.h
> index 6867512907d6..9afb30606373 100644
> --- a/include/linux/fs.h
> +++ b/include/linux/fs.h
> @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
> kgid_has_mapping(fs_userns, kgid);
> }
>
> -extern struct timespec64 current_time(struct inode *inode);
> +struct timespec64 current_time(struct inode *inode);
> +struct timespec64 inode_ctime_set_current(struct inode *inode);
> +
> +/**
> + * inode_ctime_peek - fetch the current ctime from the inode
> + * @inode: inode from which to fetch ctime
> + *
> + * Grab the current ctime from the inode and return it.
> + */
> +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
> +{
> + return inode->i_ctime;
> +}
> +
> +/**
> + * inode_ctime_set - set the ctime in the inode to the given value
> + * @inode: inode in which to set the ctime
> + * @ts: timespec value to set the ctime
> + *
> + * Set the ctime in @inode to @ts.
> + */
> +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
> +{
> + inode->i_ctime = ts;
> + return ts;
> +}
> +
> +/**
> + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
I'm curious about why you choose to split the tv_sec and tv_nsec
set_ functions. Do any callers not set them both? Wouldn't a
single call enable a more atomic behavior someday?
inode_ctime_set_sec_nsec(struct inode *, time64_t, time64_t)
(or simply initialize a timespec64 and use inode_ctime_spec() )
Tom.
> + * @inode: inode in which to set the ctime
> + * @sec: value to set the tv_sec field
> + *
> + * Set the sec field in the ctime. Returns @sec.
> + */
> +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
> +{
> + inode->i_ctime.tv_sec = sec;
> + return sec;
> +}
> +
> +/**
> + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
> + * @inode: inode in which to set the ctime
> + * @nsec: value to set the tv_nsec field
> + *
> + * Set the nsec field in the ctime. Returns @nsec.
> + */
> +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
> +{
> + inode->i_ctime.tv_nsec = nsec;
> + return nsec;
> +}
>
> /*
> * Snapshotting support.
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* [Cluster-devel] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-21 17:29 ` Tom Talpey via Ocfs2-devel
0 siblings, 0 replies; 229+ messages in thread
From: Tom Talpey @ 2023-06-21 17:29 UTC (permalink / raw)
To: cluster-devel.redhat.com
On 6/21/2023 10:45 AM, Jeff Layton wrote:
> struct timespec64 has unused bits in the tv_nsec field that can be used
> for other purposes. In future patches, we're going to change how the
> inode->i_ctime is accessed in certain inodes in order to make use of
> them. In order to do that safely though, we'll need to eradicate raw
> accesses of the inode->i_ctime field from the kernel.
>
> Add new accessor functions for the ctime that we can use to replace them.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
> ---
> fs/inode.c | 16 ++++++++++++++
> include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
> 2 files changed, 68 insertions(+), 1 deletion(-)
>
> diff --git a/fs/inode.c b/fs/inode.c
> index d37fad91c8da..c005e7328fbb 100644
> --- a/fs/inode.c
> +++ b/fs/inode.c
> @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
> }
> EXPORT_SYMBOL(current_time);
>
> +/**
> + * inode_ctime_set_current - set the ctime to current_time
> + * @inode: inode
> + *
> + * Set the inode->i_ctime to the current value for the inode. Returns
> + * the current value that was assigned to i_ctime.
> + */
> +struct timespec64 inode_ctime_set_current(struct inode *inode)
> +{
> + struct timespec64 now = current_time(inode);
> +
> + inode_set_ctime(inode, now);
> + return now;
> +}
> +EXPORT_SYMBOL(inode_ctime_set_current);
> +
> /**
> * in_group_or_capable - check whether caller is CAP_FSETID privileged
> * @idmap: idmap of the mount @inode was found from
> diff --git a/include/linux/fs.h b/include/linux/fs.h
> index 6867512907d6..9afb30606373 100644
> --- a/include/linux/fs.h
> +++ b/include/linux/fs.h
> @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
> kgid_has_mapping(fs_userns, kgid);
> }
>
> -extern struct timespec64 current_time(struct inode *inode);
> +struct timespec64 current_time(struct inode *inode);
> +struct timespec64 inode_ctime_set_current(struct inode *inode);
> +
> +/**
> + * inode_ctime_peek - fetch the current ctime from the inode
> + * @inode: inode from which to fetch ctime
> + *
> + * Grab the current ctime from the inode and return it.
> + */
> +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
> +{
> + return inode->i_ctime;
> +}
> +
> +/**
> + * inode_ctime_set - set the ctime in the inode to the given value
> + * @inode: inode in which to set the ctime
> + * @ts: timespec value to set the ctime
> + *
> + * Set the ctime in @inode to @ts.
> + */
> +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
> +{
> + inode->i_ctime = ts;
> + return ts;
> +}
> +
> +/**
> + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
I'm curious about why you choose to split the tv_sec and tv_nsec
set_ functions. Do any callers not set them both? Wouldn't a
single call enable a more atomic behavior someday?
inode_ctime_set_sec_nsec(struct inode *, time64_t, time64_t)
(or simply initialize a timespec64 and use inode_ctime_spec() )
Tom.
> + * @inode: inode in which to set the ctime
> + * @sec: value to set the tv_sec field
> + *
> + * Set the sec field in the ctime. Returns @sec.
> + */
> +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
> +{
> + inode->i_ctime.tv_sec = sec;
> + return sec;
> +}
> +
> +/**
> + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
> + * @inode: inode in which to set the ctime
> + * @nsec: value to set the tv_nsec field
> + *
> + * Set the nsec field in the ctime. Returns @nsec.
> + */
> +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
> +{
> + inode->i_ctime.tv_nsec = nsec;
> + return nsec;
> +}
>
> /*
> * Snapshotting support.
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 46/79] nfsd: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 46/79] nfsd: " Jeff Layton
@ 2023-06-21 17:36 ` Chuck Lever
0 siblings, 0 replies; 229+ messages in thread
From: Chuck Lever @ 2023-06-21 17:36 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Chuck Lever, Neil Brown, Olga Kornievskaia,
Dai Ngo, Tom Talpey, Al Viro, Jan Kara, linux-nfs, linux-kernel
On Wed, Jun 21, 2023 at 10:45:59AM -0400, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Acked-by: Chuck Lever <chuck.lever@oracle.com>
> ---
> fs/nfsd/nfsctl.c | 2 +-
> fs/nfsd/vfs.c | 2 +-
> 2 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
> index 1b8b1aab9a15..bebb65d3ad6b 100644
> --- a/fs/nfsd/nfsctl.c
> +++ b/fs/nfsd/nfsctl.c
> @@ -1131,7 +1131,7 @@ static struct inode *nfsd_get_inode(struct super_block *sb, umode_t mode)
> /* Following advice from simple_fill_super documentation: */
> inode->i_ino = iunique(sb, NFSD_MaxReserved);
> inode->i_mode = mode;
> - inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> switch (mode & S_IFMT) {
> case S_IFDIR:
> inode->i_fop = &simple_dir_operations;
> diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
> index 59b7d60ae33e..febdfdf16ec1 100644
> --- a/fs/nfsd/vfs.c
> +++ b/fs/nfsd/vfs.c
> @@ -520,7 +520,7 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp,
>
> nfsd_sanitize_attrs(inode, iap);
>
> - if (check_guard && guardtime != inode->i_ctime.tv_sec)
> + if (check_guard && guardtime != inode_ctime_peek(inode).tv_sec)
> return nfserr_notsync;
>
> /*
> --
> 2.41.0
>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 62/79] smb: switch to new ctime accessors
2023-06-21 14:46 ` [PATCH 62/79] smb: " Jeff Layton
@ 2023-06-21 17:45 ` Tom Talpey
2023-06-23 5:03 ` Sergey Senozhatsky
1 sibling, 0 replies; 229+ messages in thread
From: Tom Talpey @ 2023-06-21 17:45 UTC (permalink / raw)
To: Jeff Layton, Christian Brauner, Steve French, Paulo Alcantara,
Ronnie Sahlberg, Shyam Prasad N, Namjae Jeon, Sergey Senozhatsky
Cc: Al Viro, Jan Kara, linux-cifs, samba-technical, linux-kernel
On 6/21/2023 10:46 AM, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Acked-by: Tom Talpey <tom@talpey.com>
> ---
> fs/smb/client/file.c | 4 ++--
> fs/smb/client/fscache.h | 5 +++--
> fs/smb/client/inode.c | 15 +++++++--------
> fs/smb/client/smb2ops.c | 2 +-
> fs/smb/server/smb2pdu.c | 8 ++++----
> 5 files changed, 17 insertions(+), 17 deletions(-)
>
> diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c
> index 0a5fe8d5314b..07fed94b7501 100644
> --- a/fs/smb/client/file.c
> +++ b/fs/smb/client/file.c
> @@ -1085,7 +1085,7 @@ int cifs_close(struct inode *inode, struct file *file)
> !test_bit(CIFS_INO_CLOSE_ON_LOCK, &cinode->flags) &&
> dclose) {
> if (test_and_clear_bit(CIFS_INO_MODIFIED_ATTR, &cinode->flags)) {
> - inode->i_ctime = inode->i_mtime = current_time(inode);
> + inode->i_mtime = inode_ctime_set_current(inode);
> }
> spin_lock(&cinode->deferred_lock);
> cifs_add_deferred_close(cfile, dclose);
> @@ -2596,7 +2596,7 @@ static int cifs_partialpagewrite(struct page *page, unsigned from, unsigned to)
> write_data, to - from, &offset);
> cifsFileInfo_put(open_file);
> /* Does mm or vfs already set times? */
> - inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> if ((bytes_written > 0) && (offset))
> rc = 0;
> else if (bytes_written < 0)
> diff --git a/fs/smb/client/fscache.h b/fs/smb/client/fscache.h
> index 173999610997..1c50d286c3a4 100644
> --- a/fs/smb/client/fscache.h
> +++ b/fs/smb/client/fscache.h
> @@ -50,12 +50,13 @@ void cifs_fscache_fill_coherency(struct inode *inode,
> struct cifs_fscache_inode_coherency_data *cd)
> {
> struct cifsInodeInfo *cifsi = CIFS_I(inode);
> + struct timespec64 ctime = inode_ctime_peek(&cifsi->netfs.inode);
>
> memset(cd, 0, sizeof(*cd));
> cd->last_write_time_sec = cpu_to_le64(cifsi->netfs.inode.i_mtime.tv_sec);
> cd->last_write_time_nsec = cpu_to_le32(cifsi->netfs.inode.i_mtime.tv_nsec);
> - cd->last_change_time_sec = cpu_to_le64(cifsi->netfs.inode.i_ctime.tv_sec);
> - cd->last_change_time_nsec = cpu_to_le32(cifsi->netfs.inode.i_ctime.tv_nsec);
> + cd->last_change_time_sec = cpu_to_le64(ctime.tv_sec);
> + cd->last_change_time_nsec = cpu_to_le32(ctime.tv_nsec);
> }
>
>
> diff --git a/fs/smb/client/inode.c b/fs/smb/client/inode.c
> index 1087ac6104a9..d986b280b6c3 100644
> --- a/fs/smb/client/inode.c
> +++ b/fs/smb/client/inode.c
> @@ -172,7 +172,7 @@ cifs_fattr_to_inode(struct inode *inode, struct cifs_fattr *fattr)
> else
> inode->i_atime = fattr->cf_atime;
> inode->i_mtime = fattr->cf_mtime;
> - inode->i_ctime = fattr->cf_ctime;
> + inode_ctime_set(inode, fattr->cf_ctime);
> inode->i_rdev = fattr->cf_rdev;
> cifs_nlink_fattr_to_inode(inode, fattr);
> inode->i_uid = fattr->cf_uid;
> @@ -1744,9 +1744,9 @@ int cifs_unlink(struct inode *dir, struct dentry *dentry)
> cifs_inode = CIFS_I(inode);
> cifs_inode->time = 0; /* will force revalidate to get info
> when needed */
> - inode->i_ctime = current_time(inode);
> + inode_ctime_set_current(inode);
> }
> - dir->i_ctime = dir->i_mtime = current_time(dir);
> + dir->i_mtime = inode_ctime_set_current(dir);
> cifs_inode = CIFS_I(dir);
> CIFS_I(dir)->time = 0; /* force revalidate of dir as well */
> unlink_out:
> @@ -2060,8 +2060,8 @@ int cifs_rmdir(struct inode *inode, struct dentry *direntry)
> */
> cifsInode->time = 0;
>
> - d_inode(direntry)->i_ctime = inode->i_ctime = inode->i_mtime =
> - current_time(inode);
> + inode->i_mtime = inode_ctime_set_current(inode);
> + inode_ctime_set(d_inode(direntry), inode->i_mtime);
>
> rmdir_exit:
> free_dentry_path(page);
> @@ -2267,9 +2267,8 @@ cifs_rename2(struct mnt_idmap *idmap, struct inode *source_dir,
> /* force revalidate to go get info when needed */
> CIFS_I(source_dir)->time = CIFS_I(target_dir)->time = 0;
>
> - source_dir->i_ctime = source_dir->i_mtime = target_dir->i_ctime =
> - target_dir->i_mtime = current_time(source_dir);
> -
> + source_dir->i_mtime = inode_ctime_set_current(source_dir);
> + target_dir->i_mtime = inode_ctime_set_current(target_dir);
> cifs_rename_exit:
> kfree(info_buf_source);
> free_dentry_path(page2);
> diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c
> index 5639d8c48570..52564a5dd2b1 100644
> --- a/fs/smb/client/smb2ops.c
> +++ b/fs/smb/client/smb2ops.c
> @@ -1391,7 +1391,7 @@ smb2_close_getattr(const unsigned int xid, struct cifs_tcon *tcon,
> if (file_inf.LastWriteTime)
> inode->i_mtime = cifs_NTtimeToUnix(file_inf.LastWriteTime);
> if (file_inf.ChangeTime)
> - inode->i_ctime = cifs_NTtimeToUnix(file_inf.ChangeTime);
> + inode_ctime_set(inode, cifs_NTtimeToUnix(file_inf.ChangeTime));
> if (file_inf.LastAccessTime)
> inode->i_atime = cifs_NTtimeToUnix(file_inf.LastAccessTime);
>
> diff --git a/fs/smb/server/smb2pdu.c b/fs/smb/server/smb2pdu.c
> index d31926194ebf..7a741ab6835c 100644
> --- a/fs/smb/server/smb2pdu.c
> +++ b/fs/smb/server/smb2pdu.c
> @@ -4768,7 +4768,7 @@ static int find_file_posix_info(struct smb2_query_info_rsp *rsp,
> file_info->LastAccessTime = cpu_to_le64(time);
> time = ksmbd_UnixTimeToNT(inode->i_mtime);
> file_info->LastWriteTime = cpu_to_le64(time);
> - time = ksmbd_UnixTimeToNT(inode->i_ctime);
> + time = ksmbd_UnixTimeToNT(inode_ctime_peek(inode));
> file_info->ChangeTime = cpu_to_le64(time);
> file_info->DosAttributes = fp->f_ci->m_fattr;
> file_info->Inode = cpu_to_le64(inode->i_ino);
> @@ -5409,7 +5409,7 @@ int smb2_close(struct ksmbd_work *work)
> rsp->LastAccessTime = cpu_to_le64(time);
> time = ksmbd_UnixTimeToNT(inode->i_mtime);
> rsp->LastWriteTime = cpu_to_le64(time);
> - time = ksmbd_UnixTimeToNT(inode->i_ctime);
> + time = ksmbd_UnixTimeToNT(inode_ctime_peek(inode));
> rsp->ChangeTime = cpu_to_le64(time);
> ksmbd_fd_put(work, fp);
> } else {
> @@ -5628,7 +5628,7 @@ static int set_file_basic_info(struct ksmbd_file *fp,
> if (file_info->ChangeTime)
> attrs.ia_ctime = ksmbd_NTtimeToUnix(file_info->ChangeTime);
> else
> - attrs.ia_ctime = inode->i_ctime;
> + attrs.ia_ctime = inode_ctime_peek(inode);
>
> if (file_info->LastWriteTime) {
> attrs.ia_mtime = ksmbd_NTtimeToUnix(file_info->LastWriteTime);
> @@ -5673,7 +5673,7 @@ static int set_file_basic_info(struct ksmbd_file *fp,
> return -EACCES;
>
> inode_lock(inode);
> - inode->i_ctime = attrs.ia_ctime;
> + inode_ctime_set(inode, attrs.ia_ctime);
> attrs.ia_valid &= ~ATTR_CTIME;
> rc = notify_change(idmap, dentry, &attrs, NULL);
> inode_unlock(inode);
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 42/79] jfs: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 42/79] jfs: " Jeff Layton
@ 2023-06-21 17:47 ` Dave Kleikamp
0 siblings, 0 replies; 229+ messages in thread
From: Dave Kleikamp @ 2023-06-21 17:47 UTC (permalink / raw)
To: Jeff Layton, Christian Brauner
Cc: Al Viro, Jan Kara, jfs-discussion, linux-kernel
On 6/21/23 9:45AM, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
Looks good to me.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Acked-by: Dave Kleikamp <dave.kleikamp@oracle.com>
> ---
> fs/jfs/acl.c | 2 +-
> fs/jfs/inode.c | 2 +-
> fs/jfs/ioctl.c | 2 +-
> fs/jfs/jfs_imap.c | 8 ++++----
> fs/jfs/jfs_inode.c | 4 ++--
> fs/jfs/namei.c | 25 +++++++++++++------------
> fs/jfs/super.c | 2 +-
> fs/jfs/xattr.c | 2 +-
> 8 files changed, 24 insertions(+), 23 deletions(-)
>
> diff --git a/fs/jfs/acl.c b/fs/jfs/acl.c
> index fb96f872d207..3cd6eb00d881 100644
> --- a/fs/jfs/acl.c
> +++ b/fs/jfs/acl.c
> @@ -116,7 +116,7 @@ int jfs_set_acl(struct mnt_idmap *idmap, struct dentry *dentry,
> if (!rc) {
> if (update_mode) {
> inode->i_mode = mode;
> - inode->i_ctime = current_time(inode);
> + inode_ctime_set_current(inode);
> mark_inode_dirty(inode);
> }
> rc = txCommit(tid, 1, &inode, 0);
> diff --git a/fs/jfs/inode.c b/fs/jfs/inode.c
> index 8ac10e396050..d9f708fc6439 100644
> --- a/fs/jfs/inode.c
> +++ b/fs/jfs/inode.c
> @@ -393,7 +393,7 @@ void jfs_truncate_nolock(struct inode *ip, loff_t length)
> break;
> }
>
> - ip->i_mtime = ip->i_ctime = current_time(ip);
> + ip->i_mtime = inode_ctime_set_current(ip);
> mark_inode_dirty(ip);
>
> txCommit(tid, 1, &ip, 0);
> diff --git a/fs/jfs/ioctl.c b/fs/jfs/ioctl.c
> index ed7989bc2db1..db22929938a2 100644
> --- a/fs/jfs/ioctl.c
> +++ b/fs/jfs/ioctl.c
> @@ -96,7 +96,7 @@ int jfs_fileattr_set(struct mnt_idmap *idmap,
> jfs_inode->mode2 = flags;
>
> jfs_set_inode_flags(inode);
> - inode->i_ctime = current_time(inode);
> + inode_ctime_set_current(inode);
> mark_inode_dirty(inode);
>
> return 0;
> diff --git a/fs/jfs/jfs_imap.c b/fs/jfs/jfs_imap.c
> index 390cbfce391f..f32fb967d360 100644
> --- a/fs/jfs/jfs_imap.c
> +++ b/fs/jfs/jfs_imap.c
> @@ -3064,8 +3064,8 @@ static int copy_from_dinode(struct dinode * dip, struct inode *ip)
> ip->i_atime.tv_nsec = le32_to_cpu(dip->di_atime.tv_nsec);
> ip->i_mtime.tv_sec = le32_to_cpu(dip->di_mtime.tv_sec);
> ip->i_mtime.tv_nsec = le32_to_cpu(dip->di_mtime.tv_nsec);
> - ip->i_ctime.tv_sec = le32_to_cpu(dip->di_ctime.tv_sec);
> - ip->i_ctime.tv_nsec = le32_to_cpu(dip->di_ctime.tv_nsec);
> + inode_ctime_set_sec(ip, le32_to_cpu(dip->di_ctime.tv_sec));
> + inode_ctime_set_nsec(ip, le32_to_cpu(dip->di_ctime.tv_nsec));
> ip->i_blocks = LBLK2PBLK(ip->i_sb, le64_to_cpu(dip->di_nblocks));
> ip->i_generation = le32_to_cpu(dip->di_gen);
>
> @@ -3139,8 +3139,8 @@ static void copy_to_dinode(struct dinode * dip, struct inode *ip)
>
> dip->di_atime.tv_sec = cpu_to_le32(ip->i_atime.tv_sec);
> dip->di_atime.tv_nsec = cpu_to_le32(ip->i_atime.tv_nsec);
> - dip->di_ctime.tv_sec = cpu_to_le32(ip->i_ctime.tv_sec);
> - dip->di_ctime.tv_nsec = cpu_to_le32(ip->i_ctime.tv_nsec);
> + dip->di_ctime.tv_sec = cpu_to_le32(inode_ctime_peek(ip).tv_sec);
> + dip->di_ctime.tv_nsec = cpu_to_le32(inode_ctime_peek(ip).tv_nsec);
> dip->di_mtime.tv_sec = cpu_to_le32(ip->i_mtime.tv_sec);
> dip->di_mtime.tv_nsec = cpu_to_le32(ip->i_mtime.tv_nsec);
> dip->di_ixpxd = jfs_ip->ixpxd; /* in-memory pxd's are little-endian */
> diff --git a/fs/jfs/jfs_inode.c b/fs/jfs/jfs_inode.c
> index 9e1f02767201..5bbae1ff5129 100644
> --- a/fs/jfs/jfs_inode.c
> +++ b/fs/jfs/jfs_inode.c
> @@ -97,8 +97,8 @@ struct inode *ialloc(struct inode *parent, umode_t mode)
> jfs_inode->mode2 |= inode->i_mode;
>
> inode->i_blocks = 0;
> - inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
> - jfs_inode->otime = inode->i_ctime.tv_sec;
> + inode->i_mtime = inode->i_atime = inode_ctime_set_current(inode);
> + jfs_inode->otime = inode_ctime_peek(inode).tv_sec;
> inode->i_generation = JFS_SBI(sb)->gengen++;
>
> jfs_inode->cflag = 0;
> diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c
> index 494b9f4043cf..405b4790b916 100644
> --- a/fs/jfs/namei.c
> +++ b/fs/jfs/namei.c
> @@ -149,7 +149,7 @@ static int jfs_create(struct mnt_idmap *idmap, struct inode *dip,
>
> mark_inode_dirty(ip);
>
> - dip->i_ctime = dip->i_mtime = current_time(dip);
> + dip->i_mtime = inode_ctime_set_current(dip);
>
> mark_inode_dirty(dip);
>
> @@ -284,7 +284,7 @@ static int jfs_mkdir(struct mnt_idmap *idmap, struct inode *dip,
>
> /* update parent directory inode */
> inc_nlink(dip); /* for '..' from child directory */
> - dip->i_ctime = dip->i_mtime = current_time(dip);
> + dip->i_mtime = inode_ctime_set_current(dip);
> mark_inode_dirty(dip);
>
> rc = txCommit(tid, 2, &iplist[0], 0);
> @@ -390,7 +390,7 @@ static int jfs_rmdir(struct inode *dip, struct dentry *dentry)
> /* update parent directory's link count corresponding
> * to ".." entry of the target directory deleted
> */
> - dip->i_ctime = dip->i_mtime = current_time(dip);
> + dip->i_mtime = inode_ctime_set_current(dip);
> inode_dec_link_count(dip);
>
> /*
> @@ -512,7 +512,8 @@ static int jfs_unlink(struct inode *dip, struct dentry *dentry)
>
> ASSERT(ip->i_nlink);
>
> - ip->i_ctime = dip->i_ctime = dip->i_mtime = current_time(ip);
> + dip->i_mtime = inode_ctime_set_current(ip);
> + inode_ctime_set(dip, dip->i_mtime);
> mark_inode_dirty(dip);
>
> /* update target's inode */
> @@ -822,8 +823,8 @@ static int jfs_link(struct dentry *old_dentry,
>
> /* update object inode */
> inc_nlink(ip); /* for new link */
> - ip->i_ctime = current_time(ip);
> - dir->i_ctime = dir->i_mtime = current_time(dir);
> + inode_ctime_set_current(ip);
> + dir->i_mtime = inode_ctime_set_current(dir);
> mark_inode_dirty(dir);
> ihold(ip);
>
> @@ -1023,7 +1024,7 @@ static int jfs_symlink(struct mnt_idmap *idmap, struct inode *dip,
>
> mark_inode_dirty(ip);
>
> - dip->i_ctime = dip->i_mtime = current_time(dip);
> + dip->i_mtime = inode_ctime_set_current(dip);
> mark_inode_dirty(dip);
> /*
> * commit update of parent directory and link object
> @@ -1200,7 +1201,7 @@ static int jfs_rename(struct mnt_idmap *idmap, struct inode *old_dir,
> tblk->xflag |= COMMIT_DELETE;
> tblk->u.ip = new_ip;
> } else {
> - new_ip->i_ctime = current_time(new_ip);
> + inode_ctime_set_current(new_ip);
> mark_inode_dirty(new_ip);
> }
> } else {
> @@ -1263,10 +1264,10 @@ static int jfs_rename(struct mnt_idmap *idmap, struct inode *old_dir,
> /*
> * Update ctime on changed/moved inodes & mark dirty
> */
> - old_ip->i_ctime = current_time(old_ip);
> + inode_ctime_set_current(old_ip);
> mark_inode_dirty(old_ip);
>
> - new_dir->i_ctime = new_dir->i_mtime = current_time(new_dir);
> + new_dir->i_mtime = inode_ctime_set_current(new_dir);
> mark_inode_dirty(new_dir);
>
> /* Build list of inodes modified by this transaction */
> @@ -1278,7 +1279,7 @@ static int jfs_rename(struct mnt_idmap *idmap, struct inode *old_dir,
>
> if (old_dir != new_dir) {
> iplist[ipcount++] = new_dir;
> - old_dir->i_ctime = old_dir->i_mtime = current_time(old_dir);
> + old_dir->i_mtime = inode_ctime_set_current(old_dir);
> mark_inode_dirty(old_dir);
> }
>
> @@ -1411,7 +1412,7 @@ static int jfs_mknod(struct mnt_idmap *idmap, struct inode *dir,
>
> mark_inode_dirty(ip);
>
> - dir->i_ctime = dir->i_mtime = current_time(dir);
> + dir->i_mtime = inode_ctime_set_current(dir);
>
> mark_inode_dirty(dir);
>
> diff --git a/fs/jfs/super.c b/fs/jfs/super.c
> index d2f82cb7db1b..c810b98254dd 100644
> --- a/fs/jfs/super.c
> +++ b/fs/jfs/super.c
> @@ -818,7 +818,7 @@ static ssize_t jfs_quota_write(struct super_block *sb, int type,
> }
> if (inode->i_size < off+len-towrite)
> i_size_write(inode, off+len-towrite);
> - inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_mtime = inode_ctime_set_current(inode);
> mark_inode_dirty(inode);
> inode_unlock(inode);
> return len - towrite;
> diff --git a/fs/jfs/xattr.c b/fs/jfs/xattr.c
> index 931e50018f88..71f005bbe486 100644
> --- a/fs/jfs/xattr.c
> +++ b/fs/jfs/xattr.c
> @@ -647,7 +647,7 @@ static int ea_put(tid_t tid, struct inode *inode, struct ea_buffer *ea_buf,
> if (old_blocks)
> dquot_free_block(inode, old_blocks);
>
> - inode->i_ctime = current_time(inode);
> + inode_ctime_set_current(inode);
>
> return 0;
> }
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 56/79] pstore: switch to new ctime accessors
2023-06-21 14:46 ` [PATCH 56/79] pstore: " Jeff Layton
@ 2023-06-21 17:50 ` Kees Cook
0 siblings, 0 replies; 229+ messages in thread
From: Kees Cook @ 2023-06-21 17:50 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Tony Luck, Guilherme G. Piccoli, Al Viro,
Jan Kara, linux-hardening, linux-kernel
On Wed, Jun 21, 2023 at 10:46:09AM -0400, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Acked-by: Kees Cook <keescook@chromium.org>
--
Kees Cook
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
2023-06-21 17:29 ` [Ocfs2-devel] " Tom Talpey via Ocfs2-devel
(?)
(?)
@ 2023-06-21 18:01 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 18:01 UTC (permalink / raw)
To: Tom Talpey, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On Wed, 2023-06-21 at 13:29 -0400, Tom Talpey wrote:
> On 6/21/2023 10:45 AM, Jeff Layton wrote:
> > struct timespec64 has unused bits in the tv_nsec field that can be used
> > for other purposes. In future patches, we're going to change how the
> > inode->i_ctime is accessed in certain inodes in order to make use of
> > them. In order to do that safely though, we'll need to eradicate raw
> > accesses of the inode->i_ctime field from the kernel.
> >
> > Add new accessor functions for the ctime that we can use to replace them.
> >
> > Signed-off-by: Jeff Layton <jlayton@kernel.org>
> > ---
> > fs/inode.c | 16 ++++++++++++++
> > include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
> > 2 files changed, 68 insertions(+), 1 deletion(-)
> >
> > diff --git a/fs/inode.c b/fs/inode.c
> > index d37fad91c8da..c005e7328fbb 100644
> > --- a/fs/inode.c
> > +++ b/fs/inode.c
> > @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
> > }
> > EXPORT_SYMBOL(current_time);
> >
> > +/**
> > + * inode_ctime_set_current - set the ctime to current_time
> > + * @inode: inode
> > + *
> > + * Set the inode->i_ctime to the current value for the inode. Returns
> > + * the current value that was assigned to i_ctime.
> > + */
> > +struct timespec64 inode_ctime_set_current(struct inode *inode)
> > +{
> > + struct timespec64 now = current_time(inode);
> > +
> > + inode_set_ctime(inode, now);
> > + return now;
> > +}
> > +EXPORT_SYMBOL(inode_ctime_set_current);
> > +
> > /**
> > * in_group_or_capable - check whether caller is CAP_FSETID privileged
> > * @idmap: idmap of the mount @inode was found from
> > diff --git a/include/linux/fs.h b/include/linux/fs.h
> > index 6867512907d6..9afb30606373 100644
> > --- a/include/linux/fs.h
> > +++ b/include/linux/fs.h
> > @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
> > kgid_has_mapping(fs_userns, kgid);
> > }
> >
> > -extern struct timespec64 current_time(struct inode *inode);
> > +struct timespec64 current_time(struct inode *inode);
> > +struct timespec64 inode_ctime_set_current(struct inode *inode);
> > +
> > +/**
> > + * inode_ctime_peek - fetch the current ctime from the inode
> > + * @inode: inode from which to fetch ctime
> > + *
> > + * Grab the current ctime from the inode and return it.
> > + */
> > +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
> > +{
> > + return inode->i_ctime;
> > +}
> > +
> > +/**
> > + * inode_ctime_set - set the ctime in the inode to the given value
> > + * @inode: inode in which to set the ctime
> > + * @ts: timespec value to set the ctime
> > + *
> > + * Set the ctime in @inode to @ts.
> > + */
> > +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
> > +{
> > + inode->i_ctime = ts;
> > + return ts;
> > +}
> > +
> > +/**
> > + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
>
> I'm curious about why you choose to split the tv_sec and tv_nsec
> set_ functions. Do any callers not set them both? Wouldn't a
> single call enable a more atomic behavior someday?
>
> inode_ctime_set_sec_nsec(struct inode *, time64_t, time64_t)
>
> (or simply initialize a timespec64 and use inode_ctime_spec() )
>
Yes, quite a few places set the fields individually. For example, when
loading a value from disk that doesn't have sufficient granularity to
set the nsecs field to anything but 0.
Could I have done it by declaring a local timespec64 variable and just
use the inode_ctime_set function in these places? Absolutely.
That's a bit more difficult to handle with coccinelle though. If someone
wants to suggest a way to do that without having to change all of these
call sites manually, then I'm open to redoing the set.
That might be better left for a later cleanup though.
> > + * @inode: inode in which to set the ctime
> > + * @sec: value to set the tv_sec field
> > + *
> > + * Set the sec field in the ctime. Returns @sec.
> > + */
> > +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
> > +{
> > + inode->i_ctime.tv_sec = sec;
> > + return sec;
> > +}
> > +
> > +/**
> > + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
> > + * @inode: inode in which to set the ctime
> > + * @nsec: value to set the tv_nsec field
> > + *
> > + * Set the nsec field in the ctime. Returns @nsec.
> > + */
> > +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
> > +{
> > + inode->i_ctime.tv_nsec = nsec;
> > + return nsec;
> > +}
> >
> > /*
> > * Snapshotting support.
--
Jeff Layton <jlayton@kernel.org>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [f2fs-dev] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-21 18:01 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 18:01 UTC (permalink / raw)
To: Tom Talpey, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On Wed, 2023-06-21 at 13:29 -0400, Tom Talpey wrote:
> On 6/21/2023 10:45 AM, Jeff Layton wrote:
> > struct timespec64 has unused bits in the tv_nsec field that can be used
> > for other purposes. In future patches, we're going to change how the
> > inode->i_ctime is accessed in certain inodes in order to make use of
> > them. In order to do that safely though, we'll need to eradicate raw
> > accesses of the inode->i_ctime field from the kernel.
> >
> > Add new accessor functions for the ctime that we can use to replace them.
> >
> > Signed-off-by: Jeff Layton <jlayton@kernel.org>
> > ---
> > fs/inode.c | 16 ++++++++++++++
> > include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
> > 2 files changed, 68 insertions(+), 1 deletion(-)
> >
> > diff --git a/fs/inode.c b/fs/inode.c
> > index d37fad91c8da..c005e7328fbb 100644
> > --- a/fs/inode.c
> > +++ b/fs/inode.c
> > @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
> > }
> > EXPORT_SYMBOL(current_time);
> >
> > +/**
> > + * inode_ctime_set_current - set the ctime to current_time
> > + * @inode: inode
> > + *
> > + * Set the inode->i_ctime to the current value for the inode. Returns
> > + * the current value that was assigned to i_ctime.
> > + */
> > +struct timespec64 inode_ctime_set_current(struct inode *inode)
> > +{
> > + struct timespec64 now = current_time(inode);
> > +
> > + inode_set_ctime(inode, now);
> > + return now;
> > +}
> > +EXPORT_SYMBOL(inode_ctime_set_current);
> > +
> > /**
> > * in_group_or_capable - check whether caller is CAP_FSETID privileged
> > * @idmap: idmap of the mount @inode was found from
> > diff --git a/include/linux/fs.h b/include/linux/fs.h
> > index 6867512907d6..9afb30606373 100644
> > --- a/include/linux/fs.h
> > +++ b/include/linux/fs.h
> > @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
> > kgid_has_mapping(fs_userns, kgid);
> > }
> >
> > -extern struct timespec64 current_time(struct inode *inode);
> > +struct timespec64 current_time(struct inode *inode);
> > +struct timespec64 inode_ctime_set_current(struct inode *inode);
> > +
> > +/**
> > + * inode_ctime_peek - fetch the current ctime from the inode
> > + * @inode: inode from which to fetch ctime
> > + *
> > + * Grab the current ctime from the inode and return it.
> > + */
> > +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
> > +{
> > + return inode->i_ctime;
> > +}
> > +
> > +/**
> > + * inode_ctime_set - set the ctime in the inode to the given value
> > + * @inode: inode in which to set the ctime
> > + * @ts: timespec value to set the ctime
> > + *
> > + * Set the ctime in @inode to @ts.
> > + */
> > +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
> > +{
> > + inode->i_ctime = ts;
> > + return ts;
> > +}
> > +
> > +/**
> > + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
>
> I'm curious about why you choose to split the tv_sec and tv_nsec
> set_ functions. Do any callers not set them both? Wouldn't a
> single call enable a more atomic behavior someday?
>
> inode_ctime_set_sec_nsec(struct inode *, time64_t, time64_t)
>
> (or simply initialize a timespec64 and use inode_ctime_spec() )
>
Yes, quite a few places set the fields individually. For example, when
loading a value from disk that doesn't have sufficient granularity to
set the nsecs field to anything but 0.
Could I have done it by declaring a local timespec64 variable and just
use the inode_ctime_set function in these places? Absolutely.
That's a bit more difficult to handle with coccinelle though. If someone
wants to suggest a way to do that without having to change all of these
call sites manually, then I'm open to redoing the set.
That might be better left for a later cleanup though.
> > + * @inode: inode in which to set the ctime
> > + * @sec: value to set the tv_sec field
> > + *
> > + * Set the sec field in the ctime. Returns @sec.
> > + */
> > +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
> > +{
> > + inode->i_ctime.tv_sec = sec;
> > + return sec;
> > +}
> > +
> > +/**
> > + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
> > + * @inode: inode in which to set the ctime
> > + * @nsec: value to set the tv_nsec field
> > + *
> > + * Set the nsec field in the ctime. Returns @nsec.
> > + */
> > +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
> > +{
> > + inode->i_ctime.tv_nsec = nsec;
> > + return nsec;
> > +}
> >
> > /*
> > * Snapshotting support.
--
Jeff Layton <jlayton@kernel.org>
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [Ocfs2-devel] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-21 18:01 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton via Ocfs2-devel @ 2023-06-21 18:01 UTC (permalink / raw)
To: Tom Talpey, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On Wed, 2023-06-21 at 13:29 -0400, Tom Talpey wrote:
> On 6/21/2023 10:45 AM, Jeff Layton wrote:
> > struct timespec64 has unused bits in the tv_nsec field that can be used
> > for other purposes. In future patches, we're going to change how the
> > inode->i_ctime is accessed in certain inodes in order to make use of
> > them. In order to do that safely though, we'll need to eradicate raw
> > accesses of the inode->i_ctime field from the kernel.
> >
> > Add new accessor functions for the ctime that we can use to replace them.
> >
> > Signed-off-by: Jeff Layton <jlayton@kernel.org>
> > ---
> > fs/inode.c | 16 ++++++++++++++
> > include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
> > 2 files changed, 68 insertions(+), 1 deletion(-)
> >
> > diff --git a/fs/inode.c b/fs/inode.c
> > index d37fad91c8da..c005e7328fbb 100644
> > --- a/fs/inode.c
> > +++ b/fs/inode.c
> > @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
> > }
> > EXPORT_SYMBOL(current_time);
> >
> > +/**
> > + * inode_ctime_set_current - set the ctime to current_time
> > + * @inode: inode
> > + *
> > + * Set the inode->i_ctime to the current value for the inode. Returns
> > + * the current value that was assigned to i_ctime.
> > + */
> > +struct timespec64 inode_ctime_set_current(struct inode *inode)
> > +{
> > + struct timespec64 now = current_time(inode);
> > +
> > + inode_set_ctime(inode, now);
> > + return now;
> > +}
> > +EXPORT_SYMBOL(inode_ctime_set_current);
> > +
> > /**
> > * in_group_or_capable - check whether caller is CAP_FSETID privileged
> > * @idmap: idmap of the mount @inode was found from
> > diff --git a/include/linux/fs.h b/include/linux/fs.h
> > index 6867512907d6..9afb30606373 100644
> > --- a/include/linux/fs.h
> > +++ b/include/linux/fs.h
> > @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
> > kgid_has_mapping(fs_userns, kgid);
> > }
> >
> > -extern struct timespec64 current_time(struct inode *inode);
> > +struct timespec64 current_time(struct inode *inode);
> > +struct timespec64 inode_ctime_set_current(struct inode *inode);
> > +
> > +/**
> > + * inode_ctime_peek - fetch the current ctime from the inode
> > + * @inode: inode from which to fetch ctime
> > + *
> > + * Grab the current ctime from the inode and return it.
> > + */
> > +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
> > +{
> > + return inode->i_ctime;
> > +}
> > +
> > +/**
> > + * inode_ctime_set - set the ctime in the inode to the given value
> > + * @inode: inode in which to set the ctime
> > + * @ts: timespec value to set the ctime
> > + *
> > + * Set the ctime in @inode to @ts.
> > + */
> > +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
> > +{
> > + inode->i_ctime = ts;
> > + return ts;
> > +}
> > +
> > +/**
> > + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
>
> I'm curious about why you choose to split the tv_sec and tv_nsec
> set_ functions. Do any callers not set them both? Wouldn't a
> single call enable a more atomic behavior someday?
>
> inode_ctime_set_sec_nsec(struct inode *, time64_t, time64_t)
>
> (or simply initialize a timespec64 and use inode_ctime_spec() )
>
Yes, quite a few places set the fields individually. For example, when
loading a value from disk that doesn't have sufficient granularity to
set the nsecs field to anything but 0.
Could I have done it by declaring a local timespec64 variable and just
use the inode_ctime_set function in these places? Absolutely.
That's a bit more difficult to handle with coccinelle though. If someone
wants to suggest a way to do that without having to change all of these
call sites manually, then I'm open to redoing the set.
