Skip to contentSkip to conversation

fs: new accessors for inode->i_ctime

PATCH Applied Jeff Layton 10 participants 38 messages Jul 5, 2023 – Sep 4, 2023 5 messages not in the archive

Subsystems (33)
lore
Share & tools
lore

Jeff Layton

author maintainer
#0 lore raw

v2 cover fs: new accessors for inode->i_ctime

v2:
- prepend patches to add missing ctime updates
- add simple_rename_timestamp helper function
- rename ctime accessor functions as inode_get_ctime/inode_set_ctime_*
- drop individual inode_ctime_set_{sec,nsec} helpers

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.

The patchset starts with reposts of some missing ctime updates that I spotted in the tree. It then adds a new helper function for updating the timestamp after a successful rename, and new ctime accessor infrastructure.

The bulk of the patchset is individual conversions of different subsysteme to use the new infrastructure. Finally, the patchset renames the i_ctime field to __i_ctime to help ensure that I didn't miss anything.

This should apply cleanly to linux-next as of this morning.

Most of this conversion was done via 5 different coccinelle scripts, run in succession, with a large swath of by-hand conversions to clean up the remainder.

The coccinelle scripts that were used are below:

::::::::::::::
cocci/ctime1.cocci
::::::::::::::
// convert as much to use inode_set_ctime_current as possible @@
identifier timei;
struct inode *inode;
expression E1, E2;
@@
(
- inode->i_ctime = E1 = E2 = current_time(timei)
+ E1 = E2 = inode_set_ctime_current(inode)
|
- inode->i_ctime = E1 = current_time(timei)
+ E1 = inode_set_ctime_current(inode)
|
- E1 = inode->i_ctime = current_time(timei)
+ E1 = inode_set_ctime_current(inode)
|
- inode->i_ctime = current_time(timei)
+ inode_set_ctime_current(inode)
)

@@
struct inode *inode;
expression E1, E2, E3;
@@
(
- E1 = current_time(inode)
+ E1 = inode_set_ctime_current(inode)
|
- E1 = current_time(E3)
+ E1 = inode_set_ctime_current(inode)
)
...
(
- inode->i_ctime = E1;
|
- E2 = inode->i_ctime = E1;
+ E2 = E1;
)
::::::::::::::
cocci/ctime2.cocci
::::::::::::::
// get the places that set individual timespec64 fields @@
struct inode *inode;
expression val, val2;
@@
- inode->i_ctime.tv_sec = val
+ inode_set_ctime(inode, val, val2)
...
- inode->i_ctime.tv_nsec = val2;

// get places that just set the tv_sec
@@
struct inode *inode;
expression sec, E1, E2, E3;
@@
(
- E3 = inode->i_ctime.tv_sec = sec
+ E3 = inode_set_ctime(inode, sec, 0).tv_sec
|
- inode->i_ctime.tv_sec = sec
+ inode_set_ctime(inode, sec, 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
)
...>

::::::::::::::
cocci/ctime3.cocci
::::::::::::::
// convert places that set i_ctime to a timespec64 directly
@@
struct inode *inode;
expression ts, E1, E2;
@@
(
- inode->i_ctime = E1 = E2 = ts
+ E1 = E2 = inode_set_ctime_to_ts(inode, ts)
|
- inode->i_ctime = E1 = ts
+ E1 = inode_set_ctime_to_ts(inode, ts)
|
- inode->i_ctime = ts
+ inode_set_ctime_to_ts(inode, ts)
)
::::::::::::::
cocci/ctime4.cocci
::::::::::::::
// catch places that set the i_ctime in an inode embedded in another structure @@
expression E1, E2, E3;
@@
(
- E3.i_ctime = E1 = E2 = current_time(&E3)
+ E1 = E2 = inode_set_ctime_current(&E3)
|
- E3.i_ctime = E1 = current_time(&E3)
+ E1 = inode_set_ctime_current(&E3)
|
- E1 = E3.i_ctime = current_time(&E3)
+ E1 = inode_set_ctime_current(&E3)
|
- E3.i_ctime = current_time(&E3)
+ inode_set_ctime_current(&E3)
)
::::::::::::::
cocci/ctime5.cocci
::::::::::::::
// convert the remaining i_ctime accesses
@@
struct inode *inode;
@@
- inode->i_ctime
+ inode_get_ctime(inode)

Jeff Layton (92):
ibmvmc: update ctime in conjunction with mtime on write
bfs: update ctime in addition to mtime when adding entries
efivarfs: update ctime when mtime changes on a write
exfat: ensure that ctime is updated whenever the mtime is
apparmor: update ctime whenever the mtime changes on an inode
cifs: update the ctime on a partial page write
fs: add ctime accessors infrastructure
fs: new helper: simple_rename_timestamp
btrfs: convert to simple_rename_timestamp
ubifs: convert to simple_rename_timestamp
shmem: convert to simple_rename_timestamp
exfat: convert to simple_rename_timestamp
ntfs3: convert to simple_rename_timestamp
reiserfs: convert to simple_rename_timestamp
spufs: convert to ctime accessor functions
s390: convert to ctime accessor functions
binderfs: convert to ctime accessor functions
infiniband: convert to ctime accessor functions
ibm: convert to ctime accessor functions
usb: convert to ctime accessor functions
9p: convert to ctime accessor functions
adfs: convert to ctime accessor functions
affs: convert to ctime accessor functions
afs: convert to ctime accessor functions
fs: convert to ctime accessor functions
autofs: convert to ctime accessor functions
befs: convert to ctime accessor functions
bfs: convert to ctime accessor functions
btrfs: convert to ctime accessor functions
ceph: convert to ctime accessor functions
coda: convert to ctime accessor functions
configfs: convert to ctime accessor functions
cramfs: convert to ctime accessor functions
debugfs: convert to ctime accessor functions
devpts: convert to ctime accessor functions
ecryptfs: convert to ctime accessor functions
efivarfs: convert to ctime accessor functions
efs: convert to ctime accessor functions
erofs: convert to ctime accessor functions
exfat: convert to ctime accessor functions
ext2: convert to ctime accessor functions
ext4: convert to ctime accessor functions
f2fs: convert to ctime accessor functions
fat: convert to ctime accessor functions
freevxfs: convert to ctime accessor functions
fuse: convert to ctime accessor functions
gfs2: convert to ctime accessor functions
hfs: convert to ctime accessor functions
hfsplus: convert to ctime accessor functions
hostfs: convert to ctime accessor functions
hpfs: convert to ctime accessor functions
hugetlbfs: convert to ctime accessor functions
isofs: convert to ctime accessor functions
jffs2: convert to ctime accessor functions
jfs: convert to ctime accessor functions
kernfs: convert to ctime accessor functions
nfs: convert to ctime accessor functions
nfsd: convert to ctime accessor functions
nilfs2: convert to ctime accessor functions
ntfs: convert to ctime accessor functions
ntfs3: convert to ctime accessor functions
ocfs2: convert to ctime accessor functions
omfs: convert to ctime accessor functions
openpromfs: convert to ctime accessor functions
orangefs: convert to ctime accessor functions
overlayfs: convert to ctime accessor functions
procfs: convert to ctime accessor functions
pstore: convert to ctime accessor functions
qnx4: convert to ctime accessor functions
qnx6: convert to ctime accessor functions
ramfs: convert to ctime accessor functions
reiserfs: convert to ctime accessor functions
romfs: convert to ctime accessor functions
smb: convert to ctime accessor functions
squashfs: convert to ctime accessor functions
sysv: convert to ctime accessor functions
tracefs: convert to ctime accessor functions
ubifs: convert to ctime accessor functions
udf: convert to ctime accessor functions
ufs: convert to ctime accessor functions
vboxsf: convert to ctime accessor functions
xfs: convert to ctime accessor functions
zonefs: convert to ctime accessor functions
linux: convert to ctime accessor functions
mqueue: convert to ctime accessor functions
bpf: convert to ctime accessor functions
shmem: convert to ctime accessor functions
sunrpc: convert to ctime accessor functions
apparmor: convert to ctime accessor functions
security: convert to ctime accessor functions
selinux: convert to ctime accessor functions
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        |  3 +-
 drivers/misc/ibmasm/ibmasmfs.c            |  2 +-
 drivers/misc/ibmvmc.c                     |  2 +-
 drivers/usb/core/devio.c                  | 16 +++----
 drivers/usb/gadget/function/f_fs.c        |  3 +-
 drivers/usb/gadget/legacy/inode.c         |  3 +-
 fs/9p/vfs_inode.c                         |  4 +-
 fs/9p/vfs_inode_dotl.c                    |  8 ++--
 fs/adfs/inode.c                           |  4 +-
 fs/affs/amigaffs.c                        |  6 +--
 fs/affs/inode.c                           | 16 +++----
 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                            |  5 +--
 fs/binfmt_misc.c                          |  3 +-
 fs/btrfs/delayed-inode.c                  |  8 ++--
 fs/btrfs/file.c                           | 21 ++++-----
 fs/btrfs/inode.c                          | 54 ++++++++--------------
 fs/btrfs/ioctl.c                          |  2 +-
 fs/btrfs/reflink.c                        |  3 +-
 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                      |  3 +-
 fs/coda/dir.c                             |  2 +-
 fs/coda/file.c                            |  2 +-
 fs/coda/inode.c                           |  2 +-
 fs/configfs/inode.c                       |  7 ++-
 fs/cramfs/inode.c                         |  3 +-
 fs/debugfs/inode.c                        |  3 +-
 fs/devpts/inode.c                         |  6 +--
 fs/ecryptfs/inode.c                       |  2 +-
 fs/efivarfs/file.c                        |  2 +-
 fs/efivarfs/inode.c                       |  2 +-
 fs/efs/inode.c                            |  4 +-
 fs/erofs/inode.c                          | 15 +++----
 fs/exfat/file.c                           |  4 +-
 fs/exfat/inode.c                          |  6 +--
 fs/exfat/namei.c                          | 26 +++++------
 fs/exfat/super.c                          |  3 +-
 fs/ext2/acl.c                             |  2 +-
 fs/ext2/dir.c                             |  6 +--
 fs/ext2/ialloc.c                          |  2 +-
 fs/ext2/inode.c                           | 10 ++---
 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                            | 21 +++++++++
 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                            |  4 +-
 fs/f2fs/file.c                            | 20 ++++-----
 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                            |  7 +--
 fs/fat/misc.c                             |  3 +-
 fs/freevxfs/vxfs_inode.c                  |  3 +-
 fs/fuse/control.c                         |  2 +-
 fs/fuse/dir.c                             |  8 ++--
 fs/fuse/inode.c                           | 16 +++----
 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                           |  4 +-
 fs/hfsplus/catalog.c                      |  8 ++--
 fs/hfsplus/dir.c                          |  6 +--
 fs/hfsplus/inode.c                        | 16 ++++---
 fs/hostfs/hostfs_kern.c                   |  3 +-
 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                          |  8 ++--
 fs/isofs/rock.c                           | 16 +++----
 fs/jffs2/dir.c                            | 24 ++++++----
 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                            | 24 +++++-----
 fs/jfs/super.c                            |  2 +-
 fs/jfs/xattr.c                            |  2 +-
 fs/kernfs/inode.c                         |  5 +--
 fs/libfs.c                                | 55 +++++++++++++++--------
 fs/minix/bitmap.c                         |  2 +-
 fs/minix/dir.c                            |  6 +--
 fs/minix/inode.c                          | 10 ++---
 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                            | 20 ++++-----
 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                        |  3 +-
 fs/ntfs3/inode.c                          | 14 +++---
 fs/ntfs3/namei.c                          | 11 ++---
 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                        |  7 ++-
 fs/ocfs2/file.c                           | 16 ++++---
 fs/ocfs2/inode.c                          | 12 ++---
 fs/ocfs2/move_extents.c                   |  6 +--
 fs/ocfs2/namei.c                          | 21 ++++-----
 fs/ocfs2/refcounttree.c                   | 14 +++---
 fs/ocfs2/xattr.c                          |  6 +--
 fs/omfs/dir.c                             |  4 +-
 fs/omfs/inode.c                           |  9 ++--
 fs/openpromfs/inode.c                     |  5 +--
 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                           |  3 +-
 fs/qnx6/inode.c                           |  3 +-
 fs/ramfs/inode.c                          |  6 +--
 fs/reiserfs/inode.c                       | 12 +++--
 fs/reiserfs/ioctl.c                       |  4 +-
 fs/reiserfs/namei.c                       | 18 +++-----
 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                     | 14 +++---
 fs/smb/client/smb2ops.c                   |  3 +-
 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                           |  5 +--
 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             |  5 ++-
 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                        |  3 +-
 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                        | 49 +++++++++++++++++++-
 include/linux/fs_stack.h                  |  2 +-
 ipc/mqueue.c                              | 23 +++++-----
 kernel/bpf/inode.c                        |  6 +--
 mm/shmem.c                                | 26 +++++------
 net/sunrpc/rpc_pipe.c                     |  2 +-
 security/apparmor/apparmorfs.c            | 11 +++--
 security/apparmor/policy_unpack.c         | 11 +++--
 security/inode.c                          |  2 +-
 security/selinux/selinuxfs.c              |  2 +-
 233 files changed, 901 insertions(+), 812 deletions(-)
Signature · 1 line
2.41.0
author maintainer
#1 lore raw

PATCH v2 7/92 fs: add ctime accessors infrastructure

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 use to replace them.

---
 fs/inode.c         | 16 ++++++++++++++++
 include/linux/fs.h | 45 ++++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 60 insertions(+), 1 deletion(-)
fs/inode.c +16 -0
index d37fad91c8da..21b026d95b51 100644
@@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
 }
 EXPORT_SYMBOL(current_time);
 
+/**
+ * inode_set_ctime_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_set_ctime_current(struct inode *inode)
+{
+	struct timespec64 now = current_time(inode);
+
+	inode_set_ctime(inode, now.tv_sec, now.tv_nsec);
+	return now;
+}
+EXPORT_SYMBOL(inode_set_ctime_current);
+
 /**
  * in_group_or_capable - check whether caller is CAP_FSETID privileged
  * @idmap:	idmap of the mount @inode was found from
include/linux/fs.h +44 -1

51 lines not on this page · raw message

Signature · 1 line
2.41.0
Reviewed-by on v2 7/92
#2 lore raw

Re: [PATCH v2 07/92] fs: add ctime accessors infrastructure

Quoted 11 lines · Jeff Layton

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 use to replace them.

Looks OK to me.

Signature · 2 lines
Damien Le Moal
Western Digital Research

A message that is not in the archive (sent off-list or not yet received): 20230705185812.579118-2-jlayton-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org

Reviewed-by
#4 lore raw

Re: [PATCH v2 07/92] fs: add ctime accessors infrastructure

Quoted 11 lines · Jeff Layton

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 use to replace them.

Looks OK to me.

Signature · 2 lines
Damien Le Moal
Western Digital Research
Reviewed-by on v2 7/92
#5 lore raw

Re: [PATCH v2 07/92] fs: add ctime accessors infrastructure

In reply to Jeff Layton
Quoted 11 lines · Jeff Layton

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 use to replace them.

Looks OK to me.

Signature · 2 lines
Damien Le Moal
Western Digital Research

A message that is not in the archive (sent off-list or not yet received): 20230705185812.579118-1-jlayton-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org

#7 lore raw

PATCH v2 7/92 fs: add ctime accessors infrastructure

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 use to replace them.

---
 fs/inode.c         | 16 ++++++++++++++++
 include/linux/fs.h | 45 ++++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 60 insertions(+), 1 deletion(-)
fs/inode.c +16 -0
index d37fad91c8da..21b026d95b51 100644
@@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
 }
 EXPORT_SYMBOL(current_time);
 
+/**
+ * inode_set_ctime_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_set_ctime_current(struct inode *inode)
+{
+	struct timespec64 now = current_time(inode);
+
+	inode_set_ctime(inode, now.tv_sec, now.tv_nsec);
+	return now;
+}
+EXPORT_SYMBOL(inode_set_ctime_current);
+
 /**
  * in_group_or_capable - check whether caller is CAP_FSETID privileged
  * @idmap:	idmap of the mount @inode was found from
include/linux/fs.h +44 -1

51 lines not on this page · raw message

Signature · 1 line
2.41.0
#8 lore raw

PATCH v2 8/92 fs: new helper: simple_rename_timestamp

In reply to a message not in the archive

A rename potentially involves updating 4 different inode timestamps. Add a function that handles the details sanely, and convert the libfs.c callers to use it.

---
 fs/libfs.c         | 36 +++++++++++++++++++++++++++---------
 include/linux/fs.h |  2 ++
 2 files changed, 29 insertions(+), 9 deletions(-)
fs/libfs.c +27 -9
index a7e56baf8bbd..9ee79668c909 100644
@@ -692,6 +692,31 @@ int simple_rmdir(struct inode *dir, struct dentry *dentry)
 }
 EXPORT_SYMBOL(simple_rmdir);
 
+/**
+ * simple_rename_timestamp - update the various inode timestamps for rename
+ * @old_dir: old parent directory
+ * @old_dentry: dentry that is being renamed
+ * @new_dir: new parent directory
+ * @new_dentry: target for rename
+ *
+ * POSIX mandates that the old and new parent directories have their ctime and
+ * mtime updated, and that inodes of @old_dentry and @new_dentry (if any), have
+ * their ctime updated.
+ */
+void simple_rename_timestamp(struct inode *old_dir, struct dentry *old_dentry,
+			     struct inode *new_dir, struct dentry *new_dentry)
+{
+	struct inode *newino = d_inode(new_dentry);
+
+	old_dir->i_mtime = inode_set_ctime_current(old_dir);
+	if (new_dir != old_dir)
+		new_dir->i_mtime = inode_set_ctime_current(new_dir);
+	inode_set_ctime_current(d_inode(old_dentry));
+	if (newino)
+		inode_set_ctime_current(newino);
+}
+EXPORT_SYMBOL_GPL(simple_rename_timestamp);
+
 int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry,
 			   struct inode *new_dir, struct dentry *new_dentry)
 {
@@ -707,11 +732,7 @@ 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);
-
+	simple_rename_timestamp(old_dir, old_dentry, new_dir, new_dentry);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(simple_rename_exchange);
@@ -720,7 +741,6 @@ int simple_rename(struct mnt_idmap *idmap, struct inode *old_dir,
 		  struct dentry *old_dentry, struct inode *new_dir,
 		  struct dentry *new_dentry, unsigned int flags)
 {
-	struct inode *inode = d_inode(old_dentry);
 	int they_are_dirs = d_is_dir(old_dentry);
 
 	if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE))
@@ -743,9 +763,7 @@ 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);
-
+	simple_rename_timestamp(old_dir, old_dentry, new_dir, new_dentry);
 	return 0;
 }
 EXPORT_SYMBOL(simple_rename);
include/linux/fs.h +2 -0
index bdfbd11a5811..14e38bd900f1 100644
@@ -2979,6 +2979,8 @@ extern int simple_open(struct inode *inode, struct file *file);
 extern int simple_link(struct dentry *, struct inode *, struct dentry *);
 extern int simple_unlink(struct inode *, struct dentry *);
 extern int simple_rmdir(struct inode *, struct dentry *);
+void simple_rename_timestamp(struct inode *old_dir, struct dentry *old_dentry,
+			     struct inode *new_dir, struct dentry *new_dentry);
 extern int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry,
 				  struct inode *new_dir, struct dentry *new_dentry);
 extern int simple_rename(struct mnt_idmap *, struct inode *,
Signature · 1 line
2.41.0
#9 lore raw

PATCH v2 92/92 fs: rename i_ctime field to __i_ctime

In reply to a message not in the archive

Now that everything in-tree is converted to use the accessor functions, rename the i_ctime field in the inode to discourage direct access.

---
 include/linux/fs.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
include/linux/fs.h +3 -3
index 14e38bd900f1..b66442f91835 100644
@@ -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_set_ctime_current(struct inode *inode);
  */
 static inline struct timespec64 inode_get_ctime(const struct inode *inode)
 {
-	return inode->i_ctime;
+	return inode->__i_ctime;
 }
 
 /**
@@ -1498,7 +1498,7 @@ static inline struct timespec64 inode_get_ctime(const struct inode *inode)
 static inline struct timespec64 inode_set_ctime_to_ts(struct inode *inode,
 						      struct timespec64 ts)
 {
-	inode->i_ctime = ts;
+	inode->__i_ctime = ts;
 	return ts;
 }
 
Signature · 1 line
2.41.0
#10 lore raw
In reply to a message not in the archive
Quoted 23 lines · Jeff Layton

v2:
- prepend patches to add missing ctime updates
- add simple_rename_timestamp helper function
- rename ctime accessor functions as inode_get_ctime/inode_set_ctime_*
- drop individual inode_ctime_set_{sec,nsec} helpers

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.

The patchset starts with reposts of some missing ctime updates that I spotted in the tree. It then adds a new helper function for updating the timestamp after a successful rename, and new ctime accessor infrastructure.

The bulk of the patchset is individual conversions of different subsysteme to use the new infrastructure. Finally, the patchset renames the i_ctime field to __i_ctime to help ensure that I didn't miss anything.

Quoted 6 lines

This should apply cleanly to linux-next as of this morning.

Most of this conversion was done via 5 different coccinelle scripts, run in succession, with a large swath of by-hand conversions to clean up the remainder.

A couple of other things I should note:

If you sent me an Acked-by or Reviewed-by in the previous set, then I tried to keep it on the patch here, since the respun patches are mostly just renaming stuff from v1. Let me know if I've missed any.

I've also pushed the pile to my tree as this tag:

    https://git.kernel.org/pub/scm/linux/kernel/git/jlayton/linux.git/tag/?h=ctime.20230705

In case that's easier to work with.

Cheers,

Signature · 1 line
Jeff Layton <[email protected]>
author maintainer
#11 lore raw

PATCH v2 7/92 fs: add ctime accessors infrastructure

In reply to Jeff Layton

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 use to replace them.

---
 fs/inode.c         | 16 ++++++++++++++++
 include/linux/fs.h | 45 ++++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 60 insertions(+), 1 deletion(-)
fs/inode.c +16 -0
index d37fad91c8da..21b026d95b51 100644
@@ -2499,6 +2499,22 @@ struct timespec64 current_time(struct inode *inode)
 }
 EXPORT_SYMBOL(current_time);
 
+/**
+ * inode_set_ctime_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_set_ctime_current(struct inode *inode)
+{
+	struct timespec64 now = current_time(inode);
+
+	inode_set_ctime(inode, now.tv_sec, now.tv_nsec);
+	return now;
+}
+EXPORT_SYMBOL(inode_set_ctime_current);
+
 /**
  * in_group_or_capable - check whether caller is CAP_FSETID privileged
  * @idmap:	idmap of the mount @inode was found from
include/linux/fs.h +44 -1

51 lines not on this page · raw message

Signature · 1 line
2.41.0
author maintainer
#12 lore raw

PATCH v2 8/92 fs: new helper: simple_rename_timestamp

In reply to Jeff Layton

A rename potentially involves updating 4 different inode timestamps. Add a function that handles the details sanely, and convert the libfs.c callers to use it.

---
 fs/libfs.c         | 36 +++++++++++++++++++++++++++---------
 include/linux/fs.h |  2 ++
 2 files changed, 29 insertions(+), 9 deletions(-)
fs/libfs.c +27 -9
index a7e56baf8bbd..9ee79668c909 100644
@@ -692,6 +692,31 @@ int simple_rmdir(struct inode *dir, struct dentry *dentry)
 }
 EXPORT_SYMBOL(simple_rmdir);
 
+/**
+ * simple_rename_timestamp - update the various inode timestamps for rename
+ * @old_dir: old parent directory
+ * @old_dentry: dentry that is being renamed
+ * @new_dir: new parent directory
+ * @new_dentry: target for rename
+ *
+ * POSIX mandates that the old and new parent directories have their ctime and
+ * mtime updated, and that inodes of @old_dentry and @new_dentry (if any), have
+ * their ctime updated.
+ */
+void simple_rename_timestamp(struct inode *old_dir, struct dentry *old_dentry,
+			     struct inode *new_dir, struct dentry *new_dentry)
+{
+	struct inode *newino = d_inode(new_dentry);
+
+	old_dir->i_mtime = inode_set_ctime_current(old_dir);
+	if (new_dir != old_dir)
+		new_dir->i_mtime = inode_set_ctime_current(new_dir);
+	inode_set_ctime_current(d_inode(old_dentry));
+	if (newino)
+		inode_set_ctime_current(newino);
+}
+EXPORT_SYMBOL_GPL(simple_rename_timestamp);
+
 int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry,
 			   struct inode *new_dir, struct dentry *new_dentry)
 {
@@ -707,11 +732,7 @@ 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);
-
+	simple_rename_timestamp(old_dir, old_dentry, new_dir, new_dentry);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(simple_rename_exchange);
@@ -720,7 +741,6 @@ int simple_rename(struct mnt_idmap *idmap, struct inode *old_dir,
 		  struct dentry *old_dentry, struct inode *new_dir,
 		  struct dentry *new_dentry, unsigned int flags)
 {
-	struct inode *inode = d_inode(old_dentry);
 	int they_are_dirs = d_is_dir(old_dentry);
 
 	if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE))
@@ -743,9 +763,7 @@ 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);
-
+	simple_rename_timestamp(old_dir, old_dentry, new_dir, new_dentry);
 	return 0;
 }
 EXPORT_SYMBOL(simple_rename);
include/linux/fs.h +2 -0
index bdfbd11a5811..14e38bd900f1 100644
@@ -2979,6 +2979,8 @@ extern int simple_open(struct inode *inode, struct file *file);
 extern int simple_link(struct dentry *, struct inode *, struct dentry *);
 extern int simple_unlink(struct inode *, struct dentry *);
 extern int simple_rmdir(struct inode *, struct dentry *);
+void simple_rename_timestamp(struct inode *old_dir, struct dentry *old_dentry,
+			     struct inode *new_dir, struct dentry *new_dentry);
 extern int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry,
 				  struct inode *new_dir, struct dentry *new_dentry);
 extern int simple_rename(struct mnt_idmap *, struct inode *,
Signature · 1 line
2.41.0
#13 lore raw

Re: [PATCH v2 08/92] fs: new helper: simple_rename_timestamp

Quoted 47 lines · Jeff Layton

A rename potentially involves updating 4 different inode timestamps. Add a function that handles the details sanely, and convert the libfs.c callers to use it.

