Who broke NFS?

[Emiel Kollof]

Op dinsdag 5 december 2006 15:07, schreef Thomas E. Spanjaard:
> Emiel Kollof wrote:
> > A fix for this problem was found, thanks to TGEN and corecode. The case
> > was that nch was NULL and dereferenced by that if over there. nch gets
> > dropped before, so thatÅ› how that NULL deref was possible. Here's a patch
> > that solves the problem for me:
> > --- nfs_subs.c  2006-10-27 06:56:34.000000000 +0200
> > +++ nfs_subs.c.fixed    2006-12-05 14:31:51.000000000 +0100
> > @@ -1604,8 +1604,8 @@
> >          */
> >         if (error == 0) {
> >                 if (dvpp) {
> > +                       nch = nd->nl_nch;
> >                         if (nch.ncp->nc_parent) {
> > -                               nch = nd->nl_nch;
> >                                 nch.ncp = nch.ncp->nc_parent;
> >                                 error = cache_vget(&nch, nd->nl_cred,
> >                                                    LK_EXCLUSIVE, dvpp);
>
> If noone objects, I'll commit it this afternoon (in ~ 5 hours).
>
> Cheers,

I ran fsx from a linux box that mounted an NFS exported filesystem from a 
DFBSD box for ~30 minutes, and it didn't fall over. So I don't object :)

Cheers,
Emiel
-- 
>Ever heard of .cshrc?
That's a city in Bosnia.  Right?
(Discussion in comp.os.linux.misc on the intuitiveness of commands.)
Attachment:
pgp00006.pgp
-------------- next part --------------
A non-text attachment was scrubbed...
Name: pgp00006.pgp
Type: application/octet-stream
Size: 191 bytes
Desc: "Description: PGP signature"
URL: <http://lists.dragonflybsd.org/pipermail/kernel/attachments/20061205/e66ed921/attachment-0020.obj>