Messages in this thread | | | Date | Mon, 12 Aug 2002 10:21:03 +0100 | From | Padraig Brady <> | Subject | Re: [PATCH] Linux-2.5 fix/improve get_pid() |
| |
Alan Cox wrote: > On Mon, 2002-08-12 at 09:56, Padraig Brady wrote: > >>Anyone care to clarify which filesystems don't >>have unique inode numbers. I always thought you >>could uniquely identify any file using a device,inode >>tuple? Fair enough proc is "special" but can/should >>you not assume unique inodes within all other filesystems? > > > 2.4 functions correctly here in all the stuff I've looked at.
cool.
> I can't speak for 2.5. In the 2.4 case you must be holding the files open > during the comparison. Some file systems like MSDOS invent inodes as > they go, never issuing the same one to two objects at the same time but > happily reissuing different inode numbers the next time around.
Sure, no hardlinks so who cares what the number is, as long as it's unique.
> That breaks NFS but not much else
hmm..
thanks, Pádraig.
- 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/
| |