---
 fs/libfs.c         | 36 +++++++++++++++++++++++++++---------
 include/linux/fs.h |  2 ++
 2 files changed, 29 insertions(+), 9 deletions(-)
fs/libfs.c +25 -0
index a7e56baf8bbd..9ee79668c909 100644
@@ -692,6 +692,31 @@ int simple_rmdir(struct inode *dir, struct dentry *dentry)
 }
 EXPORT_SYMBOL(simple_rmdir);
 
+/**
+ * simple_rename_timestamp - update the various inode timestamps for rename
+ * @old_dir: old parent directory
+ * @old_dentry: dentry that is being renamed
+ * @new_dir: new parent directory
+ * @new_dentry: target for rename
+ *
+ * POSIX mandates that the old and new parent directories have their ctime and
+ * mtime updated, and that inodes of @old_dentry and @new_dentry (if any), have
+ * their ctime updated.
+ */
+void simple_rename_timestamp(struct inode *old_dir, struct dentry *old_dentry,
+			     struct inode *new_dir, struct dentry *new_dentry)
+{
+	struct inode *newino = d_inode(new_dentry);
+
+	old_dir->i_mtime = inode_set_ctime_current(old_dir);
+	if (new_dir != old_dir)
+		new_dir->i_mtime = inode_set_ctime_current(new_dir);
+	inode_set_ctime_current(d_inode(old_dentry));
+	if (newino)
+		inode_set_ctime_current(newino);
+}
+EXPORT_SYMBOL_GPL(simple_rename_timestamp);
+
 int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry,
 			   struct inode *new_dir, struct dentry *new_dentry)
 {

fs/libfs.c int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry, · line 735

Quoted 9 lines
fs/libfs.c +1 -5
@@ -707,11 +732,7 @@ 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);
-
+	simple_rename_timestamp(old_dir, old_dentry, new_dir, new_dentry);

This is somewhat changing the current behavior: before the patch, the mtime and ctime of old_dir, new_dir and the inodes associated with the dentries are always equal. But given that simple_rename_timestamp() calls inode_set_ctime_current() 4 times, the times could potentially be different.

I am not sure if that is an issue, but it seems that calling inode_set_ctime_current() once, recording the "now" time it sets and using that value to set all times may be more efficient and preserve the existing behavior.

Quoted 35 lines
fs/libfs.c +1 -4
 	return 0;
 }
 EXPORT_SYMBOL_GPL(simple_rename_exchange);
@@ -720,7 +741,6 @@ int simple_rename(struct mnt_idmap *idmap, struct inode *old_dir,
 		  struct dentry *old_dentry, struct inode *new_dir,
 		  struct dentry *new_dentry, unsigned int flags)
 {
-	struct inode *inode = d_inode(old_dentry);
 	int they_are_dirs = d_is_dir(old_dentry);
 
 	if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE))
@@ -743,9 +763,7 @@ 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);
-
+	simple_rename_timestamp(old_dir, old_dentry, new_dir, new_dentry);
 	return 0;
 }
 EXPORT_SYMBOL(simple_rename);
include/linux/fs.h +2 -0
index bdfbd11a5811..14e38bd900f1 100644
@@ -2979,6 +2979,8 @@ extern int simple_open(struct inode *inode, struct file *file);
 extern int simple_link(struct dentry *, struct inode *, struct dentry *);
 extern int simple_unlink(struct inode *, struct dentry *);
 extern int simple_rmdir(struct inode *, struct dentry *);
+void simple_rename_timestamp(struct inode *old_dir, struct dentry *old_dentry,
+			     struct inode *new_dir, struct dentry *new_dentry);
 extern int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry,
 				  struct inode *new_dir, struct dentry *new_dentry);
 extern int simple_rename(struct mnt_idmap *, struct inode *,
Signature · 2 lines
Damien Le Moal
Western Digital Research
author maintainer
#14 lore raw

Re: [PATCH v2 08/92] fs: new helper: simple_rename_timestamp

Quoted 63 lines · Damien Le Moal
Quoted 47 lines · Jeff Layton

A rename potentially involves updating 4 different inode timestamps. Add a function that handles the details sanely, and convert the libfs.c callers to use it.

---
 fs/libfs.c         | 36 +++++++++++++++++++++++++++---------
 include/linux/fs.h |  2 ++
 2 files changed, 29 insertions(+), 9 deletions(-)
fs/libfs.c +25 -0
index a7e56baf8bbd..9ee79668c909 100644
@@ -692,6 +692,31 @@ int simple_rmdir(struct inode *dir, struct dentry *dentry)
 }
 EXPORT_SYMBOL(simple_rmdir);
 
+/**
+ * simple_rename_timestamp - update the various inode timestamps for rename
+ * @old_dir: old parent directory
+ * @old_dentry: dentry that is being renamed
+ * @new_dir: new parent directory
+ * @new_dentry: target for rename
+ *
+ * POSIX mandates that the old and new parent directories have their ctime and
+ * mtime updated, and that inodes of @old_dentry and @new_dentry (if any), have
+ * their ctime updated.
+ */
+void simple_rename_timestamp(struct inode *old_dir, struct dentry *old_dentry,
+			     struct inode *new_dir, struct dentry *new_dentry)
+{
+	struct inode *newino = d_inode(new_dentry);
+
+	old_dir->i_mtime = inode_set_ctime_current(old_dir);
+	if (new_dir != old_dir)
+		new_dir->i_mtime = inode_set_ctime_current(new_dir);
+	inode_set_ctime_current(d_inode(old_dentry));
+	if (newino)
+		inode_set_ctime_current(newino);
+}
+EXPORT_SYMBOL_GPL(simple_rename_timestamp);
+
 int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry,
 			   struct inode *new_dir, struct dentry *new_dentry)
 {

fs/libfs.c int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry, · line 735

Quoted 9 lines
fs/libfs.c +1 -5
@@ -707,11 +732,7 @@ 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);
-
+	simple_rename_timestamp(old_dir, old_dentry, new_dir, new_dentry);

This is somewhat changing the current behavior: before the patch, the mtime and ctime of old_dir, new_dir and the inodes associated with the dentries are always equal. But given that simple_rename_timestamp() calls inode_set_ctime_current() 4 times, the times could potentially be different.

Quoted 4 lines

I am not sure if that is an issue, but it seems that calling inode_set_ctime_current() once, recording the "now" time it sets and using that value to set all times may be more efficient and preserve the existing behavior.

I don't believe it's an issue. I've seen nothing in the POSIX spec that mandates that timestamp updates to different inodes involved in an operation be set to the _same_ value. It just says they must be updated.

It's also hard to believe that any software would depend on this either, given that it's very inconsistent across filesystems today. AFAICT, this was mostly done in the past just as a matter of convenience.

The other problem with doing it that way is that it assumes that current_time(inode) should always return the same value when given different inodes. Is it really correct to do this?

	inode_set_ctime(dir, inode_set_ctime_current(inode));

"dir" and "inode" are different inodes, after all, and you're setting dir's timestamp to "inode"'s value. It's not a big deal today since they're always on the same sb, but the ultimate goal of these changes is to implement multigrain timestamps. That will mean that fetching a fine- grained timestamp for an update when the existing mtime or ctime value has been queried via getattr.

With that change, I think it's best that we treat updates to different inodes individually, as some of them may require updating with a fine- grained timestamp and some may not.

Quoted 36 lines
Quoted 35 lines
fs/libfs.c +1 -4
 	return 0;
 }
 EXPORT_SYMBOL_GPL(simple_rename_exchange);
@@ -720,7 +741,6 @@ int simple_rename(struct mnt_idmap *idmap, struct inode *old_dir,
 		  struct dentry *old_dentry, struct inode *new_dir,
 		  struct dentry *new_dentry, unsigned int flags)
 {
-	struct inode *inode = d_inode(old_dentry);
 	int they_are_dirs = d_is_dir(old_dentry);
 
 	if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE))
@@ -743,9 +763,7 @@ 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);
-
+	simple_rename_timestamp(old_dir, old_dentry, new_dir, new_dentry);
 	return 0;
 }
 EXPORT_SYMBOL(simple_rename);
include/linux/fs.h +2 -0
index bdfbd11a5811..14e38bd900f1 100644
@@ -2979,6 +2979,8 @@ extern int simple_open(struct inode *inode, struct file *file);
 extern int simple_link(struct dentry *, struct inode *, struct dentry *);
 extern int simple_unlink(struct inode *, struct dentry *);
 extern int simple_rmdir(struct inode *, struct dentry *);
+void simple_rename_timestamp(struct inode *old_dir, struct dentry *old_dentry,
+			     struct inode *new_dir, struct dentry *new_dentry);
 extern int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry,
 				  struct inode *new_dir, struct dentry *new_dentry);
 extern int simple_rename(struct mnt_idmap *, struct inode *,
Signature · 1 line
Jeff Layton <[email protected]>
#15 lore raw

Re: [apparmor] [PATCH v2 08/92] fs: new helper: simple_rename_timestamp

Quoted 8 lines · Jeff Layton

I don't believe it's an issue. I've seen nothing in the POSIX spec that mandates that timestamp updates to different inodes involved in an operation be set to the _same_ value. It just says they must be updated.

It's also hard to believe that any software would depend on this either, given that it's very inconsistent across filesystems today. AFAICT, this was mostly done in the past just as a matter of convenience.

I've seen this assumption in several programs:

mutt buffy.c
https://sources.debian.org/src/mutt/2.2.9-1/buffy.c/?hl=625#L625

  if (mailbox->newly_created &&
      (sb->st_ctime != sb->st_mtime || sb->st_ctime != sb->st_atime))
    mailbox->newly_created = 0;

neomutt mbox/mbox.c
https://sources.debian.org/src/neomutt/20220429+dfsg1-4.1/mbox/mbox.c/?hl=1820#L1820

  if (m->newly_created && ((st.st_ctime != st.st_mtime) || (st.st_ctime != st.st_atime)))
    m->newly_created = false;

screen logfile.c
https://sources.debian.org/src/screen/4.9.0-4/logfile.c/?hl=130#L130

  if ((!s->st_dev && !s->st_ino) ||             /* stat failed, that's new! */
      !s->st_nlink ||                           /* red alert: file unlinked */
      (s->st_size < o.st_size) ||               /*           file truncated */
      (s->st_mtime != o.st_mtime) ||            /*            file modified */
      ((s->st_ctime != o.st_ctime) &&           /*     file changed (moved) */
       !(s->st_mtime == s->st_ctime &&          /*  and it was not a change */
         o.st_ctime < s->st_ctime)))            /* due to delayed nfs write */

{

nemo libnemo-private/nemo-vfs-file.c
https://sources.debian.org/src/nemo/5.6.5-1/libnemo-private/nemo-vfs-file.c/?hl=344#L344

		/* mtime is when the contents changed; ctime is when the
		 * contents or the permissions (inc. owner/group) changed.
		 * So we can only know when the permissions changed if mtime
		 * and ctime are different.
		 */
		if (file->details->mtime == file->details->ctime) {
			return FALSE;
		}

While looking for more examples, I found a perl test that seems to suggest that at least Solaris, AFS, AmigaOS, DragonFly BSD do as you suggest:
https://sources.debian.org/src/perl/5.36.0-7/t/op/stat.t/?hl=158#L140

Thanks

author maintainer
#16 lore raw

Re: [apparmor] [PATCH v2 08/92] fs: new helper: simple_rename_timestamp

Quoted 6 lines · Seth Arnold
Quoted 5 lines · Jeff Layton

I don't believe it's an issue. I've seen nothing in the POSIX spec that mandates that timestamp updates to different inodes involved in an operation be set to the _same_ value. It just says they must be updated.

