BASE-TCONDEV test broken against Samba3??

[Volker Lendecke]

On Tue, May 16, 2006 at 01:59:28PM -0700, Jeremy Allison wrote:
> I think that Samba4 is not correct here. I remember looking
> at this. We accept what Windows sends - possibly we should
> check if *(smb_buf(inbuf) + passlen) ==0 and increment by
> one if so - but currently we are correct for Windows clients.

Just single-stepped through reply_tcon_and_X while doing a
'net view \\192.168.234.1' from XP, latest patches from
about a month ago. In line 647 I get:

(gdb) p *p at 50
$1 = "\000\\\000\\\0001\0009\0002\000.\0001\0006\0008\000.\0002\0003\0004\000.\0001\
\000\\\000I\000P\000C\000$\000\000\000?????\000"

So to me it seems that *p is indeed 0 here. My question is:
How does srvstr_pull_buf get the right string here?

Volker
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://lists.samba.org/archive/samba-technical/attachments/20060517/b542128f/attachment.bin