That might be better left for a later cleanup though.
> > + * @inode: inode in which to set the ctime
> > + * @sec: value to set the tv_sec field
> > + *
> > + * Set the sec field in the ctime. Returns @sec.
> > + */
> > +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
> > +{
> > + inode->i_ctime.tv_sec = sec;
> > + return sec;
> > +}
> > +
> > +/**
> > + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
> > + * @inode: inode in which to set the ctime
> > + * @nsec: value to set the tv_nsec field
> > + *
> > + * Set the nsec field in the ctime. Returns @nsec.
> > + */
> > +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
> > +{
> > + inode->i_ctime.tv_nsec = nsec;
> > + return nsec;
> > +}
> >
> > /*
> > * Snapshotting support.
--
Jeff Layton <jlayton@kernel.org>
_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* [Cluster-devel] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-21 18:01 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 18:01 UTC (permalink / raw)
To: cluster-devel.redhat.com
On Wed, 2023-06-21 at 13:29 -0400, Tom Talpey wrote:
> On 6/21/2023 10:45 AM, Jeff Layton wrote:
> > struct timespec64 has unused bits in the tv_nsec field that can be used
> > for other purposes. In future patches, we're going to change how the
> > inode->i_ctime is accessed in certain inodes in order to make use of
> > them. In order to do that safely though, we'll need to eradicate raw
> > accesses of the inode->i_ctime field from the kernel.
> >
> > Add new accessor functions for the ctime that we can use to replace them.
> >
> > Signed-off-by: Jeff Layton <jlayton@kernel.org>
> > ---
> > fs/inode.c | 16 ++++++++++++++
> > include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
> > 2 files changed, 68 insertions(+), 1 deletion(-)
> >
> > diff --git a/fs/inode.c b/fs/inode.c
> > index d37fad91c8da..c005e7328fbb 100644
> > --- a/fs/inode.c
> > +++ b/fs/inode.c
> > @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
> > }
> > EXPORT_SYMBOL(current_time);
> >
> > +/**
> > + * inode_ctime_set_current - set the ctime to current_time
> > + * @inode: inode
> > + *
> > + * Set the inode->i_ctime to the current value for the inode. Returns
> > + * the current value that was assigned to i_ctime.
> > + */
> > +struct timespec64 inode_ctime_set_current(struct inode *inode)
> > +{
> > + struct timespec64 now = current_time(inode);
> > +
> > + inode_set_ctime(inode, now);
> > + return now;
> > +}
> > +EXPORT_SYMBOL(inode_ctime_set_current);
> > +
> > /**
> > * in_group_or_capable - check whether caller is CAP_FSETID privileged
> > * @idmap: idmap of the mount @inode was found from
> > diff --git a/include/linux/fs.h b/include/linux/fs.h
> > index 6867512907d6..9afb30606373 100644
> > --- a/include/linux/fs.h
> > +++ b/include/linux/fs.h
> > @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
> > kgid_has_mapping(fs_userns, kgid);
> > }
> >
> > -extern struct timespec64 current_time(struct inode *inode);
> > +struct timespec64 current_time(struct inode *inode);
> > +struct timespec64 inode_ctime_set_current(struct inode *inode);
> > +
> > +/**
> > + * inode_ctime_peek - fetch the current ctime from the inode
> > + * @inode: inode from which to fetch ctime
> > + *
> > + * Grab the current ctime from the inode and return it.
> > + */
> > +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
> > +{
> > + return inode->i_ctime;
> > +}
> > +
> > +/**
> > + * inode_ctime_set - set the ctime in the inode to the given value
> > + * @inode: inode in which to set the ctime
> > + * @ts: timespec value to set the ctime
> > + *
> > + * Set the ctime in @inode to @ts.
> > + */
> > +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
> > +{
> > + inode->i_ctime = ts;
> > + return ts;
> > +}
> > +
> > +/**
> > + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
>
> I'm curious about why you choose to split the tv_sec and tv_nsec
> set_ functions. Do any callers not set them both? Wouldn't a
> single call enable a more atomic behavior someday?
>
> inode_ctime_set_sec_nsec(struct inode *, time64_t, time64_t)
>
> (or simply initialize a timespec64 and use inode_ctime_spec() )
>
Yes, quite a few places set the fields individually. For example, when
loading a value from disk that doesn't have sufficient granularity to
set the nsecs field to anything but 0.
Could I have done it by declaring a local timespec64 variable and just
use the inode_ctime_set function in these places? Absolutely.
That's a bit more difficult to handle with coccinelle though. If someone
wants to suggest a way to do that without having to change all of these
call sites manually, then I'm open to redoing the set.
That might be better left for a later cleanup though.
> > + * @inode: inode in which to set the ctime
> > + * @sec: value to set the tv_sec field
> > + *
> > + * Set the sec field in the ctime. Returns @sec.
> > + */
> > +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
> > +{
> > + inode->i_ctime.tv_sec = sec;
> > + return sec;
> > +}
> > +
> > +/**
> > + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
> > + * @inode: inode in which to set the ctime
> > + * @nsec: value to set the tv_nsec field
> > + *
> > + * Set the nsec field in the ctime. Returns @nsec.
> > + */
> > +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
> > +{
> > + inode->i_ctime.tv_nsec = nsec;
> > + return nsec;
> > +}
> >
> > /*
> > * Snapshotting support.
--
Jeff Layton <jlayton@kernel.org>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
2023-06-21 18:01 ` [f2fs-dev] " Jeff Layton
(?)
(?)
@ 2023-06-21 18:19 ` Tom Talpey
-1 siblings, 0 replies; 229+ messages in thread
From: Tom Talpey @ 2023-06-21 18:19 UTC (permalink / raw)
To: Jeff Layton, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On 6/21/2023 2:01 PM, Jeff Layton wrote:
> On Wed, 2023-06-21 at 13:29 -0400, Tom Talpey wrote:
>> On 6/21/2023 10:45 AM, Jeff Layton wrote:
>>> struct timespec64 has unused bits in the tv_nsec field that can be used
>>> for other purposes. In future patches, we're going to change how the
>>> inode->i_ctime is accessed in certain inodes in order to make use of
>>> them. In order to do that safely though, we'll need to eradicate raw
>>> accesses of the inode->i_ctime field from the kernel.
>>>
>>> Add new accessor functions for the ctime that we can use to replace them.
>>>
>>> Signed-off-by: Jeff Layton <jlayton@kernel.org>
>>> ---
>>> fs/inode.c | 16 ++++++++++++++
>>> include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
>>> 2 files changed, 68 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/fs/inode.c b/fs/inode.c
>>> index d37fad91c8da..c005e7328fbb 100644
>>> --- a/fs/inode.c
>>> +++ b/fs/inode.c
>>> @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
>>> }
>>> EXPORT_SYMBOL(current_time);
>>>
>>> +/**
>>> + * inode_ctime_set_current - set the ctime to current_time
>>> + * @inode: inode
>>> + *
>>> + * Set the inode->i_ctime to the current value for the inode. Returns
>>> + * the current value that was assigned to i_ctime.
>>> + */
>>> +struct timespec64 inode_ctime_set_current(struct inode *inode)
>>> +{
>>> + struct timespec64 now = current_time(inode);
>>> +
>>> + inode_set_ctime(inode, now);
>>> + return now;
>>> +}
>>> +EXPORT_SYMBOL(inode_ctime_set_current);
>>> +
>>> /**
>>> * in_group_or_capable - check whether caller is CAP_FSETID privileged
>>> * @idmap: idmap of the mount @inode was found from
>>> diff --git a/include/linux/fs.h b/include/linux/fs.h
>>> index 6867512907d6..9afb30606373 100644
>>> --- a/include/linux/fs.h
>>> +++ b/include/linux/fs.h
>>> @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
>>> kgid_has_mapping(fs_userns, kgid);
>>> }
>>>
>>> -extern struct timespec64 current_time(struct inode *inode);
>>> +struct timespec64 current_time(struct inode *inode);
>>> +struct timespec64 inode_ctime_set_current(struct inode *inode);
>>> +
>>> +/**
>>> + * inode_ctime_peek - fetch the current ctime from the inode
>>> + * @inode: inode from which to fetch ctime
>>> + *
>>> + * Grab the current ctime from the inode and return it.
>>> + */
>>> +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
>>> +{
>>> + return inode->i_ctime;
>>> +}
>>> +
>>> +/**
>>> + * inode_ctime_set - set the ctime in the inode to the given value
>>> + * @inode: inode in which to set the ctime
>>> + * @ts: timespec value to set the ctime
>>> + *
>>> + * Set the ctime in @inode to @ts.
>>> + */
>>> +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
>>> +{
>>> + inode->i_ctime = ts;
>>> + return ts;
>>> +}
>>> +
>>> +/**
>>> + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
>>
>> I'm curious about why you choose to split the tv_sec and tv_nsec
>> set_ functions. Do any callers not set them both? Wouldn't a
>> single call enable a more atomic behavior someday?
>>
>> inode_ctime_set_sec_nsec(struct inode *, time64_t, time64_t)
>>
>> (or simply initialize a timespec64 and use inode_ctime_spec() )
>>
>
> Yes, quite a few places set the fields individually. For example, when
> loading a value from disk that doesn't have sufficient granularity to
> set the nsecs field to anything but 0.
Well, they still need to set the tv_nsec so they could just pass 0.
But ok.
> Could I have done it by declaring a local timespec64 variable and just
> use the inode_ctime_set function in these places? Absolutely.
>
> That's a bit more difficult to handle with coccinelle though. If someone
> wants to suggest a way to do that without having to change all of these
> call sites manually, then I'm open to redoing the set.
>
> That might be better left for a later cleanup though.
Acked-by: Tom Talpey <tom@talpey.com>
>>> + * @inode: inode in which to set the ctime
>>> + * @sec: value to set the tv_sec field
>>> + *
>>> + * Set the sec field in the ctime. Returns @sec.
>>> + */
>>> +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
>>> +{
>>> + inode->i_ctime.tv_sec = sec;
>>> + return sec;
>>> +}
>>> +
>>> +/**
>>> + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
>>> + * @inode: inode in which to set the ctime
>>> + * @nsec: value to set the tv_nsec field
>>> + *
>>> + * Set the nsec field in the ctime. Returns @nsec.
>>> + */
>>> +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
>>> +{
>>> + inode->i_ctime.tv_nsec = nsec;
>>> + return nsec;
>>> +}
>>>
>>> /*
>>> * Snapshotting support.
>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [f2fs-dev] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-21 18:19 ` Tom Talpey
0 siblings, 0 replies; 229+ messages in thread
From: Tom Talpey @ 2023-06-21 18:19 UTC (permalink / raw)
To: Jeff Layton, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On 6/21/2023 2:01 PM, Jeff Layton wrote:
> On Wed, 2023-06-21 at 13:29 -0400, Tom Talpey wrote:
>> On 6/21/2023 10:45 AM, Jeff Layton wrote:
>>> struct timespec64 has unused bits in the tv_nsec field that can be used
>>> for other purposes. In future patches, we're going to change how the
>>> inode->i_ctime is accessed in certain inodes in order to make use of
>>> them. In order to do that safely though, we'll need to eradicate raw
>>> accesses of the inode->i_ctime field from the kernel.
>>>
>>> Add new accessor functions for the ctime that we can use to replace them.
>>>
>>> Signed-off-by: Jeff Layton <jlayton@kernel.org>
>>> ---
>>> fs/inode.c | 16 ++++++++++++++
>>> include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
>>> 2 files changed, 68 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/fs/inode.c b/fs/inode.c
>>> index d37fad91c8da..c005e7328fbb 100644
>>> --- a/fs/inode.c
>>> +++ b/fs/inode.c
>>> @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
>>> }
>>> EXPORT_SYMBOL(current_time);
>>>
>>> +/**
>>> + * inode_ctime_set_current - set the ctime to current_time
>>> + * @inode: inode
>>> + *
>>> + * Set the inode->i_ctime to the current value for the inode. Returns
>>> + * the current value that was assigned to i_ctime.
>>> + */
>>> +struct timespec64 inode_ctime_set_current(struct inode *inode)
>>> +{
>>> + struct timespec64 now = current_time(inode);
>>> +
>>> + inode_set_ctime(inode, now);
>>> + return now;
>>> +}
>>> +EXPORT_SYMBOL(inode_ctime_set_current);
>>> +
>>> /**
>>> * in_group_or_capable - check whether caller is CAP_FSETID privileged
>>> * @idmap: idmap of the mount @inode was found from
>>> diff --git a/include/linux/fs.h b/include/linux/fs.h
>>> index 6867512907d6..9afb30606373 100644
>>> --- a/include/linux/fs.h
>>> +++ b/include/linux/fs.h
>>> @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
>>> kgid_has_mapping(fs_userns, kgid);
>>> }
>>>
>>> -extern struct timespec64 current_time(struct inode *inode);
>>> +struct timespec64 current_time(struct inode *inode);
>>> +struct timespec64 inode_ctime_set_current(struct inode *inode);
>>> +
>>> +/**
>>> + * inode_ctime_peek - fetch the current ctime from the inode
>>> + * @inode: inode from which to fetch ctime
>>> + *
>>> + * Grab the current ctime from the inode and return it.
>>> + */
>>> +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
>>> +{
>>> + return inode->i_ctime;
>>> +}
>>> +
>>> +/**
>>> + * inode_ctime_set - set the ctime in the inode to the given value
>>> + * @inode: inode in which to set the ctime
>>> + * @ts: timespec value to set the ctime
>>> + *
>>> + * Set the ctime in @inode to @ts.
>>> + */
>>> +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
>>> +{
>>> + inode->i_ctime = ts;
>>> + return ts;
>>> +}
>>> +
>>> +/**
>>> + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
>>
>> I'm curious about why you choose to split the tv_sec and tv_nsec
>> set_ functions. Do any callers not set them both? Wouldn't a
>> single call enable a more atomic behavior someday?
>>
>> inode_ctime_set_sec_nsec(struct inode *, time64_t, time64_t)
>>
>> (or simply initialize a timespec64 and use inode_ctime_spec() )
>>
>
> Yes, quite a few places set the fields individually. For example, when
> loading a value from disk that doesn't have sufficient granularity to
> set the nsecs field to anything but 0.
Well, they still need to set the tv_nsec so they could just pass 0.
But ok.
> Could I have done it by declaring a local timespec64 variable and just
> use the inode_ctime_set function in these places? Absolutely.
>
> That's a bit more difficult to handle with coccinelle though. If someone
> wants to suggest a way to do that without having to change all of these
> call sites manually, then I'm open to redoing the set.
>
> That might be better left for a later cleanup though.
Acked-by: Tom Talpey <tom@talpey.com>
>>> + * @inode: inode in which to set the ctime
>>> + * @sec: value to set the tv_sec field
>>> + *
>>> + * Set the sec field in the ctime. Returns @sec.
>>> + */
>>> +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
>>> +{
>>> + inode->i_ctime.tv_sec = sec;
>>> + return sec;
>>> +}
>>> +
>>> +/**
>>> + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
>>> + * @inode: inode in which to set the ctime
>>> + * @nsec: value to set the tv_nsec field
>>> + *
>>> + * Set the nsec field in the ctime. Returns @nsec.
>>> + */
>>> +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
>>> +{
>>> + inode->i_ctime.tv_nsec = nsec;
>>> + return nsec;
>>> +}
>>>
>>> /*
>>> * Snapshotting support.
>
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [Ocfs2-devel] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-21 18:19 ` Tom Talpey
0 siblings, 0 replies; 229+ messages in thread
From: Tom Talpey via Ocfs2-devel @ 2023-06-21 18:19 UTC (permalink / raw)
To: Jeff Layton, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On 6/21/2023 2:01 PM, Jeff Layton wrote:
> On Wed, 2023-06-21 at 13:29 -0400, Tom Talpey wrote:
>> On 6/21/2023 10:45 AM, Jeff Layton wrote:
>>> struct timespec64 has unused bits in the tv_nsec field that can be used
>>> for other purposes. In future patches, we're going to change how the
>>> inode->i_ctime is accessed in certain inodes in order to make use of
>>> them. In order to do that safely though, we'll need to eradicate raw
>>> accesses of the inode->i_ctime field from the kernel.
>>>
>>> Add new accessor functions for the ctime that we can use to replace them.
>>>
>>> Signed-off-by: Jeff Layton <jlayton@kernel.org>
>>> ---
>>> fs/inode.c | 16 ++++++++++++++
>>> include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
>>> 2 files changed, 68 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/fs/inode.c b/fs/inode.c
>>> index d37fad91c8da..c005e7328fbb 100644
>>> --- a/fs/inode.c
>>> +++ b/fs/inode.c
>>> @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
>>> }
>>> EXPORT_SYMBOL(current_time);
>>>
>>> +/**
>>> + * inode_ctime_set_current - set the ctime to current_time
>>> + * @inode: inode
>>> + *
>>> + * Set the inode->i_ctime to the current value for the inode. Returns
>>> + * the current value that was assigned to i_ctime.
>>> + */
>>> +struct timespec64 inode_ctime_set_current(struct inode *inode)
>>> +{
>>> + struct timespec64 now = current_time(inode);
>>> +
>>> + inode_set_ctime(inode, now);
>>> + return now;
>>> +}
>>> +EXPORT_SYMBOL(inode_ctime_set_current);
>>> +
>>> /**
>>> * in_group_or_capable - check whether caller is CAP_FSETID privileged
>>> * @idmap: idmap of the mount @inode was found from
>>> diff --git a/include/linux/fs.h b/include/linux/fs.h
>>> index 6867512907d6..9afb30606373 100644
>>> --- a/include/linux/fs.h
>>> +++ b/include/linux/fs.h
>>> @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
>>> kgid_has_mapping(fs_userns, kgid);
>>> }
>>>
>>> -extern struct timespec64 current_time(struct inode *inode);
>>> +struct timespec64 current_time(struct inode *inode);
>>> +struct timespec64 inode_ctime_set_current(struct inode *inode);
>>> +
>>> +/**
>>> + * inode_ctime_peek - fetch the current ctime from the inode
>>> + * @inode: inode from which to fetch ctime
>>> + *
>>> + * Grab the current ctime from the inode and return it.
>>> + */
>>> +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
>>> +{
>>> + return inode->i_ctime;
>>> +}
>>> +
>>> +/**
>>> + * inode_ctime_set - set the ctime in the inode to the given value
>>> + * @inode: inode in which to set the ctime
>>> + * @ts: timespec value to set the ctime
>>> + *
>>> + * Set the ctime in @inode to @ts.
>>> + */
>>> +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
>>> +{
>>> + inode->i_ctime = ts;
>>> + return ts;
>>> +}
>>> +
>>> +/**
>>> + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
>>
>> I'm curious about why you choose to split the tv_sec and tv_nsec
>> set_ functions. Do any callers not set them both? Wouldn't a
>> single call enable a more atomic behavior someday?
>>
>> inode_ctime_set_sec_nsec(struct inode *, time64_t, time64_t)
>>
>> (or simply initialize a timespec64 and use inode_ctime_spec() )
>>
>
> Yes, quite a few places set the fields individually. For example, when
> loading a value from disk that doesn't have sufficient granularity to
> set the nsecs field to anything but 0.
Well, they still need to set the tv_nsec so they could just pass 0.
But ok.
> Could I have done it by declaring a local timespec64 variable and just
> use the inode_ctime_set function in these places? Absolutely.
>
> That's a bit more difficult to handle with coccinelle though. If someone
> wants to suggest a way to do that without having to change all of these
> call sites manually, then I'm open to redoing the set.
>
> That might be better left for a later cleanup though.
Acked-by: Tom Talpey <tom@talpey.com>
>>> + * @inode: inode in which to set the ctime
>>> + * @sec: value to set the tv_sec field
>>> + *
>>> + * Set the sec field in the ctime. Returns @sec.
>>> + */
>>> +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
>>> +{
>>> + inode->i_ctime.tv_sec = sec;
>>> + return sec;
>>> +}
>>> +
>>> +/**
>>> + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
>>> + * @inode: inode in which to set the ctime
>>> + * @nsec: value to set the tv_nsec field
>>> + *
>>> + * Set the nsec field in the ctime. Returns @nsec.
>>> + */
>>> +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
>>> +{
>>> + inode->i_ctime.tv_nsec = nsec;
>>> + return nsec;
>>> +}
>>>
>>> /*
>>> * Snapshotting support.
>
_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* [Cluster-devel] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-21 18:19 ` Tom Talpey
0 siblings, 0 replies; 229+ messages in thread
From: Tom Talpey @ 2023-06-21 18:19 UTC (permalink / raw)
To: cluster-devel.redhat.com
On 6/21/2023 2:01 PM, Jeff Layton wrote:
> On Wed, 2023-06-21 at 13:29 -0400, Tom Talpey wrote:
>> On 6/21/2023 10:45 AM, Jeff Layton wrote:
>>> struct timespec64 has unused bits in the tv_nsec field that can be used
>>> for other purposes. In future patches, we're going to change how the
>>> inode->i_ctime is accessed in certain inodes in order to make use of
>>> them. In order to do that safely though, we'll need to eradicate raw
>>> accesses of the inode->i_ctime field from the kernel.
>>>
>>> Add new accessor functions for the ctime that we can use to replace them.
>>>
>>> Signed-off-by: Jeff Layton <jlayton@kernel.org>
>>> ---
>>> fs/inode.c | 16 ++++++++++++++
>>> include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
>>> 2 files changed, 68 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/fs/inode.c b/fs/inode.c
>>> index d37fad91c8da..c005e7328fbb 100644
>>> --- a/fs/inode.c
>>> +++ b/fs/inode.c
>>> @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
>>> }
>>> EXPORT_SYMBOL(current_time);
>>>
>>> +/**
>>> + * inode_ctime_set_current - set the ctime to current_time
>>> + * @inode: inode
>>> + *
>>> + * Set the inode->i_ctime to the current value for the inode. Returns
>>> + * the current value that was assigned to i_ctime.
>>> + */
>>> +struct timespec64 inode_ctime_set_current(struct inode *inode)
>>> +{
>>> + struct timespec64 now = current_time(inode);
>>> +
>>> + inode_set_ctime(inode, now);
>>> + return now;
>>> +}
>>> +EXPORT_SYMBOL(inode_ctime_set_current);
>>> +
>>> /**
>>> * in_group_or_capable - check whether caller is CAP_FSETID privileged
>>> * @idmap: idmap of the mount @inode was found from
>>> diff --git a/include/linux/fs.h b/include/linux/fs.h
>>> index 6867512907d6..9afb30606373 100644
>>> --- a/include/linux/fs.h
>>> +++ b/include/linux/fs.h
>>> @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
>>> kgid_has_mapping(fs_userns, kgid);
>>> }
>>>
>>> -extern struct timespec64 current_time(struct inode *inode);
>>> +struct timespec64 current_time(struct inode *inode);
>>> +struct timespec64 inode_ctime_set_current(struct inode *inode);
>>> +
>>> +/**
>>> + * inode_ctime_peek - fetch the current ctime from the inode
>>> + * @inode: inode from which to fetch ctime
>>> + *
>>> + * Grab the current ctime from the inode and return it.
>>> + */
>>> +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
>>> +{
>>> + return inode->i_ctime;
>>> +}
>>> +
>>> +/**
>>> + * inode_ctime_set - set the ctime in the inode to the given value
>>> + * @inode: inode in which to set the ctime
>>> + * @ts: timespec value to set the ctime
>>> + *
>>> + * Set the ctime in @inode to @ts.
>>> + */
>>> +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
>>> +{
>>> + inode->i_ctime = ts;
>>> + return ts;
>>> +}
>>> +
>>> +/**
>>> + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
>>
>> I'm curious about why you choose to split the tv_sec and tv_nsec
>> set_ functions. Do any callers not set them both? Wouldn't a
>> single call enable a more atomic behavior someday?
>>
>> inode_ctime_set_sec_nsec(struct inode *, time64_t, time64_t)
>>
>> (or simply initialize a timespec64 and use inode_ctime_spec() )
>>
>
> Yes, quite a few places set the fields individually. For example, when
> loading a value from disk that doesn't have sufficient granularity to
> set the nsecs field to anything but 0.
Well, they still need to set the tv_nsec so they could just pass 0.
But ok.
> Could I have done it by declaring a local timespec64 variable and just
> use the inode_ctime_set function in these places? Absolutely.
>
> That's a bit more difficult to handle with coccinelle though. If someone
> wants to suggest a way to do that without having to change all of these
> call sites manually, then I'm open to redoing the set.
>
> That might be better left for a later cleanup though.
Acked-by: Tom Talpey <tom@talpey.com>
>>> + * @inode: inode in which to set the ctime
>>> + * @sec: value to set the tv_sec field
>>> + *
>>> + * Set the sec field in the ctime. Returns @sec.
>>> + */
>>> +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
>>> +{
>>> + inode->i_ctime.tv_sec = sec;
>>> + return sec;
>>> +}
>>> +
>>> +/**
>>> + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
>>> + * @inode: inode in which to set the ctime
>>> + * @nsec: value to set the tv_nsec field
>>> + *
>>> + * Set the nsec field in the ctime. Returns @nsec.
>>> + */
>>> +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
>>> +{
>>> + inode->i_ctime.tv_nsec = nsec;
>>> + return nsec;
>>> +}
>>>
>>> /*
>>> * Snapshotting support.
>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
2023-06-21 18:19 ` [f2fs-dev] " Tom Talpey
(?)
(?)
@ 2023-06-21 18:48 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 18:48 UTC (permalink / raw)
To: Tom Talpey, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On Wed, 2023-06-21 at 14:19 -0400, Tom Talpey wrote:
> On 6/21/2023 2:01 PM, Jeff Layton wrote:
> > On Wed, 2023-06-21 at 13:29 -0400, Tom Talpey wrote:
> > > On 6/21/2023 10:45 AM, Jeff Layton wrote:
> > > > struct timespec64 has unused bits in the tv_nsec field that can be used
> > > > for other purposes. In future patches, we're going to change how the
> > > > inode->i_ctime is accessed in certain inodes in order to make use of
> > > > them. In order to do that safely though, we'll need to eradicate raw
> > > > accesses of the inode->i_ctime field from the kernel.
> > > >
> > > > Add new accessor functions for the ctime that we can use to replace them.
> > > >
> > > > Signed-off-by: Jeff Layton <jlayton@kernel.org>
> > > > ---
> > > > fs/inode.c | 16 ++++++++++++++
> > > > include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
> > > > 2 files changed, 68 insertions(+), 1 deletion(-)
> > > >
> > > > diff --git a/fs/inode.c b/fs/inode.c
> > > > index d37fad91c8da..c005e7328fbb 100644
> > > > --- a/fs/inode.c
> > > > +++ b/fs/inode.c
> > > > @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
> > > > }
> > > > EXPORT_SYMBOL(current_time);
> > > >
> > > > +/**
> > > > + * inode_ctime_set_current - set the ctime to current_time
> > > > + * @inode: inode
> > > > + *
> > > > + * Set the inode->i_ctime to the current value for the inode. Returns
> > > > + * the current value that was assigned to i_ctime.
> > > > + */
> > > > +struct timespec64 inode_ctime_set_current(struct inode *inode)
> > > > +{
> > > > + struct timespec64 now = current_time(inode);
> > > > +
> > > > + inode_set_ctime(inode, now);
> > > > + return now;
> > > > +}
> > > > +EXPORT_SYMBOL(inode_ctime_set_current);
> > > > +
> > > > /**
> > > > * in_group_or_capable - check whether caller is CAP_FSETID privileged
> > > > * @idmap: idmap of the mount @inode was found from
> > > > diff --git a/include/linux/fs.h b/include/linux/fs.h
> > > > index 6867512907d6..9afb30606373 100644
> > > > --- a/include/linux/fs.h
> > > > +++ b/include/linux/fs.h
> > > > @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
> > > > kgid_has_mapping(fs_userns, kgid);
> > > > }
> > > >
> > > > -extern struct timespec64 current_time(struct inode *inode);
> > > > +struct timespec64 current_time(struct inode *inode);
> > > > +struct timespec64 inode_ctime_set_current(struct inode *inode);
> > > > +
> > > > +/**
> > > > + * inode_ctime_peek - fetch the current ctime from the inode
> > > > + * @inode: inode from which to fetch ctime
> > > > + *
> > > > + * Grab the current ctime from the inode and return it.
> > > > + */
> > > > +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
> > > > +{
> > > > + return inode->i_ctime;
> > > > +}
> > > > +
> > > > +/**
> > > > + * inode_ctime_set - set the ctime in the inode to the given value
> > > > + * @inode: inode in which to set the ctime
> > > > + * @ts: timespec value to set the ctime
> > > > + *
> > > > + * Set the ctime in @inode to @ts.
> > > > + */
> > > > +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
> > > > +{
> > > > + inode->i_ctime = ts;
> > > > + return ts;
> > > > +}
> > > > +
> > > > +/**
> > > > + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
> > >
> > > I'm curious about why you choose to split the tv_sec and tv_nsec
> > > set_ functions. Do any callers not set them both? Wouldn't a
> > > single call enable a more atomic behavior someday?
> > >
> > > inode_ctime_set_sec_nsec(struct inode *, time64_t, time64_t)
> > >
> > > (or simply initialize a timespec64 and use inode_ctime_spec() )
> > >
> >
> > Yes, quite a few places set the fields individually. For example, when
> > loading a value from disk that doesn't have sufficient granularity to
> > set the nsecs field to anything but 0.
>
> Well, they still need to set the tv_nsec so they could just pass 0.
> But ok.
>
Sure. The difficulty is in trying to do this in an automated way. For
instance, look at the hfsplus patch; it has separate assignments in
place already:
- result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date));
- result->i_ctime.tv_nsec = 0;
+ inode_ctime_set_sec(result,
+ result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date)));
+ inode_ctime_set_nsec(result, 0);
Granted the new code is pretty ugly, but it compiles!
Transforming that into what you're suggesting is a tougher proposition
to do with coccinelle. I didn't see a way to conditionally catch cases
like this, declare a new variable in the appropriate spot and then
transform two assignments (that may not be next to one another!) into a
single one.
Maybe it's possible, but my grasp of SMPL is not that great. The docs
and examples (including Kees' vey helpful ones!) cover fairly simple
changes well, but I didn't quite grasp how to do that complex an
evolution.
> > Could I have done it by declaring a local timespec64 variable and just
> > use the inode_ctime_set function in these places? Absolutely.
> >
> > That's a bit more difficult to handle with coccinelle though. If someone
> > wants to suggest a way to do that without having to change all of these
> > call sites manually, then I'm open to redoing the set.
> >
> > That might be better left for a later cleanup though.
>
> Acked-by: Tom Talpey <tom@talpey.com>
>
Many thanks!
> > > > + * @inode: inode in which to set the ctime
> > > > + * @sec: value to set the tv_sec field
> > > > + *
> > > > + * Set the sec field in the ctime. Returns @sec.
> > > > + */
> > > > +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
> > > > +{
> > > > + inode->i_ctime.tv_sec = sec;
> > > > + return sec;
> > > > +}
> > > > +
> > > > +/**
> > > > + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
> > > > + * @inode: inode in which to set the ctime
> > > > + * @nsec: value to set the tv_nsec field
> > > > + *
> > > > + * Set the nsec field in the ctime. Returns @nsec.
> > > > + */
> > > > +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
> > > > +{
> > > > + inode->i_ctime.tv_nsec = nsec;
> > > > + return nsec;
> > > > +}
> > > >
> > > > /*
> > > > * Snapshotting support.
> >
--
Jeff Layton <jlayton@kernel.org>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [f2fs-dev] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-21 18:48 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 18:48 UTC (permalink / raw)
To: Tom Talpey, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On Wed, 2023-06-21 at 14:19 -0400, Tom Talpey wrote:
> On 6/21/2023 2:01 PM, Jeff Layton wrote:
> > On Wed, 2023-06-21 at 13:29 -0400, Tom Talpey wrote:
> > > On 6/21/2023 10:45 AM, Jeff Layton wrote:
> > > > struct timespec64 has unused bits in the tv_nsec field that can be used
> > > > for other purposes. In future patches, we're going to change how the
> > > > inode->i_ctime is accessed in certain inodes in order to make use of
> > > > them. In order to do that safely though, we'll need to eradicate raw
> > > > accesses of the inode->i_ctime field from the kernel.
> > > >
> > > > Add new accessor functions for the ctime that we can use to replace them.
> > > >
> > > > Signed-off-by: Jeff Layton <jlayton@kernel.org>
> > > > ---
> > > > fs/inode.c | 16 ++++++++++++++
> > > > include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
> > > > 2 files changed, 68 insertions(+), 1 deletion(-)
> > > >
> > > > diff --git a/fs/inode.c b/fs/inode.c
> > > > index d37fad91c8da..c005e7328fbb 100644
> > > > --- a/fs/inode.c
> > > > +++ b/fs/inode.c
> > > > @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
> > > > }
> > > > EXPORT_SYMBOL(current_time);
> > > >
> > > > +/**
> > > > + * inode_ctime_set_current - set the ctime to current_time
> > > > + * @inode: inode
> > > > + *
> > > > + * Set the inode->i_ctime to the current value for the inode. Returns
> > > > + * the current value that was assigned to i_ctime.
> > > > + */
> > > > +struct timespec64 inode_ctime_set_current(struct inode *inode)
> > > > +{
> > > > + struct timespec64 now = current_time(inode);
> > > > +
> > > > + inode_set_ctime(inode, now);
> > > > + return now;
> > > > +}
> > > > +EXPORT_SYMBOL(inode_ctime_set_current);
> > > > +
> > > > /**
> > > > * in_group_or_capable - check whether caller is CAP_FSETID privileged
> > > > * @idmap: idmap of the mount @inode was found from
> > > > diff --git a/include/linux/fs.h b/include/linux/fs.h
> > > > index 6867512907d6..9afb30606373 100644
> > > > --- a/include/linux/fs.h
> > > > +++ b/include/linux/fs.h
> > > > @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
> > > > kgid_has_mapping(fs_userns, kgid);
> > > > }
> > > >
> > > > -extern struct timespec64 current_time(struct inode *inode);
> > > > +struct timespec64 current_time(struct inode *inode);
> > > > +struct timespec64 inode_ctime_set_current(struct inode *inode);
> > > > +
> > > > +/**
> > > > + * inode_ctime_peek - fetch the current ctime from the inode
> > > > + * @inode: inode from which to fetch ctime
> > > > + *
> > > > + * Grab the current ctime from the inode and return it.
> > > > + */
> > > > +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
> > > > +{
> > > > + return inode->i_ctime;
> > > > +}
> > > > +
> > > > +/**
> > > > + * inode_ctime_set - set the ctime in the inode to the given value
> > > > + * @inode: inode in which to set the ctime
> > > > + * @ts: timespec value to set the ctime
> > > > + *
> > > > + * Set the ctime in @inode to @ts.
> > > > + */
> > > > +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
> > > > +{
> > > > + inode->i_ctime = ts;
> > > > + return ts;
> > > > +}
> > > > +
> > > > +/**
> > > > + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
> > >
> > > I'm curious about why you choose to split the tv_sec and tv_nsec
> > > set_ functions. Do any callers not set them both? Wouldn't a
> > > single call enable a more atomic behavior someday?
> > >
> > > inode_ctime_set_sec_nsec(struct inode *, time64_t, time64_t)
> > >
> > > (or simply initialize a timespec64 and use inode_ctime_spec() )
> > >
> >
> > Yes, quite a few places set the fields individually. For example, when
> > loading a value from disk that doesn't have sufficient granularity to
> > set the nsecs field to anything but 0.
>
> Well, they still need to set the tv_nsec so they could just pass 0.
> But ok.
>
Sure. The difficulty is in trying to do this in an automated way. For
instance, look at the hfsplus patch; it has separate assignments in
place already:
- result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date));
- result->i_ctime.tv_nsec = 0;
+ inode_ctime_set_sec(result,
+ result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date)));
+ inode_ctime_set_nsec(result, 0);
Granted the new code is pretty ugly, but it compiles!
Transforming that into what you're suggesting is a tougher proposition
to do with coccinelle. I didn't see a way to conditionally catch cases
like this, declare a new variable in the appropriate spot and then
transform two assignments (that may not be next to one another!) into a
single one.
Maybe it's possible, but my grasp of SMPL is not that great. The docs
and examples (including Kees' vey helpful ones!) cover fairly simple
changes well, but I didn't quite grasp how to do that complex an
evolution.
> > Could I have done it by declaring a local timespec64 variable and just
> > use the inode_ctime_set function in these places? Absolutely.
> >
> > That's a bit more difficult to handle with coccinelle though. If someone
> > wants to suggest a way to do that without having to change all of these
> > call sites manually, then I'm open to redoing the set.
> >
> > That might be better left for a later cleanup though.
>
> Acked-by: Tom Talpey <tom@talpey.com>
>
Many thanks!
> > > > + * @inode: inode in which to set the ctime
> > > > + * @sec: value to set the tv_sec field
> > > > + *
> > > > + * Set the sec field in the ctime. Returns @sec.
> > > > + */
> > > > +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
> > > > +{
> > > > + inode->i_ctime.tv_sec = sec;
> > > > + return sec;
> > > > +}
> > > > +
> > > > +/**
> > > > + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
> > > > + * @inode: inode in which to set the ctime
> > > > + * @nsec: value to set the tv_nsec field
> > > > + *
> > > > + * Set the nsec field in the ctime. Returns @nsec.
> > > > + */
> > > > +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
> > > > +{
> > > > + inode->i_ctime.tv_nsec = nsec;
> > > > + return nsec;
> > > > +}
> > > >
> > > > /*
> > > > * Snapshotting support.
> >
--
Jeff Layton <jlayton@kernel.org>
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [Ocfs2-devel] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-21 18:48 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton via Ocfs2-devel @ 2023-06-21 18:48 UTC (permalink / raw)
To: Tom Talpey, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On Wed, 2023-06-21 at 14:19 -0400, Tom Talpey wrote:
> On 6/21/2023 2:01 PM, Jeff Layton wrote:
> > On Wed, 2023-06-21 at 13:29 -0400, Tom Talpey wrote:
> > > On 6/21/2023 10:45 AM, Jeff Layton wrote:
> > > > struct timespec64 has unused bits in the tv_nsec field that can be used
> > > > for other purposes. In future patches, we're going to change how the
> > > > inode->i_ctime is accessed in certain inodes in order to make use of
> > > > them. In order to do that safely though, we'll need to eradicate raw
> > > > accesses of the inode->i_ctime field from the kernel.
> > > >
> > > > Add new accessor functions for the ctime that we can use to replace them.
> > > >
> > > > Signed-off-by: Jeff Layton <jlayton@kernel.org>
> > > > ---
> > > > fs/inode.c | 16 ++++++++++++++
> > > > include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
> > > > 2 files changed, 68 insertions(+), 1 deletion(-)
> > > >
> > > > diff --git a/fs/inode.c b/fs/inode.c
> > > > index d37fad91c8da..c005e7328fbb 100644
> > > > --- a/fs/inode.c
> > > > +++ b/fs/inode.c
> > > > @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
> > > > }
> > > > EXPORT_SYMBOL(current_time);
> > > >
> > > > +/**
> > > > + * inode_ctime_set_current - set the ctime to current_time
> > > > + * @inode: inode
> > > > + *
> > > > + * Set the inode->i_ctime to the current value for the inode. Returns
> > > > + * the current value that was assigned to i_ctime.
> > > > + */
> > > > +struct timespec64 inode_ctime_set_current(struct inode *inode)
> > > > +{
> > > > + struct timespec64 now = current_time(inode);
> > > > +
> > > > + inode_set_ctime(inode, now);
> > > > + return now;
> > > > +}
> > > > +EXPORT_SYMBOL(inode_ctime_set_current);
> > > > +
> > > > /**
> > > > * in_group_or_capable - check whether caller is CAP_FSETID privileged
> > > > * @idmap: idmap of the mount @inode was found from
> > > > diff --git a/include/linux/fs.h b/include/linux/fs.h
> > > > index 6867512907d6..9afb30606373 100644
> > > > --- a/include/linux/fs.h
> > > > +++ b/include/linux/fs.h
> > > > @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
> > > > kgid_has_mapping(fs_userns, kgid);
> > > > }
> > > >
> > > > -extern struct timespec64 current_time(struct inode *inode);
> > > > +struct timespec64 current_time(struct inode *inode);
> > > > +struct timespec64 inode_ctime_set_current(struct inode *inode);
> > > > +
> > > > +/**
> > > > + * inode_ctime_peek - fetch the current ctime from the inode
> > > > + * @inode: inode from which to fetch ctime
> > > > + *
> > > > + * Grab the current ctime from the inode and return it.
> > > > + */
> > > > +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
> > > > +{
> > > > + return inode->i_ctime;
> > > > +}
> > > > +
> > > > +/**
> > > > + * inode_ctime_set - set the ctime in the inode to the given value
> > > > + * @inode: inode in which to set the ctime
> > > > + * @ts: timespec value to set the ctime
> > > > + *
> > > > + * Set the ctime in @inode to @ts.