Quoted 6 lines
Quoted 3 lines

It's also hard to believe that any software would depend on this either, given that it's very inconsistent across filesystems today. AFAICT, this was mostly done in the past just as a matter of convenience.

I've seen this assumption in several programs:

Thanks for looking into this!

To be clear, POSIX doesn't require that _different_ inodes ever be set to the same timestamp value. IOW, it certainly doesn't require that the source and target directories on a rename() end up with the exact same timestamp value.

Granted, POSIX is rather vague on timestamps in general, but most of the examples below involve comparing different timestamps on the _same_ inode.

Quoted 7 lines

mutt buffy.c
https://sources.debian.org/src/mutt/2.2.9-1/buffy.c/?hl=625#L625

  if (mailbox->newly_created &&
      (sb->st_ctime != sb->st_mtime || sb->st_ctime != sb->st_atime))
    mailbox->newly_created = 0;

This should be fine with this patchset. Note that this is comparing a/c/mtime on the same inode, and our usual pattern on inode instantiation is:

    inode->i_atime = inode->i_mtime = inode_set_ctime_current(inode);

...which should result in all of inode's timestamps being synchronized.

Quoted 7 lines

neomutt mbox/mbox.c
https://sources.debian.org/src/neomutt/20220429+dfsg1-4.1/mbox/mbox.c/?hl=1820#L1820

  if (m->newly_created && ((st.st_ctime != st.st_mtime) || (st.st_ctime != st.st_atime)))
    m->newly_created = false;

Ditto here.

Quoted 4 lines

screen logfile.c
https://sources.debian.org/src/screen/4.9.0-4/logfile.c/?hl=130#L130

Quoted 9 lines
  if ((!s->st_dev && !s->st_ino) ||             /* stat failed, that's new! */
      !s->st_nlink ||                           /* red alert: file unlinked */
      (s->st_size < o.st_size) ||               /*           file truncated */
      (s->st_mtime != o.st_mtime) ||            /*            file modified */
      ((s->st_ctime != o.st_ctime) &&           /*     file changed (moved) */
       !(s->st_mtime == s->st_ctime &&          /*  and it was not a change */
         o.st_ctime < s->st_ctime)))            /* due to delayed nfs write */

{

This one is really weird. You have two different struct stat's, "o" and "s". I assume though that these should be stat values from the same inode, because otherwise this comparison would make no sense:

      ((s->st_ctime != o.st_ctime) &&           /*     file changed (moved) */

In general, we can never contrive to ensure that the ctime of two different inodes are the same, since that is always set by the kernel to the current time, and you'd have to ensure that they were created within the same jiffy (at least with today's code).

Quoted 12 lines

nemo libnemo-private/nemo-vfs-file.c
https://sources.debian.org/src/nemo/5.6.5-1/libnemo-private/nemo-vfs-file.c/?hl=344#L344

		/* mtime is when the contents changed; ctime is when the
		 * contents or the permissions (inc. owner/group) changed.
		 * So we can only know when the permissions changed if mtime
		 * and ctime are different.
		 */
		if (file->details->mtime == file->details->ctime) {
			return FALSE;
		}

Ditto here with the first examples. This involves comparing timestamps on the same inode, which should be fine.

Quoted 5 lines

While looking for more examples, I found a perl test that seems to suggest that at least Solaris, AFS, AmigaOS, DragonFly BSD do as you suggest:
https://sources.debian.org/src/perl/5.36.0-7/t/op/stat.t/?hl=158#L140

(I kinda miss Perl. I wrote a bunch of stuff in it in the 90's and early naughties)

I think this test is supposed to be testing whether the mtime changes on link() ?

-----------------8<----------------

    my($nlink, $mtime, $ctime) = (stat($tmpfile))[$NLINK, $MTIME, $CTIME];

[...]

skip "Solaris tmpfs has different mtime/ctime link semantics", 2
if $Is_Solaris and $cwd =~ m#^/tmp# and
$mtime && $mtime == $ctime;
-----------------8<----------------

...again, I think this would be ok too since it's just comparing the mtime and ctime of the same inode. Granted this is a Solaris-specific test, but Linux would be fine here too.

So in conclusion, I don't think this patchset will cause problems with any of the above code.

Signature · 1 line
Jeff Layton <[email protected]>

A message that is not in the archive (sent off-list or not yet received): 3b403ef1-22e6-0220-6c9c-435e3444b4d3-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org

#18 lore raw

Re: [PATCH v2 08/92] fs: new helper: simple_rename_timestamp

Quoted 63 lines · Damien Le Moal
Quoted 47 lines · Jeff Layton

A rename potentially involves updating 4 different inode timestamps. Add a function that handles the details sanely, and convert the libfs.c callers to use it.

---
 fs/libfs.c         | 36 +++++++++++++++++++++++++++---------
 include/linux/fs.h |  2 ++
 2 files changed, 29 insertions(+), 9 deletions(-)
fs/libfs.c +25 -0
index a7e56baf8bbd..9ee79668c909 100644
@@ -692,6 +692,31 @@ int simple_rmdir(struct inode *dir, struct dentry *dentry)
 }
 EXPORT_SYMBOL(simple_rmdir);
 
+/**
+ * simple_rename_timestamp - update the various inode timestamps for rename
+ * @old_dir: old parent directory
+ * @old_dentry: dentry that is being renamed
+ * @new_dir: new parent directory
+ * @new_dentry: target for rename
+ *
+ * POSIX mandates that the old and new parent directories have their ctime and
+ * mtime updated, and that inodes of @old_dentry and @new_dentry (if any), have
+ * their ctime updated.
+ */
+void simple_rename_timestamp(struct inode *old_dir, struct dentry *old_dentry,
+			     struct inode *new_dir, struct dentry *new_dentry)
+{
+	struct inode *newino = d_inode(new_dentry);
+
+	old_dir->i_mtime = inode_set_ctime_current(old_dir);
+	if (new_dir != old_dir)
+		new_dir->i_mtime = inode_set_ctime_current(new_dir);
+	inode_set_ctime_current(d_inode(old_dentry));
+	if (newino)
+		inode_set_ctime_current(newino);
+}
+EXPORT_SYMBOL_GPL(simple_rename_timestamp);
+
 int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry,
 			   struct inode *new_dir, struct dentry *new_dentry)
 {

fs/libfs.c int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry, · line 735

Quoted 9 lines
fs/libfs.c +1 -5
@@ -707,11 +732,7 @@ 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);
-
+	simple_rename_timestamp(old_dir, old_dentry, new_dir, new_dentry);

This is somewhat changing the current behavior: before the patch, the mtime and ctime of old_dir, new_dir and the inodes associated with the dentries are always equal. But given that simple_rename_timestamp() calls inode_set_ctime_current() 4 times, the times could potentially be different.

Quoted 4 lines

I am not sure if that is an issue, but it seems that calling inode_set_ctime_current() once, recording the "now" time it sets and using that value to set all times may be more efficient and preserve the existing behavior.

I don't believe it's an issue. I've seen nothing in the POSIX spec that mandates that timestamp updates to different inodes involved in an operation be set to the _same_ value. It just says they must be updated.

It's also hard to believe that any software would depend on this either, given that it's very inconsistent across filesystems today. AFAICT, this was mostly done in the past just as a matter of convenience.

The other problem with doing it that way is that it assumes that current_time(inode) should always return the same value when given different inodes. Is it really correct to do this?

	inode_set_ctime(dir, inode_set_ctime_current(inode));

"dir" and "inode" are different inodes, after all, and you're setting dir's timestamp to "inode"'s value. It's not a big deal today since they're always on the same sb, but the ultimate goal of these changes is to implement multigrain timestamps. That will mean that fetching a fine- grained timestamp for an update when the existing mtime or ctime value has been queried via getattr.

With that change, I think it's best that we treat updates to different inodes individually, as some of them may require updating with a fine- grained timestamp and some may not.

Quoted 36 lines
Quoted 35 lines
fs/libfs.c +1 -4
 	return 0;
 }
 EXPORT_SYMBOL_GPL(simple_rename_exchange);
@@ -720,7 +741,6 @@ int simple_rename(struct mnt_idmap *idmap, struct inode *old_dir,
 		  struct dentry *old_dentry, struct inode *new_dir,
 		  struct dentry *new_dentry, unsigned int flags)
 {
-	struct inode *inode = d_inode(old_dentry);
 	int they_are_dirs = d_is_dir(old_dentry);
 
 	if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE))
@@ -743,9 +763,7 @@ 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);
-
+	simple_rename_timestamp(old_dir, old_dentry, new_dir, new_dentry);
 	return 0;
 }
 EXPORT_SYMBOL(simple_rename);
include/linux/fs.h +2 -0
index bdfbd11a5811..14e38bd900f1 100644
@@ -2979,6 +2979,8 @@ extern int simple_open(struct inode *inode, struct file *file);
 extern int simple_link(struct dentry *, struct inode *, struct dentry *);
 extern int simple_unlink(struct inode *, struct dentry *);
 extern int simple_rmdir(struct inode *, struct dentry *);
+void simple_rename_timestamp(struct inode *old_dir, struct dentry *old_dentry,
+			     struct inode *new_dir, struct dentry *new_dentry);
 extern int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry,
 				  struct inode *new_dir, struct dentry *new_dentry);
 extern int simple_rename(struct mnt_idmap *, struct inode *,
Signature · 1 line
Jeff Layton <[email protected]>
author maintainer
#19 lore raw

Re: [PATCH v2 08/92] fs: new helper: simple_rename_timestamp

In reply to Damien Le Moal
Quoted 63 lines · Damien Le Moal
Quoted 47 lines · Jeff Layton

A rename potentially involves updating 4 different inode timestamps. Add a function that handles the details sanely, and convert the libfs.c callers to use it.

---
 fs/libfs.c         | 36 +++++++++++++++++++++++++++---------
 include/linux/fs.h |  2 ++
 2 files changed, 29 insertions(+), 9 deletions(-)
fs/libfs.c +25 -0
index a7e56baf8bbd..9ee79668c909 100644
@@ -692,6 +692,31 @@ int simple_rmdir(struct inode *dir, struct dentry *dentry)
 }
 EXPORT_SYMBOL(simple_rmdir);
 
+/**
+ * simple_rename_timestamp - update the various inode timestamps for rename
+ * @old_dir: old parent directory
+ * @old_dentry: dentry that is being renamed
+ * @new_dir: new parent directory
+ * @new_dentry: target for rename
+ *
+ * POSIX mandates that the old and new parent directories have their ctime and
+ * mtime updated, and that inodes of @old_dentry and @new_dentry (if any), have
+ * their ctime updated.
+ */
+void simple_rename_timestamp(struct inode *old_dir, struct dentry *old_dentry,
+			     struct inode *new_dir, struct dentry *new_dentry)
+{
+	struct inode *newino = d_inode(new_dentry);
+
+	old_dir->i_mtime = inode_set_ctime_current(old_dir);
+	if (new_dir != old_dir)
+		new_dir->i_mtime = inode_set_ctime_current(new_dir);
+	inode_set_ctime_current(d_inode(old_dentry));
+	if (newino)
+		inode_set_ctime_current(newino);
+}
+EXPORT_SYMBOL_GPL(simple_rename_timestamp);
+
 int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry,
 			   struct inode *new_dir, struct dentry *new_dentry)
 {

fs/libfs.c int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry, · line 735

Quoted 9 lines
fs/libfs.c +1 -5
@@ -707,11 +732,7 @@ 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);
-
+	simple_rename_timestamp(old_dir, old_dentry, new_dir, new_dentry);

This is somewhat changing the current behavior: before the patch, the mtime and ctime of old_dir, new_dir and the inodes associated with the dentries are always equal. But given that simple_rename_timestamp() calls inode_set_ctime_current() 4 times, the times could potentially be different.

Quoted 4 lines

I am not sure if that is an issue, but it seems that calling inode_set_ctime_current() once, recording the "now" time it sets and using that value to set all times may be more efficient and preserve the existing behavior.

I don't believe it's an issue. I've seen nothing in the POSIX spec that mandates that timestamp updates to different inodes involved in an operation be set to the _same_ value. It just says they must be updated.

