Messages in this thread | | | Date | Sun, 12 Mar 2006 17:00:51 +0000 (GMT) | From | Anton Altaparmakov <> | Subject | Re: truncate and {m,c}time on ntfs |
| |
On Sun, 12 Mar 2006, Anton Altaparmakov wrote: > Hi Christoph, > > A patch of yours modified fs/ntfs/inode.c::ntfs_truncate() and inserted > this comment: > > [snip] > /* normally ->truncate shouldn't update ctime or mtime, > * but ntfs did before so it got a copy & paste version > * of file_update_time. one day someone should fix this > * for real. > */ > [snip] > > Did you realise that all (local) file systems in Linux kernel set both > {m,c}time in their ->truncate function. E.g. from > fs/ext3.c/inode.c::ext3_truncate(): > > inode->i_mtime = inode->i_ctime = CURRENT_TIME_SEC; > > Would you be so kind to explain what is your problem with ntfs doing it, > too? And if your statement is correct and no file system should touch > {m,c}time in their ->truncate() method, could you explain to me how the > {m,c}time would be set otherwise when open(O_TRUNC) or {f,}truncate() is > executed on a file?
Sorry, I know how open(O_TRUNC) sets it as it calls do_truncate(,, ATTR_MTIME|ATTR_CTIME,) but at present both {f,}truncate() call do_truncate(,, 0,) and I do not see anywhere else where the setting would/could be done, other than in the file system ->truncate method.
In fact if I take the setting of {m,c}time out of ntfs_truncate(), doing:
touch stat foo truncate foo 10 stat foo
Does not result in a change in the {m,c}time...
Given posix/sus3 requires setting of {m,c}time on truncate, it would appear that a file system ->truncate function needs to set {m,c}time itself...
Please explain why you added your comment to ntfs...
Thanks a lot in advance.
Best regards,
Anton -- Anton Altaparmakov <aia21 at cam.ac.uk> (replace at with @) Unix Support, Computing Service, University of Cambridge, CB2 3QH, UK Linux NTFS maintainer / IRC: #ntfs on irc.freenode.net WWW: http://linux-ntfs.sf.net/ & http://www-stu.christs.cam.ac.uk/~aia21/ - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |