[src/trunk]: src/sys/fs/tmpfs Synchronize NFS status with reality. After yam...

[jmmv <jmmv%NetBSD.org@localhost>]

details:   https://anonhg.NetBSD.org/src/rev/9905059b08bd
branches:  trunk
changeset: 584559:9905059b08bd
user:      jmmv <jmmv%NetBSD.org@localhost>
date:      Fri Sep 23 21:39:29 2005 +0000

description:
Synchronize NFS status with reality.  After yamt@ changes (I think with the
implementation of getpages and putpages and the use of UBC in the read and
write operations), the worst problem has gone away which was a panic when
a file's contents were modified in the original file system and then read
through the NFS mount point.

Also remove the entry about optimization.  While tmpfs still has room for
improvement, it has become a lot better lately, thanks to the string pools
and the changes yamt@ did in the anonymous objects handling.

diffstat:

 sys/fs/tmpfs/TODO |  15 +++++----------
 1 files changed, 5 insertions(+), 10 deletions(-)

diffs (27 lines):

diff -r ac3a7280f040 -r 9905059b08bd sys/fs/tmpfs/TODO
--- a/sys/fs/tmpfs/TODO Fri Sep 23 21:21:58 2005 +0000
+++ b/sys/fs/tmpfs/TODO Fri Sep 23 21:39:29 2005 +0000
@@ -3,18 +3,13 @@
   the system unstable by exhausting it.  Therefore, a pool allocator that
   uses anonymous memory has to be written.
 
-- Optimize for speed and memory usage.  Nodes are still quite big; their
-  size should be reduced if possible.  But most importantly, the algorithm
-  used to resize anonymous memory objects in regular files is extremely
-  inefficient.
-
 - Verify that file holes work (they should, but must be checked).  Add a
   regression test for this feature.
 
-- NFS support is broken.  The readdir operation does not work until wcc is
-  disabled (which happens as soon as doing a `touch foo' over the NFS
-  mount point).  Furthermore, the system crashes if a file is created on the
-  original file system and is read through NFS; the system spits a panic
-  about busy pages.
+- NFS support is broken.  The readdir operation does not work well; it
+  mostly works when wcc is disabled (which happens as soon as doing a `touch
+  foo' over the NFS mount point).  However, files created in the original
+  file system do not appear in the NFS mount point until another file
+  is created over it.
 
 - Fix and complete code marked with `XXX' and `TODO' tags.