It's also hard to believe that any software would depend on this either, given that it's very inconsistent across filesystems today. AFAICT, this was mostly done in the past just as a matter of convenience.

The other problem with doing it that way is that it assumes that current_time(inode) should always return the same value when given different inodes. Is it really correct to do this?

	inode_set_ctime(dir, inode_set_ctime_current(inode));

"dir" and "inode" are different inodes, after all, and you're setting dir's timestamp to "inode"'s value. It's not a big deal today since they're always on the same sb, but the ultimate goal of these changes is to implement multigrain timestamps. That will mean that fetching a fine- grained timestamp for an update when the existing mtime or ctime value has been queried via getattr.

With that change, I think it's best that we treat updates to different inodes individually, as some of them may require updating with a fine- grained timestamp and some may not.

Quoted 36 lines
Quoted 35 lines
fs/libfs.c +1 -4
 	return 0;
 }
 EXPORT_SYMBOL_GPL(simple_rename_exchange);
@@ -720,7 +741,6 @@ int simple_rename(struct mnt_idmap *idmap, struct inode *old_dir,
 		  struct dentry *old_dentry, struct inode *new_dir,
 		  struct dentry *new_dentry, unsigned int flags)
 {
-	struct inode *inode = d_inode(old_dentry);
 	int they_are_dirs = d_is_dir(old_dentry);
 
 	if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE))
@@ -743,9 +763,7 @@ 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);
-
+	simple_rename_timestamp(old_dir, old_dentry, new_dir, new_dentry);
 	return 0;
 }
 EXPORT_SYMBOL(simple_rename);
include/linux/fs.h +2 -0
index bdfbd11a5811..14e38bd900f1 100644
@@ -2979,6 +2979,8 @@ extern int simple_open(struct inode *inode, struct file *file);
 extern int simple_link(struct dentry *, struct inode *, struct dentry *);
 extern int simple_unlink(struct inode *, struct dentry *);
 extern int simple_rmdir(struct inode *, struct dentry *);
+void simple_rename_timestamp(struct inode *old_dir, struct dentry *old_dentry,
+			     struct inode *new_dir, struct dentry *new_dentry);
 extern int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry,
 				  struct inode *new_dir, struct dentry *new_dentry);
 extern int simple_rename(struct mnt_idmap *, struct inode *,
Signature · 1 line
Jeff Layton <[email protected]>

A message that is not in the archive (sent off-list or not yet received): 20230705185812.579118-3-jlayton-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org

#21 lore raw

Re: [PATCH v2 08/92] fs: new helper: simple_rename_timestamp

Quoted 47 lines · Jeff Layton

A rename potentially involves updating 4 different inode timestamps. Add a function that handles the details sanely, and convert the libfs.c callers to use it.

---
 fs/libfs.c         | 36 +++++++++++++++++++++++++++---------
 include/linux/fs.h |  2 ++
 2 files changed, 29 insertions(+), 9 deletions(-)
fs/libfs.c +25 -0
index a7e56baf8bbd..9ee79668c909 100644
@@ -692,6 +692,31 @@ int simple_rmdir(struct inode *dir, struct dentry *dentry)
 }
 EXPORT_SYMBOL(simple_rmdir);
 
+/**
+ * simple_rename_timestamp - update the various inode timestamps for rename
+ * @old_dir: old parent directory
+ * @old_dentry: dentry that is being renamed
+ * @new_dir: new parent directory
+ * @new_dentry: target for rename
+ *
+ * POSIX mandates that the old and new parent directories have their ctime and
+ * mtime updated, and that inodes of @old_dentry and @new_dentry (if any), have
+ * their ctime updated.
+ */
+void simple_rename_timestamp(struct inode *old_dir, struct dentry *old_dentry,
+			     struct inode *new_dir, struct dentry *new_dentry)
+{
+	struct inode *newino = d_inode(new_dentry);
+
+	old_dir->i_mtime = inode_set_ctime_current(old_dir);
+	if (new_dir != old_dir)
+		new_dir->i_mtime = inode_set_ctime_current(new_dir);
+	inode_set_ctime_current(d_inode(old_dentry));
+	if (newino)
+		inode_set_ctime_current(newino);
+}
+EXPORT_SYMBOL_GPL(simple_rename_timestamp);
+
 int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry,
 			   struct inode *new_dir, struct dentry *new_dentry)
 {

fs/libfs.c int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry, · line 735

Quoted 9 lines
fs/libfs.c +1 -5
@@ -707,11 +732,7 @@ 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);
-
+	simple_rename_timestamp(old_dir, old_dentry, new_dir, new_dentry);

This is somewhat changing the current behavior: before the patch, the mtime and ctime of old_dir, new_dir and the inodes associated with the dentries are always equal. But given that simple_rename_timestamp() calls inode_set_ctime_current() 4 times, the times could potentially be different.

I am not sure if that is an issue, but it seems that calling inode_set_ctime_current() once, recording the "now" time it sets and using that value to set all times may be more efficient and preserve the existing behavior.

Quoted 35 lines
fs/libfs.c +1 -4
 	return 0;
 }
 EXPORT_SYMBOL_GPL(simple_rename_exchange);
@@ -720,7 +741,6 @@ int simple_rename(struct mnt_idmap *idmap, struct inode *old_dir,
 		  struct dentry *old_dentry, struct inode *new_dir,
 		  struct dentry *new_dentry, unsigned int flags)
 {
-	struct inode *inode = d_inode(old_dentry);
 	int they_are_dirs = d_is_dir(old_dentry);
 
 	if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE))
@@ -743,9 +763,7 @@ 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);
-
+	simple_rename_timestamp(old_dir, old_dentry, new_dir, new_dentry);
 	return 0;
 }
 EXPORT_SYMBOL(simple_rename);
include/linux/fs.h +2 -0
index bdfbd11a5811..14e38bd900f1 100644
@@ -2979,6 +2979,8 @@ extern int simple_open(struct inode *inode, struct file *file);
 extern int simple_link(struct dentry *, struct inode *, struct dentry *);
 extern int simple_unlink(struct inode *, struct dentry *);
 extern int simple_rmdir(struct inode *, struct dentry *);
+void simple_rename_timestamp(struct inode *old_dir, struct dentry *old_dentry,
+			     struct inode *new_dir, struct dentry *new_dentry);
 extern int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry,
 				  struct inode *new_dir, struct dentry *new_dentry);
 extern int simple_rename(struct mnt_idmap *, struct inode *,
Signature · 2 lines
Damien Le Moal
Western Digital Research
#22 lore raw

Re: [PATCH v2 08/92] fs: new helper: simple_rename_timestamp

In reply to Jeff Layton
Quoted 47 lines · Jeff Layton

A rename potentially involves updating 4 different inode timestamps. Add a function that handles the details sanely, and convert the libfs.c callers to use it.

---
 fs/libfs.c         | 36 +++++++++++++++++++++++++++---------
 include/linux/fs.h |  2 ++
 2 files changed, 29 insertions(+), 9 deletions(-)
fs/libfs.c +25 -0
index a7e56baf8bbd..9ee79668c909 100644
@@ -692,6 +692,31 @@ int simple_rmdir(struct inode *dir, struct dentry *dentry)
 }
 EXPORT_SYMBOL(simple_rmdir);
 
+/**
+ * simple_rename_timestamp - update the various inode timestamps for rename
+ * @old_dir: old parent directory
+ * @old_dentry: dentry that is being renamed
+ * @new_dir: new parent directory
+ * @new_dentry: target for rename
+ *
+ * POSIX mandates that the old and new parent directories have their ctime and
+ * mtime updated, and that inodes of @old_dentry and @new_dentry (if any), have
+ * their ctime updated.
+ */
+void simple_rename_timestamp(struct inode *old_dir, struct dentry *old_dentry,
+			     struct inode *new_dir, struct dentry *new_dentry)
+{
+	struct inode *newino = d_inode(new_dentry);
+
+	old_dir->i_mtime = inode_set_ctime_current(old_dir);
+	if (new_dir != old_dir)
+		new_dir->i_mtime = inode_set_ctime_current(new_dir);
+	inode_set_ctime_current(d_inode(old_dentry));
+	if (newino)
+		inode_set_ctime_current(newino);
+}
+EXPORT_SYMBOL_GPL(simple_rename_timestamp);
+
 int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry,
 			   struct inode *new_dir, struct dentry *new_dentry)
 {

fs/libfs.c int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry, · line 735

Quoted 9 lines
fs/libfs.c +1 -5
@@ -707,11 +732,7 @@ 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);
-
+	simple_rename_timestamp(old_dir, old_dentry, new_dir, new_dentry);

This is somewhat changing the current behavior: before the patch, the mtime and ctime of old_dir, new_dir and the inodes associated with the dentries are always equal. But given that simple_rename_timestamp() calls inode_set_ctime_current() 4 times, the times could potentially be different.

I am not sure if that is an issue, but it seems that calling inode_set_ctime_current() once, recording the "now" time it sets and using that value to set all times may be more efficient and preserve the existing behavior.

Quoted 35 lines
fs/libfs.c +1 -4
 	return 0;
 }
 EXPORT_SYMBOL_GPL(simple_rename_exchange);
@@ -720,7 +741,6 @@ int simple_rename(struct mnt_idmap *idmap, struct inode *old_dir,
 		  struct dentry *old_dentry, struct inode *new_dir,
 		  struct dentry *new_dentry, unsigned int flags)
 {
-	struct inode *inode = d_inode(old_dentry);
 	int they_are_dirs = d_is_dir(old_dentry);
 
 	if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE))
@@ -743,9 +763,7 @@ 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);
-
+	simple_rename_timestamp(old_dir, old_dentry, new_dir, new_dentry);
 	return 0;
 }
 EXPORT_SYMBOL(simple_rename);
include/linux/fs.h +2 -0
index bdfbd11a5811..14e38bd900f1 100644
@@ -2979,6 +2979,8 @@ extern int simple_open(struct inode *inode, struct file *file);
 extern int simple_link(struct dentry *, struct inode *, struct dentry *);
 extern int simple_unlink(struct inode *, struct dentry *);
 extern int simple_rmdir(struct inode *, struct dentry *);
+void simple_rename_timestamp(struct inode *old_dir, struct dentry *old_dentry,
+			     struct inode *new_dir, struct dentry *new_dentry);
 extern int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry,
 				  struct inode *new_dir, struct dentry *new_dentry);
 extern int simple_rename(struct mnt_idmap *, struct inode *,
Signature · 2 lines
Damien Le Moal
Western Digital Research
maintainer Reviewed-by on v2 8/92
#23 lore raw

Re: [PATCH v2 08/92] fs: new helper: simple_rename_timestamp

In reply to Jeff Layton
Quoted 5 lines · Jeff Layton

A rename potentially involves updating 4 different inode timestamps. Add a function that handles the details sanely, and convert the libfs.c callers to use it.

Looks good to me. Feel free to add:

Honza

Quoted 89 lines
---
 fs/libfs.c         | 36 +++++++++++++++++++++++++++---------
 include/linux/fs.h |  2 ++
 2 files changed, 29 insertions(+), 9 deletions(-)
fs/libfs.c +27 -9
index a7e56baf8bbd..9ee79668c909 100644
@@ -692,6 +692,31 @@ int simple_rmdir(struct inode *dir, struct dentry *dentry)
 }
 EXPORT_SYMBOL(simple_rmdir);
 
+/**
+ * simple_rename_timestamp - update the various inode timestamps for rename
+ * @old_dir: old parent directory
+ * @old_dentry: dentry that is being renamed
+ * @new_dir: new parent directory
+ * @new_dentry: target for rename
+ *
+ * POSIX mandates that the old and new parent directories have their ctime and
+ * mtime updated, and that inodes of @old_dentry and @new_dentry (if any), have
+ * their ctime updated.
+ */
+void simple_rename_timestamp(struct inode *old_dir, struct dentry *old_dentry,
+			     struct inode *new_dir, struct dentry *new_dentry)
+{
+	struct inode *newino = d_inode(new_dentry);
+
+	old_dir->i_mtime = inode_set_ctime_current(old_dir);
+	if (new_dir != old_dir)
+		new_dir->i_mtime = inode_set_ctime_current(new_dir);
+	inode_set_ctime_current(d_inode(old_dentry));
+	if (newino)
+		inode_set_ctime_current(newino);
+}
+EXPORT_SYMBOL_GPL(simple_rename_timestamp);
+
 int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry,
 			   struct inode *new_dir, struct dentry *new_dentry)
 {
@@ -707,11 +732,7 @@ 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);
-
+	simple_rename_timestamp(old_dir, old_dentry, new_dir, new_dentry);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(simple_rename_exchange);
@@ -720,7 +741,6 @@ int simple_rename(struct mnt_idmap *idmap, struct inode *old_dir,
 		  struct dentry *old_dentry, struct inode *new_dir,
 		  struct dentry *new_dentry, unsigned int flags)
 {
-	struct inode *inode = d_inode(old_dentry);
 	int they_are_dirs = d_is_dir(old_dentry);
 
 	if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE))
@@ -743,9 +763,7 @@ 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);
-
+	simple_rename_timestamp(old_dir, old_dentry, new_dir, new_dentry);
 	return 0;
 }
 EXPORT_SYMBOL(simple_rename);
include/linux/fs.h +2 -0
index bdfbd11a5811..14e38bd900f1 100644
@@ -2979,6 +2979,8 @@ extern int simple_open(struct inode *inode, struct file *file);
 extern int simple_link(struct dentry *, struct inode *, struct dentry *);
 extern int simple_unlink(struct inode *, struct dentry *);
 extern int simple_rmdir(struct inode *, struct dentry *);
+void simple_rename_timestamp(struct inode *old_dir, struct dentry *old_dentry,
+			     struct inode *new_dir, struct dentry *new_dentry);
 extern int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry,
 				  struct inode *new_dir, struct dentry *new_dentry);
 extern int simple_rename(struct mnt_idmap *, struct inode *,
Signature · 1 line
2.41.0
Signature · 2 lines
Jan Kara <jack-IBi9RG/[email protected]>
SUSE Labs, CR
maintainer
#24 lore raw

Re: [PATCH v2 08/92] fs: new helper: simple_rename_timestamp

In reply to Jeff Layton
Quoted 3 lines · Jeff Layton

+ * POSIX mandates that the old and new parent directories have their ctime and
+ * mtime updated, and that inodes of @old_dentry and @new_dentry (if any), have
+ * their ctime updated.