> > > > + */
> > > > +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
> > > > +{
> > > > + inode->i_ctime = ts;
> > > > + return ts;
> > > > +}
> > > > +
> > > > +/**
> > > > + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
> > >
> > > I'm curious about why you choose to split the tv_sec and tv_nsec
> > > set_ functions. Do any callers not set them both? Wouldn't a
> > > single call enable a more atomic behavior someday?
> > >
> > > inode_ctime_set_sec_nsec(struct inode *, time64_t, time64_t)
> > >
> > > (or simply initialize a timespec64 and use inode_ctime_spec() )
> > >
> >
> > Yes, quite a few places set the fields individually. For example, when
> > loading a value from disk that doesn't have sufficient granularity to
> > set the nsecs field to anything but 0.
>
> Well, they still need to set the tv_nsec so they could just pass 0.
> But ok.
>
Sure. The difficulty is in trying to do this in an automated way. For
instance, look at the hfsplus patch; it has separate assignments in
place already:
- result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date));
- result->i_ctime.tv_nsec = 0;
+ inode_ctime_set_sec(result,
+ result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date)));
+ inode_ctime_set_nsec(result, 0);
Granted the new code is pretty ugly, but it compiles!
Transforming that into what you're suggesting is a tougher proposition
to do with coccinelle. I didn't see a way to conditionally catch cases
like this, declare a new variable in the appropriate spot and then
transform two assignments (that may not be next to one another!) into a
single one.
Maybe it's possible, but my grasp of SMPL is not that great. The docs
and examples (including Kees' vey helpful ones!) cover fairly simple
changes well, but I didn't quite grasp how to do that complex an
evolution.
> > Could I have done it by declaring a local timespec64 variable and just
> > use the inode_ctime_set function in these places? Absolutely.
> >
> > That's a bit more difficult to handle with coccinelle though. If someone
> > wants to suggest a way to do that without having to change all of these
> > call sites manually, then I'm open to redoing the set.
> >
> > That might be better left for a later cleanup though.
>
> Acked-by: Tom Talpey <tom@talpey.com>
>
Many thanks!
> > > > + * @inode: inode in which to set the ctime
> > > > + * @sec: value to set the tv_sec field
> > > > + *
> > > > + * Set the sec field in the ctime. Returns @sec.
> > > > + */
> > > > +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
> > > > +{
> > > > + inode->i_ctime.tv_sec = sec;
> > > > + return sec;
> > > > +}
> > > > +
> > > > +/**
> > > > + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
> > > > + * @inode: inode in which to set the ctime
> > > > + * @nsec: value to set the tv_nsec field
> > > > + *
> > > > + * Set the nsec field in the ctime. Returns @nsec.
> > > > + */
> > > > +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
> > > > +{
> > > > + inode->i_ctime.tv_nsec = nsec;
> > > > + return nsec;
> > > > +}
> > > >
> > > > /*
> > > > * Snapshotting support.
> >
--
Jeff Layton <jlayton@kernel.org>
_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* [Cluster-devel] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-21 18:48 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 18:48 UTC (permalink / raw)
To: cluster-devel.redhat.com
On Wed, 2023-06-21 at 14:19 -0400, Tom Talpey wrote:
> On 6/21/2023 2:01 PM, Jeff Layton wrote:
> > On Wed, 2023-06-21 at 13:29 -0400, Tom Talpey wrote:
> > > On 6/21/2023 10:45 AM, Jeff Layton wrote:
> > > > struct timespec64 has unused bits in the tv_nsec field that can be used
> > > > for other purposes. In future patches, we're going to change how the
> > > > inode->i_ctime is accessed in certain inodes in order to make use of
> > > > them. In order to do that safely though, we'll need to eradicate raw
> > > > accesses of the inode->i_ctime field from the kernel.
> > > >
> > > > Add new accessor functions for the ctime that we can use to replace them.
> > > >
> > > > Signed-off-by: Jeff Layton <jlayton@kernel.org>
> > > > ---
> > > > fs/inode.c | 16 ++++++++++++++
> > > > include/linux/fs.h | 53 +++++++++++++++++++++++++++++++++++++++++++++-
> > > > 2 files changed, 68 insertions(+), 1 deletion(-)
> > > >
> > > > diff --git a/fs/inode.c b/fs/inode.c
> > > > index d37fad91c8da..c005e7328fbb 100644
> > > > --- a/fs/inode.c
> > > > +++ b/fs/inode.c
> > > > @@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
> > > > }
> > > > EXPORT_SYMBOL(current_time);
> > > >
> > > > +/**
> > > > + * inode_ctime_set_current - set the ctime to current_time
> > > > + * @inode: inode
> > > > + *
> > > > + * Set the inode->i_ctime to the current value for the inode. Returns
> > > > + * the current value that was assigned to i_ctime.
> > > > + */
> > > > +struct timespec64 inode_ctime_set_current(struct inode *inode)
> > > > +{
> > > > + struct timespec64 now = current_time(inode);
> > > > +
> > > > + inode_set_ctime(inode, now);
> > > > + return now;
> > > > +}
> > > > +EXPORT_SYMBOL(inode_ctime_set_current);
> > > > +
> > > > /**
> > > > * in_group_or_capable - check whether caller is CAP_FSETID privileged
> > > > * @idmap: idmap of the mount @inode was found from
> > > > diff --git a/include/linux/fs.h b/include/linux/fs.h
> > > > index 6867512907d6..9afb30606373 100644
> > > > --- a/include/linux/fs.h
> > > > +++ b/include/linux/fs.h
> > > > @@ -1474,7 +1474,58 @@ static inline bool fsuidgid_has_mapping(struct super_block *sb,
> > > > kgid_has_mapping(fs_userns, kgid);
> > > > }
> > > >
> > > > -extern struct timespec64 current_time(struct inode *inode);
> > > > +struct timespec64 current_time(struct inode *inode);
> > > > +struct timespec64 inode_ctime_set_current(struct inode *inode);
> > > > +
> > > > +/**
> > > > + * inode_ctime_peek - fetch the current ctime from the inode
> > > > + * @inode: inode from which to fetch ctime
> > > > + *
> > > > + * Grab the current ctime from the inode and return it.
> > > > + */
> > > > +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
> > > > +{
> > > > + return inode->i_ctime;
> > > > +}
> > > > +
> > > > +/**
> > > > + * inode_ctime_set - set the ctime in the inode to the given value
> > > > + * @inode: inode in which to set the ctime
> > > > + * @ts: timespec value to set the ctime
> > > > + *
> > > > + * Set the ctime in @inode to @ts.
> > > > + */
> > > > +static inline struct timespec64 inode_ctime_set(struct inode *inode, struct timespec64 ts)
> > > > +{
> > > > + inode->i_ctime = ts;
> > > > + return ts;
> > > > +}
> > > > +
> > > > +/**
> > > > + * inode_ctime_set_sec - set only the tv_sec field in the inode ctime
> > >
> > > I'm curious about why you choose to split the tv_sec and tv_nsec
> > > set_ functions. Do any callers not set them both? Wouldn't a
> > > single call enable a more atomic behavior someday?
> > >
> > > inode_ctime_set_sec_nsec(struct inode *, time64_t, time64_t)
> > >
> > > (or simply initialize a timespec64 and use inode_ctime_spec() )
> > >
> >
> > Yes, quite a few places set the fields individually. For example, when
> > loading a value from disk that doesn't have sufficient granularity to
> > set the nsecs field to anything but 0.
>
> Well, they still need to set the tv_nsec so they could just pass 0.
> But ok.
>
Sure. The difficulty is in trying to do this in an automated way. For
instance, look at the hfsplus patch; it has separate assignments in
place already:
- result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date));
- result->i_ctime.tv_nsec = 0;
+ inode_ctime_set_sec(result,
+ result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date)));
+ inode_ctime_set_nsec(result, 0);
Granted the new code is pretty ugly, but it compiles!
Transforming that into what you're suggesting is a tougher proposition
to do with coccinelle. I didn't see a way to conditionally catch cases
like this, declare a new variable in the appropriate spot and then
transform two assignments (that may not be next to one another!) into a
single one.
Maybe it's possible, but my grasp of SMPL is not that great. The docs
and examples (including Kees' vey helpful ones!) cover fairly simple
changes well, but I didn't quite grasp how to do that complex an
evolution.
> > Could I have done it by declaring a local timespec64 variable and just
> > use the inode_ctime_set function in these places? Absolutely.
> >
> > That's a bit more difficult to handle with coccinelle though. If someone
> > wants to suggest a way to do that without having to change all of these
> > call sites manually, then I'm open to redoing the set.
> >
> > That might be better left for a later cleanup though.
>
> Acked-by: Tom Talpey <tom@talpey.com>
>
Many thanks!
> > > > + * @inode: inode in which to set the ctime
> > > > + * @sec: value to set the tv_sec field
> > > > + *
> > > > + * Set the sec field in the ctime. Returns @sec.
> > > > + */
> > > > +static inline time64_t inode_ctime_set_sec(struct inode *inode, time64_t sec)
> > > > +{
> > > > + inode->i_ctime.tv_sec = sec;
> > > > + return sec;
> > > > +}
> > > > +
> > > > +/**
> > > > + * inode_ctime_set_nsec - set only the tv_nsec field in the inode ctime
> > > > + * @inode: inode in which to set the ctime
> > > > + * @nsec: value to set the tv_nsec field
> > > > + *
> > > > + * Set the nsec field in the ctime. Returns @nsec.
> > > > + */
> > > > +static inline long inode_ctime_set_nsec(struct inode *inode, long nsec)
> > > > +{
> > > > + inode->i_ctime.tv_nsec = nsec;
> > > > + return nsec;
> > > > +}
> > > >
> > > > /*
> > > > * Snapshotting support.
> >
--
Jeff Layton <jlayton@kernel.org>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 00/79] fs: new accessors for inode->i_ctime
2023-06-21 14:45 ` [f2fs-dev] " Jeff Layton
` (4 preceding siblings ...)
(?)
@ 2023-06-21 19:21 ` Steven Rostedt
-1 siblings, 0 replies; 229+ messages in thread
From: Steven Rostedt @ 2023-06-21 19:21 UTC (permalink / raw)
To: Jeff Layton
Cc: Jeremy Kerr, Arnd Bergmann, Michael Ellerman, Nicholas Piggin,
Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Tom Talpey, Sergey Senozhatsky, Phillip Lougher,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On Wed, 21 Jun 2023 10:45:05 -0400
Jeff Layton <jlayton@kernel.org> wrote:
> Most of this conversion was done via coccinelle, with a few of the more
> non-standard accesses done by hand. There should be no behavioral
> changes with this set. That will come later, as we convert individual
> filesystems to use multigrain timestamps.
BTW, Linus has suggested to me that whenever a conccinelle script is used,
it should be included in the change log.
-- Steve
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [f2fs-dev] [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-21 19:21 ` Steven Rostedt
0 siblings, 0 replies; 229+ messages in thread
From: Steven Rostedt @ 2023-06-21 19:21 UTC (permalink / raw)
To: Jeff Layton
Cc: Latchesar Ionkov, Rafael J. Wysocki, Darrick J. Wong,
Anders Larsen, Carlos Llamas, Andrii Nakryiko, Hugh Dickins,
John Johansen, Seth Forshee, Alexander Gordeev,
Christoph Hellwig, Mike Marshall, Paulo Alcantara, linux-xfs,
Bart Van Assche, Michael Ellerman, John Keeping, Zhang Yi,
James Morris, Christophe Leroy, Tyler Hicks, Alan Stern,
Christian Borntraeger, devel, Shyam Prasad N, Jan Harkes,
linux-um, Nicholas Piggin, Alexander Viro, Eric Van Hensbergen,
Suren Baghdasaryan, Trond Myklebust, Anton Altaparmakov,
Christian Brauner, Wolfram Sang, Greg Kroah-Hartman,
Stephen Smalley, linux-usb, linux-kernel, Ronnie Sahlberg,
Sergey Senozhatsky, Luis Chamberlain, Chuck Lever, Sven Schnelle,
Jiri Olsa, Jan Kara, Tejun Heo, Andrew Morton,
linux-trace-kernel, linux-hardening, Dave Kleikamp,
Sandeep Dhavale, Tetsuo Handa, Mimi Zohar, linux-mm,
Joel Fernandes, Eric Dumazet, Stanislav Fomichev,
Andrzej Pietrasiewicz, Hangyu Hua, linux-s390, linux-nilfs,
Paul Moore, Leon Romanovsky, John Fastabend,
Arve Hjønnevåg, Minghao Chi, codalist, selinux,
ZhangPeng, Udipto Goswami, Yonghong Song, Iurii Zaikin,
Namjae Jeon, Masami Hiramatsu, ecryptfs, Todd Kjos,
Vasily Gorbik, Yu Zhe, linuxppc-dev, reiserfs-devel,
Miklos Szeredi, Yue Hu, Jaegeuk Kim, Aditya Garg, Martijn Coenen,
OGAWA Hirofumi, Hao Luo, Tony Luck, Theodore Ts'o,
Nicolas Pitre, linux-ntfs-dev, Muchun Song, Roberto Sassu,
linux-f2fs-devel, Guilherme G. Piccoli, Jozef Martiniak,
Eric Biederman, Anna Schumaker, xu xin, Brad Warrum,
Mike Kravetz, Jingyu Wang, linux-efi, Dan Carpenter,
Martin Brandenburg, Tom Rix, Alexei Starovoitov, Chris Mason,
linux-mtd, Matthew Wilcox (Oracle),
Marc Dionne, linux-afs, Ian Kent, Naohiro Aota, Daniel Borkmann,
Dennis Dalessandro, linux-rdma, Linyu Yuan, coda,
Viacheslav Dubeyko, Ilya Dryomov, Paolo Abeni, Alexey Dobriyan,
Serge E. Hallyn, Jens Axboe, Zeng Jingxiang, Kees Cook,
Arnd Bergmann, autofs, Yifei Liu, Damien Le Moal, Eric Paris,
ceph-devel, Gao Xiang, Jiangshan Yi, David Howells, linux-nfs,
linux-ext4, Song Liu, samba-technical, Steve French, Jeremy Kerr,
netdev, Bob Peterson, linux-fsdevel, bpf, ntfs3, linux-erofs,
David S. Miller, ocfs2-devel, jfs-discussion, Dominique Martinet,
Christian Schoenebeck, Bob Copeland, KP Singh, Oleg Kanatov,
Konstantin Komarov, Joseph Qi, Yuta Hayama, Andreas Dilger,
Mikulas Patocka, Zhengchao Shao, Chen Zhongjin, Ard Biesheuvel,
Anton Ivanov, Laurent Pinchart, Andreas Gruenbacher,
Richard Weinberger, Mark Fasheh, Stefan Roesch, cluster-devel,
Jason Gunthorpe, Jakub Kicinski, Rik van Riel, Salah Triki,
Evgeniy Dushistov, linux-cifs, Heiko Carstens, apparmor,
Josef Bacik, Liam R. Howlett, Tom Talpey, Hans de Goede,
Tigran A. Aivazian, Dave Chinner, David Sterba, Xiubo Li,
Ryusuke Konishi, Juergen Gross, Johannes Thumshirn, Ritu Agarwal,
Luis de Bethencourt, Martin KaFai Lau, v9fs,
Fabio M. De Francesco, linux-unionfs, Ruihan Li,
linux-security-module, Erez Zadok, Jeffle Xu,
Dr. David Alan Gilbert, Johannes Weiner, Phillip Lougher,
Johannes Berg, Sungjong Seo, David Woodhouse, linux-karma-devel,
linux-btrfs, Joel Becker
On Wed, 21 Jun 2023 10:45:05 -0400
Jeff Layton <jlayton@kernel.org> wrote:
> Most of this conversion was done via coccinelle, with a few of the more
> non-standard accesses done by hand. There should be no behavioral
> changes with this set. That will come later, as we convert individual
> filesystems to use multigrain timestamps.
BTW, Linus has suggested to me that whenever a conccinelle script is used,
it should be included in the change log.
-- Steve
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [Ocfs2-devel] [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-21 19:21 ` Steven Rostedt
0 siblings, 0 replies; 229+ messages in thread
From: Steven Rostedt via Ocfs2-devel @ 2023-06-21 19:21 UTC (permalink / raw)
To: Jeff Layton
Cc: Latchesar Ionkov, Rafael J. Wysocki, Anders Larsen,
Carlos Llamas, Andrii Nakryiko, Hugh Dickins, John Johansen,
Seth Forshee, Alexander Gordeev, Mike Marshall, Paulo Alcantara,
linux-xfs, Bart Van Assche, Michael Ellerman, John Keeping,
Zhang Yi, James Morris, Christophe Leroy, Tyler Hicks,
Alan Stern, Christian Borntraeger, devel, Shyam Prasad N,
Jan Harkes, linux-um, Nicholas Piggin, Alexander Viro,
Eric Van Hensbergen, Suren Baghdasaryan, Trond Myklebust,
Anton Altaparmakov, Christian Brauner, Wolfram Sang,
Greg Kroah-Hartman, Stephen Smalley, linux-usb, linux-kernel,
Ronnie Sahlberg, Sergey Senozhatsky, Luis Chamberlain,
Chuck Lever, Sven Schnelle, Jiri Olsa, Jan Kara, Tejun Heo,
linux-trace-kernel, linux-hardening, Dave Kleikamp,
Sandeep Dhavale, Tetsuo Handa, Mimi Zohar, linux-mm,
Joel Fernandes, Eric Dumazet, Stanislav Fomichev,
Andrzej Pietrasiewicz, Hangyu Hua, linux-s390, linux-nilfs,
Paul Moore, Leon Romanovsky, John Fastabend,
Arve Hjønnevåg, Minghao Chi, codalist, selinux,
ZhangPeng, Udipto Goswami, Yonghong Song, Iurii Zaikin,
Namjae Jeon, Masami Hiramatsu, ecryptfs, Todd Kjos,
Vasily Gorbik, Yu Zhe, linuxppc-dev, reiserfs-devel,
Miklos Szeredi, Yue Hu, Jaegeuk Kim, Aditya Garg, Martijn Coenen,
OGAWA Hirofumi, Hao Luo, Tony Luck, Theodore Ts'o,
Nicolas Pitre, linux-ntfs-dev, Muchun Song, Roberto Sassu,
linux-f2fs-devel, Guilherme G. Piccoli, Jozef Martiniak,
Eric Biederman, Anna Schumaker, xu xin, Brad Warrum,
Mike Kravetz, Jingyu Wang, linux-efi, Dan Carpenter,
Martin Brandenburg, Tom Rix, Alexei Starovoitov, Chris Mason,
linux-mtd, Marc Dionne, linux-afs, Ian Kent, Naohiro Aota,
Daniel Borkmann, Dennis Dalessandro, linux-rdma, Linyu Yuan,
coda, Viacheslav Dubeyko, Ilya Dryomov, Paolo Abeni,
Alexey Dobriyan, Serge E. Hallyn, Zhihao Cheng, Jens Axboe,
Zeng Jingxiang, Kees Cook, Arnd Bergmann, autofs, Yifei Liu,
Damien Le Moal, Eric Paris, ceph-devel, Gao Xiang, Jiangshan Yi,
David Howells, linux-nfs, linux-ext4, Song Liu, samba-technical,
Steve French, Jeremy Kerr, netdev, Bob Peterson, linux-fsdevel,
bpf, ntfs3, linux-erofs, David S. Miller, ocfs2-devel,
jfs-discussion, Dominique Martinet, Christian Schoenebeck,
Bob Copeland, KP Singh, Oleg Kanatov, Konstantin Komarov,
Yuta Hayama, Andreas Dilger, Mikulas Patocka, Zhengchao Shao,
Chen Zhongjin, Ard Biesheuvel, Anton Ivanov, Laurent Pinchart,
Andreas Gruenbacher, Richard Weinberger, Stefan Roesch,
cluster-devel, Jason Gunthorpe, Jakub Kicinski, Rik van Riel,
Salah Triki, Evgeniy Dushistov, linux-cifs, Heiko Carstens,
Chao Yu, apparmor, Josef Bacik, Liam R. Howlett, Tom Talpey,
Hans de Goede, Tigran A. Aivazian, Dave Chinner, David Sterba,
Xiubo Li, Ryusuke Konishi, Juergen Gross, Johannes Thumshirn,
Ritu Agarwal, Luis de Bethencourt, Martin KaFai Lau, v9fs,
Fabio M. De Francesco, linux-unionfs, Ruihan Li,
linux-security-module, Erez Zadok, Jeffle Xu,
Dr. David Alan Gilbert, Johannes Weiner, Phillip Lougher,
Johannes Berg, Sungjong Seo, David Woodhouse, linux-karma-devel,
linux-btrfs
On Wed, 21 Jun 2023 10:45:05 -0400
Jeff Layton <jlayton@kernel.org> wrote:
> Most of this conversion was done via coccinelle, with a few of the more
> non-standard accesses done by hand. There should be no behavioral
> changes with this set. That will come later, as we convert individual
> filesystems to use multigrain timestamps.
BTW, Linus has suggested to me that whenever a conccinelle script is used,
it should be included in the change log.
-- Steve
_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-21 19:21 ` Steven Rostedt
0 siblings, 0 replies; 229+ messages in thread
From: Steven Rostedt @ 2023-06-21 19:21 UTC (permalink / raw)
To: Jeff Layton
Cc: Latchesar Ionkov, Rafael J. Wysocki, Darrick J. Wong,
Anders Larsen, Carlos Llamas, Andrii Nakryiko, Hugh Dickins,
John Johansen, Seth Forshee, Alexander Gordeev,
Christoph Hellwig, Mike Marshall, Paulo Alcantara, linux-xfs,
Bart Van Assche, John Keeping, Zhang Yi, James Morris,
Tyler Hicks, Alan Stern, Christian Borntraeger, devel,
Shyam Prasad N, Jan Harkes, linux-um, Nicholas Piggin,
Alexander Viro, Eric Van Hensbergen, Suren Baghdasaryan,
Trond Myklebust, Anton Altaparmakov, Christian Brauner,
Wolfram Sang, Greg Kroah-Hartman, Stephen Smalley, linux-usb,
linux-kernel, Ronnie Sahlberg, Sergey Senozhatsky,
Luis Chamberlain, Chuck Lever, Sven Schnelle, Jiri Olsa,
Jan Kara, Tejun Heo, Andrew Morton, linux-trace-kernel,
linux-hardening, Dave Kleikamp, Sandeep Dhavale, Tetsuo Handa,
Mimi Zohar, linux-mm, Joel Fernandes, Eric Dumazet,
Stanislav Fomichev, Andrzej Pietrasiewicz, Hangyu Hua,
linux-s390, linux-nilfs, Paul Moore, Leon Romanovsky,
John Fastabend, Arve Hjønnevåg, Minghao Chi, codalist,
selinux, ZhangPeng, Udipto Goswami, Yonghong Song, Iurii Zaikin,
Namjae Jeon, Masami Hiramatsu, ecryptfs, Todd Kjos,
Vasily Gorbik, Yu Zhe, linuxppc-dev, reiserfs-devel,
Miklos Szeredi, Yue Hu, Jaegeuk Kim, Aditya Garg, Martijn Coenen,
OGAWA Hirofumi, Hao Luo, Tony Luck, Theodore Ts'o,
Nicolas Pitre, linux-ntfs-dev, Muchun Song, Roberto Sassu,
linux-f2fs-devel, Guilherme G. Piccoli, Jozef Martiniak,
Eric Biederman, Anna Schumaker, xu xin, Brad Warrum,
Mike Kravetz, Jingyu Wang, linux-efi, Dan Carpenter,
Martin Brandenburg, Tom Rix, Alexei Starovoitov, Chris Mason,
linux-mtd, Matthew Wilcox (Oracle),
Marc Dionne, linux-afs, Ian Kent, Naohiro Aota
On Wed, 21 Jun 2023 10:45:05 -0400
Jeff Layton <jlayton@kernel.org> wrote:
> Most of this conversion was done via coccinelle, with a few of the more
> non-standard accesses done by hand. There should be no behavioral
> changes with this set. That will come later, as we convert individual
> filesystems to use multigrain timestamps.
BTW, Linus has suggested to me that whenever a conccinelle script is used,
it should be included in the change log.
-- Steve
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-21 19:21 ` Steven Rostedt
0 siblings, 0 replies; 229+ messages in thread
From: Steven Rostedt @ 2023-06-21 19:21 UTC (permalink / raw)
To: Jeff Layton
Cc: Latchesar Ionkov, Rafael J. Wysocki, Darrick J. Wong,
Anders Larsen, Carlos Llamas, Andrii Nakryiko, Hugh Dickins,
John Johansen, Seth Forshee, Alexander Gordeev,
Christoph Hellwig, Mike Marshall, Paulo Alcantara, linux-xfs,
Bart Van Assche, Michael Ellerman, John Keeping, Zhang Yi,
James Morris, Christophe Leroy, Tyler Hicks, Alan Stern,
Christian Borntraeger, devel, Shyam Prasad N, Jan Harkes,
linux-um, Nicholas Piggin, Alexand er Viro, Eric Van Hensbergen,
Suren Baghdasaryan, Trond Myklebust, Anton Altaparmakov,
Christian Brauner, Wolfram Sang, Greg Kroah-Hartman,
Stephen Smalley, linux-usb, linux-kernel, Ronnie Sahlberg,
Sergey Senozhatsky, Luis Chamberlain, Chuck Lever, Sven Schnelle,
Jiri Olsa, Jan Kara, Tejun Heo, Andrew Morton,
linux-trace-kernel, linux-hardening, Dave Kleikamp, Tetsuo Handa,
Mimi Zohar, linux-mm, Joel Fernand es, Eric Dumazet,
Stanislav Fomichev, Andrzej Pietrasiewicz, Hangyu Hua,
linux-s390, linux-nilfs, Paul Moore, Leon Romanovsky,
John Fastabend, Arve Hjønnevåg, Minghao Chi, codalist,
selinux, ZhangPeng, Udipto Goswami, Yonghong Song, Iurii Zaikin,
Namjae Jeon, Masami Hiramatsu, ecryptfs, Todd Kjos,
Vasily Gorbik, Yu Zhe, linuxppc-dev, reiserfs-devel,
Miklos Szeredi, Yue Hu, Jaegeuk Kim, Aditya Garg, Martijn Coenen,
OGAWA Hirofumi, Hao Luo, Tony Luck, Theodore Ts'o,
Nicolas Pitre, linux-ntfs-dev, Muchun Song, Roberto Sassu,
linux-f2fs-devel, Guilherme G. Piccoli, Jozef Martiniak,
Eric Biederman, Anna Schumaker, xu xin, Brad Warrum,
Mike Kravetz, Jingyu Wang, linux-efi, Dan Carpenter,
Martin Brandenburg, Tom Rix, Alexei Starovoitov, Chris Mason,
linux-mtd, Matthew Wilcox (Oracle),
Marc Dionne, linux-afs, Ian Kent, Naohiro Aota, Daniel Borkmann,
Dennis Dalessandro, linux-rdma, Linyu Yuan, coda,
Viacheslav Dubeyko, Ilya Dryomov, Paolo Abeni, Alexey Dobriyan,
Serge E. Hallyn, Zhihao Cheng, Jens Axboe, Zeng Jingxiang,
Kees Cook, Arnd Bergmann, autofs, Yifei Liu, Damien Le Moal,
Eric Paris, ceph-devel, Jiangshan Yi, David Howells, linux-nfs,
linux-ext4, Song Liu, samba-technical, Steve French, Jeremy Kerr,
netdev, Bob Peterson, linux-fsdevel, bpf, ntfs3, linux-erofs,
David S. Miller, ocfs2-devel, jfs-discussion, Dominique Martinet,
Christian Schoenebeck, Bob Copeland, KP Singh, Oleg Kanatov,
Konstantin Komarov, Joseph Qi, Yuta Hayama, Andreas Dilger,
Mikulas Patocka, Zhengchao Shao, Chen Zhongjin, Ard Biesheuvel,
Anton Ivanov, Laurent Pinchart, Andreas Gruenbacher,
Richard Weinberger, Mark Fasheh, Stefan Roesch, cluster-devel,
Jason Gunthorpe, Jakub Kicinski, Rik van Riel, Salah Triki,
Evgeniy Dushistov, linux-cifs, Heiko Carstens, apparmor,
Josef Bacik, Liam R. Howlett, Tom Talpey, Hans de Goede,
Tigran A. Aivazian, Dave Chinner, David Sterba, Xiubo Li,
Ryusuke Konishi, Juergen Gross, Johannes Thumshirn, Ritu Agarwal,
Luis de Bethencourt, Martin KaFai Lau, v9fs,
Fabio M. De Francesco, linux-unionfs, Ruihan Li,
linux-security-module, Erez Zadok, Dr. David Alan Gilbert,
Johannes Weiner, Phillip Lougher, Johannes Berg, Sungjong Seo,
David Woodhouse, linux-karma-devel, linux-btrfs, Joel Becker
On Wed, 21 Jun 2023 10:45:05 -0400
Jeff Layton <jlayton@kernel.org> wrote:
> Most of this conversion was done via coccinelle, with a few of the more
> non-standard accesses done by hand. There should be no behavioral
> changes with this set. That will come later, as we convert individual
> filesystems to use multigrain timestamps.
BTW, Linus has suggested to me that whenever a conccinelle script is used,
it should be included in the change log.
-- Steve
^ permalink raw reply [flat|nested] 229+ messages in thread
* [Cluster-devel] [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-21 19:21 ` Steven Rostedt
0 siblings, 0 replies; 229+ messages in thread
From: Steven Rostedt @ 2023-06-21 19:21 UTC (permalink / raw)
To: cluster-devel.redhat.com
On Wed, 21 Jun 2023 10:45:05 -0400
Jeff Layton <jlayton@kernel.org> wrote:
> Most of this conversion was done via coccinelle, with a few of the more
> non-standard accesses done by hand. There should be no behavioral
> changes with this set. That will come later, as we convert individual
> filesystems to use multigrain timestamps.
BTW, Linus has suggested to me that whenever a conccinelle script is used,
it should be included in the change log.
-- Steve
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-21 19:21 ` Steven Rostedt
0 siblings, 0 replies; 229+ messages in thread
From: Steven Rostedt @ 2023-06-21 19:21 UTC (permalink / raw)
To: Jeff Layton
Cc: Latchesar Ionkov, Rafael J. Wysocki, Darrick J. Wong,
Anders Larsen, Carlos Llamas, Andrii Nakryiko, Hugh Dickins,
John Johansen, Seth Forshee, Alexander Gordeev,
Christoph Hellwig, Mike Marshall, Paulo Alcantara, linux-xfs,
Bart Van Assche, Michael Ellerman, John Keeping, Zhang Yi,
James Morris, Christophe Leroy, Tyler Hicks, Alan Stern,
Christian Borntraeger, devel
On Wed, 21 Jun 2023 10:45:05 -0400
Jeff Layton <jlayton@kernel.org> wrote:
> Most of this conversion was done via coccinelle, with a few of the more
> non-standard accesses done by hand. There should be no behavioral
> changes with this set. That will come later, as we convert individual
> filesystems to use multigrain timestamps.
BTW, Linus has suggested to me that whenever a conccinelle script is used,
it should be included in the change log.
-- Steve
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 05/79] qib_fs: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 05/79] qib_fs: " Jeff Layton
2023-06-21 16:35 ` Jan Kara
@ 2023-06-21 19:37 ` Dennis Dalessandro
1 sibling, 0 replies; 229+ messages in thread
From: Dennis Dalessandro @ 2023-06-21 19:37 UTC (permalink / raw)
To: Jeff Layton, Christian Brauner, Jason Gunthorpe, Leon Romanovsky
Cc: Al Viro, Jan Kara, linux-rdma, linux-kernel
On 6/21/23 10:45 AM, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
> ---
> drivers/infiniband/hw/qib/qib_fs.c | 4 +---
> 1 file changed, 1 insertion(+), 3 deletions(-)
>
> diff --git a/drivers/infiniband/hw/qib/qib_fs.c b/drivers/infiniband/hw/qib/qib_fs.c
> index a973905afd13..db008f2884e4 100644
> --- a/drivers/infiniband/hw/qib/qib_fs.c
> +++ b/drivers/infiniband/hw/qib/qib_fs.c
> @@ -64,9 +64,7 @@ static int qibfs_mknod(struct inode *dir, struct dentry *dentry,
> inode->i_uid = GLOBAL_ROOT_UID;
> inode->i_gid = GLOBAL_ROOT_GID;
> inode->i_blocks = 0;
> - inode->i_atime = current_time(inode);
> - inode->i_mtime = inode->i_atime;
> - inode->i_ctime = inode->i_atime;
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> inode->i_private = data;
> if (S_ISDIR(mode)) {
> inode->i_op = &simple_dir_inode_operations;
Acked-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 00/79] fs: new accessors for inode->i_ctime
2023-06-21 19:21 ` [f2fs-dev] " Steven Rostedt
` (4 preceding siblings ...)
(?)
@ 2023-06-21 19:52 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 19:52 UTC (permalink / raw)
To: Steven Rostedt
Cc: Jeremy Kerr, Arnd Bergmann, Michael Ellerman, Nicholas Piggin,
Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Tom Talpey, Sergey Senozhatsky, Phillip Lougher,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On Wed, 2023-06-21 at 15:21 -0400, Steven Rostedt wrote:
> On Wed, 21 Jun 2023 10:45:05 -0400
> Jeff Layton <jlayton@kernel.org> wrote:
>
> > Most of this conversion was done via coccinelle, with a few of the more
> > non-standard accesses done by hand. There should be no behavioral
> > changes with this set. That will come later, as we convert individual
> > filesystems to use multigrain timestamps.
>
> BTW, Linus has suggested to me that whenever a conccinelle script is used,
> it should be included in the change log.
>
Ok, here's what I have. I note again that my usage of coccinelle is
pretty primitive, so I ended up doing a fair bit of by-hand fixing after
applying these.
Given the way that this change is broken up into 77 patches by
subsystem, to which changelogs should I add it? I could add it to the
"infrastructure" patch, but that's the one where I _didn't_ use it.
Maybe to patch #79 (the one that renames i_ctime)?
------------------------8<------------------------------
@@
expression inode;
@@
- inode->i_ctime = current_time(inode)
+ inode_set_current_ctime(inode)
@@
expression inode;
@@
- inode->i_ctime = inode->i_mtime = current_time(inode)
+ inode->i_mtime = inode_set_current_ctime(inode)
@@
struct inode *inode;
expression value;
@@
- inode->i_ctime = value;
+ inode_set_ctime(inode, value);
@@
struct inode *inode;
expression val;
@@
- inode->i_ctime.tv_sec = val
+ inode_set_ctime_sec(inode, val)
@@
struct inode *inode;
expression val;
@@
- inode->i_ctime.tv_nsec = val
+ inode_set_ctime_nsec(inode, val)
@@
struct inode *inode;
@@
- inode->i_ctime
+ inode_ctime_peek(inode)
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [f2fs-dev] [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-21 19:52 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 19:52 UTC (permalink / raw)
To: Steven Rostedt
Cc: Latchesar Ionkov, Rafael J. Wysocki, Darrick J. Wong,
Anders Larsen, Carlos Llamas, Andrii Nakryiko, Hugh Dickins,
John Johansen, Seth Forshee, Alexander Gordeev,
Christoph Hellwig, Mike Marshall, Paulo Alcantara, linux-xfs,
Bart Van Assche, Michael Ellerman, John Keeping, Zhang Yi,
James Morris, Christophe Leroy, Tyler Hicks, Alan Stern,
Christian Borntraeger, devel, Shyam Prasad N, Jan Harkes,
linux-um, Nicholas Piggin, Alexander Viro, Eric Van Hensbergen,
Suren Baghdasaryan, Trond Myklebust, Anton Altaparmakov,
Christian Brauner, Wolfram Sang, Greg Kroah-Hartman,
Stephen Smalley, linux-usb, linux-kernel, Ronnie Sahlberg,
Sergey Senozhatsky, Luis Chamberlain, Chuck Lever, Sven Schnelle,
Jiri Olsa, Jan Kara, Tejun Heo, Andrew Morton,
linux-trace-kernel, linux-hardening, Dave Kleikamp,
Sandeep Dhavale, Tetsuo Handa, Mimi Zohar, linux-mm,
Joel Fernandes, Eric Dumazet, Stanislav Fomichev,
Andrzej Pietrasiewicz, Hangyu Hua, linux-s390, linux-nilfs,
Paul Moore, Leon Romanovsky, John Fastabend,
Arve Hjønnevåg, Minghao Chi, codalist, selinux,
ZhangPeng, Udipto Goswami, Yonghong Song, Iurii Zaikin,
Namjae Jeon, Masami Hiramatsu, ecryptfs, Todd Kjos,
Vasily Gorbik, Yu Zhe, linuxppc-dev, reiserfs-devel,
Miklos Szeredi, Yue Hu, Jaegeuk Kim, Aditya Garg, Martijn Coenen,
OGAWA Hirofumi, Hao Luo, Tony Luck, Theodore Ts'o,
Nicolas Pitre, linux-ntfs-dev, Muchun Song, Roberto Sassu,
linux-f2fs-devel, Guilherme G. Piccoli, Jozef Martiniak,
Eric Biederman, Anna Schumaker, xu xin, Brad Warrum,
Mike Kravetz, Jingyu Wang, linux-efi, Dan Carpenter,
Martin Brandenburg, Tom Rix, Alexei Starovoitov, Chris Mason,
linux-mtd, Matthew Wilcox (Oracle),
Marc Dionne, linux-afs, Ian Kent, Naohiro Aota, Daniel Borkmann,
Dennis Dalessandro, linux-rdma, Linyu Yuan, coda,
Viacheslav Dubeyko, Ilya Dryomov, Paolo Abeni, Alexey Dobriyan,
Serge E. Hallyn, Jens Axboe, Zeng Jingxiang, Kees Cook,
Arnd Bergmann, autofs, Yifei Liu, Damien Le Moal, Eric Paris,
ceph-devel, Gao Xiang, Jiangshan Yi, David Howells, linux-nfs,
linux-ext4, Song Liu, samba-technical, Steve French, Jeremy Kerr,
netdev, Bob Peterson, linux-fsdevel, bpf, ntfs3, linux-erofs,
David S. Miller, ocfs2-devel, jfs-discussion, Dominique Martinet,
Christian Schoenebeck, Bob Copeland, KP Singh, Oleg Kanatov,
Konstantin Komarov, Joseph Qi, Yuta Hayama, Andreas Dilger,
Mikulas Patocka, Zhengchao Shao, Chen Zhongjin, Ard Biesheuvel,
Anton Ivanov, Laurent Pinchart, Andreas Gruenbacher,
Richard Weinberger, Mark Fasheh, Stefan Roesch, cluster-devel,
Jason Gunthorpe, Jakub Kicinski, Rik van Riel, Salah Triki,
Evgeniy Dushistov, linux-cifs, Heiko Carstens, apparmor,
Josef Bacik, Liam R. Howlett, Tom Talpey, Hans de Goede,
Tigran A. Aivazian, Dave Chinner, David Sterba, Xiubo Li,
Ryusuke Konishi, Juergen Gross, Johannes Thumshirn, Ritu Agarwal,
Luis de Bethencourt, Martin KaFai Lau, v9fs,
Fabio M. De Francesco, linux-unionfs, Ruihan Li,
linux-security-module, Erez Zadok, Jeffle Xu,
Dr. David Alan Gilbert, Johannes Weiner, Phillip Lougher,
Johannes Berg, Sungjong Seo, David Woodhouse, linux-karma-devel,
linux-btrfs, Joel Becker
On Wed, 2023-06-21 at 15:21 -0400, Steven Rostedt wrote:
> On Wed, 21 Jun 2023 10:45:05 -0400
> Jeff Layton <jlayton@kernel.org> wrote:
>
> > Most of this conversion was done via coccinelle, with a few of the more
> > non-standard accesses done by hand. There should be no behavioral
> > changes with this set. That will come later, as we convert individual
> > filesystems to use multigrain timestamps.
>
> BTW, Linus has suggested to me that whenever a conccinelle script is used,
> it should be included in the change log.
>
Ok, here's what I have. I note again that my usage of coccinelle is
pretty primitive, so I ended up doing a fair bit of by-hand fixing after
applying these.
Given the way that this change is broken up into 77 patches by
subsystem, to which changelogs should I add it? I could add it to the
"infrastructure" patch, but that's the one where I _didn't_ use it.
Maybe to patch #79 (the one that renames i_ctime)?
