[PATCH] Re: 2.5.11 breakage
authorAlexander Viro <viro@math.psu.edu>
Mon, 29 Apr 2002 11:14:24 +0000 (04:14 -0700)
committerLinus Torvalds <torvalds@home.transmeta.com>
Mon, 29 Apr 2002 11:14:24 +0000 (04:14 -0700)
commit85d217f40107c3cfe9b7d316022beaf667dc8ef8
treedf5bfb7aeda106457ceeda0e2bd586242f2d61d7
parent7ca320475b62b1fbd94f772d9d6243b293b29f95
[PATCH] Re: 2.5.11 breakage

OK, here comes.  Patch below is an attempt to do the fastwalk
stuff in right way and so far it seems to be working.

 - dentry leak is plugged
 - locked/unlocked state of nameidata doesn't depend on history - it
   depends only on point in code.
 - LOOKUP_LOCKED is gone.
 - following mounts and .. doesn't drop dcache_lock
 - light-weight permission check distinguishes between "don't know" and
   "permission denied", so we don't call full-blown permission() unless
   we have to.
 - code that changes root/pwd holds dcache_lock _and_ write lock on
   current->fs->lock.  I.e. if we hold dcache_lock we can safely
   access our ->fs->{root,pwd}{,mnt}
 - __d_lookup() does not increment refcount; callers do dget_locked()
   if they need it (behaviour of d_lookup() didn't change, obviously).
 - link_path_walk() logics had been (somewhat) cleaned up.
fs/dcache.c
fs/namei.c
include/linux/fs.h
include/linux/fs_struct.h