APPLICATION USAGE
Some implementations mark for update the last file status change timestamp of renamed files and some do not. Applications which make use of the last file status change timestamp may behave differently with respect to renamed files unless they are designed to allow for either behavior.

So for children POSIX permits rather than mandates. Doesn't really matter; Linux behaviour had been to touch ctime on children since way back, if not since the very beginning.

author maintainer
#25 lore raw

Re: [PATCH v2 08/92] fs: new helper: simple_rename_timestamp

Quoted 12 lines · Al Viro
Quoted 3 lines · Jeff Layton

+ * POSIX mandates that the old and new parent directories have their ctime and
+ * mtime updated, and that inodes of @old_dentry and @new_dentry (if any), have
+ * their ctime updated.

APPLICATION USAGE
Some implementations mark for update the last file status change timestamp of renamed files and some do not. Applications which make use of the last file status change timestamp may behave differently with respect to renamed files unless they are designed to allow for either behavior.

Quoted 3 lines

So for children POSIX permits rather than mandates. Doesn't really matter; Linux behaviour had been to touch ctime on children since way back, if not since the very beginning.

Mea culpa. You're quite correct. I'll plan to roll a small patch to update the comment over this function.

Thanks!

Signature · 1 line
Jeff Layton <[email protected]>
author maintainer
#26 lore raw

PATCH v2 8/92 fs: new helper: simple_rename_timestamp

In reply to Jeff Layton

A rename potentially involves updating 4 different inode timestamps. Add a function that handles the details sanely, and convert the libfs.c callers to use it.

---
 fs/libfs.c         | 36 +++++++++++++++++++++++++++---------
 include/linux/fs.h |  2 ++
 2 files changed, 29 insertions(+), 9 deletions(-)
fs/libfs.c +27 -9
index a7e56baf8bbd..9ee79668c909 100644
@@ -692,6 +692,31 @@ int simple_rmdir(struct inode *dir, struct dentry *dentry)
 }
 EXPORT_SYMBOL(simple_rmdir);
 
+/**
+ * simple_rename_timestamp - update the various inode timestamps for rename
+ * @old_dir: old parent directory
+ * @old_dentry: dentry that is being renamed
+ * @new_dir: new parent directory
+ * @new_dentry: target for rename
+ *
+ * POSIX mandates that the old and new parent directories have their ctime and
+ * mtime updated, and that inodes of @old_dentry and @new_dentry (if any), have
+ * their ctime updated.
+ */
+void simple_rename_timestamp(struct inode *old_dir, struct dentry *old_dentry,
+			     struct inode *new_dir, struct dentry *new_dentry)
+{
+	struct inode *newino = d_inode(new_dentry);
+
+	old_dir->i_mtime = inode_set_ctime_current(old_dir);
+	if (new_dir != old_dir)
+		new_dir->i_mtime = inode_set_ctime_current(new_dir);
+	inode_set_ctime_current(d_inode(old_dentry));
+	if (newino)
+		inode_set_ctime_current(newino);
+}
+EXPORT_SYMBOL_GPL(simple_rename_timestamp);
+
 int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry,
 			   struct inode *new_dir, struct dentry *new_dentry)
 {
@@ -707,11 +732,7 @@ 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);
-
+	simple_rename_timestamp(old_dir, old_dentry, new_dir, new_dentry);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(simple_rename_exchange);
@@ -720,7 +741,6 @@ int simple_rename(struct mnt_idmap *idmap, struct inode *old_dir,
 		  struct dentry *old_dentry, struct inode *new_dir,
 		  struct dentry *new_dentry, unsigned int flags)
 {
-	struct inode *inode = d_inode(old_dentry);
 	int they_are_dirs = d_is_dir(old_dentry);
 
 	if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE))
@@ -743,9 +763,7 @@ 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);
-
+	simple_rename_timestamp(old_dir, old_dentry, new_dir, new_dentry);
 	return 0;
 }
 EXPORT_SYMBOL(simple_rename);
include/linux/fs.h +2 -0
index bdfbd11a5811..14e38bd900f1 100644
@@ -2979,6 +2979,8 @@ extern int simple_open(struct inode *inode, struct file *file);
 extern int simple_link(struct dentry *, struct inode *, struct dentry *);
 extern int simple_unlink(struct inode *, struct dentry *);
 extern int simple_rmdir(struct inode *, struct dentry *);
+void simple_rename_timestamp(struct inode *old_dir, struct dentry *old_dentry,
+			     struct inode *new_dir, struct dentry *new_dentry);
 extern int simple_rename_exchange(struct inode *old_dir, struct dentry *old_dentry,
 				  struct inode *new_dir, struct dentry *new_dentry);
 extern int simple_rename(struct mnt_idmap *, struct inode *,
Signature · 1 line
2.41.0
author maintainer
#27 lore raw

PATCH v2 92/92 fs: rename i_ctime field to __i_ctime

In reply to Jeff Layton

Now that everything in-tree is converted to use the accessor functions, rename the i_ctime field in the inode to discourage direct access.

---
 include/linux/fs.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
include/linux/fs.h +3 -3
index 14e38bd900f1..b66442f91835 100644
@@ -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_set_ctime_current(struct inode *inode);
  */
 static inline struct timespec64 inode_get_ctime(const struct inode *inode)
 {
-	return inode->i_ctime;
+	return inode->__i_ctime;
 }
 
 /**
@@ -1498,7 +1498,7 @@ static inline struct timespec64 inode_get_ctime(const struct inode *inode)
 static inline struct timespec64 inode_set_ctime_to_ts(struct inode *inode,
 						      struct timespec64 ts)
 {
-	inode->i_ctime = ts;
+	inode->__i_ctime = ts;
 	return ts;
 }
 
Signature · 1 line
2.41.0
Reviewed-by on v2 92/92
#28 lore raw

Re: [PATCH v2 92/92] fs: rename i_ctime field to __i_ctime

Quoted 4 lines · Jeff Layton

Now that everything in-tree is converted to use the accessor functions, rename the i_ctime field in the inode to discourage direct access.

Looks OK to me.

Signature · 2 lines
Damien Le Moal
Western Digital Research

A message that is not in the archive (sent off-list or not yet received): 20230705185812.579118-4-jlayton-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org

Reviewed-by
#30 lore raw

Re: [PATCH v2 92/92] fs: rename i_ctime field to __i_ctime

Quoted 4 lines · Jeff Layton

Now that everything in-tree is converted to use the accessor functions, rename the i_ctime field in the inode to discourage direct access.

Looks OK to me.

Signature · 2 lines
Damien Le Moal
Western Digital Research
maintainer Reviewed-by on v2 92/92
#32 lore raw

Re: [PATCH v2 92/92] fs: rename i_ctime field to __i_ctime

In reply to Jeff Layton
Quoted 4 lines · Jeff Layton

Now that everything in-tree is converted to use the accessor functions, rename the i_ctime field in the inode to discourage direct access.

Looks good. Feel free to add:

Honza

Quoted 38 lines
---
 include/linux/fs.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
include/linux/fs.h +3 -3
index 14e38bd900f1..b66442f91835 100644
@@ -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_set_ctime_current(struct inode *inode);
  */
 static inline struct timespec64 inode_get_ctime(const struct inode *inode)
 {
-	return inode->i_ctime;
+	return inode->__i_ctime;
 }
 
 /**
@@ -1498,7 +1498,7 @@ static inline struct timespec64 inode_get_ctime(const struct inode *inode)
 static inline struct timespec64 inode_set_ctime_to_ts(struct inode *inode,
 						      struct timespec64 ts)
 {
-	inode->i_ctime = ts;
+	inode->__i_ctime = ts;
 	return ts;
 }
 
Signature · 1 line
2.41.0
Signature · 2 lines
Jan Kara <[email protected]>
SUSE Labs, CR
author maintainer
#33 lore raw

PATCH v2 92/92 fs: rename i_ctime field to __i_ctime

In reply to Jeff Layton

Now that everything in-tree is converted to use the accessor functions, rename the i_ctime field in the inode to discourage direct access.

---
 include/linux/fs.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
include/linux/fs.h +3 -3
index 14e38bd900f1..b66442f91835 100644
@@ -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_set_ctime_current(struct inode *inode);
  */
 static inline struct timespec64 inode_get_ctime(const struct inode *inode)
 {
-	return inode->i_ctime;
+	return inode->__i_ctime;
 }
 
 /**
@@ -1498,7 +1498,7 @@ static inline struct timespec64 inode_get_ctime(const struct inode *inode)
 static inline struct timespec64 inode_set_ctime_to_ts(struct inode *inode,
 						      struct timespec64 ts)
 {
-	inode->i_ctime = ts;
+	inode->__i_ctime = ts;
 	return ts;
 }
 
Signature · 1 line
2.41.0
author maintainer
#34 lore raw
In reply to Jeff Layton
Quoted 23 lines · Jeff Layton

v2:
- prepend patches to add missing ctime updates
- add simple_rename_timestamp helper function
- rename ctime accessor functions as inode_get_ctime/inode_set_ctime_*
- drop individual inode_ctime_set_{sec,nsec} helpers

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.

The patchset starts with reposts of some missing ctime updates that I spotted in the tree. It then adds a new helper function for updating the timestamp after a successful rename, and new ctime accessor infrastructure.

The bulk of the patchset is individual conversions of different subsysteme to use the new infrastructure. Finally, the patchset renames the i_ctime field to __i_ctime to help ensure that I didn't miss anything.

Quoted 6 lines

This should apply cleanly to linux-next as of this morning.

Most of this conversion was done via 5 different coccinelle scripts, run in succession, with a large swath of by-hand conversions to clean up the remainder.

A couple of other things I should note:

If you sent me an Acked-by or Reviewed-by in the previous set, then I tried to keep it on the patch here, since the respun patches are mostly just renaming stuff from v1. Let me know if I've missed any.

I've also pushed the pile to my tree as this tag:

    https://git.kernel.org/pub/scm/linux/kernel/git/jlayton/linux.git/tag/?h=ctime.20230705

In case that's easier to work with.

Cheers,

Signature · 1 line
Jeff Layton <[email protected]>
#35 lore raw
Quoted 37 lines · Jeff Layton
Quoted 23 lines · Jeff Layton

v2:
- prepend patches to add missing ctime updates
- add simple_rename_timestamp helper function
- rename ctime accessor functions as inode_get_ctime/inode_set_ctime_*
- drop individual inode_ctime_set_{sec,nsec} helpers

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.

The patchset starts with reposts of some missing ctime updates that I spotted in the tree. It then adds a new helper function for updating the timestamp after a successful rename, and new ctime accessor infrastructure.

The bulk of the patchset is individual conversions of different subsysteme to use the new infrastructure. Finally, the patchset renames the i_ctime field to __i_ctime to help ensure that I didn't miss anything.

Quoted 6 lines

This should apply cleanly to linux-next as of this morning.