------------------------8<------------------------------
@@
expression inode;
@@
- inode->i_ctime = current_time(inode)
+ inode_set_current_ctime(inode)
@@
expression inode;
@@
- inode->i_ctime = inode->i_mtime = current_time(inode)
+ inode->i_mtime = inode_set_current_ctime(inode)
@@
struct inode *inode;
expression value;
@@
- inode->i_ctime = value;
+ inode_set_ctime(inode, value);
@@
struct inode *inode;
expression val;
@@
- inode->i_ctime.tv_sec = val
+ inode_set_ctime_sec(inode, val)
@@
struct inode *inode;
expression val;
@@
- inode->i_ctime.tv_nsec = val
+ inode_set_ctime_nsec(inode, val)
@@
struct inode *inode;
@@
- inode->i_ctime
+ inode_ctime_peek(inode)
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-21 19:52 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 19:52 UTC (permalink / raw)
To: Steven Rostedt
Cc: Latchesar Ionkov, Rafael J. Wysocki, Darrick J. Wong,
Anders Larsen, Carlos Llamas, Andrii Nakryiko, Hugh Dickins,
John Johansen, Seth Forshee, Alexander Gordeev,
Christoph Hellwig, Mike Marshall, Paulo Alcantara, linux-xfs,
Bart Van Assche, John Keeping, Zhang Yi, James Morris,
Tyler Hicks, Alan Stern, Christian Borntraeger, devel,
Shyam Prasad N, Jan Harkes, linux-um, Nicholas Piggin,
Alexander Viro, Eric Van Hensbergen, Suren Baghdasaryan,
Trond Myklebust, Anton Altaparmakov, Christian Brauner,
Wolfram Sang, Greg Kroah-Hartman, Stephen Smalley, linux-usb,
linux-kernel, Ronnie Sahlberg, Sergey Senozhatsky,
Luis Chamberlain, Chuck Lever, Sven Schnelle, Jiri Olsa,
Jan Kara, Tejun Heo, Andrew Morton, linux-trace-kernel,
linux-hardening, Dave Kleikamp, Sandeep Dhavale, Tetsuo Handa,
Mimi Zohar, linux-mm, Joel Fernandes, Eric Dumazet,
Stanislav Fomichev, Andrzej Pietrasiewicz, Hangyu Hua,
linux-s390, linux-nilfs, Paul Moore, Leon Romanovsky,
John Fastabend, Arve Hjønnevåg, Minghao Chi, codalist,
selinux, ZhangPeng, Udipto Goswami, Yonghong Song, Iurii Zaikin,
Namjae Jeon, Masami Hiramatsu, ecryptfs, Todd Kjos,
Vasily Gorbik, Yu Zhe, linuxppc-dev, reiserfs-devel,
Miklos Szeredi, Yue Hu, Jaegeuk Kim, Aditya Garg, Martijn Coenen,
OGAWA Hirofumi, Hao Luo, Tony Luck, Theodore Ts'o,
Nicolas Pitre, linux-ntfs-dev, Muchun Song, Roberto Sassu,
linux-f2fs-devel, Guilherme G. Piccoli, Jozef Martiniak,
Eric Biederman, Anna Schumaker, xu xin, Brad Warrum,
Mike Kravetz, Jingyu Wang, linux-efi, Dan Carpenter,
Martin Brandenburg, Tom Rix, Alexei Starovoitov, Chris Mason,
linux-mtd, Matthew Wilcox (Oracle),
Marc Dionne, linux-afs, Ian Kent, Naohiro Aota, Daniel Borkmann,
Dennis Dalessandro, linux-rdma, Linyu Yuan, coda,
Viacheslav Dubeyko, Ilya Dryomov, Paolo Abeni, Alexey Dobriyan,
Serge E. Hallyn, Zhihao Cheng, Jens Axboe, Zeng Jingxiang,
Kees Cook, Arnd Bergmann, autofs, Yifei Liu, Damien Le Moal,
Eric Paris, ceph-devel, Gao Xiang, Jiangshan Yi, David Howells,
linux-nfs, linux-ext4, Song Liu, samba-technical, Steve French,
Jeremy Kerr, netdev, Bob Peterson, linux-fsdevel, bpf, ntfs3,
linux-erofs, David S. Miller, ocfs2-devel, jfs-discussion,
Dominique Martinet, Christian Schoenebeck, Bob Copeland,
KP Singh, Oleg Kanatov, Konstantin Komarov, Joseph Qi,
Yuta Hayama, Andreas Dilger, Mikulas Patocka, Zhengchao Shao,
Chen Zhongjin, Ard Biesheuvel, Anton Ivanov, Laurent Pinchart,
Andreas Gruenbacher, Richard Weinberger, Mark Fasheh,
Stefan Roesch, cluster-devel, Jason Gunthorpe, Jakub Kicinski,
Rik van Riel, Salah Triki, Evgeniy Dushistov, linux-cifs,
Heiko Carstens, Chao Yu, apparmor, Josef Bacik, Liam R. Howlett,
Tom Talpey, Hans de Goede, Tigran A. Aivazian, Dave Chinner,
David Sterba, Xiubo Li, Ryusuke Konishi, Juergen Gross,
Johannes Thumshirn, Ritu Agarwal, Luis de Bethencourt,
Martin KaFai Lau, v9fs, Fabio M. De Francesco, linux-unionfs,
Ruihan Li, linux-security-module, Erez Zadok, Jeffle Xu,
Dr. David Alan Gilbert, Johannes Weiner, Phillip Lougher,
Johannes Berg, Sungjong Seo, David Woodhouse, linux-karma-devel,
linux-btrfs, Joel Becker
On Wed, 2023-06-21 at 15:21 -0400, Steven Rostedt wrote:
> On Wed, 21 Jun 2023 10:45:05 -0400
> Jeff Layton <jlayton@kernel.org> wrote:
>
> > Most of this conversion was done via coccinelle, with a few of the more
> > non-standard accesses done by hand. There should be no behavioral
> > changes with this set. That will come later, as we convert individual
> > filesystems to use multigrain timestamps.
>
> BTW, Linus has suggested to me that whenever a conccinelle script is used,
> it should be included in the change log.
>
Ok, here's what I have. I note again that my usage of coccinelle is
pretty primitive, so I ended up doing a fair bit of by-hand fixing after
applying these.
Given the way that this change is broken up into 77 patches by
subsystem, to which changelogs should I add it? I could add it to the
"infrastructure" patch, but that's the one where I _didn't_ use it.
Maybe to patch #79 (the one that renames i_ctime)?
------------------------8<------------------------------
@@
expression inode;
@@
- inode->i_ctime = current_time(inode)
+ inode_set_current_ctime(inode)
@@
expression inode;
@@
- inode->i_ctime = inode->i_mtime = current_time(inode)
+ inode->i_mtime = inode_set_current_ctime(inode)
@@
struct inode *inode;
expression value;
@@
- inode->i_ctime = value;
+ inode_set_ctime(inode, value);
@@
struct inode *inode;
expression val;
@@
- inode->i_ctime.tv_sec = val
+ inode_set_ctime_sec(inode, val)
@@
struct inode *inode;
expression val;
@@
- inode->i_ctime.tv_nsec = val
+ inode_set_ctime_nsec(inode, val)
@@
struct inode *inode;
@@
- inode->i_ctime
+ inode_ctime_peek(inode)
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-21 19:52 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 19:52 UTC (permalink / raw)
To: Steven Rostedt
Cc: Latchesar Ionkov, Rafael J. Wysocki, Darrick J. Wong,
Anders Larsen, Carlos Llamas, Andrii Nakryiko, Hugh Dickins,
John Johansen, Seth Forshee, Alexander Gordeev,
Christoph Hellwig, Mike Marshall, Paulo Alcantara, linux-xfs,
Bart Van Assche, Michael Ellerman, John Keeping, Zhang Yi,
James Morris, Christophe Leroy, Tyler Hicks, Alan Stern,
Christian Borntraeger, devel, Shyam Prasad N, Jan Harkes,
linux-um, Nicholas Piggin, Alexand er Viro, Eric Van Hensbergen,
Suren Baghdasaryan, Trond Myklebust, Anton Altaparmakov,
Christian Brauner, Wolfram Sang, Greg Kroah-Hartman,
Stephen Smalley, linux-usb, linux-kernel, Ronnie Sahlberg,
Sergey Senozhatsky, Luis Chamberlain, Chuck Lever, Sven Schnelle,
Jiri Olsa, Jan Kara, Tejun Heo, Andrew Morton,
linux-trace-kernel, linux-hardening, Dave Kleikamp, Tetsuo Handa,
Mimi Zohar, linux-mm, Joel Fernand es, Eric Dumazet,
Stanislav Fomichev, Andrzej Pietrasiewicz, Hangyu Hua,
linux-s390, linux-nilfs, Paul Moore, Leon Romanovsky,
John Fastabend, Arve Hjønnevåg, Minghao Chi, codalist,
selinux, ZhangPeng, Udipto Goswami, Yonghong Song, Iurii Zaikin,
Namjae Jeon, Masami Hiramatsu, ecryptfs, Todd Kjos,
Vasily Gorbik, Yu Zhe, linuxppc-dev, reiserfs-devel,
Miklos Szeredi, Yue Hu, Jaegeuk Kim, Aditya Garg, Martijn Coenen,
OGAWA Hirofumi, Hao Luo, Tony Luck, Theodore Ts'o,
Nicolas Pitre, linux-ntfs-dev, Muchun Song, Roberto Sassu,
linux-f2fs-devel, Guilherme G. Piccoli, Jozef Martiniak,
Eric Biederman, Anna Schumaker, xu xin, Brad Warrum,
Mike Kravetz, Jingyu Wang, linux-efi, Dan Carpenter,
Martin Brandenburg, Tom Rix, Alexei Starovoitov, Chris Mason,
linux-mtd, Matthew Wilcox (Oracle),
Marc Dionne, linux-afs, Ian Kent, Naohiro Aota, Daniel Borkmann,
Dennis Dalessandro, linux-rdma, Linyu Yuan, coda,
Viacheslav Dubeyko, Ilya Dryomov, Paolo Abeni, Ale xey Dobriyan,
Serge E. Hallyn, Zhihao Cheng, Jens Axboe, Zeng Jingxiang,
Kees Cook, Arnd Bergmann, autofs, Yifei Liu, Damien Le Moal,
Eric Paris, ceph-devel, Jiangshan Yi, David Howells, linux-nfs,
linux-ext4, Song Liu, samba-technical, Steve French, Jeremy Kerr,
netdev, Bob Peterson, linux-fsdevel, bpf, ntfs3, linux-erofs,
David S. Miller, ocfs2-devel, jfs-discussion, Dominique Martinet,
Christian Schoenebeck, Bob Copeland, KP Singh, Oleg Kanatov,
Konstantin Komarov, Joseph Qi, Yuta Hayama, Andreas Dilger,
Mikulas Patocka, Zhengchao Shao, Chen Zhongjin, Ard Biesheuvel,
Anton Ivanov, Laurent Pinchart, Andreas Gruenbacher,
Richard Weinberger, Mark Fasheh, Stefan Roesch, cluster-devel,
Jason Gunthorpe, Jakub Kicinski, Rik van Riel, Salah Triki,
Evgeniy Dushistov, linux-cifs, Heiko Carstens, apparmor,
Josef Bacik, Liam R. Howlett, Tom Talpey, Hans de Goede,
Tigran A. Aivazian, Dave Chinner, David Sterba, Xiubo Li,
Ryusuke Konishi, Juergen Gross, Johannes Thumshirn, Ritu Agarwal,
Luis de Bethencourt, Martin KaFai Lau, v9fs,
Fabio M. De Francesco, linux-unionfs, Ruihan Li,
linux-security-module, Erez Zadok, Dr. David Alan Gilbert,
Johannes Weiner, Phillip Lougher, Johannes Berg, Sungjong Seo,
David Woodhouse, linux-karma-devel, linux-btrfs, Joel Becker
On Wed, 2023-06-21 at 15:21 -0400, Steven Rostedt wrote:
> On Wed, 21 Jun 2023 10:45:05 -0400
> Jeff Layton <jlayton@kernel.org> wrote:
>
> > Most of this conversion was done via coccinelle, with a few of the more
> > non-standard accesses done by hand. There should be no behavioral
> > changes with this set. That will come later, as we convert individual
> > filesystems to use multigrain timestamps.
>
> BTW, Linus has suggested to me that whenever a conccinelle script is used,
> it should be included in the change log.
>
Ok, here's what I have. I note again that my usage of coccinelle is
pretty primitive, so I ended up doing a fair bit of by-hand fixing after
applying these.
Given the way that this change is broken up into 77 patches by
subsystem, to which changelogs should I add it? I could add it to the
"infrastructure" patch, but that's the one where I _didn't_ use it.
Maybe to patch #79 (the one that renames i_ctime)?
------------------------8<------------------------------
@@
expression inode;
@@
- inode->i_ctime = current_time(inode)
+ inode_set_current_ctime(inode)
@@
expression inode;
@@
- inode->i_ctime = inode->i_mtime = current_time(inode)
+ inode->i_mtime = inode_set_current_ctime(inode)
@@
struct inode *inode;
expression value;
@@
- inode->i_ctime = value;
+ inode_set_ctime(inode, value);
@@
struct inode *inode;
expression val;
@@
- inode->i_ctime.tv_sec = val
+ inode_set_ctime_sec(inode, val)
@@
struct inode *inode;
expression val;
@@
- inode->i_ctime.tv_nsec = val
+ inode_set_ctime_nsec(inode, val)
@@
struct inode *inode;
@@
- inode->i_ctime
+ inode_ctime_peek(inode)
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [Ocfs2-devel] [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-21 19:52 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton via Ocfs2-devel @ 2023-06-21 19:52 UTC (permalink / raw)
To: Steven Rostedt
Cc: Latchesar Ionkov, Rafael J. Wysocki, Anders Larsen,
Carlos Llamas, Andrii Nakryiko, Hugh Dickins, John Johansen,
Seth Forshee, Alexander Gordeev, Mike Marshall, Paulo Alcantara,
linux-xfs, Bart Van Assche, Michael Ellerman, John Keeping,
Zhang Yi, James Morris, Christophe Leroy, Tyler Hicks,
Alan Stern, Christian Borntraeger, devel, Shyam Prasad N,
Jan Harkes, linux-um, Nicholas Piggin, Alexander Viro,
Eric Van Hensbergen, Suren Baghdasaryan, Trond Myklebust,
Anton Altaparmakov, Christian Brauner, Wolfram Sang,
Greg Kroah-Hartman, Stephen Smalley, linux-usb, linux-kernel,
Ronnie Sahlberg, Sergey Senozhatsky, Luis Chamberlain,
Chuck Lever, Sven Schnelle, Jiri Olsa, Jan Kara, Tejun Heo,
linux-trace-kernel, linux-hardening, Dave Kleikamp,
Sandeep Dhavale, Tetsuo Handa, Mimi Zohar, linux-mm,
Joel Fernandes, Eric Dumazet, Stanislav Fomichev,
Andrzej Pietrasiewicz, Hangyu Hua, linux-s390, linux-nilfs,
Paul Moore, Leon Romanovsky, John Fastabend,
Arve Hjønnevåg, Minghao Chi, codalist, selinux,
ZhangPeng, Udipto Goswami, Yonghong Song, Iurii Zaikin,
Namjae Jeon, Masami Hiramatsu, ecryptfs, Todd Kjos,
Vasily Gorbik, Yu Zhe, linuxppc-dev, reiserfs-devel,
Miklos Szeredi, Yue Hu, Jaegeuk Kim, Aditya Garg, Martijn Coenen,
OGAWA Hirofumi, Hao Luo, Tony Luck, Theodore Ts'o,
Nicolas Pitre, linux-ntfs-dev, Muchun Song, Roberto Sassu,
linux-f2fs-devel, Guilherme G. Piccoli, Jozef Martiniak,
Eric Biederman, Anna Schumaker, xu xin, Brad Warrum,
Mike Kravetz, Jingyu Wang, linux-efi, Dan Carpenter,
Martin Brandenburg, Tom Rix, Alexei Starovoitov, Chris Mason,
linux-mtd, Marc Dionne, linux-afs, Ian Kent, Naohiro Aota,
Daniel Borkmann, Dennis Dalessandro, linux-rdma, Linyu Yuan,
coda, Viacheslav Dubeyko, Ilya Dryomov, Paolo Abeni,
Alexey Dobriyan, Serge E. Hallyn, Zhihao Cheng, Jens Axboe,
Zeng Jingxiang, Kees Cook, Arnd Bergmann, autofs, Yifei Liu,
Damien Le Moal, Eric Paris, ceph-devel, Gao Xiang, Jiangshan Yi,
David Howells, linux-nfs, linux-ext4, Song Liu, samba-technical,
Steve French, Jeremy Kerr, netdev, Bob Peterson, linux-fsdevel,
bpf, ntfs3, linux-erofs, David S. Miller, ocfs2-devel,
jfs-discussion, Dominique Martinet, Christian Schoenebeck,
Bob Copeland, KP Singh, Oleg Kanatov, Konstantin Komarov,
Yuta Hayama, Andreas Dilger, Mikulas Patocka, Zhengchao Shao,
Chen Zhongjin, Ard Biesheuvel, Anton Ivanov, Laurent Pinchart,
Andreas Gruenbacher, Richard Weinberger, Stefan Roesch,
cluster-devel, Jason Gunthorpe, Jakub Kicinski, Rik van Riel,
Salah Triki, Evgeniy Dushistov, linux-cifs, Heiko Carstens,
Chao Yu, apparmor, Josef Bacik, Liam R. Howlett, Tom Talpey,
Hans de Goede, Tigran A. Aivazian, Dave Chinner, David Sterba,
Xiubo Li, Ryusuke Konishi, Juergen Gross, Johannes Thumshirn,
Ritu Agarwal, Luis de Bethencourt, Martin KaFai Lau, v9fs,
Fabio M. De Francesco, linux-unionfs, Ruihan Li,
linux-security-module, Erez Zadok, Jeffle Xu,
Dr. David Alan Gilbert, Johannes Weiner, Phillip Lougher,
Johannes Berg, Sungjong Seo, David Woodhouse, linux-karma-devel,
linux-btrfs
On Wed, 2023-06-21 at 15:21 -0400, Steven Rostedt wrote:
> On Wed, 21 Jun 2023 10:45:05 -0400
> Jeff Layton <jlayton@kernel.org> wrote:
>
> > Most of this conversion was done via coccinelle, with a few of the more
> > non-standard accesses done by hand. There should be no behavioral
> > changes with this set. That will come later, as we convert individual
> > filesystems to use multigrain timestamps.
>
> BTW, Linus has suggested to me that whenever a conccinelle script is used,
> it should be included in the change log.
>
Ok, here's what I have. I note again that my usage of coccinelle is
pretty primitive, so I ended up doing a fair bit of by-hand fixing after
applying these.
Given the way that this change is broken up into 77 patches by
subsystem, to which changelogs should I add it? I could add it to the
"infrastructure" patch, but that's the one where I _didn't_ use it.
Maybe to patch #79 (the one that renames i_ctime)?
------------------------8<------------------------------
@@
expression inode;
@@
- inode->i_ctime = current_time(inode)
+ inode_set_current_ctime(inode)
@@
expression inode;
@@
- inode->i_ctime = inode->i_mtime = current_time(inode)
+ inode->i_mtime = inode_set_current_ctime(inode)
@@
struct inode *inode;
expression value;
@@
- inode->i_ctime = value;
+ inode_set_ctime(inode, value);
@@
struct inode *inode;
expression val;
@@
- inode->i_ctime.tv_sec = val
+ inode_set_ctime_sec(inode, val)
@@
struct inode *inode;
expression val;
@@
- inode->i_ctime.tv_nsec = val
+ inode_set_ctime_nsec(inode, val)
@@
struct inode *inode;
@@
- inode->i_ctime
+ inode_ctime_peek(inode)
_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* [Cluster-devel] [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-21 19:52 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 19:52 UTC (permalink / raw)
To: cluster-devel.redhat.com
On Wed, 2023-06-21 at 15:21 -0400, Steven Rostedt wrote:
> On Wed, 21 Jun 2023 10:45:05 -0400
> Jeff Layton <jlayton@kernel.org> wrote:
>
> > Most of this conversion was done via coccinelle, with a few of the more
> > non-standard accesses done by hand. There should be no behavioral
> > changes with this set. That will come later, as we convert individual
> > filesystems to use multigrain timestamps.
>
> BTW, Linus has suggested to me that whenever a conccinelle script is used,
> it should be included in the change log.
>
Ok, here's what I have. I note again that my usage of coccinelle is
pretty primitive, so I ended up doing a fair bit of by-hand fixing after
applying these.
Given the way that this change is broken up into 77 patches by
subsystem, to which changelogs should I add it? I could add it to the
"infrastructure" patch, but that's the one where I _didn't_ use it.?
Maybe to patch #79 (the one that renames i_ctime)?
------------------------8<------------------------------
@@
expression inode;
@@
- inode->i_ctime = current_time(inode)
+ inode_set_current_ctime(inode)
@@
expression inode;
@@
- inode->i_ctime = inode->i_mtime = current_time(inode)
+ inode->i_mtime = inode_set_current_ctime(inode)
@@
struct inode *inode;
expression value;
@@
- inode->i_ctime = value;
+ inode_set_ctime(inode, value);
@@
struct inode *inode;
expression val;
@@
- inode->i_ctime.tv_sec = val
+ inode_set_ctime_sec(inode, val)
@@
struct inode *inode;
expression val;
@@
- inode->i_ctime.tv_nsec = val
+ inode_set_ctime_nsec(inode, val)
@@
struct inode *inode;
@@
- inode->i_ctime
+ inode_ctime_peek(inode)
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-21 19:52 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-21 19:52 UTC (permalink / raw)
To: Steven Rostedt
Cc: Latchesar Ionkov, Rafael J. Wysocki, Darrick J. Wong,
Anders Larsen, Carlos Llamas, Andrii Nakryiko, Hugh Dickins,
John Johansen, Seth Forshee, Alexander Gordeev,
Christoph Hellwig, Mike Marshall, Paulo Alcantara, linux-xfs,
Bart Van Assche, Michael Ellerman, John Keeping, Zhang Yi,
James Morris, Christophe Leroy, Tyler Hicks, Alan Stern,
Christian Borntraeger, devel
On Wed, 2023-06-21 at 15:21 -0400, Steven Rostedt wrote:
> On Wed, 21 Jun 2023 10:45:05 -0400
> Jeff Layton <jlayton@kernel.org> wrote:
>
> > Most of this conversion was done via coccinelle, with a few of the more
> > non-standard accesses done by hand. There should be no behavioral
> > changes with this set. That will come later, as we convert individual
> > filesystems to use multigrain timestamps.
>
> BTW, Linus has suggested to me that whenever a conccinelle script is used,
> it should be included in the change log.
>
Ok, here's what I have. I note again that my usage of coccinelle is
pretty primitive, so I ended up doing a fair bit of by-hand fixing after
applying these.
Given the way that this change is broken up into 77 patches by
subsystem, to which changelogs should I add it? I could add it to the
"infrastructure" patch, but that's the one where I _didn't_ use it.
Maybe to patch #79 (the one that renames i_ctime)?
------------------------8<------------------------------
@@
expression inode;
@@
- inode->i_ctime = current_time(inode)
+ inode_set_current_ctime(inode)
@@
expression inode;
@@
- inode->i_ctime = inode->i_mtime = current_time(inode)
+ inode->i_mtime = inode_set_current_ctime(inode)
@@
struct inode *inode;
expression value;
@@
- inode->i_ctime = value;
+ inode_set_ctime(inode, value);
@@
struct inode *inode;
expression val;
@@
- inode->i_ctime.tv_sec = val
+ inode_set_ctime_sec(inode, val)
@@
struct inode *inode;
expression val;
@@
- inode->i_ctime.tv_nsec = val
+ inode_set_ctime_nsec(inode, val)
@@
struct inode *inode;
@@
- inode->i_ctime
+ inode_ctime_peek(inode)
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 02/79] spufs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
@ 2023-06-22 0:19 ` Jeremy Kerr
-1 siblings, 0 replies; 229+ messages in thread
From: Jeremy Kerr @ 2023-06-22 0:19 UTC (permalink / raw)
To: Jeff Layton, Christian Brauner, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy
Cc: Al Viro, Jan Kara, linuxppc-dev, linux-kernel
Hi Jeff,
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses
> of inode->i_ctime.
s/utilized/used/ :D
All looks good on the spufs change:
Acked-by: Jeremy Kerr <jk@ozlabs.org>
(also, thanks for including the accessors patch on the wider list, made
it much easier to review in context)
Cheers,
Jeremy
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 02/79] spufs: switch to new ctime accessors
@ 2023-06-22 0:19 ` Jeremy Kerr
0 siblings, 0 replies; 229+ messages in thread
From: Jeremy Kerr @ 2023-06-22 0:19 UTC (permalink / raw)
To: Jeff Layton, Christian Brauner, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy
Cc: linuxppc-dev, Jan Kara, Al Viro, linux-kernel
Hi Jeff,
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses
> of inode->i_ctime.
s/utilized/used/ :D
All looks good on the spufs change:
Acked-by: Jeremy Kerr <jk@ozlabs.org>
(also, thanks for including the accessors patch on the wider list, made
it much easier to review in context)
Cheers,
Jeremy
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
2023-06-21 14:45 ` [f2fs-dev] " Jeff Layton
(?)
(?)
@ 2023-06-22 0:46 ` Damien Le Moal
-1 siblings, 0 replies; 229+ messages in thread
From: Damien Le Moal @ 2023-06-22 0:46 UTC (permalink / raw)
To: Jeff Layton, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Tom Talpey, Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Naohiro Aota, Johannes Thumshirn,
Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko,
Martin KaFai Lau, Song Liu, Yonghong Song, John Fastabend,
KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa, Hugh Dickins,
Andrew Morton, David S. Miller, Eric Dumazet, Jakub Kicinski,
Paolo Abeni, John Johansen, Paul Moore, James Morris,
Serge E. Hallyn, Stephen Smalley, Eric Paris, Juergen Gross,
Ruihan Li, Laurent Pinchart, Wolfram Sang, Udipto Goswami,
Linyu Yuan, John Keeping, Andrzej Pietrasiewicz, Dan Carpenter,
Yuta Hayama, Jozef Martiniak, Jens Axboe, Alan Stern,
Sandeep Dhavale, Dave Chinner, Johannes Weiner, ZhangPeng,
Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg, Erez Zadok,
Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On 6/21/23 23:45, Jeff Layton wrote:
> struct timespec64 has unused bits in the tv_nsec field that can be used
> for other purposes. In future patches, we're going to change how the
> inode->i_ctime is accessed in certain inodes in order to make use of
> them. In order to do that safely though, we'll need to eradicate raw
> accesses of the inode->i_ctime field from the kernel.
>
> Add new accessor functions for the ctime that we can use to replace them.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
[...]
> +/**
> + * inode_ctime_peek - fetch the current ctime from the inode
> + * @inode: inode from which to fetch ctime
> + *
> + * Grab the current ctime from the inode and return it.
> + */
> +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
To be consistent with inode_ctime_set(), why not call this one inode_ctime_get()
? Also, inode_set_ctime() & inode_get_ctime() may be a little more natural. But
no strong opinion about that though.
--
Damien Le Moal
Western Digital Research
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [f2fs-dev] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-22 0:46 ` Damien Le Moal
0 siblings, 0 replies; 229+ messages in thread
From: Damien Le Moal @ 2023-06-22 0:46 UTC (permalink / raw)
To: Jeff Layton, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Tom Talpey, Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Naohiro Aota, Johannes Thumshirn,
Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko,
Martin KaFai Lau, Song Liu, Yonghong Song, John Fastabend,
KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa, Hugh Dickins,
Andrew Morton, David S. Miller, Eric Dumazet, Jakub Kicinski,
Paolo Abeni, John Johansen, Paul Moore, James Morris,
Serge E. Hallyn, Stephen Smalley, Eric Paris, Juergen Gross,
Ruihan Li, Laurent Pinchart, Wolfram Sang, Udipto Goswami,
Linyu Yuan, John Keeping, Andrzej Pietrasiewicz, Dan Carpenter,
Yuta Hayama, Jozef Martiniak, Jens Axboe, Alan Stern,
Sandeep Dhavale, Dave Chinner, Johannes Weiner, ZhangPeng,
Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg, Erez Zadok,
Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On 6/21/23 23:45, Jeff Layton wrote:
> struct timespec64 has unused bits in the tv_nsec field that can be used
> for other purposes. In future patches, we're going to change how the
> inode->i_ctime is accessed in certain inodes in order to make use of
> them. In order to do that safely though, we'll need to eradicate raw
> accesses of the inode->i_ctime field from the kernel.
>
> Add new accessor functions for the ctime that we can use to replace them.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
[...]
> +/**
> + * inode_ctime_peek - fetch the current ctime from the inode
> + * @inode: inode from which to fetch ctime
> + *
> + * Grab the current ctime from the inode and return it.
> + */
> +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
To be consistent with inode_ctime_set(), why not call this one inode_ctime_get()
? Also, inode_set_ctime() & inode_get_ctime() may be a little more natural. But
no strong opinion about that though.
--
Damien Le Moal
Western Digital Research
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [Ocfs2-devel] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-22 0:46 ` Damien Le Moal
0 siblings, 0 replies; 229+ messages in thread
From: Damien Le Moal via Ocfs2-devel @ 2023-06-22 0:46 UTC (permalink / raw)
To: Jeff Layton, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Tom Talpey, Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Naohiro Aota, Johannes Thumshirn,
Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko,
Martin KaFai Lau, Song Liu, Yonghong Song, John Fastabend,
KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa, Hugh Dickins,
Andrew Morton, David S. Miller, Eric Dumazet, Jakub Kicinski,
Paolo Abeni, John Johansen, Paul Moore, James Morris,
Serge E. Hallyn, Stephen Smalley, Eric Paris, Juergen Gross,
Ruihan Li, Laurent Pinchart, Wolfram Sang, Udipto Goswami,
Linyu Yuan, John Keeping, Andrzej Pietrasiewicz, Dan Carpenter,
Yuta Hayama, Jozef Martiniak, Jens Axboe, Alan Stern,
Sandeep Dhavale, Dave Chinner, Johannes Weiner, ZhangPeng,
Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg, Erez Zadok,
Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On 6/21/23 23:45, Jeff Layton wrote:
> struct timespec64 has unused bits in the tv_nsec field that can be used
> for other purposes. In future patches, we're going to change how the
> inode->i_ctime is accessed in certain inodes in order to make use of
> them. In order to do that safely though, we'll need to eradicate raw
> accesses of the inode->i_ctime field from the kernel.
>
> Add new accessor functions for the ctime that we can use to replace them.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
[...]
> +/**
> + * inode_ctime_peek - fetch the current ctime from the inode
> + * @inode: inode from which to fetch ctime
> + *
> + * Grab the current ctime from the inode and return it.
> + */
> +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
To be consistent with inode_ctime_set(), why not call this one inode_ctime_get()
? Also, inode_set_ctime() & inode_get_ctime() may be a little more natural. But
no strong opinion about that though.
--
Damien Le Moal
Western Digital Research
_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* [Cluster-devel] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-22 0:46 ` Damien Le Moal
0 siblings, 0 replies; 229+ messages in thread
From: Damien Le Moal @ 2023-06-22 0:46 UTC (permalink / raw)
To: cluster-devel.redhat.com
On 6/21/23 23:45, Jeff Layton wrote:
> struct timespec64 has unused bits in the tv_nsec field that can be used
> for other purposes. In future patches, we're going to change how the
> inode->i_ctime is accessed in certain inodes in order to make use of
> them. In order to do that safely though, we'll need to eradicate raw
> accesses of the inode->i_ctime field from the kernel.
>
> Add new accessor functions for the ctime that we can use to replace them.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
[...]
> +/**
> + * inode_ctime_peek - fetch the current ctime from the inode
> + * @inode: inode from which to fetch ctime
> + *
> + * Grab the current ctime from the inode and return it.
> + */
> +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
To be consistent with inode_ctime_set(), why not call this one inode_ctime_get()
? Also, inode_set_ctime() & inode_get_ctime() may be a little more natural. But
no strong opinion about that though.
--
Damien Le Moal
Western Digital Research
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 71/79] zonefs: switch to new ctime accessors
2023-06-21 14:46 ` [PATCH 71/79] zonefs: " Jeff Layton
@ 2023-06-22 0:48 ` Damien Le Moal
0 siblings, 0 replies; 229+ messages in thread
From: Damien Le Moal @ 2023-06-22 0:48 UTC (permalink / raw)
To: Jeff Layton, Christian Brauner, Naohiro Aota, Johannes Thumshirn
Cc: Al Viro, Jan Kara, linux-fsdevel, linux-kernel
On 6/21/23 23:46, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Modulo the proposed renaming for inode_ctime_peek(), looks good.
Acked-by: Damien Le Moal <dlemoal@kernel.org>
> ---
> fs/zonefs/super.c | 8 +++++---
> 1 file changed, 5 insertions(+), 3 deletions(-)
>
> diff --git a/fs/zonefs/super.c b/fs/zonefs/super.c
> index bbe44a26a8e5..75be0e039ccf 100644
> --- a/fs/zonefs/super.c
> +++ b/fs/zonefs/super.c
> @@ -658,7 +658,8 @@ static struct inode *zonefs_get_file_inode(struct inode *dir,
>
> inode->i_ino = ino;
> inode->i_mode = z->z_mode;
> - inode->i_ctime = inode->i_mtime = inode->i_atime = dir->i_ctime;
> + inode->i_mtime = inode->i_atime = inode_ctime_peek(dir);
> + inode_ctime_set(inode, inode->i_mtime);
> inode->i_uid = z->z_uid;
> inode->i_gid = z->z_gid;
> inode->i_size = z->z_wpoffset;
> @@ -694,7 +695,8 @@ static struct inode *zonefs_get_zgroup_inode(struct super_block *sb,
> inode->i_ino = ino;
> inode_init_owner(&nop_mnt_idmap, inode, root, S_IFDIR | 0555);
> inode->i_size = sbi->s_zgroup[ztype].g_nr_zones;
> - inode->i_ctime = inode->i_mtime = inode->i_atime = root->i_ctime;
> + inode->i_mtime = inode->i_atime = inode_ctime_peek(root);
> + inode_ctime_set(inode, inode->i_mtime);
> inode->i_private = &sbi->s_zgroup[ztype];
> set_nlink(inode, 2);
>
> @@ -1317,7 +1319,7 @@ static int zonefs_fill_super(struct super_block *sb, void *data, int silent)
>
> inode->i_ino = bdev_nr_zones(sb->s_bdev);
> inode->i_mode = S_IFDIR | 0555;
> - inode->i_ctime = inode->i_mtime = inode->i_atime = current_time(inode);
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> inode->i_op = &zonefs_dir_inode_operations;
> inode->i_fop = &zonefs_dir_operations;
> inode->i_size = 2;
--
Damien Le Moal
Western Digital Research
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 26/79] erofs: switch to new ctime accessors
2023-06-21 14:45 ` Jeff Layton
@ 2023-06-22 4:01 ` Gao Xiang
-1 siblings, 0 replies; 229+ messages in thread
From: Gao Xiang @ 2023-06-22 4:01 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Jan Kara, linux-kernel, Yue Hu, Al Viro, linux-erofs
On Wed, Jun 21, 2023 at 10:45:39AM -0400, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Acked-by: Gao Xiang <xiang@kernel.org>
Thanks,
Gao Xiang
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 26/79] erofs: switch to new ctime accessors
@ 2023-06-22 4:01 ` Gao Xiang
0 siblings, 0 replies; 229+ messages in thread
From: Gao Xiang @ 2023-06-22 4:01 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu,
Al Viro, Jan Kara, linux-erofs, linux-kernel
On Wed, Jun 21, 2023 at 10:45:39AM -0400, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Acked-by: Gao Xiang <xiang@kernel.org>
Thanks,
Gao Xiang
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 54/79] overlayfs: switch to new ctime accessors
2023-06-21 14:46 ` [PATCH 54/79] overlayfs: " Jeff Layton
@ 2023-06-22 6:44 ` Amir Goldstein
0 siblings, 0 replies; 229+ messages in thread
From: Amir Goldstein @ 2023-06-22 6:44 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Miklos Szeredi, Al Viro, Jan Kara,
linux-unionfs, linux-kernel
On Wed, Jun 21, 2023 at 5:54 PM Jeff Layton <jlayton@kernel.org> wrote:
>
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
> ---
Looks ok.
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Thanks,
Amir.
> fs/overlayfs/file.c | 7 +++++--
> fs/overlayfs/util.c | 2 +-
> 2 files changed, 6 insertions(+), 3 deletions(-)
>
> diff --git a/fs/overlayfs/file.c b/fs/overlayfs/file.c
> index 21245b00722a..b47013d4bd4e 100644
> --- a/fs/overlayfs/file.c
> +++ b/fs/overlayfs/file.c
> @@ -239,6 +239,7 @@ static loff_t ovl_llseek(struct file *file, loff_t offset, int whence)
> static void ovl_file_accessed(struct file *file)
> {
> struct inode *inode, *upperinode;
> + struct timespec64 ct, uct;
>
> if (file->f_flags & O_NOATIME)
> return;
> @@ -249,10 +250,12 @@ static void ovl_file_accessed(struct file *file)
> if (!upperinode)
> return;
>
> + ct = inode_ctime_peek(inode);
> + uct = inode_ctime_peek(upperinode);
> if ((!timespec64_equal(&inode->i_mtime, &upperinode->i_mtime) ||
> - !timespec64_equal(&inode->i_ctime, &upperinode->i_ctime))) {
> + !timespec64_equal(&ct, &uct))) {
> inode->i_mtime = upperinode->i_mtime;
> - inode->i_ctime = upperinode->i_ctime;
> + inode_ctime_set(inode, uct);
> }
>
> touch_atime(&file->f_path);
> diff --git a/fs/overlayfs/util.c b/fs/overlayfs/util.c
> index 7ef9e13c404a..e3746f9a202f 100644
> --- a/fs/overlayfs/util.c
> +++ b/fs/overlayfs/util.c
> @@ -1202,6 +1202,6 @@ void ovl_copyattr(struct inode *inode)
> inode->i_mode = realinode->i_mode;
> inode->i_atime = realinode->i_atime;
> inode->i_mtime = realinode->i_mtime;
> - inode->i_ctime = realinode->i_ctime;
> + inode_ctime_set(inode, inode_ctime_peek(realinode));
> i_size_write(inode, i_size_read(realinode));
> }
> --
> 2.41.0
>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 76/79] apparmor: switch to new ctime accessors
2023-06-21 14:46 ` [PATCH 76/79] apparmor: " Jeff Layton
@ 2023-06-22 9:04 ` John Johansen
0 siblings, 0 replies; 229+ messages in thread
From: John Johansen @ 2023-06-22 9:04 UTC (permalink / raw)
To: Jeff Layton, Christian Brauner, Paul Moore, James Morris,
Serge E. Hallyn
Cc: Al Viro, Jan Kara, apparmor, linux-security-module, linux-kernel
On 6/21/23 07:46, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
lgtm
Acked-by: John Johansen <john.johansen@canonical.com>
> ---
> security/apparmor/apparmorfs.c | 6 +++---
> security/apparmor/policy_unpack.c | 4 ++--
> 2 files changed, 5 insertions(+), 5 deletions(-)
>
> diff --git a/security/apparmor/apparmorfs.c b/security/apparmor/apparmorfs.c
> index 3d0d370d6ffd..8c72a94dd9e3 100644
> --- a/security/apparmor/apparmorfs.c
> +++ b/security/apparmor/apparmorfs.c
> @@ -226,7 +226,7 @@ static int __aafs_setup_d_inode(struct inode *dir, struct dentry *dentry,
>
> inode->i_ino = get_next_ino();
> inode->i_mode = mode;
> - inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> inode->i_private = data;
> if (S_ISDIR(mode)) {
> inode->i_op = iops ? iops : &simple_dir_inode_operations;
> @@ -1557,7 +1557,7 @@ void __aafs_profile_migrate_dents(struct aa_profile *old,
> if (new->dents[i]) {
> struct inode *inode = d_inode(new->dents[i]);
>
> - inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_mtime = inode_ctime_set_current(inode);
> }
> old->dents[i] = NULL;
> }
> @@ -2546,7 +2546,7 @@ static int aa_mk_null_file(struct dentry *parent)
>
> inode->i_ino = get_next_ino();
> inode->i_mode = S_IFCHR | S_IRUGO | S_IWUGO;
> - inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> init_special_inode(inode, S_IFCHR | S_IRUGO | S_IWUGO,
> MKDEV(MEM_MAJOR, 3));
> d_instantiate(dentry, inode);
> diff --git a/security/apparmor/policy_unpack.c b/security/apparmor/policy_unpack.c
> index 4c188a44d65c..334d69b118a9 100644
> --- a/security/apparmor/policy_unpack.c
> +++ b/security/apparmor/policy_unpack.c
> @@ -89,10 +89,10 @@ void __aa_loaddata_update(struct aa_loaddata *data, long revision)
> struct inode *inode;
>
> inode = d_inode(data->dents[AAFS_LOADDATA_DIR]);
> - inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_mtime = inode_ctime_set_current(inode);
>
> inode = d_inode(data->dents[AAFS_LOADDATA_REVISION]);
> - inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_mtime = inode_ctime_set_current(inode);
> }
> }
>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
2023-06-22 0:46 ` [f2fs-dev] " Damien Le Moal
(?)
(?)
