It is unexpected to block reading of frozen filesystem because of atime update. Also handling blocking on frozen filesystem because of atime update would make locking more complex than it already is. So just skip atime update when filesystem is frozen like we skip it when filesystem is remounted read-only. BugLink: https://bugs.launchpad.net/bugs/897421 Tested-by: Kamal Mostafa <kamal@xxxxxxxxxxxxx> Tested-by: Peter M. Petrakis <peter.petrakis@xxxxxxxxxxxxx> Tested-by: Dann Frazier <dann.frazier@xxxxxxxxxxxxx> Tested-by: Massimo Morana <massimo.morana@xxxxxxxxxxxxx> Signed-off-by: Jan Kara <jack@xxxxxxx> --- fs/inode.c | 9 +++++++-- 1 files changed, 7 insertions(+), 2 deletions(-) diff --git a/fs/inode.c b/fs/inode.c index 095828c..ff3b385 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -1535,12 +1535,17 @@ void touch_atime(struct vfsmount *mnt, struct dentry *dentry) if (timespec_equal(&inode->i_atime, &now)) return; - if (mnt_want_write(mnt)) + if (!sb_start_write_trylock(inode->i_sb)) return; + if (__mnt_want_write(mnt)) + goto skip_update; + inode->i_atime = now; mark_inode_dirty_sync(inode); - mnt_drop_write(mnt); + __mnt_drop_write(mnt); +skip_update: + sb_end_write(inode->i_sb); } EXPORT_SYMBOL(touch_atime); -- 1.7.1 -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