Most of this conversion was done via 5 different coccinelle scripts, run in succession, with a large swath of by-hand conversions to clean up the remainder.

A couple of other things I should note:

If you sent me an Acked-by or Reviewed-by in the previous set, then I tried to keep it on the patch here, since the respun patches are mostly just renaming stuff from v1. Let me know if I've missed any.

Quoted 5 lines

I've also pushed the pile to my tree as this tag:

    https://git.kernel.org/pub/scm/linux/kernel/git/jlayton/linux.git/tag/?h=ctime.20230705

In case that's easier to work with.

Are there any preliminary patches showing what you want your introduced accessors to turn into? It is hard to judge the sanity of the introduction of wrappers without seeing what the wrappers are ultimately going to do.

Eric

author maintainer
#36 lore raw
Quoted 43 lines · Eric W. Biederman
Quoted 41 lines · Jeff Layton
Quoted 23 lines · Jeff Layton

v2:
- prepend patches to add missing ctime updates
- add simple_rename_timestamp helper function
- rename ctime accessor functions as inode_get_ctime/inode_set_ctime_*
- drop individual inode_ctime_set_{sec,nsec} helpers

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.

The patchset starts with reposts of some missing ctime updates that I spotted in the tree. It then adds a new helper function for updating the timestamp after a successful rename, and new ctime accessor infrastructure.

The bulk of the patchset is individual conversions of different subsysteme to use the new infrastructure. Finally, the patchset renames the i_ctime field to __i_ctime to help ensure that I didn't miss anything.

Quoted 6 lines

This should apply cleanly to linux-next as of this morning.

Most of this conversion was done via 5 different coccinelle scripts, run in succession, with a large swath of by-hand conversions to clean up the remainder.

A couple of other things I should note:

If you sent me an Acked-by or Reviewed-by in the previous set, then I tried to keep it on the patch here, since the respun patches are mostly just renaming stuff from v1. Let me know if I've missed any.

I've also pushed the pile to my tree as this tag:

    https://git.kernel.org/pub/scm/linux/kernel/git/jlayton/linux.git/tag/?h=ctime.20230705
Quoted 8 lines
Quoted 1 line

In case that's easier to work with.

Are there any preliminary patches showing what you want your introduced accessors to turn into? It is hard to judge the sanity of the introduction of wrappers without seeing what the wrappers are ultimately going to do.

Eric

I have a draft version of the multigrain patches on top of the wrapper conversion I've already posted in my "mgctime-experimental" branch:

https://git.kernel.org/pub/scm/linux/kernel/git/jlayton/linux.git/log/?h=mgctime-experimental

The rationale is best explained in this changelog:

    https://git.kernel.org/pub/scm/linux/kernel/git/jlayton/linux.git/commit/?h=mgctime-experimental&id=face437a144d3375afb7f70c233b0644b4edccba

The idea will be to enable this on a per-fs basis.

Signature · 1 line
Jeff Layton <[email protected]>
author maintainer
#37 lore raw
In reply to Jeff Layton
Quoted 23 lines · Jeff Layton

v2:
- prepend patches to add missing ctime updates
- add simple_rename_timestamp helper function
- rename ctime accessor functions as inode_get_ctime/inode_set_ctime_*
- drop individual inode_ctime_set_{sec,nsec} helpers

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.

The patchset starts with reposts of some missing ctime updates that I spotted in the tree. It then adds a new helper function for updating the timestamp after a successful rename, and new ctime accessor infrastructure.

The bulk of the patchset is individual conversions of different subsysteme to use the new infrastructure. Finally, the patchset renames the i_ctime field to __i_ctime to help ensure that I didn't miss anything.

Quoted 6 lines

This should apply cleanly to linux-next as of this morning.

Most of this conversion was done via 5 different coccinelle scripts, run in succession, with a large swath of by-hand conversions to clean up the remainder.

A couple of other things I should note:

If you sent me an Acked-by or Reviewed-by in the previous set, then I tried to keep it on the patch here, since the respun patches are mostly just renaming stuff from v1. Let me know if I've missed any.

I've also pushed the pile to my tree as this tag:

    https://git.kernel.org/pub/scm/linux/kernel/git/jlayton/linux.git/tag/?h=ctime.20230705

In case that's easier to work with.

Cheers,

Signature · 1 line
Jeff Layton <[email protected]>
author maintainer
#38 lore raw
In reply to Jeff Layton
Quoted 6 lines · Jeff Layton

v2:
- prepend patches to add missing ctime updates
- add simple_rename_timestamp helper function
- rename ctime accessor functions as inode_get_ctime/inode_set_ctime_*
- drop individual inode_ctime_set_{sec,nsec} helpers

After review by Jan and others, and Jan's ext4 rework, the diff on top of the series I posted a couple of days ago is below. I don't really want to spam everyone with another ~100 patch v3 series, but I can if you think that's best.

Christian, what would you like me to do here?

fs/ceph/inode.c +1 -2
index bcdb1a0beccf..5f6e93714f5a 100644
@@ -699,8 +699,7 @@ void ceph_fill_file_time(struct inode *inode, int issued,
 		if (ci->i_version == 0 ||
 		    timespec64_compare(ctime, &ictime) > 0) {
 			dout("ctime %lld.%09ld -> %lld.%09ld inc w/ cap\n",
-			     inode_get_ctime(inode).tv_sec,
-			     inode_get_ctime(inode).tv_nsec,
+			     ictime.tv_sec, ictime.tv_nsec,
 			     ctime->tv_sec, ctime->tv_nsec);
 			inode_set_ctime_to_ts(inode, *ctime);
 		}
fs/erofs/inode.c +1 -4
index 806374d866d1..567c0d305ea4 100644
@@ -175,10 +175,7 @@ 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_get_ctime(inode).tv_sec;
-	inode->i_atime.tv_sec = inode_get_ctime(inode).tv_sec;
-	inode->i_mtime.tv_nsec = inode_get_ctime(inode).tv_nsec;
-	inode->i_atime.tv_nsec = inode_get_ctime(inode).tv_nsec;
+	inode->i_mtime = inode->i_atime = inode_get_ctime(inode);
 
 	inode->i_flags &= ~S_DAX;
 	if (test_opt(&sbi->opt, DAX_ALWAYS) && S_ISREG(inode->i_mode) &&
fs/exfat/namei.c +1 -1
index c007de6ac1c7..1b9f587f6cca 100644
@@ -1351,7 +1351,7 @@ static int exfat_rename(struct mnt_idmap *idmap,
 			exfat_warn(sb, "abnormal access to an inode dropped");
 			WARN_ON(new_inode->i_nlink == 0);
 		}
-		EXFAT_I(new_inode)->i_crtime = inode_set_ctime_current(new_inode);
+		EXFAT_I(new_inode)->i_crtime = current_time(new_inode);
 	}
 
 unlock:
fs/ext4/ext4.h +38 -60

129 lines not on this page · raw message

fs/ext4/inode-test.c +3 -3
index 7935ea6cf92c..f0c0fd507fbc 100644
@@ -245,9 +245,9 @@ static void inode_test_xtimestamp_decoding(struct kunit *test)
 	struct timestamp_expectation *test_param =
 			(struct timestamp_expectation *)(test->param_value);
 
-	timestamp.tv_sec = get_32bit_time(test_param);
-	ext4_decode_extra_time(&timestamp,
-			       cpu_to_le32(test_param->extra_bits));
+	timestamp = ext4_decode_extra_time(
+				cpu_to_le32(get_32bit_time(test_param)),
+				cpu_to_le32(test_param->extra_bits));
 
 	KUNIT_EXPECT_EQ_MSG(test,
 			    test_param->expected.tv_sec,
fs/ext4/inode.c +3 -3
index bbc57954dfd3..c6a837b90af4 100644
@@ -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_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_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_get_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);
fs/fat/inode.c +1 -3
index 2be40ff8a74f..cdd39b6020f3 100644
@@ -1407,9 +1407,7 @@ 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_set_ctime(inode,
-									0, 0).tv_sec;
-	inode->i_mtime.tv_nsec = inode->i_atime.tv_nsec = 0;
+	inode->i_mtime = inode->i_atime = inode_set_ctime(inode, 0, 0);
 	set_nlink(inode, fat_subdirs(inode)+2);
 
 	return 0;
fs/hpfs/namei.c +9 -22
index 36babb78b510..f4eb8d6f5989 100644
@@ -15,8 +15,7 @@ static void hpfs_update_directory_times(struct inode *dir)
 	if (t == dir->i_mtime.tv_sec &&
 	    t == inode_get_ctime(dir).tv_sec)
 		return;
-	dir->i_mtime.tv_sec = inode_set_ctime(dir, t, 0).tv_sec;
-	dir->i_mtime.tv_nsec = 0;
+	dir->i_mtime = inode_set_ctime(dir, t, 0);
 	hpfs_write_inode_nolock(dir);
 }
 
@@ -59,11 +58,8 @@ 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;
-	inode_set_ctime(result,
-			result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date)),
-			0);
-	result->i_mtime.tv_nsec = 0; 
-	result->i_atime.tv_nsec = 0; 
+	result->i_mtime = result->i_atime =
+		inode_set_ctime(result, local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date)), 0);
 	hpfs_i(result)->i_ea_size = 0;
 	result->i_mode |= S_IFDIR;
 	result->i_op = &hpfs_dir_iops;
@@ -168,11 +164,8 @@ 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;
-	inode_set_ctime(result,
-			result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date)),
-			0);
-	result->i_mtime.tv_nsec = 0;
-	result->i_atime.tv_nsec = 0;
+	result->i_mtime = result->i_atime =
+		inode_set_ctime(result, local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date)), 0);
 	hpfs_i(result)->i_ea_size = 0;
 	if (dee.read_only)
 		result->i_mode &= ~0222;
@@ -252,11 +245,8 @@ 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;
-	inode_set_ctime(result,
-			result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date)),
-			0);
-	result->i_mtime.tv_nsec = 0;
-	result->i_atime.tv_nsec = 0;
+	result->i_mtime = result->i_atime =
+		inode_set_ctime(result, local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date)), 0);
 	hpfs_i(result)->i_ea_size = 0;
 	result->i_uid = current_fsuid();
 	result->i_gid = current_fsgid();
@@ -329,11 +319,8 @@ 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;
-	inode_set_ctime(result,
-			result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date)),
-			0);
-	result->i_mtime.tv_nsec = 0;
-	result->i_atime.tv_nsec = 0;
+	result->i_mtime = result->i_atime =
+		inode_set_ctime(result, local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date)), 0);
 	hpfs_i(result)->i_ea_size = 0;
 	result->i_mode = S_IFLNK | 0777;
 	result->i_uid = current_fsuid();
fs/isofs/inode.c +2 -7
index 98a78200cff1..2ee21286ac8f 100644
@@ -1422,13 +1422,8 @@ static int isofs_read_inode(struct inode *inode, int relocated)
 			inode->i_ino, de->flags[-high_sierra]);
 	}
 #endif
-
-	inode->i_mtime.tv_sec =
-	inode->i_atime.tv_sec = inode_set_ctime(inode,
-						iso_date(de->date, high_sierra),
-						0).tv_sec;
-	inode->i_mtime.tv_nsec =
-	inode->i_atime.tv_nsec = 0;
+	inode->i_mtime = inode->i_atime =
+		inode_set_ctime(inode, iso_date(de->date, high_sierra), 0);
 
 	ei->i_first_extent = (isonum_733(de->extent) +
 			isonum_711(de->ext_attr_length));
fs/minix/inode.c +1 -5
index 3715a3940bd4..8a4fc9420b36 100644
@@ -501,11 +501,7 @@ 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_set_ctime(inode,
-									raw_inode->i_time,
-									0).tv_sec;
-	inode->i_mtime.tv_nsec = 0;
-	inode->i_atime.tv_nsec = 0;
+	inode->i_mtime = inode->i_atime = inode_set_ctime(inode, raw_inode->i_time, 0);
 	inode->i_blocks = 0;
 	for (i = 0; i < 9; i++)
 		minix_inode->u.i1_data[i] = raw_inode->i_zone[i];
fs/overlayfs/file.c +1 -1
index 7acd3e3fe790..7e7876aae01c 100644
@@ -255,7 +255,7 @@ static void ovl_file_accessed(struct file *file)
 	if ((!timespec64_equal(&inode->i_mtime, &upperinode->i_mtime) ||
 	     !timespec64_equal(&ctime, &uctime))) {
 		inode->i_mtime = upperinode->i_mtime;
-		inode_set_ctime_to_ts(inode, inode_get_ctime(upperinode));
+		inode_set_ctime_to_ts(inode, uctime);
 	}
 
 	touch_atime(&file->f_path);
fs/romfs/super.c +1 -2
index 961b9d342e0e..d89739655f9e 100644
@@ -322,8 +322,7 @@ 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 = inode_set_ctime(i, 0, 0).tv_sec;
-	i->i_mtime.tv_nsec = i->i_atime.tv_nsec = 0;
+	i->i_mtime = i->i_atime = inode_set_ctime(i, 0, 0);
 
 	/* set up mode and ops */
 	mode = romfs_modemap[nextfh & ROMFH_TYPE];
fs/smb/client/fscache.h +1 -1
index a228964bc2ce..84f3b09367d2 100644
@@ -56,7 +56,7 @@ void cifs_fscache_fill_coherency(struct inode *inode,
 	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(ctime.tv_sec);
-	cd->last_change_time_nsec  = cpu_to_le64(ctime.tv_nsec);
+	cd->last_change_time_nsec = cpu_to_le32(ctime.tv_nsec);
 }
 
 
Signature · 1 line
Jeff Layton <[email protected]>
#39 lore raw
Quoted 8 lines · Jeff Layton
Quoted 6 lines · Jeff Layton

v2:
- prepend patches to add missing ctime updates
- add simple_rename_timestamp helper function
- rename ctime accessor functions as inode_get_ctime/inode_set_ctime_*
- drop individual inode_ctime_set_{sec,nsec} helpers

Quoted 6 lines

After review by Jan and others, and Jan's ext4 rework, the diff on top of the series I posted a couple of days ago is below. I don't really want to spam everyone with another ~100 patch v3 series, but I can if you think that's best.