@ 2023-06-22 10:14 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-22 10:14 UTC (permalink / raw)
To: Damien Le Moal, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Tom Talpey, Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Naohiro Aota, Johannes Thumshirn,
Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko,
Martin KaFai Lau, Song Liu, Yonghong Song, John Fastabend,
KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa, Hugh Dickins,
Andrew Morton, David S. Miller, Eric Dumazet, Jakub Kicinski,
Paolo Abeni, John Johansen, Paul Moore, James Morris,
Serge E. Hallyn, Stephen Smalley, Eric Paris, Juergen Gross,
Ruihan Li, Laurent Pinchart, Wolfram Sang, Udipto Goswami,
Linyu Yuan, John Keeping, Andrzej Pietrasiewicz, Dan Carpenter,
Yuta Hayama, Jozef Martiniak, Jens Axboe, Alan Stern,
Sandeep Dhavale, Dave Chinner, Johannes Weiner, ZhangPeng,
Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg, Erez Zadok,
Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On Thu, 2023-06-22 at 09:46 +0900, Damien Le Moal wrote:
> On 6/21/23 23:45, Jeff Layton wrote:
> > struct timespec64 has unused bits in the tv_nsec field that can be used
> > for other purposes. In future patches, we're going to change how the
> > inode->i_ctime is accessed in certain inodes in order to make use of
> > them. In order to do that safely though, we'll need to eradicate raw
> > accesses of the inode->i_ctime field from the kernel.
> >
> > Add new accessor functions for the ctime that we can use to replace them.
> >
> > Signed-off-by: Jeff Layton <jlayton@kernel.org>
>
> [...]
>
> > +/**
> > + * inode_ctime_peek - fetch the current ctime from the inode
> > + * @inode: inode from which to fetch ctime
> > + *
> > + * Grab the current ctime from the inode and return it.
> > + */
> > +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
>
> To be consistent with inode_ctime_set(), why not call this one inode_ctime_get()
In later patches fetching the ctime for presentation may have side
effects on certain filesystems. Using "peek" here is a hint that we want
to avoid those side effects in these calls.
> ? Also, inode_set_ctime() & inode_get_ctime() may be a little more natural. But
> no strong opinion about that though.
>
I like the consistency of the inode_ctime_* prefix. It makes it simpler
to find these calls when grepping, etc.
That said, my opinions on naming are pretty loosely-held, so if the
consensus is that the names should as you suggest, I'll go along with
it.
--
Jeff Layton <jlayton@kernel.org>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [f2fs-dev] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-22 10:14 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-22 10:14 UTC (permalink / raw)
To: Damien Le Moal, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Tom Talpey, Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Naohiro Aota, Johannes Thumshirn,
Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko,
Martin KaFai Lau, Song Liu, Yonghong Song, John Fastabend,
KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa, Hugh Dickins,
Andrew Morton, David S. Miller, Eric Dumazet, Jakub Kicinski,
Paolo Abeni, John Johansen, Paul Moore, James Morris,
Serge E. Hallyn, Stephen Smalley, Eric Paris, Juergen Gross,
Ruihan Li, Laurent Pinchart, Wolfram Sang, Udipto Goswami,
Linyu Yuan, John Keeping, Andrzej Pietrasiewicz, Dan Carpenter,
Yuta Hayama, Jozef Martiniak, Jens Axboe, Alan Stern,
Sandeep Dhavale, Dave Chinner, Johannes Weiner, ZhangPeng,
Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg, Erez Zadok,
Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On Thu, 2023-06-22 at 09:46 +0900, Damien Le Moal wrote:
> On 6/21/23 23:45, Jeff Layton wrote:
> > struct timespec64 has unused bits in the tv_nsec field that can be used
> > for other purposes. In future patches, we're going to change how the
> > inode->i_ctime is accessed in certain inodes in order to make use of
> > them. In order to do that safely though, we'll need to eradicate raw
> > accesses of the inode->i_ctime field from the kernel.
> >
> > Add new accessor functions for the ctime that we can use to replace them.
> >
> > Signed-off-by: Jeff Layton <jlayton@kernel.org>
>
> [...]
>
> > +/**
> > + * inode_ctime_peek - fetch the current ctime from the inode
> > + * @inode: inode from which to fetch ctime
> > + *
> > + * Grab the current ctime from the inode and return it.
> > + */
> > +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
>
> To be consistent with inode_ctime_set(), why not call this one inode_ctime_get()
In later patches fetching the ctime for presentation may have side
effects on certain filesystems. Using "peek" here is a hint that we want
to avoid those side effects in these calls.
> ? Also, inode_set_ctime() & inode_get_ctime() may be a little more natural. But
> no strong opinion about that though.
>
I like the consistency of the inode_ctime_* prefix. It makes it simpler
to find these calls when grepping, etc.
That said, my opinions on naming are pretty loosely-held, so if the
consensus is that the names should as you suggest, I'll go along with
it.
--
Jeff Layton <jlayton@kernel.org>
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [Ocfs2-devel] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-22 10:14 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton via Ocfs2-devel @ 2023-06-22 10:14 UTC (permalink / raw)
To: Damien Le Moal, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Tom Talpey, Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Naohiro Aota, Johannes Thumshirn,
Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko,
Martin KaFai Lau, Song Liu, Yonghong Song, John Fastabend,
KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa, Hugh Dickins,
Andrew Morton, David S. Miller, Eric Dumazet, Jakub Kicinski,
Paolo Abeni, John Johansen, Paul Moore, James Morris,
Serge E. Hallyn, Stephen Smalley, Eric Paris, Juergen Gross,
Ruihan Li, Laurent Pinchart, Wolfram Sang, Udipto Goswami,
Linyu Yuan, John Keeping, Andrzej Pietrasiewicz, Dan Carpenter,
Yuta Hayama, Jozef Martiniak, Jens Axboe, Alan Stern,
Sandeep Dhavale, Dave Chinner, Johannes Weiner, ZhangPeng,
Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg, Erez Zadok,
Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On Thu, 2023-06-22 at 09:46 +0900, Damien Le Moal wrote:
> On 6/21/23 23:45, Jeff Layton wrote:
> > struct timespec64 has unused bits in the tv_nsec field that can be used
> > for other purposes. In future patches, we're going to change how the
> > inode->i_ctime is accessed in certain inodes in order to make use of
> > them. In order to do that safely though, we'll need to eradicate raw
> > accesses of the inode->i_ctime field from the kernel.
> >
> > Add new accessor functions for the ctime that we can use to replace them.
> >
> > Signed-off-by: Jeff Layton <jlayton@kernel.org>
>
> [...]
>
> > +/**
> > + * inode_ctime_peek - fetch the current ctime from the inode
> > + * @inode: inode from which to fetch ctime
> > + *
> > + * Grab the current ctime from the inode and return it.
> > + */
> > +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
>
> To be consistent with inode_ctime_set(), why not call this one inode_ctime_get()
In later patches fetching the ctime for presentation may have side
effects on certain filesystems. Using "peek" here is a hint that we want
to avoid those side effects in these calls.
> ? Also, inode_set_ctime() & inode_get_ctime() may be a little more natural. But
> no strong opinion about that though.
>
I like the consistency of the inode_ctime_* prefix. It makes it simpler
to find these calls when grepping, etc.
That said, my opinions on naming are pretty loosely-held, so if the
consensus is that the names should as you suggest, I'll go along with
it.
--
Jeff Layton <jlayton@kernel.org>
_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* [Cluster-devel] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-22 10:14 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-22 10:14 UTC (permalink / raw)
To: cluster-devel.redhat.com
On Thu, 2023-06-22 at 09:46 +0900, Damien Le Moal wrote:
> On 6/21/23 23:45, Jeff Layton wrote:
> > struct timespec64 has unused bits in the tv_nsec field that can be used
> > for other purposes. In future patches, we're going to change how the
> > inode->i_ctime is accessed in certain inodes in order to make use of
> > them. In order to do that safely though, we'll need to eradicate raw
> > accesses of the inode->i_ctime field from the kernel.
> >
> > Add new accessor functions for the ctime that we can use to replace them.
> >
> > Signed-off-by: Jeff Layton <jlayton@kernel.org>
>
> [...]
>
> > +/**
> > + * inode_ctime_peek - fetch the current ctime from the inode
> > + * @inode: inode from which to fetch ctime
> > + *
> > + * Grab the current ctime from the inode and return it.
> > + */
> > +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
>
> To be consistent with inode_ctime_set(), why not call this one inode_ctime_get()
In later patches fetching the ctime for presentation may have side
effects on certain filesystems. Using "peek" here is a hint that we want
to avoid those side effects in these calls.
> ? Also, inode_set_ctime() & inode_get_ctime() may be a little more natural. But
> no strong opinion about that though.
>
I like the consistency of the inode_ctime_* prefix. It makes it simpler
to find these calls when grepping, etc.
That said, my opinions on naming are pretty loosely-held, so if the
consensus is that the names should as you suggest, I'll go along with
it.
--
Jeff Layton <jlayton@kernel.org>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
2023-06-22 0:46 ` [f2fs-dev] " Damien Le Moal
` (4 preceding siblings ...)
(?)
@ 2023-06-22 10:14 ` Jeff Layton
-1 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-22 10:14 UTC (permalink / raw)
To: Damien Le Moal, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan
On Thu, 2023-06-22 at 09:46 +0900, Damien Le Moal wrote:
> On 6/21/23 23:45, Jeff Layton wrote:
> > struct timespec64 has unused bits in the tv_nsec field that can be used
> > for other purposes. In future patches, we're going to change how the
> > inode->i_ctime is accessed in certain inodes in order to make use of
> > them. In order to do that safely though, we'll need to eradicate raw
> > accesses of the inode->i_ctime field from the kernel.
> >
> > Add new accessor functions for the ctime that we can use to replace them.
> >
> > Signed-off-by: Jeff Layton <jlayton@kernel.org>
>
> [...]
>
> > +/**
> > + * inode_ctime_peek - fetch the current ctime from the inode
> > + * @inode: inode from which to fetch ctime
> > + *
> > + * Grab the current ctime from the inode and return it.
> > + */
> > +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
>
> To be consistent with inode_ctime_set(), why not call this one inode_ctime_get()
In later patches fetching the ctime for presentation may have side
effects on certain filesystems. Using "peek" here is a hint that we want
to avoid those side effects in these calls.
> ? Also, inode_set_ctime() & inode_get_ctime() may be a little more natural. But
> no strong opinion about that though.
>
I like the consistency of the inode_ctime_* prefix. It makes it simpler
to find these calls when grepping, etc.
That said, my opinions on naming are pretty loosely-held, so if the
consensus is that the names should as you suggest, I'll go along with
it.
--
Jeff Layton <jlayton@kernel.org>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
[not found] ` <99b3c749-23d9-6f09-fb75-6a84f3d1b066-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
@ 2023-06-22 10:14 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-22 10:14 UTC (permalink / raw)
To: Damien Le Moal, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan
On Thu, 2023-06-22 at 09:46 +0900, Damien Le Moal wrote:
> On 6/21/23 23:45, Jeff Layton wrote:
> > struct timespec64 has unused bits in the tv_nsec field that can be used
> > for other purposes. In future patches, we're going to change how the
> > inode->i_ctime is accessed in certain inodes in order to make use of
> > them. In order to do that safely though, we'll need to eradicate raw
> > accesses of the inode->i_ctime field from the kernel.
> >
> > Add new accessor functions for the ctime that we can use to replace them.
> >
> > Signed-off-by: Jeff Layton <jlayton-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
>
> [...]
>
> > +/**
> > + * inode_ctime_peek - fetch the current ctime from the inode
> > + * @inode: inode from which to fetch ctime
> > + *
> > + * Grab the current ctime from the inode and return it.
> > + */
> > +static inline struct timespec64 inode_ctime_peek(const struct inode *inode)
>
> To be consistent with inode_ctime_set(), why not call this one inode_ctime_get()
In later patches fetching the ctime for presentation may have side
effects on certain filesystems. Using "peek" here is a hint that we want
to avoid those side effects in these calls.
> ? Also, inode_set_ctime() & inode_get_ctime() may be a little more natural. But
> no strong opinion about that though.
>
I like the consistency of the inode_ctime_* prefix. It makes it simpler
to find these calls when grepping, etc.
That said, my opinions on naming are pretty loosely-held, so if the
consensus is that the names should as you suggest, I'll go along with
it.
--
Jeff Layton <jlayton-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 10/79] affs: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 10/79] affs: " Jeff Layton
2023-06-21 16:39 ` Jan Kara
@ 2023-06-22 11:37 ` David Sterba
1 sibling, 0 replies; 229+ messages in thread
From: David Sterba @ 2023-06-22 11:37 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, David Sterba, Al Viro, Jan Kara,
linux-fsdevel, linux-kernel
On Wed, Jun 21, 2023 at 10:45:23AM -0400, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Acked-by: David Sterba <dsterba@suse.com>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 16/79] btrfs: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 16/79] btrfs: " Jeff Layton
@ 2023-06-22 11:42 ` David Sterba
0 siblings, 0 replies; 229+ messages in thread
From: David Sterba @ 2023-06-22 11:42 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Chris Mason, Josef Bacik, David Sterba,
Al Viro, Jan Kara, linux-btrfs, linux-kernel
On Wed, Jun 21, 2023 at 10:45:29AM -0400, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
> ---
> fs/btrfs/delayed-inode.c | 10 +++---
> fs/btrfs/file.c | 24 +++++----------
> fs/btrfs/inode.c | 66 ++++++++++++++++------------------------
> fs/btrfs/ioctl.c | 2 +-
> fs/btrfs/reflink.c | 7 ++---
> fs/btrfs/transaction.c | 3 +-
> fs/btrfs/tree-log.c | 4 +--
> fs/btrfs/xattr.c | 4 +--
> 8 files changed, 48 insertions(+), 72 deletions(-)
>
> diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c
> index c0a6a1784697..ae493a4dc206 100644
> --- a/fs/btrfs/delayed-inode.c
> +++ b/fs/btrfs/delayed-inode.c
> @@ -1808,9 +1808,9 @@ static void fill_stack_inode_item(struct btrfs_trans_handle *trans,
> inode->i_mtime.tv_nsec);
>
> btrfs_set_stack_timespec_sec(&inode_item->ctime,
> - inode->i_ctime.tv_sec);
> + inode_ctime_peek(inode).tv_sec);
> btrfs_set_stack_timespec_nsec(&inode_item->ctime,
> - inode->i_ctime.tv_nsec);
> + inode_ctime_peek(inode).tv_nsec);
>
> btrfs_set_stack_timespec_sec(&inode_item->otime,
> BTRFS_I(inode)->i_otime.tv_sec);
> @@ -1861,8 +1861,10 @@ int btrfs_fill_inode(struct inode *inode, u32 *rdev)
> inode->i_mtime.tv_sec = btrfs_stack_timespec_sec(&inode_item->mtime);
> inode->i_mtime.tv_nsec = btrfs_stack_timespec_nsec(&inode_item->mtime);
>
> - inode->i_ctime.tv_sec = btrfs_stack_timespec_sec(&inode_item->ctime);
> - inode->i_ctime.tv_nsec = btrfs_stack_timespec_nsec(&inode_item->ctime);
> + inode_ctime_set_sec(inode,
> + btrfs_stack_timespec_sec(&inode_item->ctime));
> + inode_ctime_set_nsec(inode,
> + btrfs_stack_timespec_nsec(&inode_item->ctime));
>
> BTRFS_I(inode)->i_otime.tv_sec =
> btrfs_stack_timespec_sec(&inode_item->otime);
> diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
> index fd03e689a6be..b4082b322b41 100644
> --- a/fs/btrfs/file.c
> +++ b/fs/btrfs/file.c
> @@ -1108,17 +1108,10 @@ void btrfs_check_nocow_unlock(struct btrfs_inode *inode)
>
> static void update_time_for_write(struct inode *inode)
> {
> - struct timespec64 now;
> -
> if (IS_NOCMTIME(inode))
> return;
>
> - now = current_time(inode);
> - if (!timespec64_equal(&inode->i_mtime, &now))
> - inode->i_mtime = now;
> -
> - if (!timespec64_equal(&inode->i_ctime, &now))
> - inode->i_ctime = now;
> + inode->i_mtime = inode_ctime_set_current(inode);
>
> if (IS_I_VERSION(inode))
> inode_inc_iversion(inode);
> @@ -2460,8 +2453,9 @@ int btrfs_replace_file_extents(struct btrfs_inode *inode,
> inode_inc_iversion(&inode->vfs_inode);
>
> if (!extent_info || extent_info->update_times) {
> - inode->vfs_inode.i_mtime = current_time(&inode->vfs_inode);
> - inode->vfs_inode.i_ctime = inode->vfs_inode.i_mtime;
> + struct inode *vinode = &inode->vfs_inode;
Please use vfs_inode for the variable name.
> +
> + vinode->i_mtime = inode_ctime_set_current(vinode);
> }
>
> ret = btrfs_update_inode(trans, root, inode);
> @@ -2703,8 +2697,7 @@ static int btrfs_punch_hole(struct file *file, loff_t offset, loff_t len)
>
> ASSERT(trans != NULL);
> inode_inc_iversion(inode);
> - inode->i_mtime = current_time(inode);
> - inode->i_ctime = inode->i_mtime;
> + inode->i_mtime = inode_ctime_set_current(inode);
> ret = btrfs_update_inode(trans, root, BTRFS_I(inode));
> updated_inode = true;
> btrfs_end_transaction(trans);
> @@ -2721,11 +2714,8 @@ static int btrfs_punch_hole(struct file *file, loff_t offset, loff_t len)
> * for detecting, at fsync time, if the inode isn't yet in the
> * log tree or it's there but not up to date.
> */
> - struct timespec64 now = current_time(inode);
> -
> inode_inc_iversion(inode);
> - inode->i_mtime = now;
> - inode->i_ctime = now;
> + inode->i_mtime = inode_ctime_set_current(inode);
> trans = btrfs_start_transaction(root, 1);
> if (IS_ERR(trans)) {
> ret = PTR_ERR(trans);
> @@ -2796,7 +2786,7 @@ static int btrfs_fallocate_update_isize(struct inode *inode,
> if (IS_ERR(trans))
> return PTR_ERR(trans);
>
> - inode->i_ctime = current_time(inode);
> + inode_ctime_set_current(inode);
> i_size_write(inode, end);
> btrfs_inode_safe_disk_i_size_write(BTRFS_I(inode), 0);
> ret = btrfs_update_inode(trans, root, BTRFS_I(inode));
> diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
> index 601fdc956484..4bbb6c6a7516 100644
> --- a/fs/btrfs/inode.c
> +++ b/fs/btrfs/inode.c
> @@ -3901,8 +3901,10 @@ static int btrfs_read_locked_inode(struct inode *inode,
> inode->i_mtime.tv_sec = btrfs_timespec_sec(leaf, &inode_item->mtime);
> inode->i_mtime.tv_nsec = btrfs_timespec_nsec(leaf, &inode_item->mtime);
>
> - inode->i_ctime.tv_sec = btrfs_timespec_sec(leaf, &inode_item->ctime);
> - inode->i_ctime.tv_nsec = btrfs_timespec_nsec(leaf, &inode_item->ctime);
> + inode_ctime_set_sec(inode,
> + btrfs_timespec_sec(leaf, &inode_item->ctime));
> + inode_ctime_set_nsec(inode,
> + btrfs_timespec_nsec(leaf, &inode_item->ctime));
>
> BTRFS_I(inode)->i_otime.tv_sec =
> btrfs_timespec_sec(leaf, &inode_item->otime);
> @@ -4073,9 +4075,9 @@ static void fill_inode_item(struct btrfs_trans_handle *trans,
> inode->i_mtime.tv_nsec);
>
> btrfs_set_token_timespec_sec(&token, &item->ctime,
> - inode->i_ctime.tv_sec);
> + inode_ctime_peek(inode).tv_sec);
> btrfs_set_token_timespec_nsec(&token, &item->ctime,
> - inode->i_ctime.tv_nsec);
> + inode_ctime_peek(inode).tv_nsec);
>
> btrfs_set_token_timespec_sec(&token, &item->otime,
> BTRFS_I(inode)->i_otime.tv_sec);
> @@ -4273,9 +4275,8 @@ static int __btrfs_unlink_inode(struct btrfs_trans_handle *trans,
> btrfs_i_size_write(dir, dir->vfs_inode.i_size - name->len * 2);
> inode_inc_iversion(&inode->vfs_inode);
> inode_inc_iversion(&dir->vfs_inode);
> - inode->vfs_inode.i_ctime = current_time(&inode->vfs_inode);
> - dir->vfs_inode.i_mtime = inode->vfs_inode.i_ctime;
> - dir->vfs_inode.i_ctime = inode->vfs_inode.i_ctime;
> + inode_ctime_set_current(&inode->vfs_inode);
> + dir->vfs_inode.i_mtime = inode_ctime_set_current(&dir->vfs_inode);
> ret = btrfs_update_inode(trans, root, dir);
> out:
> return ret;
> @@ -4448,8 +4449,7 @@ static int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
>
> btrfs_i_size_write(dir, dir->vfs_inode.i_size - fname.disk_name.len * 2);
> inode_inc_iversion(&dir->vfs_inode);
> - dir->vfs_inode.i_mtime = current_time(&dir->vfs_inode);
> - dir->vfs_inode.i_ctime = dir->vfs_inode.i_mtime;
> + dir->vfs_inode.i_mtime = inode_ctime_set_current(&dir->vfs_inode);
> ret = btrfs_update_inode_fallback(trans, root, dir);
> if (ret)
> btrfs_abort_transaction(trans, ret);
> @@ -5090,10 +5090,8 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr)
> */
> if (newsize != oldsize) {
> inode_inc_iversion(inode);
> - if (!(mask & (ATTR_CTIME | ATTR_MTIME))) {
> - inode->i_mtime = current_time(inode);
> - inode->i_ctime = inode->i_mtime;
> - }
> + if (!(mask & (ATTR_CTIME | ATTR_MTIME)))
> + inode->i_mtime = inode_ctime_set_current(inode);
> }
>
> if (newsize > oldsize) {
> @@ -5736,9 +5734,7 @@ static struct inode *new_simple_dir(struct super_block *s,
> inode->i_opflags &= ~IOP_XATTR;
> inode->i_fop = &simple_dir_operations;
> inode->i_mode = S_IFDIR | S_IRUGO | S_IWUSR | S_IXUGO;
> - inode->i_mtime = current_time(inode);
> - inode->i_atime = inode->i_mtime;
> - inode->i_ctime = inode->i_mtime;
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
Please don't use chained assignments in btrfs code.
inode->i_mtime = inode_ctime_set_current(inode);
inode->i_atime = inode->i_mtime;
> BTRFS_I(inode)->i_otime = inode->i_mtime;
>
> return inode;
> @@ -6075,7 +6071,7 @@ static int btrfs_update_time(struct inode *inode, struct timespec64 *now,
> if (flags & S_VERSION)
> dirty |= inode_maybe_inc_iversion(inode, dirty);
> if (flags & S_CTIME)
> - inode->i_ctime = *now;
> + inode_ctime_set(inode, *now);
> if (flags & S_MTIME)
> inode->i_mtime = *now;
> if (flags & S_ATIME)
> @@ -6378,9 +6374,7 @@ int btrfs_create_new_inode(struct btrfs_trans_handle *trans,
> goto discard;
> }
>
> - inode->i_mtime = current_time(inode);
> - inode->i_atime = inode->i_mtime;
> - inode->i_ctime = inode->i_mtime;
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
inode->i_mtime = inode_ctime_set_current(inode);
inode->i_atime = inode->i_mtime;
> BTRFS_I(inode)->i_otime = inode->i_mtime;
>
> /*
> @@ -6545,12 +6539,9 @@ int btrfs_add_link(struct btrfs_trans_handle *trans,
> * log replay procedure is responsible for setting them to their correct
> * values (the ones it had when the fsync was done).
> */
> - if (!test_bit(BTRFS_FS_LOG_RECOVERING, &root->fs_info->flags)) {
> - struct timespec64 now = current_time(&parent_inode->vfs_inode);
> + if (!test_bit(BTRFS_FS_LOG_RECOVERING, &root->fs_info->flags))
> + parent_inode->vfs_inode.i_mtime = inode_ctime_set_current(&parent_inode->vfs_inode);
>
> - parent_inode->vfs_inode.i_mtime = now;
> - parent_inode->vfs_inode.i_ctime = now;
> - }
> ret = btrfs_update_inode(trans, root, parent_inode);
> if (ret)
> btrfs_abort_transaction(trans, ret);
> @@ -6690,7 +6681,7 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
> BTRFS_I(inode)->dir_index = 0ULL;
> inc_nlink(inode);
> inode_inc_iversion(inode);
> - inode->i_ctime = current_time(inode);
> + inode_ctime_set_current(inode);
> ihold(inode);
> set_bit(BTRFS_INODE_COPY_EVERYTHING, &BTRFS_I(inode)->runtime_flags);
>
> @@ -8778,7 +8769,6 @@ static int btrfs_rename_exchange(struct inode *old_dir,
> struct btrfs_root *dest = BTRFS_I(new_dir)->root;
> struct inode *new_inode = new_dentry->d_inode;
> struct inode *old_inode = old_dentry->d_inode;
> - struct timespec64 ctime = current_time(old_inode);
> struct btrfs_rename_ctx old_rename_ctx;
> struct btrfs_rename_ctx new_rename_ctx;
> u64 old_ino = btrfs_ino(BTRFS_I(old_inode));
> @@ -8909,12 +8899,10 @@ static int btrfs_rename_exchange(struct inode *old_dir,
> inode_inc_iversion(new_dir);
> inode_inc_iversion(old_inode);
> inode_inc_iversion(new_inode);
> - old_dir->i_mtime = ctime;
> - old_dir->i_ctime = ctime;
> - new_dir->i_mtime = ctime;
> - new_dir->i_ctime = ctime;
> - old_inode->i_ctime = ctime;
> - new_inode->i_ctime = ctime;
> + old_dir->i_mtime = inode_ctime_set_current(old_dir);
> + new_dir->i_mtime = inode_ctime_set_current(new_dir);
> + inode_ctime_set_current(old_inode);
> + inode_ctime_set_current(new_inode);
>
> if (old_dentry->d_parent != new_dentry->d_parent) {
> btrfs_record_unlink_dir(trans, BTRFS_I(old_dir),
> @@ -9178,11 +9166,9 @@ static int btrfs_rename(struct mnt_idmap *idmap,
> inode_inc_iversion(old_dir);
> inode_inc_iversion(new_dir);
> inode_inc_iversion(old_inode);
> - old_dir->i_mtime = current_time(old_dir);
> - old_dir->i_ctime = old_dir->i_mtime;
> - new_dir->i_mtime = old_dir->i_mtime;
> - new_dir->i_ctime = old_dir->i_mtime;
> - old_inode->i_ctime = old_dir->i_mtime;
> + old_dir->i_mtime = inode_ctime_set_current(old_dir);
> + new_dir->i_mtime = inode_ctime_set_current(new_dir);
> + inode_ctime_set_current(old_inode);
>
> if (old_dentry->d_parent != new_dentry->d_parent)
> btrfs_record_unlink_dir(trans, BTRFS_I(old_dir),
> @@ -9204,7 +9190,7 @@ static int btrfs_rename(struct mnt_idmap *idmap,
>
> if (new_inode) {
> inode_inc_iversion(new_inode);
> - new_inode->i_ctime = current_time(new_inode);
> + inode_ctime_set_current(new_inode);
> if (unlikely(btrfs_ino(BTRFS_I(new_inode)) ==
> BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)) {
> ret = btrfs_unlink_subvol(trans, BTRFS_I(new_dir), new_dentry);
> @@ -9744,7 +9730,7 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode,
> *alloc_hint = ins.objectid + ins.offset;
>
> inode_inc_iversion(inode);
> - inode->i_ctime = current_time(inode);
> + inode_ctime_set_current(inode);
> BTRFS_I(inode)->flags |= BTRFS_INODE_PREALLOC;
> if (!(mode & FALLOC_FL_KEEP_SIZE) &&
> (actual_len > inode->i_size) &&
> diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
> index a895d105464b..3d50bd67ec85 100644
> --- a/fs/btrfs/ioctl.c
> +++ b/fs/btrfs/ioctl.c
> @@ -384,7 +384,7 @@ int btrfs_fileattr_set(struct mnt_idmap *idmap,
> binode->flags = binode_flags;
> btrfs_sync_inode_flags_to_i_flags(inode);
> inode_inc_iversion(inode);
> - inode->i_ctime = current_time(inode);
> + inode_ctime_set_current(inode);
> ret = btrfs_update_inode(trans, root, BTRFS_I(inode));
>
> out_end_trans:
> diff --git a/fs/btrfs/reflink.c b/fs/btrfs/reflink.c
> index 0474bbe39da7..3cf4716dea46 100644
> --- a/fs/btrfs/reflink.c
> +++ b/fs/btrfs/reflink.c
> @@ -29,10 +29,9 @@ static int clone_finish_inode_update(struct btrfs_trans_handle *trans,
> int ret;
>
> inode_inc_iversion(inode);
> - if (!no_time_update) {
> - inode->i_mtime = current_time(inode);
> - inode->i_ctime = inode->i_mtime;
> - }
> + if (!no_time_update)
> + inode->i_mtime = inode_ctime_set_current(inode);
> +
> /*
> * We round up to the block size at eof when determining which
> * extents to clone above, but shouldn't round up the file size.
> diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
> index e7cfc992e02a..138f919646e2 100644
> --- a/fs/btrfs/transaction.c
> +++ b/fs/btrfs/transaction.c
> @@ -1831,8 +1831,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
>
> btrfs_i_size_write(BTRFS_I(parent_inode), parent_inode->i_size +
> fname.disk_name.len * 2);
> - parent_inode->i_mtime = current_time(parent_inode);
> - parent_inode->i_ctime = parent_inode->i_mtime;
> + parent_inode->i_mtime = inode_ctime_set_current(parent_inode);
> ret = btrfs_update_inode_fallback(trans, parent_root, BTRFS_I(parent_inode));
> if (ret) {
> btrfs_abort_transaction(trans, ret);
> diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
> index 365a1cc0a3c3..c862d1450365 100644
> --- a/fs/btrfs/tree-log.c
> +++ b/fs/btrfs/tree-log.c
> @@ -4148,9 +4148,9 @@ static void fill_inode_item(struct btrfs_trans_handle *trans,
> inode->i_mtime.tv_nsec);
>
> btrfs_set_token_timespec_sec(&token, &item->ctime,
> - inode->i_ctime.tv_sec);
> + inode_ctime_peek(inode).tv_sec);
> btrfs_set_token_timespec_nsec(&token, &item->ctime,
> - inode->i_ctime.tv_nsec);
> + inode_ctime_peek(inode).tv_nsec);
>
> /*
> * We do not need to set the nbytes field, in fact during a fast fsync
> diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c
> index fc4b20c2688a..a2d331436963 100644
> --- a/fs/btrfs/xattr.c
> +++ b/fs/btrfs/xattr.c
> @@ -264,7 +264,7 @@ int btrfs_setxattr_trans(struct inode *inode, const char *name,
> goto out;
>
> inode_inc_iversion(inode);
> - inode->i_ctime = current_time(inode);
> + inode_ctime_set_current(inode);
> ret = btrfs_update_inode(trans, root, BTRFS_I(inode));
> if (ret)
> btrfs_abort_transaction(trans, ret);
> @@ -407,7 +407,7 @@ static int btrfs_xattr_handler_set_prop(const struct xattr_handler *handler,
> ret = btrfs_set_prop(trans, inode, name, value, size, flags);
> if (!ret) {
> inode_inc_iversion(inode);
> - inode->i_ctime = current_time(inode);
> + inode_ctime_set_current(inode);
> ret = btrfs_update_inode(trans, root, BTRFS_I(inode));
> if (ret)
> btrfs_abort_transaction(trans, ret);
> --
> 2.41.0
>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 15/79] bfs: switch to new ctime accessors
2023-06-21 16:57 ` Jeff Layton
@ 2023-06-22 12:30 ` Jan Kara
2023-06-22 12:51 ` Jeff Layton
0 siblings, 1 reply; 229+ messages in thread
From: Jan Kara @ 2023-06-22 12:30 UTC (permalink / raw)
To: Jeff Layton
Cc: Jan Kara, Christian Brauner, Tigran A. Aivazian, Al Viro, linux-kernel
[-- Attachment #1: Type: text/plain, Size: 2294 bytes --]
On Wed 21-06-23 12:57:19, Jeff Layton wrote:
> On Wed, 2023-06-21 at 18:48 +0200, Jan Kara wrote:
> > On Wed 21-06-23 10:45:28, Jeff Layton wrote:
> > > In later patches, we're going to change how the ctime.tv_nsec field is
> > > utilized. Switch to using accessor functions instead of raw accesses of
> > > inode->i_ctime.
> > >
> > > Signed-off-by: Jeff Layton <jlayton@kernel.org>
> >
> > ...
> >
> > > diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c
> > > index 1926bec2c850..c964316be32b 100644
> > > --- a/fs/bfs/inode.c
> > > +++ b/fs/bfs/inode.c
> > > @@ -82,10 +82,10 @@ struct inode *bfs_iget(struct super_block *sb, unsigned long ino)
> > > inode->i_blocks = BFS_FILEBLOCKS(di);
> > > inode->i_atime.tv_sec = le32_to_cpu(di->i_atime);
> > > inode->i_mtime.tv_sec = le32_to_cpu(di->i_mtime);
> > > - inode->i_ctime.tv_sec = le32_to_cpu(di->i_ctime);
> > > + inode_ctime_set_sec(inode, le32_to_cpu(di->i_ctime));
> > > inode->i_atime.tv_nsec = 0;
> > > inode->i_mtime.tv_nsec = 0;
> > > - inode->i_ctime.tv_nsec = 0;
> > > + inode_ctime_set_nsec(inode, 0);
> >
> > So I'm somewhat wondering here - in other filesystem you construct
> > timespec64 and then use inode_ctime_set(). Here you use
> > inode_ctime_set_sec() + inode_ctime_set_nsec(). What's the benefit? It
> > seems these two functions are not used that much some maybe we could just
> > live with just inode_ctime_set() and constructing timespec64 when needed?
> >
> > Honza
>
> The main advantage is that by using that, I didn't need to do quite so
> much of this conversion by hand. My coccinelle skills are pretty
> primitive. I went with whatever conversion was going to give minimal
> changes, to the existing accesses for the most part.
>
> We could certainly do it the way you suggest, it just means having to
> re-touch a lot of this code by hand, or someone with better coccinelle
> chops suggesting a way to declare a temporary variables in place.
Well, maybe temporary variables aren't that convenient but we could provide
function setting ctime from sec & nsec value without having to declare
temporary timespec64? Attached is a semantic patch that should deal with
that - at least it seems to handle all the cases I've found.
Honza
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
[-- Attachment #2: ctime.cocci --]
[-- Type: text/plain, Size: 450 bytes --]
@@
struct inode *inode;
expression val, E1, E2;
@@
- inode->i_ctime.tv_sec = val
+ inode_set_ctime(inode, val, 0)
...
(
- inode->i_ctime.tv_nsec = 0;
|
- E1 = inode->i_ctime.tv_nsec = 0
+ E1 = 0
|
- inode->i_ctime.tv_nsec = E1 = 0
+ E1 = 0
|
- inode->i_ctime.tv_nsec = E1 = E2 = 0
+ E1 = E2 = 0
)
@@
struct inode *inode;
expression val, val2;
@@
- inode->i_ctime.tv_sec = val
+ inode_set_ctime(inode, val, val2)
...
- inode->i_ctime.tv_nsec = val2;
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 15/79] bfs: switch to new ctime accessors
2023-06-22 12:30 ` Jan Kara
@ 2023-06-22 12:51 ` Jeff Layton
2023-06-22 14:57 ` Jan Kara
0 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-22 12:51 UTC (permalink / raw)
To: Jan Kara
Cc: Christian Brauner, Tigran A. Aivazian, Al Viro, linux-kernel,
damien.lemoal
On Thu, 2023-06-22 at 14:30 +0200, Jan Kara wrote:
> On Wed 21-06-23 12:57:19, Jeff Layton wrote:
> > On Wed, 2023-06-21 at 18:48 +0200, Jan Kara wrote:
> > > On Wed 21-06-23 10:45:28, Jeff Layton wrote:
> > > > In later patches, we're going to change how the ctime.tv_nsec field is
> > > > utilized. Switch to using accessor functions instead of raw accesses of
> > > > inode->i_ctime.
> > > >
> > > > Signed-off-by: Jeff Layton <jlayton@kernel.org>
> > >
> > > ...
> > >
> > > > diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c
> > > > index 1926bec2c850..c964316be32b 100644
> > > > --- a/fs/bfs/inode.c
> > > > +++ b/fs/bfs/inode.c
> > > > @@ -82,10 +82,10 @@ struct inode *bfs_iget(struct super_block *sb, unsigned long ino)
> > > > inode->i_blocks = BFS_FILEBLOCKS(di);
> > > > inode->i_atime.tv_sec = le32_to_cpu(di->i_atime);
> > > > inode->i_mtime.tv_sec = le32_to_cpu(di->i_mtime);
> > > > - inode->i_ctime.tv_sec = le32_to_cpu(di->i_ctime);
> > > > + inode_ctime_set_sec(inode, le32_to_cpu(di->i_ctime));
> > > > inode->i_atime.tv_nsec = 0;
> > > > inode->i_mtime.tv_nsec = 0;
> > > > - inode->i_ctime.tv_nsec = 0;
> > > > + inode_ctime_set_nsec(inode, 0);
> > >
> > > So I'm somewhat wondering here - in other filesystem you construct
> > > timespec64 and then use inode_ctime_set(). Here you use
> > > inode_ctime_set_sec() + inode_ctime_set_nsec(). What's the benefit? It
> > > seems these two functions are not used that much some maybe we could just
> > > live with just inode_ctime_set() and constructing timespec64 when needed?
> > >
> > > Honza
> >
> > The main advantage is that by using that, I didn't need to do quite so
> > much of this conversion by hand. My coccinelle skills are pretty
> > primitive. I went with whatever conversion was going to give minimal
> > changes, to the existing accesses for the most part.
> >
> > We could certainly do it the way you suggest, it just means having to
> > re-touch a lot of this code by hand, or someone with better coccinelle
> > chops suggesting a way to declare a temporary variables in place.
>
> Well, maybe temporary variables aren't that convenient but we could provide
> function setting ctime from sec & nsec value without having to declare
> temporary timespec64? Attached is a semantic patch that should deal with
> that - at least it seems to handle all the cases I've found.
>
Ok, let me try respinning this with your cocci script and see how it
looks.
Damien also suggested in a reply to the zonefs patch a preference for
the naming style you have above. Should I also rename these like?
inode_ctime_peek -> inode_get_ctime
inode_ctime_set -> inode_set_ctime
This would be the time to change it if that's preferred.
--
Jeff Layton <jlayton@kernel.org>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 15/79] bfs: switch to new ctime accessors
2023-06-22 12:51 ` Jeff Layton
@ 2023-06-22 14:57 ` Jan Kara
2023-06-23 12:33 ` Christian Brauner
0 siblings, 1 reply; 229+ messages in thread
From: Jan Kara @ 2023-06-22 14:57 UTC (permalink / raw)
To: Jeff Layton
Cc: Jan Kara, Christian Brauner, Tigran A. Aivazian, Al Viro,
linux-kernel, damien.lemoal
On Thu 22-06-23 08:51:58, Jeff Layton wrote:
> On Thu, 2023-06-22 at 14:30 +0200, Jan Kara wrote:
> > On Wed 21-06-23 12:57:19, Jeff Layton wrote:
> > > On Wed, 2023-06-21 at 18:48 +0200, Jan Kara wrote:
> > > > On Wed 21-06-23 10:45:28, Jeff Layton wrote:
> > > > > In later patches, we're going to change how the ctime.tv_nsec field is
> > > > > utilized. Switch to using accessor functions instead of raw accesses of
> > > > > inode->i_ctime.
> > > > >
> > > > > Signed-off-by: Jeff Layton <jlayton@kernel.org>
> > > >
> > > > ...
> > > >
> > > > > diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c
> > > > > index 1926bec2c850..c964316be32b 100644
> > > > > --- a/fs/bfs/inode.c
> > > > > +++ b/fs/bfs/inode.c
> > > > > @@ -82,10 +82,10 @@ struct inode *bfs_iget(struct super_block *sb, unsigned long ino)
> > > > > inode->i_blocks = BFS_FILEBLOCKS(di);
> > > > > inode->i_atime.tv_sec = le32_to_cpu(di->i_atime);
> > > > > inode->i_mtime.tv_sec = le32_to_cpu(di->i_mtime);
> > > > > - inode->i_ctime.tv_sec = le32_to_cpu(di->i_ctime);
> > > > > + inode_ctime_set_sec(inode, le32_to_cpu(di->i_ctime));
> > > > > inode->i_atime.tv_nsec = 0;
> > > > > inode->i_mtime.tv_nsec = 0;
> > > > > - inode->i_ctime.tv_nsec = 0;
> > > > > + inode_ctime_set_nsec(inode, 0);
> > > >
> > > > So I'm somewhat wondering here - in other filesystem you construct
> > > > timespec64 and then use inode_ctime_set(). Here you use
> > > > inode_ctime_set_sec() + inode_ctime_set_nsec(). What's the benefit? It
> > > > seems these two functions are not used that much some maybe we could just
> > > > live with just inode_ctime_set() and constructing timespec64 when needed?
> > > >
> > > > Honza
> > >
> > > The main advantage is that by using that, I didn't need to do quite so
> > > much of this conversion by hand. My coccinelle skills are pretty
> > > primitive. I went with whatever conversion was going to give minimal
> > > changes, to the existing accesses for the most part.
> > >
> > > We could certainly do it the way you suggest, it just means having to
> > > re-touch a lot of this code by hand, or someone with better coccinelle
> > > chops suggesting a way to declare a temporary variables in place.
> >
> > Well, maybe temporary variables aren't that convenient but we could provide
> > function setting ctime from sec & nsec value without having to declare
> > temporary timespec64? Attached is a semantic patch that should deal with
> > that - at least it seems to handle all the cases I've found.
> >
>
> Ok, let me try respinning this with your cocci script and see how it
> looks.
>
> Damien also suggested in a reply to the zonefs patch a preference for
> the naming style you have above. Should I also rename these like?
>
> inode_ctime_peek -> inode_get_ctime
> inode_ctime_set -> inode_set_ctime
>
> This would be the time to change it if that's preferred.
I don't really care much so whatever you decide is better :)
Honza
--
Jan Kara <jack@suse.com>
SUSE Labs, CR
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 03/79] s390: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 03/79] s390: " Jeff Layton
2023-06-21 16:34 ` Jan Kara
@ 2023-06-22 17:35 ` Alexander Gordeev
2023-06-22 17:51 ` Jeff Layton
1 sibling, 1 reply; 229+ messages in thread
From: Alexander Gordeev @ 2023-06-22 17:35 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Heiko Carstens, Vasily Gorbik,
Christian Borntraeger, Sven Schnelle, Al Viro, Jan Kara,
linux-s390, linux-kernel
On Wed, Jun 21, 2023 at 10:45:16AM -0400, Jeff Layton wrote:
Hi Jeff,
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
> ---
> arch/s390/hypfs/inode.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c
> index ee919bfc8186..30fa336ec63e 100644
> --- a/arch/s390/hypfs/inode.c
> +++ b/arch/s390/hypfs/inode.c
> @@ -53,7 +53,7 @@ static void hypfs_update_update(struct super_block *sb)
> struct inode *inode = d_inode(sb_info->update_file);
>
> sb_info->last_update = ktime_get_seconds();
> - inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> }
>
> /* directory tree removal functions */
> @@ -101,7 +101,7 @@ static struct inode *hypfs_make_inode(struct super_block *sb, umode_t mode)
> ret->i_mode = mode;
> ret->i_uid = hypfs_info->uid;
> ret->i_gid = hypfs_info->gid;
> - ret->i_atime = ret->i_mtime = ret->i_ctime = current_time(ret);
> + ret->i_atime = ret->i_mtime = inode_ctime_set_current(ret);
> if (S_ISDIR(mode))
> set_nlink(ret, 2);
> }
I guess, inode_set_ctime() called from inode_ctime_set_current()
updates i_ctime and is part of some other series?
Thanks!
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 03/79] s390: switch to new ctime accessors
2023-06-22 17:35 ` Alexander Gordeev
@ 2023-06-22 17:51 ` Jeff Layton
2023-06-22 18:22 ` Alexander Gordeev
0 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-06-22 17:51 UTC (permalink / raw)
To: Alexander Gordeev
Cc: Christian Brauner, Heiko Carstens, Vasily Gorbik,
Christian Borntraeger, Sven Schnelle, Al Viro, Jan Kara,
linux-s390, linux-kernel
On Thu, 2023-06-22 at 19:35 +0200, Alexander Gordeev wrote:
> On Wed, Jun 21, 2023 at 10:45:16AM -0400, Jeff Layton wrote:
>
> Hi Jeff,
> > In later patches, we're going to change how the ctime.tv_nsec field is
> > utilized. Switch to using accessor functions instead of raw accesses of
> > inode->i_ctime.
> >
> > Signed-off-by: Jeff Layton <jlayton@kernel.org>
> > ---
> > arch/s390/hypfs/inode.c | 4 ++--
> > 1 file changed, 2 insertions(+), 2 deletions(-)
> >
> > diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c
> > index ee919bfc8186..30fa336ec63e 100644
> > --- a/arch/s390/hypfs/inode.c
> > +++ b/arch/s390/hypfs/inode.c
> > @@ -53,7 +53,7 @@ static void hypfs_update_update(struct super_block *sb)
> > struct inode *inode = d_inode(sb_info->update_file);
> >
> > sb_info->last_update = ktime_get_seconds();
> > - inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
> > + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> > }
> >
> > /* directory tree removal functions */
> > @@ -101,7 +101,7 @@ static struct inode *hypfs_make_inode(struct super_block *sb, umode_t mode)
> > ret->i_mode = mode;
> > ret->i_uid = hypfs_info->uid;
> > ret->i_gid = hypfs_info->gid;
> > - ret->i_atime = ret->i_mtime = ret->i_ctime = current_time(ret);
> > + ret->i_atime = ret->i_mtime = inode_ctime_set_current(ret);
> > if (S_ISDIR(mode))
> > set_nlink(ret, 2);
> > }
>
> I guess, inode_set_ctime() called from inode_ctime_set_current()
> updates i_ctime and is part of some other series?
>
No, that gets added in patch #1 of this series.
You should have gotten cc'ed on that one, though the postings to vger
mailing lists of patches 1, 2, and 79 bounced because the mail header
length on those was >8k.
--
Jeff Layton <jlayton@kernel.org>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 03/79] s390: switch to new ctime accessors
2023-06-22 17:51 ` Jeff Layton
@ 2023-06-22 18:22 ` Alexander Gordeev
2023-06-22 18:46 ` Jeff Layton
0 siblings, 1 reply; 229+ messages in thread
From: Alexander Gordeev @ 2023-06-22 18:22 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Heiko Carstens, Vasily Gorbik,
Christian Borntraeger, Sven Schnelle, Al Viro, Jan Kara,
linux-s390, linux-kernel
On Thu, Jun 22, 2023 at 01:51:33PM -0400, Jeff Layton wrote:
> On Thu, 2023-06-22 at 19:35 +0200, Alexander Gordeev wrote:
> > On Wed, Jun 21, 2023 at 10:45:16AM -0400, Jeff Layton wrote:
> >
> > Hi Jeff,
> > > In later patches, we're going to change how the ctime.tv_nsec field is
> > > utilized. Switch to using accessor functions instead of raw accesses of
> > > inode->i_ctime.
> > >
> > > Signed-off-by: Jeff Layton <jlayton@kernel.org>
> > > ---
> > > arch/s390/hypfs/inode.c | 4 ++--
> > > 1 file changed, 2 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c
> > > index ee919bfc8186..30fa336ec63e 100644
> > > --- a/arch/s390/hypfs/inode.c
> > > +++ b/arch/s390/hypfs/inode.c
> > > @@ -53,7 +53,7 @@ static void hypfs_update_update(struct super_block *sb)
> > > struct inode *inode = d_inode(sb_info->update_file);
> > >
> > > sb_info->last_update = ktime_get_seconds();
> > > - inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
> > > + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> > > }
> > >
> > > /* directory tree removal functions */
> > > @@ -101,7 +101,7 @@ static struct inode *hypfs_make_inode(struct super_block *sb, umode_t mode)
> > > ret->i_mode = mode;
> > > ret->i_uid = hypfs_info->uid;
> > > ret->i_gid = hypfs_info->gid;
> > > - ret->i_atime = ret->i_mtime = ret->i_ctime = current_time(ret);
> > > + ret->i_atime = ret->i_mtime = inode_ctime_set_current(ret);
> > > if (S_ISDIR(mode))
> > > set_nlink(ret, 2);
> > > }
> >
> > I guess, inode_set_ctime() called from inode_ctime_set_current()
> > updates i_ctime and is part of some other series?
> >
>
> No, that gets added in patch #1 of this series.
>
> You should have gotten cc'ed on that one, though the postings to vger
> mailing lists of patches 1, 2, and 79 bounced because the mail header
> length on those was >8k.
I actually received #1, if we're talking about this one:
https://lore.kernel.org/all/20230621144507.55591-2-jlayton@kernel.org/
I see inode_set_ctime() gets called, but nowere defined.
> Jeff Layton <jlayton@kernel.org>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 03/79] s390: switch to new ctime accessors
2023-06-22 18:22 ` Alexander Gordeev
@ 2023-06-22 18:46 ` Jeff Layton
0 siblings, 0 replies; 229+ messages in thread
From: Jeff Layton @ 2023-06-22 18:46 UTC (permalink / raw)
To: Alexander Gordeev
Cc: Christian Brauner, Heiko Carstens, Vasily Gorbik,
Christian Borntraeger, Sven Schnelle, Al Viro, Jan Kara,
linux-s390, linux-kernel
On Thu, 2023-06-22 at 20:22 +0200, Alexander Gordeev wrote:
> On Thu, Jun 22, 2023 at 01:51:33PM -0400, Jeff Layton wrote:
> > On Thu, 2023-06-22 at 19:35 +0200, Alexander Gordeev wrote:
> > > On Wed, Jun 21, 2023 at 10:45:16AM -0400, Jeff Layton wrote:
> > >
> > > Hi Jeff,
> > > > In later patches, we're going to change how the ctime.tv_nsec field is
> > > > utilized. Switch to using accessor functions instead of raw accesses of
> > > > inode->i_ctime.
> > > >
> > > > Signed-off-by: Jeff Layton <jlayton@kernel.org>
> > > > ---
> > > > arch/s390/hypfs/inode.c | 4 ++--
> > > > 1 file changed, 2 insertions(+), 2 deletions(-)
> > > >
> > > > diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c
> > > > index ee919bfc8186..30fa336ec63e 100644
> > > > --- a/arch/s390/hypfs/inode.c
> > > > +++ b/arch/s390/hypfs/inode.c
> > > > @@ -53,7 +53,7 @@ static void hypfs_update_update(struct super_block *sb)
> > > > struct inode *inode = d_inode(sb_info->update_file);
> > > >
> > > > sb_info->last_update = ktime_get_seconds();
> > > > - inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
> > > > + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> > > > }
> > > >
> > > > /* directory tree removal functions */
> > > > @@ -101,7 +101,7 @@ static struct inode *hypfs_make_inode(struct super_block *sb, umode_t mode)
> > > > ret->i_mode = mode;
> > > > ret->i_uid = hypfs_info->uid;
> > > > ret->i_gid = hypfs_info->gid;
> > > > - ret->i_atime = ret->i_mtime = ret->i_ctime = current_time(ret);
> > > > + ret->i_atime = ret->i_mtime = inode_ctime_set_current(ret);
> > > > if (S_ISDIR(mode))
> > > > set_nlink(ret, 2);
> > > > }
> > >
> > > I guess, inode_set_ctime() called from inode_ctime_set_current()
> > > updates i_ctime and is part of some other series?
> > >
> >
> > No, that gets added in patch #1 of this series.
> >
> > You should have gotten cc'ed on that one, though the postings to vger
> > mailing lists of patches 1, 2, and 79 bounced because the mail header
> > length on those was >8k.
>
> I actually received #1, if we're talking about this one:
> https://lore.kernel.org/all/20230621144507.55591-2-jlayton@kernel.org/
>
> I see inode_set_ctime() gets called, but nowere defined.
>
>
That's a bug -- that should be calling inode_ctime_set() instead. It
gets fixed up in a later patch in the series, but that should be fixed.
I'm already respinning this now with an updated coccinelle patch. I'll
make sure that's correct on the next posting.
Thanks,
--
Jeff Layton <jlayton@kernel.org>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 62/79] smb: switch to new ctime accessors
2023-06-21 14:46 ` [PATCH 62/79] smb: " Jeff Layton
2023-06-21 17:45 ` Tom Talpey
@ 2023-06-23 5:03 ` Sergey Senozhatsky
1 sibling, 0 replies; 229+ messages in thread
From: Sergey Senozhatsky @ 2023-06-23 5:03 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Steve French, Paulo Alcantara,
Ronnie Sahlberg, Shyam Prasad N, Tom Talpey, Namjae Jeon,
Sergey Senozhatsky, Al Viro, Jan Kara, linux-cifs,
samba-technical, linux-kernel
On (23/06/21 10:46), Jeff Layton wrote:
>
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: Sergey Senozhatsky <senozhatsky@chromium.org>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 51/79] omfs: switch to new ctime accessors
2023-06-21 14:46 ` [PATCH 51/79] omfs: " Jeff Layton
@ 2023-06-23 12:12 ` Bob Copeland
0 siblings, 0 replies; 229+ messages in thread
From: Bob Copeland @ 2023-06-23 12:12 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Al Viro, Jan Kara, linux-karma-devel, linux-kernel
On Wed, Jun 21, 2023 at 10:46:04AM -0400, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Acked-by: Bob Copeland <me@bobcopeland.com>
--
Bob Copeland %% https://bobcopeland.com/
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 15/79] bfs: switch to new ctime accessors
2023-06-22 14:57 ` Jan Kara
@ 2023-06-23 12:33 ` Christian Brauner
2023-07-03 10:12 ` Christian Brauner
0 siblings, 1 reply; 229+ messages in thread
From: Christian Brauner @ 2023-06-23 12:33 UTC (permalink / raw)
To: Jan Kara
Cc: Jeff Layton, Tigran A. Aivazian, Al Viro, linux-kernel, damien.lemoal
On Thu, Jun 22, 2023 at 04:57:47PM +0200, Jan Kara wrote:
> On Thu 22-06-23 08:51:58, Jeff Layton wrote:
> > On Thu, 2023-06-22 at 14:30 +0200, Jan Kara wrote:
> > > On Wed 21-06-23 12:57:19, Jeff Layton wrote:
> > > > On Wed, 2023-06-21 at 18:48 +0200, Jan Kara wrote:
> > > > > On Wed 21-06-23 10:45:28, Jeff Layton wrote:
> > > > > > In later patches, we're going to change how the ctime.tv_nsec field is
> > > > > > utilized. Switch to using accessor functions instead of raw accesses of
> > > > > > inode->i_ctime.
> > > > > >
> > > > > > Signed-off-by: Jeff Layton <jlayton@kernel.org>
> > > > >
> > > > > ...
> > > > >
> > > > > > diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c
> > > > > > index 1926bec2c850..c964316be32b 100644
> > > > > > --- a/fs/bfs/inode.c
> > > > > > +++ b/fs/bfs/inode.c
> > > > > > @@ -82,10 +82,10 @@ struct inode *bfs_iget(struct super_block *sb, unsigned long ino)
> > > > > > inode->i_blocks = BFS_FILEBLOCKS(di);
> > > > > > inode->i_atime.tv_sec = le32_to_cpu(di->i_atime);
> > > > > > inode->i_mtime.tv_sec = le32_to_cpu(di->i_mtime);
> > > > > > - inode->i_ctime.tv_sec = le32_to_cpu(di->i_ctime);
> > > > > > + inode_ctime_set_sec(inode, le32_to_cpu(di->i_ctime));
> > > > > > inode->i_atime.tv_nsec = 0;
> > > > > > inode->i_mtime.tv_nsec = 0;
> > > > > > - inode->i_ctime.tv_nsec = 0;
> > > > > > + inode_ctime_set_nsec(inode, 0);
> > > > >
> > > > > So I'm somewhat wondering here - in other filesystem you construct
> > > > > timespec64 and then use inode_ctime_set(). Here you use
> > > > > inode_ctime_set_sec() + inode_ctime_set_nsec(). What's the benefit? It
> > > > > seems these two functions are not used that much some maybe we could just
> > > > > live with just inode_ctime_set() and constructing timespec64 when needed?
> > > > >
> > > > > Honza
> > > >
> > > > The main advantage is that by using that, I didn't need to do quite so
> > > > much of this conversion by hand. My coccinelle skills are pretty
> > > > primitive. I went with whatever conversion was going to give minimal
> > > > changes, to the existing accesses for the most part.
> > > >
> > > > We could certainly do it the way you suggest, it just means having to
> > > > re-touch a lot of this code by hand, or someone with better coccinelle
> > > > chops suggesting a way to declare a temporary variables in place.
> > >
> > > Well, maybe temporary variables aren't that convenient but we could provide
> > > function setting ctime from sec & nsec value without having to declare
> > > temporary timespec64? Attached is a semantic patch that should deal with
> > > that - at least it seems to handle all the cases I've found.
> > >
> >
> > Ok, let me try respinning this with your cocci script and see how it
> > looks.
> >
> > Damien also suggested in a reply to the zonefs patch a preference for
> > the naming style you have above. Should I also rename these like?
> >
> > inode_ctime_peek -> inode_get_ctime
> > inode_ctime_set -> inode_set_ctime
> >
> > This would be the time to change it if that's preferred.
>
> I don't really care much so whatever you decide is better :)
I have a mild preference for inode_{get,set}_ctime().
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 00/79] fs: new accessors for inode->i_ctime
2023-06-21 19:52 ` [f2fs-dev] " Jeff Layton
` (4 preceding siblings ...)
(?)
@ 2023-06-23 12:41 ` Christian Brauner
-1 siblings, 0 replies; 229+ messages in thread
From: Christian Brauner @ 2023-06-23 12:41 UTC (permalink / raw)
To: Jeff Layton
Cc: Steven Rostedt, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Luis Chamberlain,
Iurii Zaikin, Tony Luck, Guilherme G. Piccoli, Anders Larsen,
Steve French, Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N,
Tom Talpey, Sergey Senozhatsky, Phillip Lougher,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On Wed, Jun 21, 2023 at 03:52:27PM -0400, Jeff Layton wrote:
> On Wed, 2023-06-21 at 15:21 -0400, Steven Rostedt wrote:
> > On Wed, 21 Jun 2023 10:45:05 -0400
> > Jeff Layton <jlayton@kernel.org> wrote:
> >
> > > Most of this conversion was done via coccinelle, with a few of the more
> > > non-standard accesses done by hand. There should be no behavioral
> > > changes with this set. That will come later, as we convert individual
> > > filesystems to use multigrain timestamps.
> >
> > BTW, Linus has suggested to me that whenever a conccinelle script is used,
> > it should be included in the change log.
> >
>
> Ok, here's what I have. I note again that my usage of coccinelle is
> pretty primitive, so I ended up doing a fair bit of by-hand fixing after
> applying these.
>
> Given the way that this change is broken up into 77 patches by
> subsystem, to which changelogs should I add it? I could add it to the
> "infrastructure" patch, but that's the one where I _didn't_ use it.
>
> Maybe to patch #79 (the one that renames i_ctime)?
That works. I can also put this into a merge commit or pr message.
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [f2fs-dev] [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-23 12:41 ` Christian Brauner
0 siblings, 0 replies; 229+ messages in thread
From: Christian Brauner @ 2023-06-23 12:41 UTC (permalink / raw)
To: Jeff Layton
Cc: Latchesar Ionkov, Rafael J. Wysocki, Darrick J. Wong,
Anders Larsen, Carlos Llamas, Andrii Nakryiko, Hugh Dickins,
John Johansen, Seth Forshee, Alexander Gordeev,
Christoph Hellwig, Mike Marshall, Paulo Alcantara, linux-xfs,
Bart Van Assche, Michael Ellerman, John Keeping, Zhang Yi,
James Morris, Christophe Leroy, Tyler Hicks, Alan Stern,
Christian Borntraeger, devel, Shyam Prasad N, Jan Harkes,
linux-um, Nicholas Piggin, Joel Becker, Eric Van Hensbergen,
Suren Baghdasaryan, Trond Myklebust, Anton Altaparmakov,
Wolfram Sang, Greg Kroah-Hartman, Stephen Smalley, linux-usb,
linux-kernel, Ronnie Sahlberg, Sergey Senozhatsky,
Luis Chamberlain, Chuck Lever, Sven Schnelle, Jiri Olsa,
Jan Kara, Tejun Heo, Andrew Morton, linux-trace-kernel,
linux-hardening, Dave Kleikamp, Sandeep Dhavale, Tetsuo Handa,
Mimi Zohar, linux-mm, Joel Fernandes, Eric Dumazet,
Stanislav Fomichev, Andrzej Pietrasiewicz, Hangyu Hua,
linux-s390, linux-nilfs, Paul Moore, Leon Romanovsky,
John Fastabend, Arve Hjønnevåg, Minghao Chi, codalist,
selinux, ZhangPeng, Udipto Goswami, Yonghong Song, Iurii Zaikin,
Namjae Jeon, Masami Hiramatsu, ecryptfs, Todd Kjos,
Vasily Gorbik, Yu Zhe, linuxppc-dev, reiserfs-devel,
Miklos Szeredi, Yue Hu, Jaegeuk Kim, Aditya Garg, Martijn Coenen,
OGAWA Hirofumi, Hao Luo, Tony Luck, Theodore Ts'o,
Nicolas Pitre, linux-ntfs-dev, Muchun Song, Roberto Sassu,
linux-f2fs-devel, Guilherme G. Piccoli, Jozef Martiniak,
Eric Biederman, Anna Schumaker, xu xin, Brad Warrum,
Mike Kravetz, Jingyu Wang, linux-efi, Dan Carpenter,
Martin Brandenburg, Tom Rix, Alexei Starovoitov, Chris Mason,
linux-mtd, Matthew Wilcox (Oracle),
Marc Dionne, linux-afs, Ian Kent, Naohiro Aota, Daniel Borkmann,
Dennis Dalessandro, linux-rdma, Linyu Yuan, coda,
Viacheslav Dubeyko, Ilya Dryomov, Paolo Abeni, Alexey Dobriyan,
Serge E. Hallyn, Jens Axboe, Zeng Jingxiang, Kees Cook,
Arnd Bergmann, autofs, Steven Rostedt, Yifei Liu, Damien Le Moal,
Eric Paris, ceph-devel, Gao Xiang, Jiangshan Yi, David Howells,
linux-nfs, linux-ext4, Song Liu, samba-technical, Steve French,
Jeremy Kerr, netdev, Bob Peterson, linux-fsdevel, bpf, ntfs3,
linux-erofs, David S. Miller, ocfs2-devel, jfs-discussion,
Dominique Martinet, Christian Schoenebeck, Bob Copeland,
KP Singh, Oleg Kanatov, Konstantin Komarov, Joseph Qi,
Yuta Hayama, Andreas Dilger, Mikulas Patocka, Zhengchao Shao,
Chen Zhongjin, Ard Biesheuvel, Anton Ivanov, Laurent Pinchart,
Andreas Gruenbacher, Richard Weinberger, Mark Fasheh,
Stefan Roesch, cluster-devel, Jason Gunthorpe, Jakub Kicinski,
Rik van Riel, Salah Triki, Evgeniy Dushistov, linux-cifs,
Heiko Carstens, apparmor, Josef Bacik, Liam R. Howlett,
Tom Talpey, Hans de Goede, Tigran A. Aivazian, Dave Chinner,
David Sterba, Xiubo Li, Ryusuke Konishi, Juergen Gross,
Johannes Thumshirn, Ritu Agarwal, Luis de Bethencourt,
Martin KaFai Lau, v9fs, Fabio M. De Francesco, linux-unionfs,
Ruihan Li, linux-security-module, Erez Zadok, Jeffle Xu,
Dr. David Alan Gilbert, Johannes Weiner, Phillip Lougher,
Johannes Berg, Sungjong Seo, David Woodhouse, linux-karma-devel,
linux-btrfs, Alexander Viro
On Wed, Jun 21, 2023 at 03:52:27PM -0400, Jeff Layton wrote:
> On Wed, 2023-06-21 at 15:21 -0400, Steven Rostedt wrote:
> > On Wed, 21 Jun 2023 10:45:05 -0400
> > Jeff Layton <jlayton@kernel.org> wrote:
> >
> > > Most of this conversion was done via coccinelle, with a few of the more
> > > non-standard accesses done by hand. There should be no behavioral
> > > changes with this set. That will come later, as we convert individual
> > > filesystems to use multigrain timestamps.
> >
> > BTW, Linus has suggested to me that whenever a conccinelle script is used,
> > it should be included in the change log.
> >
>
> Ok, here's what I have. I note again that my usage of coccinelle is
> pretty primitive, so I ended up doing a fair bit of by-hand fixing after
> applying these.
>
> Given the way that this change is broken up into 77 patches by
> subsystem, to which changelogs should I add it? I could add it to the
> "infrastructure" patch, but that's the one where I _didn't_ use it.
>
> Maybe to patch #79 (the one that renames i_ctime)?
That works. I can also put this into a merge commit or pr message.
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [Ocfs2-devel] [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-23 12:41 ` Christian Brauner
0 siblings, 0 replies; 229+ messages in thread
From: Christian Brauner via Ocfs2-devel @ 2023-06-23 12:41 UTC (permalink / raw)
To: Jeff Layton
Cc: Latchesar Ionkov, Rafael J. Wysocki, Anders Larsen,
Carlos Llamas, Andrii Nakryiko, Hugh Dickins, John Johansen,
Seth Forshee, Alexander Gordeev, Mike Marshall, Paulo Alcantara,
linux-xfs, Bart Van Assche, Michael Ellerman, John Keeping,
Zhang Yi, James Morris, Christophe Leroy, Tyler Hicks,
Alan Stern, Christian Borntraeger, devel, Shyam Prasad N,
Jan Harkes, linux-um, Nicholas Piggin, Eric Van Hensbergen,
Suren Baghdasaryan, Trond Myklebust, Anton Altaparmakov,
Wolfram Sang, Greg Kroah-Hartman, Stephen Smalley, linux-usb,
linux-kernel, Ronnie Sahlberg, Sergey Senozhatsky,
Luis Chamberlain, Chuck Lever, Sven Schnelle, Jiri Olsa,
Jan Kara, Tejun Heo, linux-trace-kernel, linux-hardening,
Dave Kleikamp, Sandeep Dhavale, Tetsuo Handa, Mimi Zohar,
linux-mm, Joel Fernandes, Eric Dumazet, Stanislav Fomichev,
Andrzej Pietrasiewicz, Hangyu Hua, linux-s390, linux-nilfs,
Paul Moore, Leon Romanovsky, John Fastabend,
Arve Hjønnevåg, Minghao Chi, codalist, selinux,
ZhangPeng, Udipto Goswami, Yonghong Song, Iurii Zaikin,
Namjae Jeon, Masami Hiramatsu, ecryptfs, Todd Kjos,
Vasily Gorbik, Yu Zhe, linuxppc-dev, reiserfs-devel,
Miklos Szeredi, Yue Hu, Jaegeuk Kim, Aditya Garg, Martijn Coenen,
OGAWA Hirofumi, Hao Luo, Tony Luck, Theodore Ts'o,
Nicolas Pitre, linux-ntfs-dev, Muchun Song, Roberto Sassu,
linux-f2fs-devel, Guilherme G. Piccoli, Jozef Martiniak,
Eric Biederman, Anna Schumaker, xu xin, Brad Warrum,
Mike Kravetz, Jingyu Wang, linux-efi, Dan Carpenter,
Martin Brandenburg, Tom Rix, Alexei Starovoitov, Chris Mason,
linux-mtd, Marc Dionne, linux-afs, Ian Kent, Naohiro Aota,
Daniel Borkmann, Dennis Dalessandro, linux-rdma, Linyu Yuan,
coda, Viacheslav Dubeyko, Ilya Dryomov, Paolo Abeni,
Alexey Dobriyan, Serge E. Hallyn, Zhihao Cheng, Jens Axboe,
Zeng Jingxiang, Kees Cook, Arnd Bergmann, autofs, Steven Rostedt,
Yifei Liu, Damien Le Moal, Eric Paris, ceph-devel, Gao Xiang,
Jiangshan Yi, David Howells, linux-nfs, linux-ext4, Song Liu,
samba-technical, Steve French, Jeremy Kerr, netdev, Bob Peterson,
linux-fsdevel, bpf, ntfs3, linux-erofs, David S. Miller,
ocfs2-devel, jfs-discussion, Dominique Martinet,
Christian Schoenebeck, Bob Copeland, KP Singh, Oleg Kanatov,
Konstantin Komarov, Yuta Hayama, Andreas Dilger, Mikulas Patocka,
Zhengchao Shao, Chen Zhongjin, Ard Biesheuvel, Anton Ivanov,
Laurent Pinchart, Andreas Gruenbacher, Richard Weinberger,
Stefan Roesch, cluster-devel, Jason Gunthorpe, Jakub Kicinski,
Rik van Riel, Salah Triki, Evgeniy Dushistov, linux-cifs,
Heiko Carstens, Chao Yu, apparmor, Josef Bacik, Liam R. Howlett,
Tom Talpey, Hans de Goede, Tigran A. Aivazian, Dave Chinner,
David Sterba, Xiubo Li, Ryusuke Konishi, Juergen Gross,
Johannes Thumshirn, Ritu Agarwal, Luis de Bethencourt,
Martin KaFai Lau, v9fs, Fabio M. De Francesco, linux-unionfs,
Ruihan Li, linux-security-module, Erez Zadok, Jeffle Xu,
Dr. David Alan Gilbert, Johannes Weiner, Phillip Lougher,
Johannes Berg, Sungjong Seo, David Woodhouse, linux-karma-devel,
linux-btrfs, Alexander Viro
On Wed, Jun 21, 2023 at 03:52:27PM -0400, Jeff Layton wrote:
> On Wed, 2023-06-21 at 15:21 -0400, Steven Rostedt wrote:
> > On Wed, 21 Jun 2023 10:45:05 -0400
> > Jeff Layton <jlayton@kernel.org> wrote:
> >
> > > Most of this conversion was done via coccinelle, with a few of the more
> > > non-standard accesses done by hand. There should be no behavioral
> > > changes with this set. That will come later, as we convert individual
> > > filesystems to use multigrain timestamps.
> >
> > BTW, Linus has suggested to me that whenever a conccinelle script is used,
> > it should be included in the change log.
> >
>
> Ok, here's what I have. I note again that my usage of coccinelle is
> pretty primitive, so I ended up doing a fair bit of by-hand fixing after
> applying these.
>
> Given the way that this change is broken up into 77 patches by
> subsystem, to which changelogs should I add it? I could add it to the
> "infrastructure" patch, but that's the one where I _didn't_ use it.
>
> Maybe to patch #79 (the one that renames i_ctime)?
That works. I can also put this into a merge commit or pr message.
_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-23 12:41 ` Christian Brauner
0 siblings, 0 replies; 229+ messages in thread
From: Christian Brauner @ 2023-06-23 12:41 UTC (permalink / raw)
To: Jeff Layton
Cc: Latchesar Ionkov, Rafael J. Wysocki, Darrick J. Wong,
Anders Larsen, Carlos Llamas, Andrii Nakryiko, Hugh Dickins,
John Johansen, Seth Forshee, Alexander Gordeev,
Christoph Hellwig, Mike Marshall, Paulo Alcantara, linux-xfs,
Bart Van Assche, John Keeping, Zhang Yi, James Morris,
Tyler Hicks, Alan Stern, Christian Borntraeger, devel,
Shyam Prasad N, Jan Harkes, linux-um, Nicholas Piggin,
Joel Becker, Eric Van Hen sbergen, Suren Baghdasaryan,
Trond Myklebust, Anton Altaparmakov, Wolfram Sang,
Greg Kroah-Hartman, Stephen Smalley, linux-usb, linux-kernel,
Ronnie Sahlberg, Sergey Senozhatsky, Luis Chamberlain,
Chuck Lever, Sven Schnelle, Jiri Olsa, Jan Kara, Tejun Heo,
Andrew Morton, linux-trace-kernel, linux-hardening,
Dave Kleikamp, Sandeep Dhavale, Tetsuo Handa, Mimi Zohar,
linux-mm, Joel Fernandes
On Wed, Jun 21, 2023 at 03:52:27PM -0400, Jeff Layton wrote:
> On Wed, 2023-06-21 at 15:21 -0400, Steven Rostedt wrote:
> > On Wed, 21 Jun 2023 10:45:05 -0400
> > Jeff Layton <jlayton@kernel.org> wrote:
> >
> > > Most of this conversion was done via coccinelle, with a few of the more
> > > non-standard accesses done by hand. There should be no behavioral
> > > changes with this set. That will come later, as we convert individual
> > > filesystems to use multigrain timestamps.
> >
> > BTW, Linus has suggested to me that whenever a conccinelle script is used,
> > it should be included in the change log.
> >
>
> Ok, here's what I have. I note again that my usage of coccinelle is
> pretty primitive, so I ended up doing a fair bit of by-hand fixing after
> applying these.
>
> Given the way that this change is broken up into 77 patches by
> subsystem, to which changelogs should I add it? I could add it to the
> "infrastructure" patch, but that's the one where I _didn't_ use it.
>
> Maybe to patch #79 (the one that renames i_ctime)?
That works. I can also put this into a merge commit or pr message.
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-23 12:41 ` Christian Brauner
0 siblings, 0 replies; 229+ messages in thread
From: Christian Brauner @ 2023-06-23 12:41 UTC (permalink / raw)
To: Jeff Layton
Cc: Latchesar Ionkov, Rafael J. Wysocki, Darrick J. Wong,
Anders Larsen, Carlos Llamas, Andrii Nakryiko, Hugh Dickins,
John Johansen, Seth Forshee, Alexander Gordeev,
Christoph Hellwig, Mike Marshall, Paulo Alcantara, linux-xfs,
Bart Van Assche, Michael Ellerman, John Keeping, Zhang Yi,
James Morris, Christophe Leroy, Tyler Hicks, Alan Stern,
Christian Borntraeger, devel, Shyam Prasad N, Jan Harkes,
linux-um
On Wed, Jun 21, 2023 at 03:52:27PM -0400, Jeff Layton wrote:
> On Wed, 2023-06-21 at 15:21 -0400, Steven Rostedt wrote:
> > On Wed, 21 Jun 2023 10:45:05 -0400
> > Jeff Layton <jlayton@kernel.org> wrote:
> >
> > > Most of this conversion was done via coccinelle, with a few of the more
> > > non-standard accesses done by hand. There should be no behavioral
> > > changes with this set. That will come later, as we convert individual
> > > filesystems to use multigrain timestamps.
> >
> > BTW, Linus has suggested to me that whenever a conccinelle script is used,
> > it should be included in the change log.
> >
>
> Ok, here's what I have. I note again that my usage of coccinelle is
> pretty primitive, so I ended up doing a fair bit of by-hand fixing after
> applying these.
>
> Given the way that this change is broken up into 77 patches by
> subsystem, to which changelogs should I add it? I could add it to the
> "infrastructure" patch, but that's the one where I _didn't_ use it.
>
> Maybe to patch #79 (the one that renames i_ctime)?
That works. I can also put this into a merge commit or pr message.
^ permalink raw reply [flat|nested] 229+ messages in thread
* [Cluster-devel] [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-23 12:41 ` Christian Brauner
0 siblings, 0 replies; 229+ messages in thread
From: Christian Brauner @ 2023-06-23 12:41 UTC (permalink / raw)
To: cluster-devel.redhat.com
On Wed, Jun 21, 2023 at 03:52:27PM -0400, Jeff Layton wrote:
> On Wed, 2023-06-21 at 15:21 -0400, Steven Rostedt wrote:
> > On Wed, 21 Jun 2023 10:45:05 -0400
> > Jeff Layton <jlayton@kernel.org> wrote:
> >
> > > Most of this conversion was done via coccinelle, with a few of the more
> > > non-standard accesses done by hand. There should be no behavioral
> > > changes with this set. That will come later, as we convert individual
> > > filesystems to use multigrain timestamps.
> >
> > BTW, Linus has suggested to me that whenever a conccinelle script is used,
> > it should be included in the change log.
> >
>
> Ok, here's what I have. I note again that my usage of coccinelle is
> pretty primitive, so I ended up doing a fair bit of by-hand fixing after
> applying these.
>
> Given the way that this change is broken up into 77 patches by
> subsystem, to which changelogs should I add it? I could add it to the
> "infrastructure" patch, but that's the one where I _didn't_ use it.?
>
> Maybe to patch #79 (the one that renames i_ctime)?
That works. I can also put this into a merge commit or pr message.
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-23 12:41 ` Christian Brauner
0 siblings, 0 replies; 229+ messages in thread
From: Christian Brauner @ 2023-06-23 12:41 UTC (permalink / raw)
To: Jeff Layton
Cc: Latchesar Ionkov, Rafael J. Wysocki, Darrick J. Wong,
Anders Larsen, Carlos Llamas, Andrii Nakryiko, Hugh Dickins,
John Johansen, Seth Forshee, Alexander Gordeev,
Christoph Hellwig, Mike Marshall, Paulo Alcantara, linux-xfs,
Bart Van Assche, Michael Ellerman, John Keeping, Zhang Yi,
James Morris, Christophe Leroy, Tyler Hicks, Alan Stern,
Christian Borntraeger, devel
On Wed, Jun 21, 2023 at 03:52:27PM -0400, Jeff Layton wrote:
> On Wed, 2023-06-21 at 15:21 -0400, Steven Rostedt wrote:
> > On Wed, 21 Jun 2023 10:45:05 -0400
> > Jeff Layton <jlayton@kernel.org> wrote:
> >
> > > Most of this conversion was done via coccinelle, with a few of the more
> > > non-standard accesses done by hand. There should be no behavioral
> > > changes with this set. That will come later, as we convert individual
> > > filesystems to use multigrain timestamps.
> >
> > BTW, Linus has suggested to me that whenever a conccinelle script is used,
> > it should be included in the change log.
> >
>
> Ok, here's what I have. I note again that my usage of coccinelle is
> pretty primitive, so I ended up doing a fair bit of by-hand fixing after
> applying these.
>
> Given the way that this change is broken up into 77 patches by
> subsystem, to which changelogs should I add it? I could add it to the
> "infrastructure" patch, but that's the one where I _didn't_ use it.Â
>
> Maybe to patch #79 (the one that renames i_ctime)?
That works. I can also put this into a merge commit or pr message.
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 77/79] security: switch to new ctime accessors
2023-06-21 14:46 ` [PATCH 77/79] security: " Jeff Layton
@ 2023-06-23 14:15 ` Paul Moore
0 siblings, 0 replies; 229+ messages in thread
From: Paul Moore @ 2023-06-23 14:15 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, James Morris, Serge E. Hallyn, Al Viro,
Jan Kara, linux-security-module, linux-kernel
On Wed, Jun 21, 2023 at 10:49 AM Jeff Layton <jlayton@kernel.org> wrote:
>
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
> ---
> security/inode.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/security/inode.c b/security/inode.c
> index 6c326939750d..086280390793 100644
> --- a/security/inode.c
> +++ b/security/inode.c
> @@ -145,7 +145,7 @@ static struct dentry *securityfs_create_dentry(const char *name, umode_t mode,
>
> inode->i_ino = get_next_ino();
> inode->i_mode = mode;
> - inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
In the process of looking at inode_ctime_set_current() I ran into the
same bug others noticed regarding inode_set_ctime()/inode_ctime_set();
assuming that gets fixed this looks fine to me.
Acked-by: Paul Moore <paul@paul-moore.com>
> inode->i_private = data;
> if (S_ISDIR(mode)) {
> inode->i_op = &simple_dir_inode_operations;
> --
> 2.41.0
--
paul-moore.com
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 78/79] selinux: switch to new ctime accessors
2023-06-21 14:46 ` [PATCH 78/79] selinux: " Jeff Layton
@ 2023-06-23 14:17 ` Paul Moore
0 siblings, 0 replies; 229+ messages in thread
From: Paul Moore @ 2023-06-23 14:17 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Stephen Smalley, Eric Paris, Al Viro,
Jan Kara, selinux, linux-kernel
On Wed, Jun 21, 2023 at 10:49 AM Jeff Layton <jlayton@kernel.org> wrote:
>
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
> ---
> security/selinux/selinuxfs.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/security/selinux/selinuxfs.c b/security/selinux/selinuxfs.c
> index bad1f6b685fd..d3908baddb30 100644
> --- a/security/selinux/selinuxfs.c
> +++ b/security/selinux/selinuxfs.c
> @@ -1197,7 +1197,7 @@ static struct inode *sel_make_inode(struct super_block *sb, int mode)
>
> if (ret) {
> ret->i_mode = mode;
> - ret->i_atime = ret->i_mtime = ret->i_ctime = current_time(ret);
> + ret->i_atime = ret->i_mtime = inode_ctime_set_current(ret);
Same issue as in 77/79.
