aboutsummaryrefslogtreecommitdiffstats
path: root/packet-smb.c
diff options
context:
space:
mode:
authorGuy Harris <guy@alum.mit.edu>2003-06-10 05:28:02 +0000
committerGuy Harris <guy@alum.mit.edu>2003-06-10 05:28:02 +0000
commita048cbebed8e95d9b4a864dd7b6e8e13b12ceaa4 (patch)
tree23ab31b551051a587f158a7b5ddc0d4b06ee83ba /packet-smb.c
parent00bfd16e5f2dcfa969ab73d49ffe64c5ba150105 (diff)
Show the upper 2 bytes of the formerly-reserved area as a "High Part of
PID" even for SMB-over-IPX. svn path=/trunk/; revision=7823
Diffstat (limited to 'packet-smb.c')
-rw-r--r--packet-smb.c28
1 files changed, 13 insertions, 15 deletions
diff --git a/packet-smb.c b/packet-smb.c
index 21bd7ae2ac..b216338d24 100644
--- a/packet-smb.c
+++ b/packet-smb.c
@@ -3,7 +3,7 @@
* Copyright 1999, Richard Sharpe <rsharpe@ns.aus.com>
* 2001 Rewrite by Ronnie Sahlberg and Guy Harris
*
- * $Id: packet-smb.c,v 1.351 2003/06/09 22:59:11 guy Exp $
+ * $Id: packet-smb.c,v 1.352 2003/06/10 05:28:02 guy Exp $
*
* Ethereal - Network traffic analyzer
* By Gerald Combs <gerald@ethereal.com>
@@ -16217,15 +16217,21 @@ dissect_smb(tvbuff_t *tvb, packet_info *pinfo, proto_tree *parent_tree)
* Network Monitor 2.x dissects the four bytes before the Session ID
* as a "Key", and the two bytes after the SequenceNumber as
* a "Group ID".
+ *
+ * The "High Part of PID" has been seen in calls other than NT
+ * Create and X, although most of them appear to be I/O on DCE RPC
+ * pipes opened with the NT Create and X in question.
*/
+ proto_tree_add_item(htree, hf_smb_pid_high, tvb, offset, 2, TRUE);
+ offset += 2;
+
if (pinfo->ptype == PT_IPX &&
(pinfo->match_port == IPX_SOCKET_NWLINK_SMB_SERVER ||
pinfo->match_port == IPX_SOCKET_NWLINK_SMB_REDIR ||
pinfo->match_port == IPX_SOCKET_NWLINK_SMB_MESSENGER)) {
/*
* This is SMB-over-IPX.
- * XXX - high part of pid?
- * XXX - doe we have to worry about "sequenced commands",
+ * XXX - do we have to worry about "sequenced commands",
* as per the Samba document? They say that for
* "unsequenced commands" (with a sequence number of 0),
* the Mid must be unique, but perhaps the Mid doesn't
@@ -16233,10 +16239,6 @@ dissect_smb(tvbuff_t *tvb, packet_info *pinfo, proto_tree *parent_tree)
* one capture with SMB-over-IPX, however, the Mids
* are unique even for sequenced commands.
*/
- proto_tree_add_item(htree, hf_smb_reserved, tvb, offset, 2,
- TRUE);
- offset += 2;
-
/* Key */
proto_tree_add_item(htree, hf_smb_key, tvb, offset, 4,
TRUE);
@@ -16258,15 +16260,11 @@ dissect_smb(tvbuff_t *tvb, packet_info *pinfo, proto_tree *parent_tree)
offset += 2;
} else {
/*
- * 12 reserved bytes.
- * XXX - high part of pid?
- * According to http://ubiqx.org/cifs/SMB.html#SMB.4.2.1
- * and http://ubiqx.org/cifs/SMB.html#SMB.5.5.1 these are
- * a two byte pid-high, and an 8-byte signature ...
+ * According to http://ubiqx.org/cifs/SMB.html#SMB.4.2.1
+ * and http://ubiqx.org/cifs/SMB.html#SMB.5.5.1 the 8
+ * bytes after the "High part of PID" are an 8-byte
+ * signature ...
*/
- proto_tree_add_item(htree, hf_smb_pid_high, tvb, offset, 2, TRUE);
- offset += 2;
-
proto_tree_add_item(htree, hf_smb_sig, tvb, offset, 8, TRUE);
offset += 8;