Christian, what would you like me to do here?

I picked up the series from the list and folded the fixups you posted here into the respective fs conversion patches. I hope that helps you avoid a resend. You should have received a separate "thank you" mail for all of this.

To each patch that I folded one of the fixlets from below into I added a git note that records a link to your mail here and the respective patch hunk from this mail that I folded into the patch. git.kernel.org will show notes by default. For example,
https://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs.git/commit/?h=vfs.ctime&id=8b0e3c2e99004609a16ba145bcbdfdddb78e220e
should show you the note I added. You can also fetch them via git fetch $remote refs/notes/*:refs/notes/*
(You probably know that ofc but jic.) if you're interested.

Based on v6.5-rc1 as of today.

Btw, both b4 and patchwork somehow treat the series in weird was. IOW, based on the message id of the cover letter I was able to pull most messages except for:

[07/92] fs: add ctime accessors infrastructure
[08/92] fs: new helper: simple_rename_timestamp
[92/92] fs: rename i_ctime field to __i_ctime

which I pulled in separately. Not sure what the cause of this is.

author maintainer
#40 lore raw
Quoted 30 lines · Christian Brauner
Quoted 8 lines · Jeff Layton
Quoted 6 lines · Jeff Layton

v2:
- prepend patches to add missing ctime updates
- add simple_rename_timestamp helper function
- rename ctime accessor functions as inode_get_ctime/inode_set_ctime_*
- drop individual inode_ctime_set_{sec,nsec} helpers

Quoted 6 lines

After review by Jan and others, and Jan's ext4 rework, the diff on top of the series I posted a couple of days ago is below. I don't really want to spam everyone with another ~100 patch v3 series, but I can if you think that's best.

Christian, what would you like me to do here?

I picked up the series from the list and folded the fixups you posted here into the respective fs conversion patches. I hope that helps you avoid a resend. You should have received a separate "thank you" mail for all of this.

To each patch that I folded one of the fixlets from below into I added a git note that records a link to your mail here and the respective patch hunk from this mail that I folded into the patch. git.kernel.org will show notes by default. For example,
https://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs.git/commit/?h=vfs.ctime&id=8b0e3c2e99004609a16ba145bcbdfdddb78e220e
should show you the note I added. You can also fetch them via git fetch $remote refs/notes/*:refs/notes/*
(You probably know that ofc but jic.) if you're interested.

Quoted 2 lines

Based on v6.5-rc1 as of today.

Many thanks!!! I'll get to work rebasing the multigrain timestamp series on top of that.

Quoted 11 lines

Btw, both b4 and patchwork somehow treat the series in weird was. IOW, based on the message id of the cover letter I was able to pull most messages except for:

[07/92] fs: add ctime accessors infrastructure
[08/92] fs: new helper: simple_rename_timestamp
[92/92] fs: rename i_ctime field to __i_ctime

which I pulled in separately. Not sure what the cause of

this is.

Good to know.

I ended up doing the send in two phases: one for the cover letter and infrastructure patches that went to everyone, and one for the per- subsystem patches that went do individual maintainers and lists.

I suspect that screwed up the message IDs somehow. Hopefully I won't need to do a posting like that again soon, but I'll pay closer attention to the message id handling next time.

Thanks again!

Signature · 1 line
Jeff Layton <[email protected]>
maintainer Applied
#41 lore raw
In reply to Jeff Layton
Quoted 6 lines · Jeff Layton

v2:
- prepend patches to add missing ctime updates
- add simple_rename_timestamp helper function
- rename ctime accessor functions as inode_get_ctime/inode_set_ctime_*
- drop individual inode_ctime_set_{sec,nsec} helpers

Quoted 8 lines

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.

[...]

Applied to the vfs.ctime branch of the vfs/vfs.git tree. Patches in the vfs.ctime branch should appear in linux-next soon.

Please report any outstanding bugs that were missed during review in a new review to the original patch series allowing us to drop it.

It's encouraged to provide Acked-bys and Reviewed-bys even though the patch has now been applied. If possible patch trailers will be updated.

Note that commit hashes shown below are subject to change due to rebase, trailer updates or similar. If in doubt, please check the listed branch.

tree:   https://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs.git
branch: vfs.ctime

[01/92] ibmvmc: update ctime in conjunction with mtime on write

        https://git.kernel.org/vfs/vfs/c/ead310563ad2
[02/92] bfs: update ctime in addition to mtime when adding entries
        https://git.kernel.org/vfs/vfs/c/f42faf14b838
[03/92] efivarfs: update ctime when mtime changes on a write
        https://git.kernel.org/vfs/vfs/c/d8d026e0d1f2
[04/92] exfat: ensure that ctime is updated whenever the mtime is
        https://git.kernel.org/vfs/vfs/c/d84bd8fa48d7
[05/92] apparmor: update ctime whenever the mtime changes on an inode
        https://git.kernel.org/vfs/vfs/c/73955caedfae
[06/92] cifs: update the ctime on a partial page write
        https://git.kernel.org/vfs/vfs/c/c2f784379c99
[07/92] fs: add ctime accessors infrastructure
        https://git.kernel.org/vfs/vfs/c/64f0367de800
[08/92] fs: new helper: simple_rename_timestamp
        https://git.kernel.org/vfs/vfs/c/54ced54a0239
[09/92] btrfs: convert to simple_rename_timestamp
        https://git.kernel.org/vfs/vfs/c/218e0f662fee
[10/92] ubifs: convert to simple_rename_timestamp
        https://git.kernel.org/vfs/vfs/c/caac4f65568d
[11/92] shmem: convert to simple_rename_timestamp
        https://git.kernel.org/vfs/vfs/c/d3d11e9927b6
[12/92] exfat: convert to simple_rename_timestamp
        https://git.kernel.org/vfs/vfs/c/71534b484c63
[13/92] ntfs3: convert to simple_rename_timestamp
        https://git.kernel.org/vfs/vfs/c/140880821ce0
[14/92] reiserfs: convert to simple_rename_timestamp
        https://git.kernel.org/vfs/vfs/c/1a1a4df5e8fc
[15/92] spufs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/784e5a93c9cf
[16/92] s390: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/1cece1f8e5c2
[17/92] binderfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/0bcd830a76f3
[18/92] infiniband: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/811f97f80b01
[19/92] ibm: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/b447ed7597f0
[20/92] usb: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/2557dc7f2dde
[21/92] 9p: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/4cd4b11385ef
[22/92] adfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/e257d7ade66e
[23/92] affs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/770619f19a77
[24/92] afs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/758506e44668
[25/92] fs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/a0a5a9810b37
[26/92] autofs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/d7d1363cc3f6
[27/92] befs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/d6218773de2d
[28/92] bfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/368b313ac2ab
[29/92] btrfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/d3d15221956a
[30/92] ceph: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/818fc6e0129a
[31/92] coda: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/4e0b22fbc012
[32/92] configfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/69c977798a6a
[33/92] cramfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/911f086eae23
[34/92] debugfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/634a50390dbb
[35/92] devpts: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/92bb29a24707
[36/92] ecryptfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/16d21856dfd6
[37/92] efivarfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/cfeee05a50e1
[38/92] efs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/3a30b097319f
[39/92] erofs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/e3594996216f
[40/92] exfat: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/8bd562d6f46d
[41/92] ext2: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/7483252e8894
[42/92] ext4: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/f2ddb05870fb
[43/92] 9afc475653af f2fs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/f2ddb05870fb
[44/92] fat: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/8a0c417b695b
[45/92] freevxfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/7affaeb5b914
[46/92] fuse: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/688279761436
[47/92] gfs2: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/9e5b114b5ee4
[48/92] hfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/d41f5876a177
[49/92] hfsplus: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/147f3dd171d2
[50/92] hostfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/2ceaa835b4f5
[51/92] hpfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/e6fd7f49daa7
[52/92] hugetlbfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/f5950f079b1a
[53/92] isofs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/53f2bb3567f0
[54/92] jffs2: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/7e8dc4ab1afb
[55/92] jfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/77737373dbb3
[56/92] kernfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/8b0e3c2e9900
[57/92] nfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/12844cb15dc6
[58/92] nfsd: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/f297728268bf
[59/92] nilfs2: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/1e9f083bc9cd
[60/92] ntfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/3cc66672eaa5
[61/92] ntfs3: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/9438de01396e
[62/92] ocfs2: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/da5b97da32e7
[63/92] omfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/563d772c8d70
[64/92] openpromfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/5f0978a6f0a6
[65/92] orangefs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/6a83804b4325
[66/92] overlayfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/60dcee636746
[67/92] procfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/85e0a6b3b8cf
[68/92] pstore: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/2b8125b5e7c6
[69/92] qnx4: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/77eb00659cb5
[70/92] qnx6: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/af1acd38df36
[71/92] ramfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/0eb8012f4b0b
[72/92] reiserfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/e3e5f5f70292
[73/92] romfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/b6058a9af143
[74/92] smb: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/d5c263f2187d
[75/92] squashfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/eaace9c73ba8
[76/92] sysv: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/41b6f4fbbe32
[77/92] tracefs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/5f69a5364568
[78/92] ubifs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/5bb225ba81c0
[79/92] udf: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/e251f0e98433
[80/92] ufs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/376ef9f6cab1
[81/92] vboxsf: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/9f06612951d5
[82/92] xfs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/4e8c1361146f
[83/92] zonefs: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/cbdc6aa5f65d
[84/92] linux: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/ff12abb4a71a
[85/92] mqueue: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/a6b5a0055142
[86/92] bpf: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/d2b6a0a3863a
[87/92] shmem: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/ffcd778237d3
[88/92] sunrpc: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/ccf1c9002d71
[89/92] apparmor: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/ff91aaa76f1a
[90/92] security: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/701071f9ad33
[91/92] selinux: convert to ctime accessor functions
        https://git.kernel.org/vfs/vfs/c/cb6dfffdc7e9
[92/92] fs: rename i_ctime field to __i_ctime
        https://git.kernel.org/vfs/vfs/c/c06d4bf5e207
#42 lore raw
In reply to Jeff Layton
This series was applied to jaegeuk/f2fs.git (dev) by Christian Brauner <[email protected]>:

Hello:

This series was applied to jaegeuk/f2fs.git (dev)
by Christian Brauner <[email protected]>:

Quoted 6 lines · On Wed, 5 Jul 2023 14:58:09 -0400 you wrote:

v2:
- prepend patches to add missing ctime updates
- add simple_rename_timestamp helper function
- rename ctime accessor functions as inode_get_ctime/inode_set_ctime_*
- drop individual inode_ctime_set_{sec,nsec} helpers

Quoted 8 lines

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.

[...]

Here is the summary with links:
- [f2fs-dev,v2,07/92] fs: add ctime accessors infrastructure

    https://git.kernel.org/jaegeuk/f2fs/c/9b6304c1d537
  - [f2fs-dev,v2,08/92] fs: new helper: simple_rename_timestamp
    https://git.kernel.org/jaegeuk/f2fs/c/0c4767923ed6
  - [f2fs-dev,v2,92/92] fs: rename i_ctime field to __i_ctime
    https://git.kernel.org/jaegeuk/f2fs/c/13bc24457850

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html