Acked-by: Paul Moore <paul@paul-moore.com>
--
paul-moore.com
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 57/79] qnx4: switch to new ctime accessors
2023-06-21 14:46 ` [PATCH 57/79] qnx4: " Jeff Layton
@ 2023-06-23 14:26 ` Anders Larsen
0 siblings, 0 replies; 229+ messages in thread
From: Anders Larsen @ 2023-06-23 14:26 UTC (permalink / raw)
To: Jeff Layton; +Cc: Christian Brauner, Al Viro, Jan Kara, linux-kernel
On Wednesday, 2023-06-21 16:46 Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
> ---
> fs/qnx4/inode.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c
> index 391ea402920d..3fde90b3f99b 100644
> --- a/fs/qnx4/inode.c
> +++ b/fs/qnx4/inode.c
> @@ -305,8 +305,8 @@ struct inode *qnx4_iget(struct super_block *sb, unsigned long ino)
> inode->i_mtime.tv_nsec = 0;
> inode->i_atime.tv_sec = le32_to_cpu(raw_inode->di_atime);
> inode->i_atime.tv_nsec = 0;
> - inode->i_ctime.tv_sec = le32_to_cpu(raw_inode->di_ctime);
> - inode->i_ctime.tv_nsec = 0;
> + inode_ctime_set_sec(inode, le32_to_cpu(raw_inode->di_ctime));
> + inode_ctime_set_nsec(inode, 0);
> inode->i_blocks = le32_to_cpu(raw_inode->di_first_xtnt.xtnt_size);
>
> memcpy(qnx4_inode, raw_inode, QNX4_DIR_ENTRY_SIZE);
Acked-by: Anders Larsen <al@alarsen.net>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 17/79] ceph: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 17/79] ceph: " Jeff Layton
@ 2023-06-26 0:56 ` Xiubo Li
0 siblings, 0 replies; 229+ messages in thread
From: Xiubo Li @ 2023-06-26 0:56 UTC (permalink / raw)
To: Jeff Layton, Christian Brauner, Ilya Dryomov
Cc: Al Viro, Jan Kara, ceph-devel, linux-kernel
On 6/21/23 22:45, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
> ---
> fs/ceph/acl.c | 2 +-
> fs/ceph/caps.c | 2 +-
> fs/ceph/inode.c | 17 ++++++++++-------
> fs/ceph/snap.c | 2 +-
> fs/ceph/xattr.c | 2 +-
> 5 files changed, 14 insertions(+), 11 deletions(-)
>
> diff --git a/fs/ceph/acl.c b/fs/ceph/acl.c
> index 6945a938d396..a3de2b9c3a68 100644
> --- a/fs/ceph/acl.c
> +++ b/fs/ceph/acl.c
> @@ -93,7 +93,7 @@ int ceph_set_acl(struct mnt_idmap *idmap, struct dentry *dentry,
> char *value = NULL;
> struct iattr newattrs;
> struct inode *inode = d_inode(dentry);
> - struct timespec64 old_ctime = inode->i_ctime;
> + struct timespec64 old_ctime = inode_ctime_peek(inode);
> umode_t new_mode = inode->i_mode, old_mode = inode->i_mode;
>
> if (ceph_snap(inode) != CEPH_NOSNAP) {
> diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
> index 2321e5ddb664..c144a07e334e 100644
> --- a/fs/ceph/caps.c
> +++ b/fs/ceph/caps.c
> @@ -1400,7 +1400,7 @@ static void __prep_cap(struct cap_msg_args *arg, struct ceph_cap *cap,
>
> arg->mtime = inode->i_mtime;
> arg->atime = inode->i_atime;
> - arg->ctime = inode->i_ctime;
> + arg->ctime = inode_ctime_peek(inode);
> arg->btime = ci->i_btime;
> arg->change_attr = inode_peek_iversion_raw(inode);
>
> diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
> index 8e5f41d45283..f0b3b11d695e 100644
> --- a/fs/ceph/inode.c
> +++ b/fs/ceph/inode.c
> @@ -100,7 +100,7 @@ struct inode *ceph_get_snapdir(struct inode *parent)
> inode->i_uid = parent->i_uid;
> inode->i_gid = parent->i_gid;
> inode->i_mtime = parent->i_mtime;
> - inode->i_ctime = parent->i_ctime;
> + inode_ctime_set(inode, inode_ctime_peek(parent));
> inode->i_atime = parent->i_atime;
> ci->i_rbytes = 0;
> ci->i_btime = ceph_inode(parent)->i_btime;
> @@ -695,12 +695,14 @@ void ceph_fill_file_time(struct inode *inode, int issued,
> CEPH_CAP_FILE_BUFFER|
> CEPH_CAP_AUTH_EXCL|
> CEPH_CAP_XATTR_EXCL)) {
> + struct timespec64 ictime = inode_ctime_peek(inode);
> +
> if (ci->i_version == 0 ||
> - timespec64_compare(ctime, &inode->i_ctime) > 0) {
> + timespec64_compare(ctime, &ictime) > 0) {
> dout("ctime %lld.%09ld -> %lld.%09ld inc w/ cap\n",
> - inode->i_ctime.tv_sec, inode->i_ctime.tv_nsec,
> + ictime.tv_sec, ictime.tv_nsec,
> ctime->tv_sec, ctime->tv_nsec);
> - inode->i_ctime = *ctime;
> + inode_ctime_set(inode, *ctime);
> }
> if (ci->i_version == 0 ||
> ceph_seq_cmp(time_warp_seq, ci->i_time_warp_seq) > 0) {
> @@ -738,7 +740,7 @@ void ceph_fill_file_time(struct inode *inode, int issued,
> } else {
> /* we have no write|excl caps; whatever the MDS says is true */
> if (ceph_seq_cmp(time_warp_seq, ci->i_time_warp_seq) >= 0) {
> - inode->i_ctime = *ctime;
> + inode_ctime_set(inode, *ctime);
> inode->i_mtime = *mtime;
> inode->i_atime = *atime;
> ci->i_time_warp_seq = time_warp_seq;
> @@ -2166,7 +2168,8 @@ int __ceph_setattr(struct inode *inode, struct iattr *attr)
> bool only = (ia_valid & (ATTR_SIZE|ATTR_MTIME|ATTR_ATIME|
> ATTR_MODE|ATTR_UID|ATTR_GID)) == 0;
> dout("setattr %p ctime %lld.%ld -> %lld.%ld (%s)\n", inode,
> - inode->i_ctime.tv_sec, inode->i_ctime.tv_nsec,
> + inode_ctime_peek(inode).tv_sec,
> + inode_ctime_peek(inode).tv_nsec,
> attr->ia_ctime.tv_sec, attr->ia_ctime.tv_nsec,
> only ? "ctime only" : "ignored");
> if (only) {
> @@ -2191,7 +2194,7 @@ int __ceph_setattr(struct inode *inode, struct iattr *attr)
> if (dirtied) {
> inode_dirty_flags = __ceph_mark_dirty_caps(ci, dirtied,
> &prealloc_cf);
> - inode->i_ctime = attr->ia_ctime;
> + inode_ctime_set(inode, attr->ia_ctime);
> inode_inc_iversion_raw(inode);
> }
>
> diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c
> index 2e73ba62bd7a..f02df070fa84 100644
> --- a/fs/ceph/snap.c
> +++ b/fs/ceph/snap.c
> @@ -660,7 +660,7 @@ int __ceph_finish_cap_snap(struct ceph_inode_info *ci,
> capsnap->size = i_size_read(inode);
> capsnap->mtime = inode->i_mtime;
> capsnap->atime = inode->i_atime;
> - capsnap->ctime = inode->i_ctime;
> + capsnap->ctime = inode_ctime_peek(inode);
> capsnap->btime = ci->i_btime;
> capsnap->change_attr = inode_peek_iversion_raw(inode);
> capsnap->time_warp_seq = ci->i_time_warp_seq;
> diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c
> index 806183959c47..8e217f7f58bd 100644
> --- a/fs/ceph/xattr.c
> +++ b/fs/ceph/xattr.c
> @@ -1238,7 +1238,7 @@ int __ceph_setxattr(struct inode *inode, const char *name,
> dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_XATTR_EXCL,
> &prealloc_cf);
> ci->i_xattrs.dirty = true;
> - inode->i_ctime = current_time(inode);
> + inode_ctime_set_current(inode);
> }
>
> spin_unlock(&ci->i_ceph_lock);
Just back from PTO.
LGTM.
Reviewed-by: Xiubo Li <xiubli@redhat.com>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 47/79] nilfs2: switch to new ctime accessors
@ 2023-06-26 15:26 ` Ryusuke Konishi
0 siblings, 0 replies; 229+ messages in thread
From: Ryusuke Konishi @ 2023-06-26 15:26 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Al Viro, Jan Kara, linux-nilfs, linux-kernel
On Wed, Jun 21, 2023 at 11:48 PM Jeff Layton wrote:
>
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
> ---
> fs/nilfs2/dir.c | 6 +++---
> fs/nilfs2/inode.c | 12 ++++++------
> fs/nilfs2/ioctl.c | 2 +-
> fs/nilfs2/namei.c | 8 ++++----
> 4 files changed, 14 insertions(+), 14 deletions(-)
Acked-by: Ryusuke Konishi <konishi.ryusuke@gmail.com>
As already mentioned in the s390 patch comment, the implementation of
inode_ctime_set_current() needs to be rewritten to use
inode_ctime_set() instead of inode_set_ctime(), but I agree with this
conversion patch for nilfs2 itself.
Thank you for your efforts.
Ryusuke Konishi
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 47/79] nilfs2: switch to new ctime accessors
@ 2023-06-26 15:26 ` Ryusuke Konishi
0 siblings, 0 replies; 229+ messages in thread
From: Ryusuke Konishi @ 2023-06-26 15:26 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Al Viro, Jan Kara,
linux-nilfs-u79uwXL29TY76Z2rM5mHXA,
linux-kernel-u79uwXL29TY76Z2rM5mHXA
On Wed, Jun 21, 2023 at 11:48 PM Jeff Layton wrote:
>
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
> ---
> fs/nilfs2/dir.c | 6 +++---
> fs/nilfs2/inode.c | 12 ++++++------
> fs/nilfs2/ioctl.c | 2 +-
> fs/nilfs2/namei.c | 8 ++++----
> 4 files changed, 14 insertions(+), 14 deletions(-)
Acked-by: Ryusuke Konishi <konishi.ryusuke-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
As already mentioned in the s390 patch comment, the implementation of
inode_ctime_set_current() needs to be rewritten to use
inode_ctime_set() instead of inode_set_ctime(), but I agree with this
conversion patch for nilfs2 itself.
Thank you for your efforts.
Ryusuke Konishi
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 13/79] autofs: switch to new ctime accessors
2023-06-21 14:45 ` [PATCH 13/79] autofs: " Jeff Layton
2023-06-21 16:43 ` Jan Kara
@ 2023-06-27 1:48 ` Ian Kent
1 sibling, 0 replies; 229+ messages in thread
From: Ian Kent @ 2023-06-27 1:48 UTC (permalink / raw)
To: Jeff Layton, Christian Brauner; +Cc: Al Viro, Jan Kara, autofs, linux-kernel
On 21/623 22:45, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Looks fine to me.
Please feel free to add my Acked-by or even Signed-off-by as
you need dictates.
Ian
> ---
> fs/autofs/inode.c | 2 +-
> fs/autofs/root.c | 6 +++---
> 2 files changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/fs/autofs/inode.c b/fs/autofs/inode.c
> index affa70360b1f..47e3054b29dc 100644
> --- a/fs/autofs/inode.c
> +++ b/fs/autofs/inode.c
> @@ -370,7 +370,7 @@ struct inode *autofs_get_inode(struct super_block *sb, umode_t mode)
> inode->i_uid = d_inode(sb->s_root)->i_uid;
> inode->i_gid = d_inode(sb->s_root)->i_gid;
> }
> - inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode);
> + inode->i_atime = inode->i_mtime = inode_ctime_set_current(inode);
> inode->i_ino = get_next_ino();
>
> if (S_ISDIR(mode)) {
> diff --git a/fs/autofs/root.c b/fs/autofs/root.c
> index 93046c9dc461..4c0fc0f8d688 100644
> --- a/fs/autofs/root.c
> +++ b/fs/autofs/root.c
> @@ -600,7 +600,7 @@ static int autofs_dir_symlink(struct mnt_idmap *idmap,
> p_ino = autofs_dentry_ino(dentry->d_parent);
> p_ino->count++;
>
> - dir->i_mtime = dir->i_ctime = current_time(dir);
> + dir->i_mtime = inode_ctime_set_current(dir);
>
> return 0;
> }
> @@ -633,7 +633,7 @@ static int autofs_dir_unlink(struct inode *dir, struct dentry *dentry)
> d_inode(dentry)->i_size = 0;
> clear_nlink(d_inode(dentry));
>
> - dir->i_mtime = dir->i_ctime = current_time(dir);
> + dir->i_mtime = inode_ctime_set_current(dir);
>
> spin_lock(&sbi->lookup_lock);
> __autofs_add_expiring(dentry);
> @@ -749,7 +749,7 @@ static int autofs_dir_mkdir(struct mnt_idmap *idmap,
> p_ino = autofs_dentry_ino(dentry->d_parent);
> p_ino->count++;
> inc_nlink(dir);
> - dir->i_mtime = dir->i_ctime = current_time(dir);
> + dir->i_mtime = inode_ctime_set_current(dir);
>
> return 0;
> }
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 00/79] fs: new accessors for inode->i_ctime
2023-06-21 19:21 ` [f2fs-dev] " Steven Rostedt
` (3 preceding siblings ...)
(?)
@ 2023-06-30 22:11 ` Luis Chamberlain
-1 siblings, 0 replies; 229+ messages in thread
From: Luis Chamberlain @ 2023-06-30 22:11 UTC (permalink / raw)
To: Steven Rostedt, Julia Lawall, Takashi Iwai
Cc: Jeff Layton, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Iurii Zaikin, Tony Luck,
Guilherme G. Piccoli, Anders Larsen, Steve French,
Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N, Tom Talpey,
Sergey Senozhatsky, Phillip Lougher, Masami Hiramatsu,
Evgeniy Dushistov, Hans de Goede, Darrick J. Wong,
Damien Le Moal, Naohiro Aota, Johannes Thumshirn,
Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko,
Martin KaFai Lau, Song Liu, Yonghong Song, John Fastabend,
KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa, Hugh Dickins,
Andrew Morton, David S. Miller, Eric Dumazet, Jakub Kicinski,
Paolo Abeni, John Johansen, Paul Moore, James Morris,
Serge E. Hallyn, Stephen Smalley, Eric Paris, Juergen Gross,
Ruihan Li, Laurent Pinchart, Wolfram Sang, Udipto Goswami,
Linyu Yuan, John Keeping, Andrzej Pietrasiewicz, Dan Carpenter,
Yuta Hayama, Jozef Martiniak, Jens Axboe, Alan Stern,
Sandeep Dhavale, Dave Chinner, Johannes Weiner, ZhangPeng,
Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg, Erez Zadok,
Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On Wed, Jun 21, 2023 at 03:21:41PM -0400, Steven Rostedt wrote:
> On Wed, 21 Jun 2023 10:45:05 -0400
> Jeff Layton <jlayton@kernel.org> wrote:
>
> > Most of this conversion was done via coccinelle, with a few of the more
> > non-standard accesses done by hand. There should be no behavioral
> > changes with this set. That will come later, as we convert individual
> > filesystems to use multigrain timestamps.
>
> BTW, Linus has suggested to me that whenever a conccinelle script is used,
> it should be included in the change log.
Sometimes people like the coccinelle included in the commit, sometimes
people don't [0], it really ends up being up to a subjective maintainer
preference. A compromise could be to use git notes as these are
optional, however if we want to go down that path we should try to make
a general consensus on it so we can send a consistent message.
[0] https://lore.kernel.org/all/20230512073100.GC32559@twin.jikos.cz/
Luis
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [f2fs-dev] [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-30 22:11 ` Luis Chamberlain
0 siblings, 0 replies; 229+ messages in thread
From: Luis Chamberlain @ 2023-06-30 22:11 UTC (permalink / raw)
To: Steven Rostedt, Julia Lawall, Takashi Iwai
Cc: Latchesar Ionkov, Rafael J. Wysocki, Darrick J. Wong,
Anders Larsen, Carlos Llamas, Hugh Dickins, John Johansen,
Seth Forshee, Alexander Gordeev, Christoph Hellwig,
Mike Marshall, Paulo Alcantara, linux-xfs, Bart Van Assche,
Michael Ellerman, John Keeping, Zhang Yi, James Morris,
Christophe Leroy, Tyler Hicks, Alan Stern, Christian Borntraeger,
devel, Shyam Prasad N, Jan Harkes, linux-um, Nicholas Piggin,
Alexander Viro, Eric Van Hensbergen, Suren Baghdasaryan,
Trond Myklebust, Anton Altaparmakov, Christian Brauner,
Wolfram Sang, Greg Kroah-Hartman, Stephen Smalley, linux-usb,
linux-kernel, Ronnie Sahlberg, Sergey Senozhatsky,
Arve Hjønnevåg, Chuck Lever, Sven Schnelle, Jiri Olsa,
Jan Kara, Tejun Heo, Andrew Morton, linux-trace-kernel,
linux-hardening, Dave Kleikamp, Sandeep Dhavale, Tetsuo Handa,
samba-technical, Mimi Zohar, linux-mm, Joel Fernandes,
Eric Dumazet, Stanislav Fomichev, Andrzej Pietrasiewicz,
Hangyu Hua, linux-s390, linux-nilfs, Paul Moore, Leon Romanovsky,
John Fastabend, Andrii Nakryiko, Minghao Chi, codalist, selinux,
ZhangPeng, Udipto Goswami, Yonghong Song, Iurii Zaikin,
Namjae Jeon, Masami Hiramatsu, ecryptfs, Todd Kjos,
Vasily Gorbik, Yu Zhe, linuxppc-dev, reiserfs-devel,
Miklos Szeredi, Yue Hu, Jaegeuk Kim, Aditya Garg, Martijn Coenen,
OGAWA Hirofumi, Hao Luo, Tony Luck, Theodore Ts'o,
Nicolas Pitre, linux-ntfs-dev, Muchun Song, Roberto Sassu,
linux-f2fs-devel, Guilherme G. Piccoli, Jozef Martiniak,
Eric Biederman, Anna Schumaker, xu xin, Brad Warrum,
Mike Kravetz, Jingyu Wang, linux-efi, Dan Carpenter,
Martin Brandenburg, Tom Rix, Alexei Starovoitov, Chris Mason,
linux-mtd, Matthew Wilcox (Oracle),
Marc Dionne, linux-afs, Ian Kent, Naohiro Aota, Daniel Borkmann,
Dennis Dalessandro, linux-rdma, Linyu Yuan, coda,
Viacheslav Dubeyko, Ilya Dryomov, Paolo Abeni, Alexey Dobriyan,
Serge E. Hallyn, Jens Axboe, Zeng Jingxiang, Kees Cook,
Arnd Bergmann, autofs, Yifei Liu, Damien Le Moal, Eric Paris,
ceph-devel, Gao Xiang, Jiangshan Yi, David Howells, linux-nfs,
linux-ext4, Song Liu, Jeff Layton, Steve French, Jeremy Kerr,
netdev, Bob Peterson, linux-fsdevel, bpf, ntfs3, linux-erofs,
David S. Miller, ocfs2-devel, jfs-discussion, Dominique Martinet,
Christian Schoenebeck, Bob Copeland, KP Singh, Oleg Kanatov,
Konstantin Komarov, Joseph Qi, Yuta Hayama, Andreas Dilger,
Mikulas Patocka, Zhengchao Shao, Chen Zhongjin, Ard Biesheuvel,
Anton Ivanov, Laurent Pinchart, Andreas Gruenbacher,
Richard Weinberger, Mark Fasheh, Stefan Roesch, cluster-devel,
Jason Gunthorpe, Jakub Kicinski, Rik van Riel, Salah Triki,
Evgeniy Dushistov, linux-cifs, Heiko Carstens, apparmor,
Josef Bacik, Liam R. Howlett, Tom Talpey, Hans de Goede,
Tigran A. Aivazian, Dave Chinner, David Sterba, Xiubo Li,
Ryusuke Konishi, Juergen Gross, Johannes Thumshirn, Ritu Agarwal,
Luis de Bethencourt, Martin KaFai Lau, v9fs,
Fabio M. De Francesco, linux-unionfs, Ruihan Li,
linux-security-module, Erez Zadok, Jeffle Xu,
Dr. David Alan Gilbert, Johannes Weiner, Phillip Lougher,
Johannes Berg, Sungjong Seo, David Woodhouse, linux-karma-devel,
linux-btrfs, Joel Becker
On Wed, Jun 21, 2023 at 03:21:41PM -0400, Steven Rostedt wrote:
> On Wed, 21 Jun 2023 10:45:05 -0400
> Jeff Layton <jlayton@kernel.org> wrote:
>
> > Most of this conversion was done via coccinelle, with a few of the more
> > non-standard accesses done by hand. There should be no behavioral
> > changes with this set. That will come later, as we convert individual
> > filesystems to use multigrain timestamps.
>
> BTW, Linus has suggested to me that whenever a conccinelle script is used,
> it should be included in the change log.
Sometimes people like the coccinelle included in the commit, sometimes
people don't [0], it really ends up being up to a subjective maintainer
preference. A compromise could be to use git notes as these are
optional, however if we want to go down that path we should try to make
a general consensus on it so we can send a consistent message.
[0] https://lore.kernel.org/all/20230512073100.GC32559@twin.jikos.cz/
Luis
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-30 22:11 ` Luis Chamberlain
0 siblings, 0 replies; 229+ messages in thread
From: Luis Chamberlain @ 2023-06-30 22:11 UTC (permalink / raw)
To: Steven Rostedt, Julia Lawall, Takashi Iwai
Cc: Latchesar Ionkov, Rafael J. Wysocki, Darrick J. Wong,
Anders Larsen, Carlos Llamas, Hugh Dickins, John Johansen,
Seth Forshee, Alexander Gordeev, Christoph Hellwig,
Mike Marshall, Paulo Alcantara, linux-xfs, Bart Van Assche,
John Keeping, Zhang Yi, James Morris, Tyler Hicks, Alan Stern,
Christian Borntraeger, devel, Shyam Prasad N, Jan Harkes,
linux-um, Nicholas Piggin, Alexander Viro, Eric Van Hensbergen,
Suren Baghdasaryan, Trond Myklebust, Anton Altaparmakov,
Christian Brauner, Wolfram Sang, Greg Kroah-Hartman,
Stephen Smalley, linux-usb, linux-kernel, Ronnie Sahlberg,
Sergey Senozhatsky, Arve Hjønnevåg, Chuck Lever,
Sven Schnelle, Jiri Olsa, Jan Kara, Tejun Heo, Andrew Morton,
linux-trace-kernel, linux-hardening, Dave Kleikamp,
Sandeep Dhavale, Tetsuo Handa, samba-technical, Mimi Zohar,
linux-mm, Joel Fernandes, Eric Dumazet, Stanislav Fomichev,
Andrzej Pietrasiewicz, Hangyu Hua, linux-s390, linux-nilfs,
Paul Moore, Leon Romanovsky, John Fastabend, Andrii Nakryiko,
Minghao Chi, codalist, selinux, ZhangPeng, Udipto Goswami,
Yonghong Song, Iurii Zaikin, Namjae Jeon, Masami Hiramatsu,
ecryptfs, Todd Kjos, Vasily Gorbik, Yu Zhe, linuxppc-dev,
reiserfs-devel, Miklos Szeredi, Yue Hu, Jaegeuk Kim,
Adit ya Garg, Martijn Coenen, OGAWA Hirofumi, Hao Luo, Tony Luck,
Theodore Ts'o, Nicolas Pitre, linux-ntfs-dev, Muchun Song,
Roberto Sassu, linux-f2fs-devel, Guilherme G. Piccoli,
Jozef Martiniak, Eric Biederman, Anna Schumaker, xu xin,
Brad Warrum, Mike Kravetz, Jingyu Wang, linux-efi, Dan Carpenter,
Martin Brandenburg, Tom Rix, Alexei Starovoitov, Chris Mason,
linux-mtd, Matthew Wilcox (Oracle)
On Wed, Jun 21, 2023 at 03:21:41PM -0400, Steven Rostedt wrote:
> On Wed, 21 Jun 2023 10:45:05 -0400
> Jeff Layton <jlayton@kernel.org> wrote:
>
> > Most of this conversion was done via coccinelle, with a few of the more
> > non-standard accesses done by hand. There should be no behavioral
> > changes with this set. That will come later, as we convert individual
> > filesystems to use multigrain timestamps.
>
> BTW, Linus has suggested to me that whenever a conccinelle script is used,
> it should be included in the change log.
Sometimes people like the coccinelle included in the commit, sometimes
people don't [0], it really ends up being up to a subjective maintainer
preference. A compromise could be to use git notes as these are
optional, however if we want to go down that path we should try to make
a general consensus on it so we can send a consistent message.
[0] https://lore.kernel.org/all/20230512073100.GC32559@twin.jikos.cz/
Luis
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-30 22:11 ` Luis Chamberlain
0 siblings, 0 replies; 229+ messages in thread
From: Luis Chamberlain @ 2023-06-30 22:11 UTC (permalink / raw)
To: Steven Rostedt, Julia Lawall, Takashi Iwai
Cc: Latchesar Ionkov, Rafael J. Wysocki, Darrick J. Wong,
Anders Larsen, Carlos Llamas, Hugh Dickins, John Johansen,
Seth Forshee, Alexander Gordeev, Christoph Hellwig,
Mike Marshall, Paulo Alcantara, linux-xfs, Bart Van Assche,
Michael Ellerman, John Keeping, Zhang Yi, James Morris,
Christophe Leroy, Tyler Hicks, Alan Stern, Christian Borntraeger,
devel, Shyam Prasad N, Jan Harkes, linux-um, Nicholas Piggin
On Wed, Jun 21, 2023 at 03:21:41PM -0400, Steven Rostedt wrote:
> On Wed, 21 Jun 2023 10:45:05 -0400
> Jeff Layton <jlayton@kernel.org> wrote:
>
> > Most of this conversion was done via coccinelle, with a few of the more
> > non-standard accesses done by hand. There should be no behavioral
> > changes with this set. That will come later, as we convert individual
> > filesystems to use multigrain timestamps.
>
> BTW, Linus has suggested to me that whenever a conccinelle script is used,
> it should be included in the change log.
Sometimes people like the coccinelle included in the commit, sometimes
people don't [0], it really ends up being up to a subjective maintainer
preference. A compromise could be to use git notes as these are
optional, however if we want to go down that path we should try to make
a general consensus on it so we can send a consistent message.
[0] https://lore.kernel.org/all/20230512073100.GC32559@twin.jikos.cz/
Luis
^ permalink raw reply [flat|nested] 229+ messages in thread
* [Cluster-devel] [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-30 22:11 ` Luis Chamberlain
0 siblings, 0 replies; 229+ messages in thread
From: Luis Chamberlain @ 2023-06-30 22:11 UTC (permalink / raw)
To: cluster-devel.redhat.com
On Wed, Jun 21, 2023 at 03:21:41PM -0400, Steven Rostedt wrote:
> On Wed, 21 Jun 2023 10:45:05 -0400
> Jeff Layton <jlayton@kernel.org> wrote:
>
> > Most of this conversion was done via coccinelle, with a few of the more
> > non-standard accesses done by hand. There should be no behavioral
> > changes with this set. That will come later, as we convert individual
> > filesystems to use multigrain timestamps.
>
> BTW, Linus has suggested to me that whenever a conccinelle script is used,
> it should be included in the change log.
Sometimes people like the coccinelle included in the commit, sometimes
people don't [0], it really ends up being up to a subjective maintainer
preference. A compromise could be to use git notes as these are
optional, however if we want to go down that path we should try to make
a general consensus on it so we can send a consistent message.
[0] https://lore.kernel.org/all/20230512073100.GC32559 at twin.jikos.cz/
Luis
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 00/79] fs: new accessors for inode->i_ctime
@ 2023-06-30 22:11 ` Luis Chamberlain
0 siblings, 0 replies; 229+ messages in thread
From: Luis Chamberlain @ 2023-06-30 22:11 UTC (permalink / raw)
To: Steven Rostedt, Julia Lawall, Takashi Iwai
Cc: Jeff Layton, Jeremy Kerr, Arnd Bergmann, Michael Ellerman,
Nicholas Piggin, Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan
On Wed, Jun 21, 2023 at 03:21:41PM -0400, Steven Rostedt wrote:
> On Wed, 21 Jun 2023 10:45:05 -0400
> Jeff Layton <jlayton@kernel.org> wrote:
>
> > Most of this conversion was done via coccinelle, with a few of the more
> > non-standard accesses done by hand. There should be no behavioral
> > changes with this set. That will come later, as we convert individual
> > filesystems to use multigrain timestamps.
>
> BTW, Linus has suggested to me that whenever a conccinelle script is used,
> it should be included in the change log.
Sometimes people like the coccinelle included in the commit, sometimes
people don't [0], it really ends up being up to a subjective maintainer
preference. A compromise could be to use git notes as these are
optional, however if we want to go down that path we should try to make
a general consensus on it so we can send a consistent message.
[0] https://lore.kernel.org/all/20230512073100.GC32559@twin.jikos.cz/
Luis
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
2023-06-21 14:45 ` [f2fs-dev] " Jeff Layton
` (3 preceding siblings ...)
(?)
@ 2023-06-30 22:12 ` Luis Chamberlain
-1 siblings, 0 replies; 229+ messages in thread
From: Luis Chamberlain @ 2023-06-30 22:12 UTC (permalink / raw)
To: Jeff Layton
Cc: Jeremy Kerr, Arnd Bergmann, Michael Ellerman, Nicholas Piggin,
Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro, Jason Gunthorpe,
Leon Romanovsky, Brad Warrum, Ritu Agarwal, Eric Van Hensbergen,
Latchesar Ionkov, Dominique Martinet, Christian Schoenebeck,
David Sterba, David Howells, Marc Dionne, Alexander Viro,
Ian Kent, Luis de Bethencourt, Salah Triki, Tigran A. Aivazian,
Eric Biederman, Kees Cook, Chris Mason, Josef Bacik, Xiubo Li,
Ilya Dryomov, Jan Harkes, coda, Joel Becker, Christoph Hellwig,
Nicolas Pitre, Rafael J. Wysocki, Tyler Hicks, Ard Biesheuvel,
Gao Xiang, Chao Yu, Yue Hu, Jeffle Xu, Namjae Jeon, Sungjong Seo,
Jan Kara, Theodore Ts'o, Andreas Dilger, Jaegeuk Kim,
OGAWA Hirofumi, Miklos Szeredi, Bob Peterson,
Andreas Gruenbacher, Richard Weinberger, Anton Ivanov,
Johannes Berg, Mikulas Patocka, Mike Kravetz, Muchun Song,
David Woodhouse, Dave Kleikamp, Tejun Heo, Trond Myklebust,
Anna Schumaker, Chuck Lever, Ryusuke Konishi, Anton Altaparmakov,
Konstantin Komarov, Mark Fasheh, Joseph Qi, Bob Copeland,
Mike Marshall, Martin Brandenburg, Iurii Zaikin, Tony Luck,
Guilherme G. Piccoli, Anders Larsen, Steve French,
Paulo Alcantara, Ronnie Sahlberg, Shyam Prasad N, Tom Talpey,
Sergey Senozhatsky, Phillip Lougher, Steven Rostedt,
Masami Hiramatsu, Evgeniy Dushistov, Hans de Goede,
Darrick J. Wong, Damien Le Moal, Naohiro Aota,
Johannes Thumshirn, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa,
Hugh Dickins, Andrew Morton, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, John Johansen, Paul Moore,
James Morris, Serge E. Hallyn, Stephen Smalley, Eric Paris,
Juergen Gross, Ruihan Li, Laurent Pinchart, Wolfram Sang,
Udipto Goswami, Linyu Yuan, John Keeping, Andrzej Pietrasiewicz,
Dan Carpenter, Yuta Hayama, Jozef Martiniak, Jens Axboe,
Alan Stern, Sandeep Dhavale, Dave Chinner, Johannes Weiner,
ZhangPeng, Viacheslav Dubeyko, Tetsuo Handa, Aditya Garg,
Erez Zadok, Yifei Liu, Yu Zhe, Matthew Wilcox (Oracle),
Oleg Kanatov, Dr. David Alan Gilbert, Jiangshan Yi, xu xin,
Stefan Roesch, Zhihao Cheng, Liam R. Howlett, Alexey Dobriyan,
Minghao Chi, Seth Forshee, Zeng Jingxiang, Bart Van Assche,
Mimi Zohar, Roberto Sassu, Zhang Yi, Tom Rix,
Fabio M. De Francesco, Chen Zhongjin, Zhengchao Shao,
Rik van Riel, Jingyu Wang, Hangyu Hua, linuxppc-dev,
linux-kernel, linux-s390, linux-rdma, linux-usb, v9fs,
linux-fsdevel, linux-afs, autofs, linux-mm, linux-btrfs,
ceph-devel, codalist, ecryptfs, linux-efi, linux-erofs,
linux-ext4, linux-f2fs-devel, cluster-devel, linux-um, linux-mtd,
jfs-discussion, linux-nfs, linux-nilfs, linux-ntfs-dev, ntfs3,
ocfs2-devel, linux-karma-devel, devel, linux-unionfs,
linux-hardening, reiserfs-devel, linux-cifs, samba-technical,
linux-trace-kernel, linux-xfs, bpf, netdev, apparmor,
linux-security-module, selinux
On Wed, Jun 21, 2023 at 10:45:06AM -0400, Jeff Layton wrote:
> struct timespec64 has unused bits in the tv_nsec field that can be used
> for other purposes. In future patches, we're going to change how the
> inode->i_ctime is accessed in certain inodes in order to make use of
> them. In order to do that safely though, we'll need to eradicate raw
> accesses of the inode->i_ctime field from the kernel.
>
> Add new accessor functions for the ctime that we can use to replace them.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Luis
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [f2fs-dev] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-30 22:12 ` Luis Chamberlain
0 siblings, 0 replies; 229+ messages in thread
From: Luis Chamberlain @ 2023-06-30 22:12 UTC (permalink / raw)
To: Jeff Layton
Cc: Latchesar Ionkov, Rafael J. Wysocki, Darrick J. Wong,
Anders Larsen, Carlos Llamas, Hugh Dickins, John Johansen,
Seth Forshee, Alexander Gordeev, Christoph Hellwig,
Mike Marshall, Paulo Alcantara, linux-xfs, Bart Van Assche,
Michael Ellerman, John Keeping, Zhang Yi, James Morris,
Christophe Leroy, Tyler Hicks, Alan Stern, Christian Borntraeger,
devel, Shyam Prasad N, Jan Harkes, linux-um, Nicholas Piggin,
Alexander Viro, Eric Van Hensbergen, Suren Baghdasaryan,
Trond Myklebust, Anton Altaparmakov, Christian Brauner,
Wolfram Sang, Greg Kroah-Hartman, Stephen Smalley, linux-usb,
linux-kernel, Ronnie Sahlberg, Sergey Senozhatsky,
Arve Hjønnevåg, Chuck Lever, Sven Schnelle, Jiri Olsa,
Jan Kara, Tejun Heo, Andrew Morton, linux-trace-kernel,
linux-hardening, Dave Kleikamp, Sandeep Dhavale, Tetsuo Handa,
Mimi Zohar, linux-mm, Joel Fernandes, Eric Dumazet,
Stanislav Fomichev, Andrzej Pietrasiewicz, Hangyu Hua,
linux-s390, linux-nilfs, Paul Moore, Leon Romanovsky,
John Fastabend, Andrii Nakryiko, Minghao Chi, codalist, selinux,
ZhangPeng, Udipto Goswami, Yonghong Song, Iurii Zaikin,
Namjae Jeon, Masami Hiramatsu, ecryptfs, Todd Kjos,
Vasily Gorbik, Yu Zhe, linuxppc-dev, reiserfs-devel,
Miklos Szeredi, Yue Hu, Jaegeuk Kim, Aditya Garg, Martijn Coenen,
OGAWA Hirofumi, Hao Luo, Tony Luck, Theodore Ts'o,
Nicolas Pitre, linux-ntfs-dev, Muchun Song, Roberto Sassu,
linux-f2fs-devel, Guilherme G. Piccoli, Jozef Martiniak,
Eric Biederman, Anna Schumaker, xu xin, Brad Warrum,
Mike Kravetz, Jingyu Wang, linux-efi, Dan Carpenter,
Martin Brandenburg, Tom Rix, Alexei Starovoitov, Chris Mason,
linux-mtd, Matthew Wilcox (Oracle),
Marc Dionne, linux-afs, Ian Kent, Naohiro Aota, Daniel Borkmann,
Dennis Dalessandro, linux-rdma, Linyu Yuan, coda,
Viacheslav Dubeyko, Ilya Dryomov, Paolo Abeni, Alexey Dobriyan,
Serge E. Hallyn, Jens Axboe, Zeng Jingxiang, Kees Cook,
Arnd Bergmann, autofs, Steven Rostedt, Yifei Liu, Damien Le Moal,
Eric Paris, ceph-devel, Gao Xiang, Jiangshan Yi, David Howells,
linux-nfs, linux-ext4, Song Liu, samba-technical, Steve French,
Jeremy Kerr, netdev, Bob Peterson, linux-fsdevel, bpf, ntfs3,
linux-erofs, David S. Miller, ocfs2-devel, jfs-discussion,
Dominique Martinet, Christian Schoenebeck, Bob Copeland,
KP Singh, Oleg Kanatov, Konstantin Komarov, Joseph Qi,
Yuta Hayama, Andreas Dilger, Mikulas Patocka, Zhengchao Shao,
Chen Zhongjin, Ard Biesheuvel, Anton Ivanov, Laurent Pinchart,
Andreas Gruenbacher, Richard Weinberger, Mark Fasheh,
Stefan Roesch, cluster-devel, Jason Gunthorpe, Jakub Kicinski,
Rik van Riel, Salah Triki, Evgeniy Dushistov, linux-cifs,
Heiko Carstens, apparmor, Josef Bacik, Liam R. Howlett,
Tom Talpey, Hans de Goede, Tigran A. Aivazian, Dave Chinner,
David Sterba, Xiubo Li, Ryusuke Konishi, Juergen Gross,
Johannes Thumshirn, Ritu Agarwal, Luis de Bethencourt,
Martin KaFai Lau, v9fs, Fabio M. De Francesco, linux-unionfs,
Ruihan Li, linux-security-module, Erez Zadok, Jeffle Xu,
Dr. David Alan Gilbert, Johannes Weiner, Phillip Lougher,
Johannes Berg, Sungjong Seo, David Woodhouse, linux-karma-devel,
linux-btrfs, Joel Becker
On Wed, Jun 21, 2023 at 10:45:06AM -0400, Jeff Layton wrote:
> struct timespec64 has unused bits in the tv_nsec field that can be used
> for other purposes. In future patches, we're going to change how the
> inode->i_ctime is accessed in certain inodes in order to make use of
> them. In order to do that safely though, we'll need to eradicate raw
> accesses of the inode->i_ctime field from the kernel.
>
> Add new accessor functions for the ctime that we can use to replace them.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Luis
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-30 22:12 ` Luis Chamberlain
0 siblings, 0 replies; 229+ messages in thread
From: Luis Chamberlain @ 2023-06-30 22:12 UTC (permalink / raw)
To: Jeff Layton
Cc: Latchesar Ionkov, Rafael J. Wysocki, Darrick J. Wong,
Anders Larsen, Carlos Llamas, Hugh Dickins, John Johansen,
Seth Forshee, Alexander Gordeev, Christoph Hellwig,
Mike Marshall, Paulo Alcantara, linux-xfs, Bart Van Assche,
John Keeping, Zhang Yi, James Morris, Tyler Hicks, Alan Stern,
Christian Borntraeger, devel, Shyam Prasad N, Jan Harkes,
linux-um, Nicholas Piggin, Alexander Viro, Eric Van Hensbergen,
Suren Baghdasaryan, Trond Myklebust, Anton Altaparmakov,
Christian Brauner, Wolfram Sang, Greg Kroah-Hartman,
Stephen Smalley, linux-usb, linux-kernel, Ronnie Sahlberg,
Sergey Senozhatsky, Arve Hjønnevåg, Chuck Lever,
Sven Schnelle, Jiri Olsa, Jan Kara, Tejun Heo, Andrew Morton,
linux-trace-kernel, linux-hardening, Dave Kleikamp,
Sandeep Dhavale, Tetsuo Handa, Mimi Zohar, linux-mm,
Joel Fernandes
On Wed, Jun 21, 2023 at 10:45:06AM -0400, Jeff Layton wrote:
> struct timespec64 has unused bits in the tv_nsec field that can be used
> for other purposes. In future patches, we're going to change how the
> inode->i_ctime is accessed in certain inodes in order to make use of
> them. In order to do that safely though, we'll need to eradicate raw
> accesses of the inode->i_ctime field from the kernel.
