diff options
author | Steve French <sfrench@us.ibm.com> | 2007-02-16 23:30:54 -0500 |
---|---|---|
committer | Steve French <sfrench@us.ibm.com> | 2007-02-16 23:30:54 -0500 |
commit | 1b2b212603ceb47d60aff571dcfffc846fcfa336 (patch) | |
tree | a46aa9356cf8142b5f78520bdfc8efba8ecdbe1b /fs/cifs/link.c | |
parent | c14e894bd40868d6d1f2379705b68acf5288ba27 (diff) |
[CIFS] mtime bounces from local to remote when cifs nocmtime i_flags overwritten
atime flag was also overwritten. Noticed by Shirish when he was debugging
an atime problem. Should help performance a bit too.
cifs should be getting time stamps from the server (that was the original
intent too)
Signed-off-by: Steve French <sfrench@us.ibm.com>
Diffstat (limited to 'fs/cifs/link.c')
-rw-r--r-- | fs/cifs/link.c | 3 |
1 files changed, 2 insertions, 1 deletions
diff --git a/fs/cifs/link.c b/fs/cifs/link.c index 8e259969354b..6baea85d726e 100644 --- a/fs/cifs/link.c +++ b/fs/cifs/link.c | |||
@@ -77,7 +77,8 @@ cifs_hardlink(struct dentry *old_file, struct inode *inode, | |||
77 | cifsInode = CIFS_I(old_file->d_inode); | 77 | cifsInode = CIFS_I(old_file->d_inode); |
78 | if(rc == 0) { | 78 | if(rc == 0) { |
79 | old_file->d_inode->i_nlink++; | 79 | old_file->d_inode->i_nlink++; |
80 | old_file->d_inode->i_ctime = CURRENT_TIME; | 80 | /* BB should we make this contingent on superblock flag NOATIME? */ |
81 | /* old_file->d_inode->i_ctime = CURRENT_TIME;*/ | ||
81 | /* parent dir timestamps will update from srv | 82 | /* parent dir timestamps will update from srv |
82 | within a second, would it really be worth it | 83 | within a second, would it really be worth it |
83 | to set the parent dir cifs inode time to zero | 84 | to set the parent dir cifs inode time to zero |