On Wednesday 27 June 2007 12:58, Kyle Moffett wrote: > I seem to recall you could actually end up racing and building a path > to the file in those directories as "a/d/0/3" or some other path at > which it never even remotely existed. I'd love to be wrong, Cheer up, you recall wrong. > but I can't help but see this problem in any reverse-pathname-generation > proposal which gets the locking right. Have a look at how __d_path() is implemented (with the fixes): It takes the dcache_lock, and the vfsmount_lock where necessary, and this ensures that the pathname can't change under it, neither because of a rename nor unlink nor remount. The pathname computed is *exactly* the name the file has at that specific point time. A few more details about how pathnames work are explained in the tech doc at: http://forge.novell.com/modules/xfcontent/downloads.php/apparmor/LKML_Submission-May_07 Andreas - 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