>
> Add new accessor functions for the ctime that we can use to replace them.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Luis
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-30 22:12 ` Luis Chamberlain
0 siblings, 0 replies; 229+ messages in thread
From: Luis Chamberlain @ 2023-06-30 22:12 UTC (permalink / raw)
To: Jeff Layton
Cc: Latchesar Ionkov, Rafael J. Wysocki, Darrick J. Wong,
Anders Larsen, Carlos Llamas, Hugh Dickins, John Johansen,
Seth Forshee, Alexander Gordeev, Christoph Hellwig,
Mike Marshall, Paulo Alcantara, linux-xfs, Bart Van Assche,
Michael Ellerman, John Keeping, Zhang Yi, James Morris,
Christophe Leroy, Tyler Hicks, Alan Stern, Christian Borntraeger,
devel, Shyam Prasad N, Jan Harkes, linux-um, Nicholas Piggin
On Wed, Jun 21, 2023 at 10:45:06AM -0400, Jeff Layton wrote:
> struct timespec64 has unused bits in the tv_nsec field that can be used
> for other purposes. In future patches, we're going to change how the
> inode->i_ctime is accessed in certain inodes in order to make use of
> them. In order to do that safely though, we'll need to eradicate raw
> accesses of the inode->i_ctime field from the kernel.
>
> Add new accessor functions for the ctime that we can use to replace them.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Luis
^ permalink raw reply [flat|nested] 229+ messages in thread
* [Cluster-devel] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-30 22:12 ` Luis Chamberlain
0 siblings, 0 replies; 229+ messages in thread
From: Luis Chamberlain @ 2023-06-30 22:12 UTC (permalink / raw)
To: cluster-devel.redhat.com
On Wed, Jun 21, 2023 at 10:45:06AM -0400, Jeff Layton wrote:
> struct timespec64 has unused bits in the tv_nsec field that can be used
> for other purposes. In future patches, we're going to change how the
> inode->i_ctime is accessed in certain inodes in order to make use of
> them. In order to do that safely though, we'll need to eradicate raw
> accesses of the inode->i_ctime field from the kernel.
>
> Add new accessor functions for the ctime that we can use to replace them.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Luis
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-06-30 22:12 ` Luis Chamberlain
0 siblings, 0 replies; 229+ messages in thread
From: Luis Chamberlain @ 2023-06-30 22:12 UTC (permalink / raw)
To: Jeff Layton
Cc: Jeremy Kerr, Arnd Bergmann, Michael Ellerman, Nicholas Piggin,
Christophe Leroy, Heiko Carstens, Vasily Gorbik,
Alexander Gordeev, Christian Borntraeger, Sven Schnelle,
Greg Kroah-Hartman, Arve Hjønnevåg, Todd Kjos,
Martijn Coenen, Joel Fernandes, Christian Brauner, Carlos Llamas,
Suren Baghdasaryan, Dennis Dalessandro
On Wed, Jun 21, 2023 at 10:45:06AM -0400, Jeff Layton wrote:
> struct timespec64 has unused bits in the tv_nsec field that can be used
> for other purposes. In future patches, we're going to change how the
> inode->i_ctime is accessed in certain inodes in order to make use of
> them. In order to do that safely though, we'll need to eradicate raw
> accesses of the inode->i_ctime field from the kernel.
>
> Add new accessor functions for the ctime that we can use to replace them.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Luis
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 55/79] proc: switch to new ctime accessors
2023-06-21 14:46 ` [PATCH 55/79] proc: " Jeff Layton
@ 2023-06-30 22:13 ` Luis Chamberlain
0 siblings, 0 replies; 229+ messages in thread
From: Luis Chamberlain @ 2023-06-30 22:13 UTC (permalink / raw)
To: Jeff Layton
Cc: Christian Brauner, Kees Cook, Iurii Zaikin, Al Viro, Jan Kara,
linux-kernel, linux-fsdevel
On Wed, Jun 21, 2023 at 10:46:08AM -0400, Jeff Layton wrote:
> In later patches, we're going to change how the ctime.tv_nsec field is
> utilized. Switch to using accessor functions instead of raw accesses of
> inode->i_ctime.
>
> Signed-off-by: Jeff Layton <jlayton@kernel.org>
Acked-by: Luis Chamberlain <mcgrof@kernel.org>
Luis
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 15/79] bfs: switch to new ctime accessors
2023-06-23 12:33 ` Christian Brauner
@ 2023-07-03 10:12 ` Christian Brauner
2023-07-03 10:46 ` Jeff Layton
0 siblings, 1 reply; 229+ messages in thread
From: Christian Brauner @ 2023-07-03 10:12 UTC (permalink / raw)
To: Jeff Layton
Cc: Kara, Tigran A. Aivazian, Al Viro, linux-kernel, damien.lemoal,
linux-fsdevel
On Fri, Jun 23, 2023 at 02:33:26PM +0200, Christian Brauner wrote:
> On Thu, Jun 22, 2023 at 04:57:47PM +0200, Jan Kara wrote:
> > On Thu 22-06-23 08:51:58, Jeff Layton wrote:
> > > On Thu, 2023-06-22 at 14:30 +0200, Jan Kara wrote:
> > > > On Wed 21-06-23 12:57:19, Jeff Layton wrote:
> > > > > On Wed, 2023-06-21 at 18:48 +0200, Jan Kara wrote:
> > > > > > On Wed 21-06-23 10:45:28, Jeff Layton wrote:
> > > > > > > In later patches, we're going to change how the ctime.tv_nsec field is
> > > > > > > utilized. Switch to using accessor functions instead of raw accesses of
> > > > > > > inode->i_ctime.
> > > > > > >
> > > > > > > Signed-off-by: Jeff Layton <jlayton@kernel.org>
> > > > > >
> > > > > > ...
> > > > > >
> > > > > > > diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c
> > > > > > > index 1926bec2c850..c964316be32b 100644
> > > > > > > --- a/fs/bfs/inode.c
> > > > > > > +++ b/fs/bfs/inode.c
> > > > > > > @@ -82,10 +82,10 @@ struct inode *bfs_iget(struct super_block *sb, unsigned long ino)
> > > > > > > inode->i_blocks = BFS_FILEBLOCKS(di);
> > > > > > > inode->i_atime.tv_sec = le32_to_cpu(di->i_atime);
> > > > > > > inode->i_mtime.tv_sec = le32_to_cpu(di->i_mtime);
> > > > > > > - inode->i_ctime.tv_sec = le32_to_cpu(di->i_ctime);
> > > > > > > + inode_ctime_set_sec(inode, le32_to_cpu(di->i_ctime));
> > > > > > > inode->i_atime.tv_nsec = 0;
> > > > > > > inode->i_mtime.tv_nsec = 0;
> > > > > > > - inode->i_ctime.tv_nsec = 0;
> > > > > > > + inode_ctime_set_nsec(inode, 0);
> > > > > >
> > > > > > So I'm somewhat wondering here - in other filesystem you construct
> > > > > > timespec64 and then use inode_ctime_set(). Here you use
> > > > > > inode_ctime_set_sec() + inode_ctime_set_nsec(). What's the benefit? It
> > > > > > seems these two functions are not used that much some maybe we could just
> > > > > > live with just inode_ctime_set() and constructing timespec64 when needed?
> > > > > >
> > > > > > Honza
> > > > >
> > > > > The main advantage is that by using that, I didn't need to do quite so
> > > > > much of this conversion by hand. My coccinelle skills are pretty
> > > > > primitive. I went with whatever conversion was going to give minimal
> > > > > changes, to the existing accesses for the most part.
> > > > >
> > > > > We could certainly do it the way you suggest, it just means having to
> > > > > re-touch a lot of this code by hand, or someone with better coccinelle
> > > > > chops suggesting a way to declare a temporary variables in place.
> > > >
> > > > Well, maybe temporary variables aren't that convenient but we could provide
> > > > function setting ctime from sec & nsec value without having to declare
> > > > temporary timespec64? Attached is a semantic patch that should deal with
> > > > that - at least it seems to handle all the cases I've found.
> > > >
> > >
> > > Ok, let me try respinning this with your cocci script and see how it
> > > looks.
> > >
> > > Damien also suggested in a reply to the zonefs patch a preference for
> > > the naming style you have above. Should I also rename these like?
> > >
> > > inode_ctime_peek -> inode_get_ctime
> > > inode_ctime_set -> inode_set_ctime
> > >
> > > This would be the time to change it if that's preferred.
> >
> > I don't really care much so whatever you decide is better :)
>
> I have a mild preference for inode_{get,set}_ctime().
Jeff, did you plan on sending a v2 with this renamed or do you want me
to pick this up now?
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 15/79] bfs: switch to new ctime accessors
2023-07-03 10:12 ` Christian Brauner
@ 2023-07-03 10:46 ` Jeff Layton
2023-07-03 10:57 ` Christian Brauner
0 siblings, 1 reply; 229+ messages in thread
From: Jeff Layton @ 2023-07-03 10:46 UTC (permalink / raw)
To: Christian Brauner
Cc: Kara, Tigran A. Aivazian, Al Viro, linux-kernel, damien.lemoal,
linux-fsdevel
On Mon, 2023-07-03 at 12:12 +0200, Christian Brauner wrote:
> On Fri, Jun 23, 2023 at 02:33:26PM +0200, Christian Brauner wrote:
> > On Thu, Jun 22, 2023 at 04:57:47PM +0200, Jan Kara wrote:
> > > On Thu 22-06-23 08:51:58, Jeff Layton wrote:
> > > > On Thu, 2023-06-22 at 14:30 +0200, Jan Kara wrote:
> > > > > On Wed 21-06-23 12:57:19, Jeff Layton wrote:
> > > > > > On Wed, 2023-06-21 at 18:48 +0200, Jan Kara wrote:
> > > > > > > On Wed 21-06-23 10:45:28, Jeff Layton wrote:
> > > > > > > > In later patches, we're going to change how the ctime.tv_nsec field is
> > > > > > > > utilized. Switch to using accessor functions instead of raw accesses of
> > > > > > > > inode->i_ctime.
> > > > > > > >
> > > > > > > > Signed-off-by: Jeff Layton <jlayton@kernel.org>
> > > > > > >
> > > > > > > ...
> > > > > > >
> > > > > > > > diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c
> > > > > > > > index 1926bec2c850..c964316be32b 100644
> > > > > > > > --- a/fs/bfs/inode.c
> > > > > > > > +++ b/fs/bfs/inode.c
> > > > > > > > @@ -82,10 +82,10 @@ struct inode *bfs_iget(struct super_block *sb, unsigned long ino)
> > > > > > > > inode->i_blocks = BFS_FILEBLOCKS(di);
> > > > > > > > inode->i_atime.tv_sec = le32_to_cpu(di->i_atime);
> > > > > > > > inode->i_mtime.tv_sec = le32_to_cpu(di->i_mtime);
> > > > > > > > - inode->i_ctime.tv_sec = le32_to_cpu(di->i_ctime);
> > > > > > > > + inode_ctime_set_sec(inode, le32_to_cpu(di->i_ctime));
> > > > > > > > inode->i_atime.tv_nsec = 0;
> > > > > > > > inode->i_mtime.tv_nsec = 0;
> > > > > > > > - inode->i_ctime.tv_nsec = 0;
> > > > > > > > + inode_ctime_set_nsec(inode, 0);
> > > > > > >
> > > > > > > So I'm somewhat wondering here - in other filesystem you construct
> > > > > > > timespec64 and then use inode_ctime_set(). Here you use
> > > > > > > inode_ctime_set_sec() + inode_ctime_set_nsec(). What's the benefit? It
> > > > > > > seems these two functions are not used that much some maybe we could just
> > > > > > > live with just inode_ctime_set() and constructing timespec64 when needed?
> > > > > > >
> > > > > > > Honza
> > > > > >
> > > > > > The main advantage is that by using that, I didn't need to do quite so
> > > > > > much of this conversion by hand. My coccinelle skills are pretty
> > > > > > primitive. I went with whatever conversion was going to give minimal
> > > > > > changes, to the existing accesses for the most part.
> > > > > >
> > > > > > We could certainly do it the way you suggest, it just means having to
> > > > > > re-touch a lot of this code by hand, or someone with better coccinelle
> > > > > > chops suggesting a way to declare a temporary variables in place.
> > > > >
> > > > > Well, maybe temporary variables aren't that convenient but we could provide
> > > > > function setting ctime from sec & nsec value without having to declare
> > > > > temporary timespec64? Attached is a semantic patch that should deal with
> > > > > that - at least it seems to handle all the cases I've found.
> > > > >
> > > >
> > > > Ok, let me try respinning this with your cocci script and see how it
> > > > looks.
> > > >
> > > > Damien also suggested in a reply to the zonefs patch a preference for
> > > > the naming style you have above. Should I also rename these like?
> > > >
> > > > inode_ctime_peek -> inode_get_ctime
> > > > inode_ctime_set -> inode_set_ctime
> > > >
> > > > This would be the time to change it if that's preferred.
> > >
> > > I don't really care much so whatever you decide is better :)
> >
> > I have a mild preference for inode_{get,set}_ctime().
>
> Jeff, did you plan on sending a v2 with this renamed or do you want me
> to pick this up now?
I'm working on a new set that I'll send out in a few days. Sorry it has
taken a while, I spent quite a bit of time trying to improve my
coccinelle chops for this.
Thanks,
--
Jeff Layton <jlayton@kernel.org>
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 15/79] bfs: switch to new ctime accessors
2023-07-03 10:46 ` Jeff Layton
@ 2023-07-03 10:57 ` Christian Brauner
0 siblings, 0 replies; 229+ messages in thread
From: Christian Brauner @ 2023-07-03 10:57 UTC (permalink / raw)
To: Jeff Layton
Cc: Kara, Tigran A. Aivazian, Al Viro, linux-kernel, damien.lemoal,
linux-fsdevel
On Mon, Jul 03, 2023 at 06:46:33AM -0400, Jeff Layton wrote:
> On Mon, 2023-07-03 at 12:12 +0200, Christian Brauner wrote:
> > On Fri, Jun 23, 2023 at 02:33:26PM +0200, Christian Brauner wrote:
> > > On Thu, Jun 22, 2023 at 04:57:47PM +0200, Jan Kara wrote:
> > > > On Thu 22-06-23 08:51:58, Jeff Layton wrote:
> > > > > On Thu, 2023-06-22 at 14:30 +0200, Jan Kara wrote:
> > > > > > On Wed 21-06-23 12:57:19, Jeff Layton wrote:
> > > > > > > On Wed, 2023-06-21 at 18:48 +0200, Jan Kara wrote:
> > > > > > > > On Wed 21-06-23 10:45:28, Jeff Layton wrote:
> > > > > > > > > In later patches, we're going to change how the ctime.tv_nsec field is
> > > > > > > > > utilized. Switch to using accessor functions instead of raw accesses of
> > > > > > > > > inode->i_ctime.
> > > > > > > > >
> > > > > > > > > Signed-off-by: Jeff Layton <jlayton@kernel.org>
> > > > > > > >
> > > > > > > > ...
> > > > > > > >
> > > > > > > > > diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c
> > > > > > > > > index 1926bec2c850..c964316be32b 100644
> > > > > > > > > --- a/fs/bfs/inode.c
> > > > > > > > > +++ b/fs/bfs/inode.c
> > > > > > > > > @@ -82,10 +82,10 @@ struct inode *bfs_iget(struct super_block *sb, unsigned long ino)
> > > > > > > > > inode->i_blocks = BFS_FILEBLOCKS(di);
> > > > > > > > > inode->i_atime.tv_sec = le32_to_cpu(di->i_atime);
> > > > > > > > > inode->i_mtime.tv_sec = le32_to_cpu(di->i_mtime);
> > > > > > > > > - inode->i_ctime.tv_sec = le32_to_cpu(di->i_ctime);
> > > > > > > > > + inode_ctime_set_sec(inode, le32_to_cpu(di->i_ctime));
> > > > > > > > > inode->i_atime.tv_nsec = 0;
> > > > > > > > > inode->i_mtime.tv_nsec = 0;
> > > > > > > > > - inode->i_ctime.tv_nsec = 0;
> > > > > > > > > + inode_ctime_set_nsec(inode, 0);
> > > > > > > >
> > > > > > > > So I'm somewhat wondering here - in other filesystem you construct
> > > > > > > > timespec64 and then use inode_ctime_set(). Here you use
> > > > > > > > inode_ctime_set_sec() + inode_ctime_set_nsec(). What's the benefit? It
> > > > > > > > seems these two functions are not used that much some maybe we could just
> > > > > > > > live with just inode_ctime_set() and constructing timespec64 when needed?
> > > > > > > >
> > > > > > > > Honza
> > > > > > >
> > > > > > > The main advantage is that by using that, I didn't need to do quite so
> > > > > > > much of this conversion by hand. My coccinelle skills are pretty
> > > > > > > primitive. I went with whatever conversion was going to give minimal
> > > > > > > changes, to the existing accesses for the most part.
> > > > > > >
> > > > > > > We could certainly do it the way you suggest, it just means having to
> > > > > > > re-touch a lot of this code by hand, or someone with better coccinelle
> > > > > > > chops suggesting a way to declare a temporary variables in place.
> > > > > >
> > > > > > Well, maybe temporary variables aren't that convenient but we could provide
> > > > > > function setting ctime from sec & nsec value without having to declare
> > > > > > temporary timespec64? Attached is a semantic patch that should deal with
> > > > > > that - at least it seems to handle all the cases I've found.
> > > > > >
> > > > >
> > > > > Ok, let me try respinning this with your cocci script and see how it
> > > > > looks.
> > > > >
> > > > > Damien also suggested in a reply to the zonefs patch a preference for
> > > > > the naming style you have above. Should I also rename these like?
> > > > >
> > > > > inode_ctime_peek -> inode_get_ctime
> > > > > inode_ctime_set -> inode_set_ctime
> > > > >
> > > > > This would be the time to change it if that's preferred.
> > > >
> > > > I don't really care much so whatever you decide is better :)
> > >
> > > I have a mild preference for inode_{get,set}_ctime().
> >
> > Jeff, did you plan on sending a v2 with this renamed or do you want me
> > to pick this up now?
>
> I'm working on a new set that I'll send out in a few days. Sorry it has
> taken a while, I spent quite a bit of time trying to improve my
> coccinelle chops for this.
Absolutely no problem of course. I just wanted to check that I didn't
pointlessly delay you by not taking care of this.
Thanks!
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
2023-06-21 14:45 ` [f2fs-dev] " Jeff Layton
(?)
(?)
@ 2023-07-12 15:31 ` Randy Dunlap
-1 siblings, 0 replies; 229+ messages in thread
From: Randy Dunlap @ 2023-07-12 15:31 UTC (permalink / raw)
To: Jeff Layton, linux-kernel, Linux FS-devel Mailing List, linux-um
Hi Jeff,
On arch/um/, (subarch i386 or x86_64), hostfs build fails with:
../fs/hostfs/hostfs_kern.c:520:36: error: incompatible type for arg
ument 2 of 'inode_set_ctime_to_ts'
../include/linux/fs.h:1499:73: note: expected 'struct timespec64' b
ut argument is of type 'const struct hostfs_timespec *'
--
~Randy
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-07-12 15:31 ` Randy Dunlap
0 siblings, 0 replies; 229+ messages in thread
From: Randy Dunlap @ 2023-07-12 15:31 UTC (permalink / raw)
To: Jeff Layton, linux-kernel, Linux FS-devel Mailing List, linux-um
Hi Jeff,
On arch/um/, (subarch i386 or x86_64), hostfs build fails with:
../fs/hostfs/hostfs_kern.c:520:36: error: incompatible type for arg
ument 2 of 'inode_set_ctime_to_ts'
../include/linux/fs.h:1499:73: note: expected 'struct timespec64' b
ut argument is of type 'const struct hostfs_timespec *'
--
~Randy
_______________________________________________
linux-um mailing list
linux-um@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-um
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-07-12 15:31 ` Randy Dunlap
0 siblings, 0 replies; 229+ messages in thread
From: Randy Dunlap @ 2023-07-12 15:31 UTC (permalink / raw)
To: Jeff Layton, linux-kernel, Linux FS-devel Mailing List, linux-um
Hi Jeff,
On arch/um/, (subarch i386 or x86_64), hostfs build fails with:
../fs/hostfs/hostfs_kern.c:520:36: error: incompatible type for arg
ument 2 of 'inode_set_ctime_to_ts'
../include/linux/fs.h:1499:73: note: expected 'struct timespec64' b
ut argument is of type 'const struct hostfs_timespec *'
--
~Randy
______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/
^ permalink raw reply [flat|nested] 229+ messages in thread
* [Cluster-devel] [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-07-12 15:31 ` Randy Dunlap
0 siblings, 0 replies; 229+ messages in thread
From: Randy Dunlap @ 2023-07-12 15:31 UTC (permalink / raw)
To: cluster-devel.redhat.com
Hi Jeff,
On arch/um/, (subarch i386 or x86_64), hostfs build fails with:
../fs/hostfs/hostfs_kern.c:520:36: error: incompatible type for arg
ument 2 of 'inode_set_ctime_to_ts'
../include/linux/fs.h:1499:73: note: expected 'struct timespec64' b
ut argument is of type 'const struct hostfs_timespec *'
--
~Randy
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
2023-06-21 14:45 ` [f2fs-dev] " Jeff Layton
` (8 preceding siblings ...)
(?)
@ 2023-07-12 15:31 ` Randy Dunlap
-1 siblings, 0 replies; 229+ messages in thread
From: Randy Dunlap @ 2023-07-12 15:31 UTC (permalink / raw)
To: Jeff Layton, linux-kernel, Linux FS-devel Mailing List, linux-um
Hi Jeff,
On arch/um/, (subarch i386 or x86_64), hostfs build fails with:
../fs/hostfs/hostfs_kern.c:520:36: error: incompatible type for arg
ument 2 of 'inode_set_ctime_to_ts'
../include/linux/fs.h:1499:73: note: expected 'struct timespec64' b
ut argument is of type 'const struct hostfs_timespec *'
--
~Randy
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
[not found] ` <20230621144507.55591-2-jlayton-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
@ 2023-07-12 15:31 ` Randy Dunlap
0 siblings, 0 replies; 229+ messages in thread
From: Randy Dunlap @ 2023-07-12 15:31 UTC (permalink / raw)
To: Jeff Layton, linux-kernel-u79uwXL29TY76Z2rM5mHXA,
Linux FS-devel Mailing List, linux-um
Hi Jeff,
On arch/um/, (subarch i386 or x86_64), hostfs build fails with:
../fs/hostfs/hostfs_kern.c:520:36: error: incompatible type for arg
ument 2 of 'inode_set_ctime_to_ts'
../include/linux/fs.h:1499:73: note: expected 'struct timespec64' b
ut argument is of type 'const struct hostfs_timespec *'
--
~Randy
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
2023-07-12 15:31 ` Randy Dunlap
@ 2023-07-12 15:34 ` Christian Brauner
-1 siblings, 0 replies; 229+ messages in thread
From: Christian Brauner @ 2023-07-12 15:34 UTC (permalink / raw)
To: Randy Dunlap
Cc: Jeff Layton, linux-kernel, Linux FS-devel Mailing List, linux-um
On Wed, Jul 12, 2023 at 08:31:55AM -0700, Randy Dunlap wrote:
> Hi Jeff,
>
> On arch/um/, (subarch i386 or x86_64), hostfs build fails with:
>
> ../fs/hostfs/hostfs_kern.c:520:36: error: incompatible type for arg
> ument 2 of 'inode_set_ctime_to_ts'
> ../include/linux/fs.h:1499:73: note: expected 'struct timespec64' b
> ut argument is of type 'const struct hostfs_timespec *'
Thanks for notifying us! We fixed this earlier today. The vfs.ctime
branch has been updated with the fix folded into the hostfs patch.
^ permalink raw reply [flat|nested] 229+ messages in thread
* Re: [PATCH 01/79] fs: add ctime accessors infrastructure
@ 2023-07-12 15:34 ` Christian Brauner
0 siblings, 0 replies; 229+ messages in thread
From: Christian Brauner @ 2023-07-12 15:34 UTC (permalink / raw)
To: Randy Dunlap
Cc: Jeff Layton, linux-kernel, Linux FS-devel Mailing List, linux-um
On Wed, Jul 12, 2023 at 08:31:55AM -0700, Randy Dunlap wrote:
> Hi Jeff,
>
> On arch/um/, (subarch i386 or x86_64), hostfs build fails with:
>
> ../fs/hostfs/hostfs_kern.c:520:36: error: incompatible type for arg
> ument 2 of 'inode_set_ctime_to_ts'
> ../include/linux/fs.h:1499:73: note: expected 'struct timespec64' b
> ut argument is of type 'const struct hostfs_timespec *'
Thanks for notifying us! We fixed this earlier today. The vfs.ctime
branch has been updated with the fix folded into the hostfs patch.
_______________________________________________
linux-um mailing list
linux-um@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-um
^ permalink raw reply [flat|nested] 229+ messages in thread
end of thread, other threads:[~2023-07-12 15:34 UTC | newest]
Thread overview: 229+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-06-21 14:45 [PATCH 00/79] fs: new accessors for inode->i_ctime Jeff Layton
2023-06-21 14:45 ` [Cluster-devel] " Jeff Layton
2023-06-21 14:45 ` [Ocfs2-devel] " Jeff Layton via Ocfs2-devel
2023-06-21 14:45 ` [f2fs-dev] " Jeff Layton
2023-06-21 14:45 ` [PATCH 01/79] fs: add ctime accessors infrastructure Jeff Layton
2023-06-21 14:45 ` [Cluster-devel] " Jeff Layton
2023-06-21 14:45 ` [Ocfs2-devel] " Jeff Layton via Ocfs2-devel
2023-06-21 14:45 ` [f2fs-dev] " Jeff Layton
2023-06-21 16:34 ` Jan Kara
2023-06-21 16:34 ` Jan Kara
2023-06-21 16:34 ` [Cluster-devel] " Jan Kara
2023-06-21 16:34 ` Jan Kara
2023-06-21 16:34 ` Jan Kara
2023-06-21 16:34 ` [Ocfs2-devel] " Jan Kara via Ocfs2-devel
2023-06-21 16:34 ` [f2fs-dev] " Jan Kara
2023-06-21 17:29 ` Tom Talpey
2023-06-21 17:29 ` [Cluster-devel] " Tom Talpey
2023-06-21 17:29 ` [f2fs-dev] " Tom Talpey
2023-06-21 17:29 ` [Ocfs2-devel] " Tom Talpey via Ocfs2-devel
2023-06-21 18:01 ` Jeff Layton
2023-06-21 18:01 ` [Cluster-devel] " Jeff Layton
2023-06-21 18:01 ` [Ocfs2-devel] " Jeff Layton via Ocfs2-devel
2023-06-21 18:01 ` [f2fs-dev] " Jeff Layton
2023-06-21 18:19 ` Tom Talpey
2023-06-21 18:19 ` [Cluster-devel] " Tom Talpey
2023-06-21 18:19 ` [Ocfs2-devel] " Tom Talpey via Ocfs2-devel
2023-06-21 18:19 ` [f2fs-dev] " Tom Talpey
2023-06-21 18:48 ` Jeff Layton
2023-06-21 18:48 ` [Cluster-devel] " Jeff Layton
2023-06-21 18:48 ` [Ocfs2-devel] " Jeff Layton via Ocfs2-devel
2023-06-21 18:48 ` [f2fs-dev] " Jeff Layton
2023-06-22 0:46 ` Damien Le Moal
2023-06-22 0:46 ` [Cluster-devel] " Damien Le Moal
2023-06-22 0:46 ` [Ocfs2-devel] " Damien Le Moal via Ocfs2-devel
2023-06-22 0:46 ` [f2fs-dev] " Damien Le Moal
2023-06-22 10:14 ` Jeff Layton
2023-06-22 10:14 ` [Cluster-devel] " Jeff Layton
2023-06-22 10:14 ` [Ocfs2-devel] " Jeff Layton via Ocfs2-devel
2023-06-22 10:14 ` [f2fs-dev] " Jeff Layton
[not found] ` <99b3c749-23d9-6f09-fb75-6a84f3d1b066-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
2023-06-22 10:14 ` Jeff Layton
2023-06-22 10:14 ` Jeff Layton
2023-06-30 22:12 ` Luis Chamberlain
2023-06-30 22:12 ` Luis Chamberlain
2023-06-30 22:12 ` [Cluster-devel] " Luis Chamberlain
2023-06-30 22:12 ` Luis Chamberlain
2023-06-30 22:12 ` Luis Chamberlain
2023-06-30 22:12 ` [f2fs-dev] " Luis Chamberlain
[not found] ` <20230621144507.55591-2-jlayton-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
2023-07-12 15:31 ` Randy Dunlap
2023-07-12 15:31 ` Randy Dunlap
2023-07-12 15:31 ` [Cluster-devel] " Randy Dunlap
2023-07-12 15:31 ` Randy Dunlap
2023-07-12 15:31 ` Randy Dunlap
2023-07-12 15:34 ` Christian Brauner
2023-07-12 15:34 ` Christian Brauner
2023-07-12 15:31 ` Randy Dunlap
2023-06-21 14:45 ` [PATCH 02/79] spufs: switch to new ctime accessors Jeff Layton
2023-06-21 14:45 ` Jeff Layton
2023-06-21 14:45 ` [PATCH 03/79] s390: " Jeff Layton
2023-06-21 16:34 ` Jan Kara
2023-06-22 17:35 ` Alexander Gordeev
2023-06-22 17:51 ` Jeff Layton
2023-06-22 18:22 ` Alexander Gordeev
2023-06-22 18:46 ` Jeff Layton
2023-06-21 14:45 ` [PATCH 04/79] binderfs: " Jeff Layton
2023-06-21 15:29 ` Greg Kroah-Hartman
2023-06-21 16:35 ` Jan Kara
2023-06-21 14:45 ` [PATCH 05/79] qib_fs: " Jeff Layton
2023-06-21 16:35 ` Jan Kara
2023-06-21 19:37 ` Dennis Dalessandro
2023-06-21 14:45 ` [PATCH 06/79] ibm: " Jeff Layton
2023-06-21 15:29 ` Greg Kroah-Hartman
2023-06-21 16:35 ` Jan Kara
2023-06-21 14:45 ` [PATCH 07/79] usb: " Jeff Layton
2023-06-21 15:29 ` Greg Kroah-Hartman
2023-06-21 16:36 ` Jan Kara
2023-06-21 14:45 ` [PATCH 08/79] 9p: " Jeff Layton
2023-06-21 16:37 ` Jan Kara
2023-06-21 14:45 ` [PATCH 09/79] adfs: " Jeff Layton
2023-06-21 16:38 ` Jan Kara
2023-06-21 14:45 ` [PATCH 10/79] affs: " Jeff Layton
2023-06-21 16:39 ` Jan Kara
2023-06-22 11:37 ` David Sterba
2023-06-21 14:45 ` [PATCH 11/79] afs: " Jeff Layton
2023-06-21 16:43 ` Jan Kara
2023-06-21 14:45 ` [PATCH 12/79] fs: " Jeff Layton
2023-06-21 16:42 ` Jan Kara
2023-06-21 14:45 ` [PATCH 13/79] autofs: " Jeff Layton
2023-06-21 16:43 ` Jan Kara
2023-06-27 1:48 ` Ian Kent
2023-06-21 14:45 ` [PATCH 14/79] befs: " Jeff Layton
2023-06-21 16:44 ` Jan Kara
2023-06-21 14:45 ` [PATCH 15/79] bfs: " Jeff Layton
2023-06-21 16:48 ` Jan Kara
2023-06-21 16:57 ` Jeff Layton
2023-06-22 12:30 ` Jan Kara
2023-06-22 12:51 ` Jeff Layton
2023-06-22 14:57 ` Jan Kara
2023-06-23 12:33 ` Christian Brauner
2023-07-03 10:12 ` Christian Brauner
2023-07-03 10:46 ` Jeff Layton
2023-07-03 10:57 ` Christian Brauner
2023-06-21 14:45 ` [PATCH 16/79] btrfs: " Jeff Layton
2023-06-22 11:42 ` David Sterba
2023-06-21 14:45 ` [PATCH 17/79] ceph: " Jeff Layton
2023-06-26 0:56 ` Xiubo Li
2023-06-21 14:45 ` [PATCH 18/79] coda: " Jeff Layton
2023-06-21 14:45 ` [PATCH 19/79] configfs: " Jeff Layton
2023-06-21 14:45 ` [PATCH 20/79] cramfs: " Jeff Layton
2023-06-21 15:29 ` Nicolas Pitre
2023-06-21 14:45 ` [PATCH 21/79] debugfs: " Jeff Layton
2023-06-21 15:29 ` Greg Kroah-Hartman
2023-06-21 14:45 ` [PATCH 22/79] devpts: " Jeff Layton
2023-06-21 14:45 ` [PATCH 23/79] ecryptfs: " Jeff Layton
2023-06-21 14:45 ` [PATCH 24/79] efivarfs: " Jeff Layton
2023-06-21 14:45 ` [PATCH 25/79] efs: " Jeff Layton
2023-06-21 14:45 ` [PATCH 26/79] erofs: " Jeff Layton
2023-06-21 14:45 ` Jeff Layton
2023-06-22 4:01 ` Gao Xiang
2023-06-22 4:01 ` Gao Xiang
2023-06-21 14:45 ` [PATCH 27/79] exfat: " Jeff Layton
2023-06-21 14:45 ` [PATCH 28/79] ext2: " Jeff Layton
2023-06-21 14:45 ` [PATCH 29/79] ext4: " Jeff Layton
2023-06-21 14:45 ` [f2fs-dev] [PATCH 30/79] f2fs: " Jeff Layton
2023-06-21 14:45 ` Jeff Layton
2023-06-21 14:45 ` [PATCH 31/79] fat: " Jeff Layton
2023-06-21 14:45 ` [PATCH 32/79] freevxfs: " Jeff Layton
2023-06-21 14:45 ` [PATCH 33/79] fuse: " Jeff Layton
2023-06-21 14:45 ` [PATCH 34/79] gfs2: " Jeff Layton
2023-06-21 14:45 ` [Cluster-devel] " Jeff Layton
2023-06-21 14:45 ` [PATCH 35/79] hfs: " Jeff Layton
2023-06-21 14:45 ` [PATCH 36/79] hfsplus: " Jeff Layton
2023-06-21 14:45 ` [PATCH 37/79] hostfs: " Jeff Layton
2023-06-21 14:45 ` Jeff Layton
2023-06-21 14:45 ` [PATCH 38/79] hpfs: " Jeff Layton
2023-06-21 14:45 ` [PATCH 39/79] hugetlbfs: " Jeff Layton
2023-06-21 14:45 ` [PATCH 40/79] isofs: " Jeff Layton
2023-06-21 14:45 ` [PATCH 41/79] jffs2: " Jeff Layton
2023-06-21 14:45 ` Jeff Layton
2023-06-21 14:45 ` [PATCH 42/79] jfs: " Jeff Layton
2023-06-21 17:47 ` Dave Kleikamp
2023-06-21 14:45 ` [PATCH 43/79] kernfs: " Jeff Layton
2023-06-21 15:30 ` Greg Kroah-Hartman
2023-06-21 14:45 ` [PATCH 44/79] minix: " Jeff Layton
2023-06-21 14:45 ` [PATCH 45/79] nfs: " Jeff Layton
2023-06-21 14:45 ` [PATCH 46/79] nfsd: " Jeff Layton
2023-06-21 17:36 ` Chuck Lever
2023-06-21 14:46 ` [PATCH 47/79] nilfs2: " Jeff Layton
2023-06-21 14:46 ` Jeff Layton
2023-06-26 15:26 ` Ryusuke Konishi
2023-06-26 15:26 ` Ryusuke Konishi
2023-06-21 14:46 ` [PATCH 48/79] ntfs: " Jeff Layton
2023-06-21 14:46 ` [PATCH 49/79] ntfs3: " Jeff Layton
2023-06-21 14:46 ` [Ocfs2-devel] [PATCH 50/79] ocfs2: " Jeff Layton via Ocfs2-devel
2023-06-21 14:46 ` Jeff Layton
2023-06-21 14:46 ` [PATCH 51/79] omfs: " Jeff Layton
2023-06-23 12:12 ` Bob Copeland
2023-06-21 14:46 ` [PATCH 52/79] openpromfs: " Jeff Layton
2023-06-21 14:46 ` [PATCH 53/79] orangefs: " Jeff Layton
2023-06-21 14:46 ` [PATCH 54/79] overlayfs: " Jeff Layton
2023-06-22 6:44 ` Amir Goldstein
2023-06-21 14:46 ` [PATCH 55/79] proc: " Jeff Layton
2023-06-30 22:13 ` Luis Chamberlain
2023-06-21 14:46 ` [PATCH 56/79] pstore: " Jeff Layton
2023-06-21 17:50 ` Kees Cook
2023-06-21 14:46 ` [PATCH 57/79] qnx4: " Jeff Layton
2023-06-23 14:26 ` Anders Larsen
2023-06-21 14:46 ` [PATCH 58/79] qnx6: " Jeff Layton
2023-06-21 14:46 ` [PATCH 59/79] ramfs: " Jeff Layton
2023-06-21 14:46 ` [PATCH 60/79] reiserfs: " Jeff Layton
2023-06-21 14:46 ` [PATCH 61/79] romfs: " Jeff Layton
2023-06-21 14:46 ` [PATCH 62/79] smb: " Jeff Layton
2023-06-21 17:45 ` Tom Talpey
2023-06-23 5:03 ` Sergey Senozhatsky
2023-06-21 14:46 ` [PATCH 63/79] squashfs: " Jeff Layton
2023-06-21 14:46 ` [PATCH 64/79] sysv: " Jeff Layton
2023-06-21 14:46 ` [PATCH 65/79] tracefs: " Jeff Layton
2023-06-21 14:46 ` [PATCH 66/79] ubifs: " Jeff Layton
2023-06-21 14:46 ` Jeff Layton
2023-06-21 14:46 ` [PATCH 67/79] udf: " Jeff Layton
2023-06-21 14:46 ` [PATCH 68/79] ufs: " Jeff Layton
2023-06-21 14:46 ` [PATCH 69/79] vboxsf: " Jeff Layton
2023-06-21 14:46 ` [PATCH 70/79] xfs: " Jeff Layton
2023-06-21 14:46 ` [PATCH 71/79] zonefs: " Jeff Layton
2023-06-22 0:48 ` Damien Le Moal
2023-06-21 14:46 ` [PATCH 72/79] mqueue: " Jeff Layton
2023-06-21 14:46 ` [PATCH 73/79] bpf: " Jeff Layton
2023-06-21 14:46 ` [PATCH 74/79] shmem: " Jeff Layton
2023-06-21 14:46 ` [PATCH 75/79] rpc_pipefs: " Jeff Layton
2023-06-21 14:46 ` [PATCH 76/79] apparmor: " Jeff Layton
2023-06-22 9:04 ` John Johansen
2023-06-21 14:46 ` [PATCH 77/79] security: " Jeff Layton
2023-06-23 14:15 ` Paul Moore
2023-06-21 14:46 ` [PATCH 78/79] selinux: " Jeff Layton
2023-06-23 14:17 ` Paul Moore
2023-06-21 16:34 ` [PATCH 02/79] spufs: " Jan Kara
2023-06-21 16:34 ` Jan Kara
2023-06-22 0:19 ` Jeremy Kerr
2023-06-22 0:19 ` Jeremy Kerr
2023-06-21 14:49 ` [PATCH 79/79] fs: rename i_ctime field to __i_ctime Jeff Layton
2023-06-21 14:49 ` [Cluster-devel] " Jeff Layton
2023-06-21 14:49 ` [Ocfs2-devel] " Jeff Layton via Ocfs2-devel
2023-06-21 14:49 ` [f2fs-dev] " Jeff Layton
2023-06-21 19:21 ` [PATCH 00/79] fs: new accessors for inode->i_ctime Steven Rostedt
2023-06-21 19:21 ` Steven Rostedt
2023-06-21 19:21 ` [Cluster-devel] " Steven Rostedt
2023-06-21 19:21 ` Steven Rostedt
2023-06-21 19:21 ` Steven Rostedt
2023-06-21 19:21 ` [Ocfs2-devel] " Steven Rostedt via Ocfs2-devel
2023-06-21 19:21 ` [f2fs-dev] " Steven Rostedt
2023-06-21 19:52 ` Jeff Layton
2023-06-21 19:52 ` Jeff Layton
2023-06-21 19:52 ` [Cluster-devel] " Jeff Layton
2023-06-21 19:52 ` [Ocfs2-devel] " Jeff Layton via Ocfs2-devel
2023-06-21 19:52 ` Jeff Layton
2023-06-21 19:52 ` Jeff Layton
2023-06-21 19:52 ` [f2fs-dev] " Jeff Layton
2023-06-23 12:41 ` Christian Brauner
2023-06-23 12:41 ` Christian Brauner
2023-06-23 12:41 ` [Cluster-devel] " Christian Brauner
2023-06-23 12:41 ` Christian Brauner
2023-06-23 12:41 ` Christian Brauner
2023-06-23 12:41 ` [Ocfs2-devel] " Christian Brauner via Ocfs2-devel
2023-06-23 12:41 ` [f2fs-dev] " Christian Brauner
2023-06-30 22:11 ` Luis Chamberlain
2023-06-30 22:11 ` Luis Chamberlain
2023-06-30 22:11 ` [Cluster-devel] " Luis Chamberlain
2023-06-30 22:11 ` Luis Chamberlain
2023-06-30 22:11 ` Luis Chamberlain
2023-06-30 22:11 ` [f2fs-dev] " Luis Chamberlain
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.